CN116635693A

CN116635693A - System and method for corrosion and erosion monitoring of pipes and vessels

Info

Publication number: CN116635693A
Application number: CN202080107799.8A
Authority: CN
Inventors: 萨查·什克; 丹尼尔·路托福-卡罗尔
Original assignee: Molex LLC
Current assignee: Molex LLC
Priority date: 2020-12-08
Filing date: 2020-12-08
Publication date: 2023-08-22
Also published as: CA3201619A1; EP4260057A1; EP4260057A4; WO2022125059A1; JP2023552461A; JP7553716B2

Abstract

The present disclosure relates to the field of corrosion and erosion monitoring of pipes and vessels. More specifically, the present disclosure relates to a system and method for corrosion and erosion monitoring of pipes and vessels, wherein the system/method combines ultrasonic thickness monitoring using longitudinal waves with ultrasonic area monitoring using one or more guided waves, whereby representative thickness measurements are complemented by an area monitoring feature to detect localized corrosion/erosion between representative thickness measurement sites. In another embodiment, a system and method for optimizing asset health monitoring is disclosed that includes an analysis scheme.

Description

System and method for corrosion and erosion monitoring of pipes and vessels

RELATED APPLICATIONS

The present application is related to U.S. provisional patent application US62/982751 (docket No. MX-2020-PAT-0029-US-PRO) entitled "system and method for corrosion and erosion monitoring of pipes and vessels" filed on 28, 2, 2020. The aforementioned patent application is incorporated by reference in its entirety.

Technical Field

The present disclosure relates to the field of corrosion and erosion monitoring of pipes and vessels. In particular, the present disclosure relates to a corrosion and/or erosion monitoring system including mechanical components, hardware, software, analysis, and/or combinations thereof. In one embodiment, the mechanical components and hardware may include one or more ultrasonic transducers, base units (base units), gateways, and/or combinations thereof. The system may also include a software platform for remote monitoring. In some embodiments, the system may further include analysis tools for front-end and back-end services for remote monitoring and/or diagnostics. More specifically, in some embodiments, the present disclosure may relate to a system and method for corrosion and erosion monitoring of pipes and vessels, wherein the system/method combines ultrasonic thickness monitoring with longitudinal waves with ultrasonic area (area) monitoring with one or more guided waves, whereby representative thickness measurements are supplemented (completed) by an area monitoring feature to detect (detect) localized corrosion/erosion between representative thickness measurement sites (locations). In another embodiment, a system and method for optimizing asset health monitoring is disclosed that includes an analysis scheme.

Background

The use of ultrasonic transducers for ultrasonic monitoring of the condition and integrity of structural assets including piping and pressure vessels such as those employed in the oil, gas and power industries is well known. Currently, corrosion and erosion monitoring systems and techniques involving inclusion/use of ultrasonic transducers are known to include thickness monitoring and area monitoring at a site (also known as guided wave inspection (guided wave inspection)). However, these two systems and techniques are typically independent of each other. In addition, in addition to Ultrasonic (UT) testing, internal corrosion of the piping system is sometimes monitored using a Radial (RT) thickness test to measure the wall thickness of selected components at prescribed intervals over the life of the system.

Thickness monitoring ultrasonic transducers and systems employing thickness monitoring ultrasonic transducers typically measure a thickness of a wall of a pipe/vessel at the point (spot) where the ultrasonic transducer is disposed, in other words, it does not provide any information about the thickness of the wall of the pipe/vessel at a location around the exact point where the ultrasonic transducer is disposed. Thus, if corrosion/erosion occurs at a location outside of where the ultrasound transducer is located, it is likely that no corrosion/erosion will be detected unless thickness monitoring is accompanied by ultrasound transducer mapping (mapping). Of course, ultrasound transducer mapping increases inspection costs. However, these ultrasonic transducers and systems are beneficial for permanent installation on the pipe/vessel.

Conversely, area monitoring ultrasonic transducers and systems employing area monitoring ultrasonic transducers typically measure the thickness of the wall of a pipe/vessel across a relatively large area of the wall of the pipe/vessel, which area typically exceeds the location at which the thickness monitoring ultrasonic transducer is disposed on the pipe/vessel. Such area monitoring ultrasonic transducers and systems employing the same will typically develop a thickness map (thickness map) of the walls of the pipe/vessel across the area being measured. Theoretically, such generated thickness maps are beneficial, but at present such guided wave inspection is extremely complex, because the general hardware in that section generates ten to twenty different guided wave modes, and the number of guided wave modes and complex analysis negatively affects the reliability of the inspection results. Furthermore, guided wave inspection is typically not permanently installed on pipes and vessels. Furthermore, highly localized corrosion cannot be reliably detected using a temporarily installed guided wave system as illustrated by API 574 (API 574, inspection practices for pipeline system components, fourth edition, 2016).

In addition, existing permanently installed corrosion monitoring systems fail to employ sufficient data to determine the placement of sensors in an industrial facility, such as a refinery and petrochemical plant, that uses pipeline systems to transport fluids. The pipeline system may transport the fluid to one or more tanks and/or chemical processing units. Some pipeline systems handle dedicated fluids at specified temperatures and/or pressures; these piping systems can carry highly corrosive fluids at high temperatures and pressures.

In addition, many industrial facilities face health and safety issues. They may transport fluids that may be flammable and/or toxic. As such, a fault in the piping system may result in leakage to the atmosphere and/or exposure to plant personnel. In addition, some facilities operate without planned (scheduled) downtime for several years. Thus, the reliability of the pipeline system and its components is important.

In addition to health and safety issues, unscheduled outages due to pipeline system failures are also problematic from a business outcome perspective. In view of the potential safety, health, environmental and business risks associated with pipeline faults, the condition of the pipeline system is monitored to accurately predict (project) its remaining life and to determine a safe repair or replacement date.

For the reasons stated above, certain individuals would appreciate improvements in systems and methods for corrosion and erosion monitoring of pipes and vessels.

Disclosure of Invention

In the following description of the various illustrative embodiments, reference is made to the accompanying drawings that form a part hereof, and in which are shown by way of illustration various embodiments in which aspects of the disclosure may be practiced. It is to be understood that other embodiments may be utilized and structural and functional modifications may be made without departing from the scope of the present disclosure. Note that various connections between elements are discussed in the following description. Note that these connections are generic and may be direct or indirect, wired or wireless, unless stated otherwise, and the specification is not intended to be limiting in this respect.

A system of one or more computers can be configured such that the system performs a particular operation or action by virtue of having software, firmware, hardware, or a combination thereof installed on the system that when operated causes the system to perform the action. One or more computer programs can be configured such that the computer programs perform particular operations or acts by virtue of including instructions that when executed by data processing apparatus cause the apparatus to perform the acts. One general aspect includes a method for screening from a probe assembly mounted on a pipeline system. The method further comprises the steps of: before training a model, a group sensitivity superparameter, a threshold measurement superparameter, and a group size superparameter are set for the model. The method further comprises the steps of: a first set of probe assemblies is grouped based on at least historical pipe wall thickness measurements acquired from probe assemblies installed on the pipeline system over a period of time by executing the model on a processor. The method further comprises the steps of: a unique group ID is assigned to each group of probe assemblies. The method further comprises the steps of: after training the model, an optimization function is selected from a plurality of optimization functions for the model by the model. The method further comprises the steps of: a single probe assembly corresponding to each set ID is identified by the model for use in pipe wall thickness monitoring of the pipeline system. The method further comprises the steps of: a tube wall thickness measurement from the individual probe assemblies of each set ID is sent for verification by a thickness monitoring controller associated with the tubing system. Other embodiments of this aspect include corresponding computer systems, apparatus, and computer programs recorded on one or more computer memory devices, each configured to perform the actions of the methods.

Implementations may include one or more of the following features. The method may include one or more steps to disregard all remaining probe assemblies in each group ID during inspection, except for a single probe assembly from each group ID. The grouping of the first set of probe assemblies is further based at least on inspection information provided to the system and historical pipe wall thickness measurements acquired from probe assemblies installed on the pipeline system over a period of time. The tubing system may include a tank, and wherein a first probe assembly of the probe assemblies is configured to measure a wall thickness of the tank. The method may further comprise the steps of: storing in a computer memory communicatively coupled to the processor historical pipe wall thickness measurements acquired from the probe assembly mounted on the pipeline system over an extended period of time; and training, by the processor, the model using at least the historical pipe wall thickness measurements stored in the computer memory. The model may include an artificial neural network. Implementations of the described technology may include hardware, a method or process, or computer software on a computer-accessible medium.

One general aspect includes a system for detecting general corrosion (e.g., no localized corrosion) of a plurality of components transporting material across a distance. The system may further include a plurality of probe assemblies attached to one or more components, wherein the probe assemblies may include: at least one thickness monitoring ultrasonic transducer and one area monitoring ultrasonic transducer are configured to detect corrosion (e.g., general corrosion and/or localized corrosion) of the component. The system may further comprise: a data memory configured to store historical wall thickness measurements acquired over a period of time from measurements performed by the probe assembly. The system may further comprise: a model is trained on the historical wall thickness measurements in the data store, and the superparameters may include a packet sensitivity superparameter, a threshold measurement superparameter, and a packet size superparameter. The system may further comprise a monitoring device, which may include a processor and a memory storing computer-executable instructions that, when executed by the processor, cause the system to perform steps, which may further comprise: grouping a first set of probe assemblies based on the model; assigning a unique group ID to each group of probe assemblies; selecting an optimization function from a plurality of optimization functions based on the model; identifying a probe assembly corresponding to each set of IDs for wall thickness monitoring of the component based on the model and the selected optimization function; and transmitting, by a thickness monitoring controller associated with the component, a wall thickness measurement of the probe assembly from each set of IDs for verification. In another embodiment, the system may output a list corresponding to the unique identifier of any group ID in lieu of sending wall thickness measurements for verification. As discussed in various embodiments disclosed herein, an inspector can receive the output of the system and react accordingly. Other embodiments of this aspect include corresponding computer systems, apparatus, and computer programs recorded on one or more computer memory devices, each configured to perform the actions of the methods.

Implementations may include one or more of the following features. In the system, wherein the probe assemblies identified from each group ID may include more than one probe assembly of the plurality of probe assemblies, and wherein the memory of the monitoring device stores computer-executable instructions that, when executed by the processor, cause the system to perform steps that may include: during inspection, not considering all remaining probe assemblies in each group ID except for the more than one probe assembly from each group ID; verifying that the wall thickness measurements of the more than one probe assembly for each set of IDs are general corrosion rather than localized corrosion. The wall thickness measurement of the probe assembly from a first set of IDs may include a wall thickness of a pipe component at the probe assembly. The wall thickness measurement of the probe assembly from a first set of IDs may comprise a thickness of a wall of a tank component at the probe assembly. The method may include verifying that the pipe wall thickness measurement of the single probe assembly is general corrosion (e.g., no localized corrosion) by: (i) generating a probability map of all pipe wall thickness measurements associated with the pipeline system, (ii) grouping drawn pipe wall thickness measurements by nominal thickness, and (iii) identifying a non-linear relationship on the probability map of pipe wall thickness measurements grouped by nominal thickness to confirm general corrosion (e.g., no localized corrosion). The pipe wall thickness monitoring may include the steps of: the original wall thickness, loss of wall thickness over time, calibration errors, and measurement site repeatability errors are analyzed by the probe assembly. Implementations of the described technology may include hardware, a method or process, or computer software on a computer-accessible medium.

Implementations may include one or more of the following features. The method may further comprise the steps of: the single probe assembly pipe wall thickness measurement is verified to be general corrosion (e.g., no localized corrosion) by: generating a probability map of all pipe wall thickness measurements associated with the pipeline system, grouping the plotted pipe wall thickness measurements by nominal thickness, and identifying a non-linear relationship on the probability map of pipe wall thickness measurements grouped by nominal thickness to confirm general corrosion (e.g., no localized corrosion). The pipe wall thickness monitoring may comprise the steps of: the probe assembly is used for analyzing the original wall thickness, the loss of the wall thickness with time, the calibration error and the repeatability error of the measuring position. Implementations of the described technology may include hardware, a method or process, or computer software on a computer-accessible medium.

Drawings

The present disclosure is illustrated by way of example and not limitation in the figures of the accompanying drawings, in which like references indicate similar elements, and in which:

FIG. 1 is a diagram of a system for corrosion/erosion monitoring;

FIG. 2 is a schematic representation of a thickness monitoring controller and a piezoelectric assembly of the system of FIG. 1;

FIG. 3 is a schematic representation of the thickness monitoring controller of FIG. 2;

FIG. 4 is a diagram of a switch assembly forming portion forming the piezoelectric assembly of FIG. 2;

FIG. 5 is a schematic representation of the piezoelectric assembly of FIG. 2;

FIGS. 6, 7 and 8 are illustrations of a method for corrosion/erosion monitoring;

fig. 9, 10, 11 and 12 are diagrams showing signal modulation;

fig. 13A and 13B (collectively "fig. 13") are diagrams of one illustrative pipeline with an installed MUT sensor, in accordance with one or more aspects of features disclosed herein;

FIG. 14 is an illustrative network architecture of an industrial facility in accordance with aspects of the present disclosure;

FIG. 15 is an illustrative diagram of a grouping of probe assemblies in one embodiment of the present disclosure;

fig. 16A, 16B, and 16C (collectively, "fig. 16") show curves on a graph. FIG. 16A is a graph showing probability curves for measurements used to verify the comparison of general corrosion to localized corrosion. Fig. 16B is a graph of plotted risk levels versus TMLs according to various aspects disclosed herein.

Fig. 16C illustrates movement of curves giving risk levels versus TMLs after screening according to various aspects disclosed herein;

FIG. 17 is a graph showing cumulative thickness distribution of a pipe corroded by naphthenic acid;

FIG. 18A is a corrosion sensor analysis showing TML measurements by date in one embodiment of the present disclosure;

FIG. 18B is another corrosion sensor analysis chart showing the measurement of TML by date but with a higher group sensitivity setting like FIG. 8A;

FIG. 18C is a further corrosion sensor analysis chart showing the measurement of TML by date but with an even higher grouping sensitivity setting like FIG. 8A;

19A and 19B are diagrams in accordance with one or more aspects of the present disclosure;

FIGS. 20A and 20B are also diagrams according to one or more aspects of the present disclosure;

FIG. 21 shows an illustrative artificial neural network configured to operate in conjunction with the systems, methods, and algorithms disclosed herein; and

FIG. 22 is a flowchart showing illustrative steps of a method performed in accordance with some embodiments disclosed herein;

FIG. 23 is a diagram of a simplified pipeline meter flow chart (PID) corresponding to an illustrative corrosion/erosion monitoring system shown in FIG. 1, according to some embodiments disclosed herein.

Detailed Description

While this disclosure may be susceptible of embodiment in different forms, there is shown in the drawings and will herein be described in detail specific embodiments with the understanding that the present disclosure is to be considered an exemplification of the principles of the disclosure and is not intended to limit the disclosure to that as illustrated and described herein. Thus, unless otherwise indicated, features disclosed herein may be combined together to form additional combinations that are not given for the sake of brevity. It will further be appreciated that, in some embodiments, one or more elements shown by way of example in the figures may be removed and/or replaced with alternative elements within the scope of the present disclosure.

Aspects of the present disclosure relate to monitoring (monitoring) and detecting (detecting) corrosion and/or erosion of pipes, vessels, and other components in an industrial facility. The monitoring system may include a software platform for remote monitoring and analysis of historical measurements acquired by a plurality of sensors attached to the pipe and components. The monitoring system may include analytical tools for monitoring, diagnosis and/or prediction of localized and/or general corrosion (general corrosion). By employing the analysis system disclosed herein, the thickness monitoring sites (TML) can be optimized to reduce the number of measurement sites, i.e., screening (down-selection), among other things, without risk of compromise. As explained in this disclosure, by strategically reducing the number of probe assemblies that need to be sampled during a test, the amount of time/cost of a test is reduced while maintaining (or even reducing) the risk profile of the industrial facility.

Fig. 1 and 2 illustrate a system 100 for monitoring corrosion and erosion of a pipe/vessel. The system 100 includes a data analysis and visualization platform 110, an optional gateway 120, a thickness monitoring controller 130, a thickness monitoring ultrasound transducer 140 for standardization purposes, and at least one probe assembly 150. Each probe assembly 150 includes a switch assembly 160, at least one thickness monitoring ultrasonic transducer 170, and at least one area monitoring ultrasonic transducer 180.

The data analysis and visualization platform 110 includes a data analysis portion 112 and a visualization portion 114. The data analysis portion 112 is typically a cloud-based driven piece of software configured to receive (typically wirelessly) signals from one or both of the gateway 120 or the thickness monitoring controller 130. These signals are analyzed by the data analysis portion 112, which converts them into visual data (visual) for display on the visualization portion 114. The visualization portion 114 may be any suitable device, such as a computer monitor (computer monitor), a tablet, a cell phone, etc., that is one type of information that will assist the personal monitoring platform 110 in understanding the corrosion/erosion identified with respect to the system 100. An individual may also have the ability to change the image/information on the visualization portion 114 by providing further input to the software.

Gateway 120 may be configured to receive (typically wirelessly) signals from thickness monitoring controller 130 and transmit (typically wirelessly) such signals to platform 110. For example, rather than having the thickness monitoring controller 130 have its own data arrangement (data plan) at a facility, it may be more economical to employ the gateway 120 to establish a cellular connection. In this case, thickness monitoring controller 130 would communicate with gateway 120 using, for example, the XBE protocol. In another example, if there is no good cellular connection at the location of thickness monitoring controller 130, gateway 120 may be installed at a higher location to establish the cellular connection and thickness monitoring controller 130 will submit data to gateway 120 using, for example, the XBee protocol.

As best shown in fig. 3, the thickness monitor controller 130 includes a modem 131, a microprocessor 132, a pulser 133, an analog-to-digital converter (ADC) 134, an adjustable gain amplifier 135, a transmit channel 136, and a receive channel 137. Modem 131 is configured to communicate with one or both of platform 110 and gateway 120. Modem 131 may employ any suitable communication options including, but not limited to, XBee 915MHz and LTE-M/NB. The modem 131 is configured to communicate with the microprocessor 132. The microprocessor 132 may be any type of microprocessor that will provide the desired functionality. One such microprocessor 132 is the LPC4370 manufactured and sold by Enzhps semiconductor corporation (NXP Semiconductors). Microprocessor 132 is configured to communicate with both pulser 133 and ADC 134. The pulser 133 is preferably a high voltage pulser capacitor. The ADC 134 is preferably a 16-bit, 2msps (millions of samples per second), but other types of ADCs may be provided as appropriate. The ADC 134 is configured to communicate with an adjustable gain amplifier 135 (sometimes also commonly referred to as a variable gain amplifier). The adjustable gain amplifier 135 preferably has a dB range of 26-54dB and a frequency range of 10kHZ to 300kHz, although other ranges may be set as appropriate. Pulser 133 is configured to communicate with transmit channel 136 to transmit signals to transmit channel 136. The adjustable gain amplifier 135 is configured to communicate with the receive channel 137 to receive signals from the receive channel 137. The thickness monitoring controller 130 is preferably configured to accommodate a desired number of amplitude sweeps (or waveform displays). In the illustrated embodiment, the controller 130 is configured to accommodate 16 a-scans (one from the thickness measuring ultrasound transducer 140, five from three different probe assemblies 150). Of course, it should be understood that the controller 130 can be configured to accommodate more or less than 16A-scans as appropriate, as the number of probe assemblies 150 varies and/or according to the number of ultrasound transducers 170/180 included in each probe assembly 150 (as will be discussed in further detail below).

The thickness monitoring ultrasonic transducer 140 is configured to receive signals from the transmit channel 136 of the thickness monitoring controller 130 and is also configured to transmit signals to the receive channel 137 of the thickness monitoring controller 130. As previously described, the thickness monitoring ultrasonic transducer 140 is used for standardization purposes and, therefore, functions to calibrate the measurement system when a set of ultrasonic transducers (in this example, at least one thickness monitoring ultrasonic transducer 170 and at least one area monitoring ultrasonic transducer 180) are employed. The standardized thickness monitoring ultrasonic transducer 140 operates to ensure that the system 100 always performs the same way and functions properly, as required by industry standards. In the illustrated embodiment, the standardized thickness monitoring ultrasonic transducer 140 is configured to perform a single a-scan. In practice, the thickness monitoring ultrasonic transducer 140 is typically placed on a standardized block or a thickness calibrated metal sheet as a standardized transducer.

As shown in fig. 1, the system 100 includes three distinct probe assemblies 150A, 150B, 150C (each also referred to as probe assembly 150). Depending on the system 100, the number of probe assemblies 150 disposed in the system 100 may be less than three (e.g., one or two) or may be more than three (e.g., four, five, etc.), as appropriate. Those of ordinary skill in the art will appreciate that minor variations/modifications to the system 100 may be required depending on the number of probe assemblies 150 provided in the system 100.

As described above, each probe assembly 150 includes a switch assembly 160. As best shown in fig. 4, the switch assembly 160 includes a power supply 161, a transmit switch 162, a microcontroller 163, a memory 164, a receive switch 165, an amplifier 166, and an optional Resistance Temperature Detector (RTD) interface 167. The power source 161 communicates with the transmit channel 136 of the thickness monitoring controller 130. The transmit switch 162 communicates with the transmit channel 136 of the thickness monitoring controller 130. The transmit switch 162 preferably has five "switch" channels 162a, 162b, 162c, 162d, 162e, each of which has a purpose and function that will be discussed herein. The microcontroller 163 communicates with the transmit channel 136, transmit switch 162, memory 164, and receive switch 165 of the thickness monitoring controller 130. The microcontroller 163 may be any type of microcontroller that will provide the desired functionality. One such microcontroller 163 is the PIC18 produced and sold by microchip technology company (Microchip Technology). The memory 164 is preferably a non-volatile memory. The receive switch 165 preferably has four "switch" channels 165a, 165b, 165c, 165d, the purpose and function of each of which will be discussed below. The amplifier 166 communicates with the receive channel 137 and the receive switch 165 of the thickness monitoring controller 130. The amplifier 166 preferably has a magnification (amplification) of 26-48dB and a frequency range of 10kHz-300kHz, although other levels/ranges may be set as appropriate. The amplifier 166 is preferably a two-stage amplifier, with 26dB of amplification being set for a single stage option and 48dB of amplification being set for a two-stage option, which can be selected by filling (boost) or deleting (depolate) components on an amplifying board (amplification board). If at least one thickness monitoring ultrasonic transducer 170 comprises an RTD 171 (as described below), an optional RTD interface 167 is provided. In the illustrated embodiment, each switch assembly 160 is instructed by the controller 130 to acquire five a-scans (one from each of the thickness monitoring ultrasonic transducers 170 and one from each of the four area monitoring ultrasonic transducers 180).

As described above, each probe assembly 150 includes at least one thickness monitoring ultrasonic transducer 170. As shown in fig. 1, each probe assembly 150 includes a thickness monitoring ultrasonic transducer 170. Depending on the system 100 and the probe assemblies 150, the number of thickness monitoring ultrasonic transducers 170 disposed in each probe assembly 150 may be more than one (e.g., two, three, four, etc.) as appropriate. Those of ordinary skill in the art will appreciate that minor deformation/modification of the probe assemblies 150 and/or the system 100 may be required depending on the number of thickness monitoring ultrasonic transducers 170 provided in each probe assembly 150. Each thickness monitoring ultrasonic transducer 170 may optionally have an RTD 171 associated therewith to measure the temperature at or near the point where the thickness measurement of the pipe/vessel occurs. Each thickness monitoring ultrasonic transducer 170 communicates with a fifth "on-off" channel 162e of transmit switch 162, and if thickness monitoring ultrasonic transducer 170 includes RTD 171, thickness monitoring ultrasonic transducer 170 also communicates with RTD interface 167.

The thickness monitoring ultrasonic transducer 170 (and the thickness monitoring ultrasonic transducer 140) operates by generating high frequency ultrasonic waves (e.g., 5 MHz). These ultrasonic waves are commonly referred to as Longitudinal Waves (LW), and as such, the thickness monitoring ultrasonic transducer 170 may also be referred to as an LW transducer. In the illustrated embodiment, each thickness monitoring ultrasonic transducer 170 is configured to perform a single a-scan. Unlike the thickness monitoring ultrasonic transducer 140, the thickness monitoring ultrasonic transducer 170 is not placed on a standardized block or a thickness calibrated metal sheet, but is placed on the pipe/vessel to measure the thickness of the pipe/vessel at the location where the thickness monitoring ultrasonic transducer 170 is installed.

As described above, each probe assembly 150 includes at least one area monitoring ultrasonic transducer 180. As shown in fig. 1, 2, 3, 4, and 5, each probe assembly 150 includes four area monitoring ultrasonic transducers 180A, 180B, 180C, 180D (each also referred to as an area monitoring ultrasonic transducer 180). Depending on the system 100 and the probe assemblies 150, the number of area monitoring ultrasonic transducers 180 disposed in each probe assembly 150 may be less than four (e.g., one, two, or three) or more than four (e.g., five, six, etc.), as appropriate. Those of ordinary skill in the art will appreciate that minor variations/modifications to the probe assemblies 150 and/or the system 100 may be required depending on the number of area monitoring ultrasonic transducers 180 disposed in each probe assembly 150. The first area monitoring ultrasonic transducer 180A communicates with a first "on-off" channel 162a of the transmit switch 162 and a first "on-off" channel 165a of the receive switch 165. The second area monitoring ultrasonic transducer 180B communicates with a second "on-off" channel 162B of the transmit switch 162 and a second "on-off" channel 165B of the receive switch 165. The third area monitoring ultrasonic transducer 180C communicates with a third "on-off" channel 162C of the transmit switch 162 and a third "on-off" channel 165C of the receive switch 165. The fourth area monitoring ultrasonic transducer 180D communicates with a fourth "on-off" channel 162D of the transmit switch 162 and a fourth "on-off" channel 165D of the receive switch 165.

In one embodiment, the probe assembly 150 may include a thickness transducer 170 and a set of area transducers 180 wired to a switch/preamplifier (preamplifier) assembly 160, respectively. In a different embodiment, the thickness transducer 170 and the area transducer 180 may be combined in a single larger probe that is wired to the switch/preamplifier assembly 160 via a single multi-conductor cable. In another embodiment, it may also be a set of larger probes (thickness +2 areas, area + area, etc.).

The area monitoring ultrasonic transducer 180 operates by generating low frequency ultrasonic waves (e.g., 50kHz to 500 kHz). These ultrasonic waves are commonly referred to as Guided Waves (GW), and as such, the area monitoring ultrasonic transducer 180 is alsoMay be referred to as a GW transducer. One such type of guided wave from a GW transducer, zero order shear horizontal wave (called SH in the plate ₀ Referred to as T (0, 1) in the pipeline), is of interest because of its non-dispersive (non-dispersive) behaviour. In the illustrated embodiment, each area monitoring ultrasonic transducer 180 is configured to perform a single a-scan.

GW transducer 180 is preferably in the form of a piezoelectric patch transducer, but may alternatively be in other forms, such as a face-shear (face-shear) piezoelectric element. In a preferred embodiment, as best shown in fig. 1 and 5, GW transducers 180A, 180B, 180C, 180D are positioned in a rectangular configuration around LW transducer 170, with GW transducer 180A positioned above left of LW transducer 170, GW transducer 180B positioned below left of LW transducer 170, GW transducer 180C positioned below right of LW transducer 170, and GW transducer 180D positioned above right of LW transducer 170. When applied to a pipe/vessel, a line from GW transducer 180A to GW transducer 180B is parallel to a line from GW transducer 180C to GW transducer 180D, and a line from GW transducer 180A to GW transducer 180D is parallel to a line from GW transducer 180B to GW transducer 180C. Further, when applied to a pipe/vessel, a line from GW transducer 180A to GW transducer 180C intersects LW transducer 170 and a line from GW transducer 180B to GW transducer 180D intersects LW transducer 170, such that an "X" shaped configuration is provided.

When the system 100 is associated with a pipe/vessel, the system 100 may be used to measure corrosion/erosion of the pipe/vessel. In one embodiment, a method 200 of measuring corrosion/erosion of a pipe/vessel is described below and shown in fig. 6, 7, and 8.

The method 200 comprises a step 205 of manually measuring the actual longitudinal speed and temperature of the pipe/vessel to be inspected.

The method 200 includes a step 210 of manually measuring the actual guided wave speed and temperature of the pipe/vessel to be inspected.

Method 200 includes a step 215 of performing a thickness normalization measurement using normalization thickness monitoring ultrasonic transducer 140 and RTD 171 (it should be understood that normalization thickness monitoring ultrasonic transducer 140 may optionally include RTD 171 as well as thickness monitoring ultrasonic transducer 170).

The method 200 includes a step 220 of taking measurements using the probe assembly 150A. Step 220 includes sub-step 220a of taking a thickness measurement using thickness monitoring ultrasonic transducer 170 and RTD 171. Step 220 includes sub-step 220B of monitoring an area thickness using the area monitoring ultrasonic transducers 180A, 180B, 180C, 180D at a first frequency. Sub-step 220B includes sub-step 220B1 of performing an axial scan whereby area monitoring ultrasonic sensor 180A is energized (calibrated) and data is recorded by area monitoring ultrasonic sensor 180B. The measurements made in sub-step 220b1 are repeated again and again (of) as specified in the configuration settings and averaged over multiple a scans. Sub-step 220b includes sub-step 220b2 of performing an axial scan whereby the area monitoring ultrasonic transducer 180C is energized and data is recorded by the area monitoring ultrasonic transducer 180D. The measurements made in sub-step 220b2 are repeated again and again as specified in the configuration settings and averaged over multiple a-scans. Sub-step 220b includes sub-step 220b3 of performing a circumferential scan whereby the area monitoring ultrasonic transducer 180A is energized and data is recorded by the area monitoring ultrasonic transducer 180D. The measurements made in sub-step 220b3 are repeated again and again as specified in the configuration settings and averaged over multiple a-scans. Sub-step 220b includes sub-step 220b4 of performing a circumferential scan whereby the area monitoring ultrasonic transducer 180C is energized and data is recorded by the area monitoring ultrasonic transducer 180C. The measurements made in sub-step 220b4 are repeated again and again as specified in the configuration settings and averaged over multiple a-scans. Thus, channels 162a, 162C (which are associated with GW transducers 180A, 180C) act as guided wave transmit channels, while channels 162B, 162D (which are associated with GW transducers 180B, 180D) act as guided wave receive channels. The receive path further proceeds via amplifier 166 to a receive channel 137 of thickness monitoring controller 130.

Step 220 includes sub-step 220c of repeating sub-step 220b at a second frequency, the second frequency being different from the first frequency.

Step 220 includes sub-step 220d of repeating sub-step 220b at a third frequency, the third frequency being different from both the first frequency and the second frequency.

The method 200 includes a step 225, the step 225 including repeating the step 220 to make a measurement with the probe assembly 150B.

The method 200 includes a step 230, the step 230 including repeating the step 220 to make a measurement with the probe assembly 150C.

Thus, the method 200 combines ultrasonic thickness monitoring with longitudinal waves with ultrasonic area monitoring with guided waves; also, in a preferred embodiment, there is only one particular non-dispersive shear wave mode (SH ₀ Or T (0, 1)). Rather than attempting to develop a thickness map that would be complemented by an area monitoring feature to detect localized corrosion/erosion between representative thickness measurement sites, the method 200 takes representative thickness measurements. The system 100 employs new electronics that employ a single circuit to deliver two distinct (distinctive different) excitation signals (e.g., high frequency ultrasound (5 MHz) for thickness monitoring and low frequency ultrasound (50-500 kHz) for area monitoring) from two different types of ultrasound transducers (e.g., LW transducer 170 and GW transducer 180). Each excitation signal needs to be generated and processed in a different way (differential). More specifically, the pulser 133 of the controller 130 is a digital switch capable of delivering only a predetermined fixed voltage level: high voltage, low voltage, and zero voltage. The high and low voltage levels are typically adjustable in the range of 5V to 90V and-5V to-90V, but different voltage levels are also permissible. The microprocessor 132 signals the pulser 133 to cause the pulser 133 to output one of the fixed voltage levels to the transmit channel 136: for example, a high voltage for a specified time period (time of period). An example of a pulse for energizing LW transducer 170: the processor 130 instructs the pulser 133 to output 0V, followed by a high voltage for a period (period) of 100ns, followed by a low voltage for a period of 100ns, followed by 0V. The illustrated sequence will produce a bipolar square wave of 5MHz frequency suitable for exciting the LW transducer 170. For GW transducer 180, different frequencies and signal amplitudes are required.

As best shown in fig. 9, 10, 11, and 12, the waveforms required to excite GW transducer 180 may have a fairly complex shape, such as: the 5 periodic sine waves are superimposed on the Hanning window signal (e.g., half-period cosine) as indicated at 330 in fig. 12, which will allow for smoother transitions from no signal to signal states. To produce a suitable combination of waveforms for GW transducer 180 with respect to the digital output of a pulser 133, as shown in waveform 300 of fig. 9, 10, 11 and 12, the series resistance of transmit channel 136 and the impedance of GW transducer 180 are employed. The impedance of GW transducer 180 in a frequency range (50-500 kHz) for generating GW waves is typically composed mainly of capacitance. The capacitance and the mentioned series resistance of the transmission channel 136 form a low-pass filter. Pulser 133 generates a high frequency (typically in the range of tens of MHz) digital waveform 300 upon command from microprocessor 132, which high frequency digital waveform 300, when passing through transmit channel 136 and the capacitance of GW transducer 180, generates a waveform 310 (shown in fig. 10) that is different from the initial output from pulser 133. Changing the high frequency digital waveform from the pulser 133 can produce a series of analog waveforms once through the series resistance of the transmit channel 136 and the capacitance of the transducer 180, such as: a sine without hanning window as indicated at 320 in fig. 11, or a sine with hanning window as indicated at 330 in fig. 12, chirp (frequency change during pulse duration), rise (ramp-up), saw tooth, etc. Of course, other waveforms than those illustrated and shown may be generated.

In one embodiment, a chirp signal may be used to excite multiple frequencies from a channel simultaneously. Suitable software filtering can decode individual (differential) frequency responses from a single a-scan.

By employing the system 100 and method 200, the time of flight and amplitude of the reflected echo at a defect (defect) on the pipe/vessel can be estimated. More specifically, by sending an excitation signal from GW transducer 180C and receiving it by GW transducers 180B, 180D, the reflected echo will be earlier on the time trace of GW transducers 180B, 180D closer to the lesion (damage) (in the case where two GW transducers 180B, 180D are positioned as shown in fig. 1 and 5, for example, GW transducer 180B if the lesion is to the left of two GW transducers 180B, 180D, and GW transducer 180D if the lesion is to the right of two GW transducers 180B, 180D). Defects such as holes (holes) or erosion/corrosion patches (patches) typically increase in size over time. Thus, the amplitude of the echo reflected at the defect increases with time. Thus, permanently installed systems allow one to monitor changes in amplitude over the (next to) time of flight. Monitoring changes in the a-scan, e.g., after baseline subtraction and digital filtering, reduces the complexity of the analysis and increases the confidence in the test results. Next to the (next to) baseline subtraction, other digital signal processing tools or machine learning algorithms may be used for feature extraction or pattern recognition (pattern recognition), which additionally increases the confidence level and helps to detect changes earlier in time.

FIG. 23 shows a simplified pipeline meter flow diagram (PID) corresponding to an illustrative corrosion/erosion monitoring system shown in FIG. 1, according to some embodiments disclosed herein. Simplified PID 2300 includes a number of probe assemblies, shown in circles, numbered 19 through 84. For example, three different/distinct probe assemblies 150A, 150B, 150C are shown. Of course, the number of probe assemblies in the PID 2300 can be any number as appropriate. In one example, a human operator/inspector may focus the inspection on a list of TMLs being screened, as explained herein. These screened TMLs may represent more effective candidate measurement sites to capture overall corrosion behavior of the entire asset while still being able to verify against localized corrosion. For example, by screening the number of TMLs, a significant amount of time/energy and cost can be saved so that only those probe assemblies that are most likely to detect localized corrosion will be reviewed by a human operator/inspector. Instead of checking all probe assemblies from 19 th to 84 th probe assemblies, or even randomly checking less than all probe assemblies from 19 th to 84 th probe assemblies, the TMLs being screened is a more optimal identification of which TMLs to measure. In some examples, an inspector can employ a hand-held or other manual device to measure the wall thickness at numbered locations on the simplified PID 2300. In other examples, the same class (of solvents) of kits (rig) or kits (harnesss) may be pre-installed at numbered locations on the simplified PID 2300 to allow an inspector to measure the wall thickness at each thickness measurement location. In yet another example, the inspector may be an automated machine that makes measurements at the TMLs being screened at specific time intervals. Screening TMLs is advantageous even in an automated measurement system because it reduces the processing power and network bandwidth used by measurement data generated by a measurement device at each numbered site on the simplified PID 2300. For example, some large industrial facilities may have thousands of probe assemblies, which may result in an excessive amount of data being generated. Furthermore, once any localized corrosion is identified and repaired, the human operator can indicate as much (as much), and any model can be updated to reflect the new wall thickness values. Additionally, in some examples, if a localized corrosion is falsely identified, the supervisor inputs into a machine learning or neural network executing in a digital analysis platform, the alarms and models of which may be refined accordingly.

Fig. 13A shows an illustrative pipeline with multiple sensors 1301 mounted on the pipeline according to one or more aspects of the features disclosed herein. The conduit may have a flow of liquid in the direction indicated by the arrow. During an inspection, an approach may be to inspect and measure each of the sensors 1-6 shown in FIG. 13A. In another example, a randomly selected sensor may be inspected and measured. In accordance with several of the systems and methods disclosed herein, in another example, multiple Thickness Monitoring Sites (TMLs) shown at respective sensors 1-6 are intelligently contemplated, and a smaller/narrower set of TMLs may be screened for testing. Furthermore, in accordance with several systems and methods disclosed herein, TMLs may be grouped based on one or more criteria (criterion) in the process of screening TMLs. In a simplified example, the screening criteria may identify and exclude those sensors that historically measure only general corrosion in their area (e.g., sensor 1 and sensor 3). Thus, by screening, the system 100 avoids using clusters (clusters), but instead uses groupings to screen some sensors that are redundant to the estimation of the health of the mechanical components. Thus, time and resources are saved. In contrast, some existing systems attempt to reduce risk by adding more TMLs and testing of these TMLs. However, the risk-based verification (RBI) approach described in various aspects of the present disclosure provides a better process and system. An RBI practice may also employ a model that takes into account other criteria such as the type of fluid being transported in the pipeline system, the temperature inside and outside of the pipeline/components, the elbow/configuration of the pipeline components, and other criteria. For example, measurements at a bend may be weighted to be more likely to be selected as part of screening in a group, because historically, locations in the pipeline near a bend would be places of more turbulence and friction, and thus corrosion and acidity (acidity) may be higher.

Referring to fig. 13B, probe assembly 1302 may include a tethered device that captures accurate point measurements of the thickness of the component. In another embodiment, the probe assembly may include a tethered device that captures accurate point measurements and area monitoring. For example, the apparatus of FIG. 13B or equivalent apparatus may be used to capture area monitoring of the thickness of a pipe component. In yet another embodiment, the probe assembly 2 may include a wireless device that captures accurate point measurements without direct contact with a plumbing component that requires thickness monitoring. The probe assembly may include one or more thickness monitoring ultrasonic transducers, area monitoring ultrasonic transducers, and/or combinations thereof configured to verify (validate) general corrosion in the pipeline system (e.g., confirm that no localized corrosion is detected).

Fig. 13B is a diagram of an illustrative conduit with installed sensors. Sensor 1302 can be any of a variety of types of sensors configured to measure a thickness of the pipeline at or near the point of installation of sensor 1320 on the pipeline. Sensor 1302 is typically mounted in a permanent location and remains attached to the pipeline for an extended period of time, such as for the life span of that circuit of the pipeline, for more than five years, for more than three years, or other time period. Although the sensor 1302 shown in fig. 13B is mounted to the exterior of a pipeline and is tethered with a wire, in some examples according to one or more aspects of the present disclosure, the sensor may not be tethered and wirelessly communicate data to one or more wireless receiver/transceiver devices. In addition, although the sensor shown in fig. 13B is shown in a straight line pattern along the longitudinal direction (longitude) of the pipe, the present disclosure contemplates sensors mounted in any of several different patterns. For example, the density of the installed sensors may be based on the direction of gravity and the type of material being transported in the pipeline. For example, assume that the pipeline in FIG. 13B transports a liquid from left to right along the length of the pipeline in one example, where the bottom of the pipeline is the portion of the pipeline where sensor 1302 is mounted. In such examples, the sensors mounted on the pipeline may be distributed around the circumference of the pipeline, taking into account that climatic conditions (e.g., rain, hail, sun) may expose a portion of the pipeline to more likely deterioration while internal conditions of the pipeline (e.g., more liquid contacting the bottom of the pipeline than the top of the pipeline) may cause a portion of the interior of the pipeline to more likely deterioration.

FIG. 14 is an illustrative network architecture of an industrial facility with sensors, communication components, and other components in accordance with aspects of the present disclosure. The data analysis platform 112 may be communicatively coupled to one or more networked components via a network, such as a local area network 1408. For example, the data analysis platform 112 may output to a visualization platform 114 for generating one or more illustrative charts contained herein. A monitoring system may include a software platform 112 to remotely monitor and analyze historical measurements acquired by a plurality of sensors attached to the pipe and components. The monitoring system may include analytical tools for monitoring, diagnosing, and/or predicting areas for which localized corrosion is a candidate (e.g., because the system is unable to confirm general corrosion of the area). By employing the analytical system disclosed herein, TML can be optimized to reduce the number of measurement sites, i.e., screening, among other things, without compromising risk.

In another example, the data analysis platform 112 may trigger the generation of an alert at a remote alert device 1410. The remote alert device 1410 may cause an immediate inspection of one or more components or cause particular pipeline components to be prioritized for a subsequent inspection of the facility.

As measurements and other data are collected by the system 1400, the data may be stored in a data store 1406, which data store 1406 is communicatively coupled to the data analysis platform 112 and has access to the data analysis platform 112. In some examples, the data may be stored in computer memory 1404, however, the amount of computer memory required may be high. Instead (instead), in some examples, a model 1412, such as a machine learning artificial neural network, may be stored at computer memory 1404 for execution by a processor 1402, while historical data and other data may be stored at a data memory 1406. In some examples, the data store may be moved into the platform 112, but for illustrative purposes the data store is shown in communication with the platform 112 through the local area network 1408.

FIG. 15 is an illustrative diagram of a grouping of multiple sensors (e.g., probe assemblies) in an embodiment of the present disclosure. As shown in fig. 15, each probe assembly may be assigned a unique TML identifier (TML ID). The TML ID may be any unique letter, character or other identifier to uniquely identify each TML (i.e., probe assembly). In fig. 15, a grouping of probe assemblies is given around a bold rectangular box selecting a TML ID number. In 1502, at 3-7-2007, the system has grouped probe assemblies 4, 5, 6, and 7 into one group based on a rule or rules. At 1504, a graphical representation of the data stored in computer memory 1404 gives that system 1400 has adjusted the grouping to include/exclude one or more TMLs at 3-7-2008. In 1504, the model may suggest that the probe assembly corresponding to TML ID number 4 should no longer be part of the group ID corresponding to the bold rectangular box in 1504. As a result, one or more probe assemblies screened for that group ID may also change. Finally, at 1506, the graphical illustration shows that system 1400 has further tuned grouping probe assemblies 5 and 6 to now be grouped into a first set of IDs and probe assemblies 7 and 8 to be grouped into a second, different/independent set of IDs at 3-7-2009. As a result, as discussed below in fig. 16, the screening and risk conditions for the overall system 100 will vary.

In one example, grouping TMLs into a set of IDs may be accomplished in one of several different ways. For example, the initial grouping of circuits for multiple components at a facility may be based on measured data levels. For each date that a probe assembly takes a measurement, a new group may be triggered if a probe assembly meets any of the following conditions: (i) if the probe assembly is the first TML of the circuit; (ii) If the (absolute) difference between the measured value and the measured value of the previous TML exceeds the standard deviation (standard deviation) of about 0.5 to 3.0 for all measurements on that date, the value of this parameter may be decreased for more conservative groupings, or the value of this parameter may be increased for more aggressive groupings; (iii) If the nominal wall thickness measurement of the TML is different than the nominal wall thickness measurement of the previous TML, or if the TML has historically only one measurement (across all dates). In another example, grouping of TMLs may be accomplished in a multi-step process. In a first step, all measurements made in a set of connected components (e.g., a circuit) on a particular date (or any other predetermined period of time, e.g., within an hour window of time, within the same week, or otherwise) can be compared to determine how many pairs (or tuples) are measured on the particular date. In one example, any TML pair group that is lower than a predetermined percentage (e.g., 70%, 80%, 60%, 75%, or other percentage) of the total measurements over the investigation year (or other time period) is deleted. Next, the smallest measure of all TMLs may be identified and all TMLs of a group that would be paired with that TML in an earlier (e.g., first) step are assigned to the same group ID. Other examples of rules for TMLs packets will be apparent to those skilled in the art after reading the entire disclosure herein.

Still other illustrative rules for grouping TMLs are also contemplated in this disclosure. For example, in some rules, packets may be reassigned based on the percentage of TML versus groups. For a circuit having multiple components of at least two measurement dates, groups of TMLs grouped together are at least a predetermined threshold percentage of the two times to remain in the same group, but TMLs that do not meet this threshold may be assigned individually to different groups using one or more rules. In another example, measurement dates that do not have sufficient TMLs may be discarded (drop). For each circuit of the plurality of components, in some examples, system 1400 may consider only those measurement dates having a percentage of a predetermined threshold value of at least the maximum number of TMLs for any one date. TMLs appearing in dates that do not meet the threshold may be individually assigned to separate groups.

In some examples, system 1400 may discard (e.g., discard) the measurements that appear invalid based on lack of historical data and continue to repackage TMLs based on one or more rules described herein. The threshold used is a super parameter (hyperparameters) that can be adjusted based on diversity (diversity) and quality (quality) of the data set. The adjustment may be made at the end of the data validation and verification process. In one example, the percentage of threshold may be set to 75%, but for some TMLs, previous measurements may not have occurred over the past many years. In some embodiments, a hyper-grid (hyper-grid) may be generated and used to adjust parameters and/or hyper-parameters of the system 1400. In some examples, the threshold setting may be strongly correlated to how many TML measurements have been collected by a system 1400 for each TML ID. Thus, the threshold may be adjusted up or down based on how much data is available to the system 1400.

Fig. 16 and 17 illustrate graphs of various data collected and/or analyzed by system 1400. Fig. 16A shows a probability map of the measured values (normal) 95%) with the percentages on the Y-axis and the measured values on the X-axis. The system 1400 defaults to assuming that general corrosion has been detected, except when the curve shows that it is not traveling vertically (such as near the top of the curve in fig. 16A). The data analysis platform 112 may verify that the pipe wall thickness measurements of the probe assembly are general corrosion rather than localized corrosion by performing one or more steps. For example, in some embodiments, the verification is performed by: a probability map of all pipe wall thickness measurements associated with the pipeline system is generated, the mapped (plotted) pipe wall thickness measurements are then grouped by nominal thickness (nominal thickness), and a non-linear relationship is identified in the probability map of pipe wall thickness measurements grouped by nominal thickness to confirm that corrosion may not be general corrosion. Meanwhile, if a linear relationship is shown in the figure, TMLs corresponding to those data points in the figure exhibit general corrosion. This approach is an improvement over systems that have employed standard deviations to build a normal probability map (normal probability plot). In addition, the verification step adds a further assurance that the system 1400 is accurately detecting general corrosion and accordingly acting to screen the appropriate probe assemblies mounted on the components in the facility. The system 1400 should not generate an alarm (e.g., from the device 1410) for general corrosion, as general corrosion is ubiquitous and is typically not of primary concern during inspection. Conversely, general corrosion is a concern in planning and scheduling the replacement of components in a facility in large numbers.

Referring to fig. 16B and 16C, these diagrams show the relationship between the risk of false identification of general corrosion and the number of Thickness Measurement Sites (TMLs). Although the risk amount (amountof risk) is progressive to a minimum amount of risk threshold 1602, regardless of the number of measurement sites, fig. 16B shows that the risk level plotted against the number of probe assemblies (i.e., TMLs) decreases with more TMLs additions. Meanwhile, the effect of the systems and methods disclosed herein is shown in fig. 16C, fig. 16C showing a shift (shift) after screening in the plotted curve of risk level versus TMLs number. Fig. 16B and 16C will be described in more detail below in connection with the method steps shown in the flow chart of fig. 22. Meanwhile, fig. 17 is a graph showing a cumulative thickness distribution of the naphthenic acid etched tubes in the prior art system known in the art.

FIG. 18A is a graph showing a corrosion sensor analysis of TML measurements by date for a particular circuit ID (or asset ID). The X-axis corresponds to the TML identifier. For practical purposes, a probe assembly mounted on a pipeline system may assign identifiers in a sequential or other ordered (ordered) order along the circuit formed by the pipeline system. Each TML may have an ID that gives its location upstream or downstream on the pipe. Other data cleansing (cleaning) and/or flushing (flushing) of the TML based on the location data may be performed to reconcile (harmony)/standardize (standard ize) measured data for analysis. Each TML may be assigned a nominal thickness from the time the pipe is first installed. One or more publicly accessible databases (e.g., a Meridian database) may provide data, including measurements and specifications of nominal thickness. Meanwhile, as shown in the legend (legend) on the right hand side of fig. 18A, measurements may be made over a period of time so that historical data spanning at least several years (i.e., an extended period of time) may be stored and analyzed. In this example, wall thickness measurement data of approximately 25 years is stored, analyzed, and plotted in fig. 18A. The graph plot 1802 in FIG. 18A corresponds to the measurement taken 2015-08-03. Meanwhile, the other graphs in this figure correspond to thickness measurements made by each TML on a corresponding date spanning approximately 25 years (e.g., an extended period of time) rearward.

The data analysis platform 112 may set one or more hyper-parameters for the model 1412 corresponding to the graph depicted in fig. 18A. A hyper-parameter is typically set before the training/learning process begins on the model; in contrast, the values of the other parameters are obtained by training of the model. In fig. 18A, a graphical user interface for adjusting the group sensitivity super-parameters is displayed on top. The visualization platform 114 may include a graphics tool/slider through which the super parameters may be adjusted. In fig. 18A, the packet sensitivity (grouping_sensitivity) super parameter shows the setting to the "standard" setting. Meanwhile, in fig. 18B, which shows another illustration of the mode 1412, the packet sensitivity superparameter shows the setting to a "medium" setting. As a result, the number of groups is only 61 in fig. 18B, and 97 in fig. 18A. In addition, the graph 1812 drawn in fig. 18B is slightly different from the graph 1802 in fig. 18A due to the change in the super parameter setting and TML selection (selection) method. Further, the graph 1822 plotted in fig. 18C is increasingly different from fig. 18A and 18B for the packet sensitivity super parameter being set to "high" in fig. 18C. The number of groups is about seventy-five, while the total number of TMLs remains constant at one hundred fifty-five.

The packet sensitivity superparameter refers to the sensitivity or aggressiveness (or aggressiveness) of the TML packet, and can be applied in the initial packet phase. In some examples, a TML may be assigned to a new group when the (absolute) difference between the measured value of the TML and the measured value of the previous TML is greater than 1 Standard Deviation (SD) for all measurements on that date. This threshold can be adjusted for more conservative or more aggressive groupings. A threshold of less than 1SD will render the packet more sensitive to changes in the measurement and will result in a more conservative packet. On the other hand, a threshold value greater than 1SD will render the packet less sensitive to changes in the measurement and result in a more aggressive packet (e.g., higher packet rate). In one example, five different packet sensitivities may be implemented, as shown in fig. 18C, decreasing from most conservative to most aggressive sensitivities as follows: high (0.5 SD), standard (1 SD), medium (1.5 SD), low (2 SD) and very low (3 SD). In another example, more or less than the five groupings described above may be employed to provide finer or coarser sensitivity. Since the packet sensitivity superparameter is applied in the initial packet phase, as is the case with superparameter (case hyperparameters), all subsequent packet steps can be re-run based on the initial packet result, i.e. the whole packet cycle (cycle) is repeated five times, once for each of the five packet sensitivity levels.

Note that fig. 18A, 18B, and 18C (collectively, "fig. 18") list a number of TML selection methods that can be applied to measurements to optimize the grouping and mapping of data points. Although FIG. 18 lists three optimization functions, namely a median (medium_TML_within_groupID) of TMLs within a group ID, a minimum average (minimum_average_TML_within_groupID) of TMLs within a group ID, and a minimum deviation (minimum_variation_from_mean) from the mean, other optimization functions may be used in accordance with one or more aspects of the present disclosure. For example, a TML location (TML_position) optimization function may be employed, and if one TML is to be selected, the TML located at the center of the group is selected. If two TMLs are to be selected, the group will be divided into two subgroups and the TMLs at the center of each subgroup are selected, and so on. Other examples of TML selection methods are also contemplated herein. For example, the optimization function may be a minimum_average_tml_witin_groupid optimization function. In the minimum_average_tml_within_groupid method for deciding which TMLs to choose from among the groups, the method selects the TML with the lowest average measurement in the groups (date of crossing). For example, in one illustrative system employing a minimum_average_tml_within_groupid optimization function, the system may calculate an average measure (across dates) of each TML, rank (e.g., rise) the TMLs in each group by the average measure, and pick the first n TMLs based on the number (n) of TMLs to be selected from each group. Likewise, the media_tml_within_groupid optimization function is similar to the minimum_average_tml_within_groupid optimization function, but based on a median instead of a minimum average.

In another example, the optimization function may be a minimum_variation_from_mean optimization function. In the minimum_variation_from_mean method for deciding which TMLs to choose from among the groups, the method selects the TML with the lowest mean deviation (lowerest average vairation) compared to the mean measurement (mean measurement) of the group. For example, in one illustrative system employing a minimum_variation_from_mean optimization function, the system may calculate a mean group measurement for each date (means group measurement). Then, for each TML, an absolute difference from the mean is calculated for each date, and for each TML, an average deviation (e.g., an absolute difference from the mean) is calculated. Next, the minimum_variation_from_mean optimization function sorts (e.g., increases) the TMLs in each group by average bias and selects the first n TMLs based on the number (n) of TMLs to be selected from each group.

After the finalizing (grouping), the system 1400 determines a TML candidate selection method and the number of required probe assemblies per group. The candidates for each group may be another hyper-parameter. By default, the system 1400 can employ 1% or 1's for each TML group. If more than one TML candidate is to be selected, the TML group may be divided into equally large subgroups while preserving the ordering of TMLs. The system 1400 may then apply a TML candidate selection method according to one or more scenarios described herein.

Fig. 19A is a graph 1902 showing measured TC thickness (millimeters) versus time for a component. In addition, the graph may give a temperature calibration value (degrees celsius), a temperature coefficient (e.g., 1%), a corrosion rate ST (millimeters per year), a corrosion rate LT (millimeters per year), a remaining life (years) of the component, a remaining half life (also years), and an actual thickness (millimeters).

In addition, fig. 19B is a graph showing a correction (corrected) of FSH (percent) value and thickness value (millimeters or other units). The figure also shows the thickness ranges of a Gate (Gate) 1904 and B Gate 1906. Alternatively, the plot may be plotted with mV values instead of FSH. Further, in some examples, the map may be displayed as HF instead of corrected.

Fig. 20A is a graph 2002 of an acid battery facility showing measured TC thickness (millimeters) versus time for a component. In addition, the graph may give a temperature calibration value (degrees celsius), a temperature coefficient (e.g., 1%), a corrosion rate ST (millimeters per year), a corrosion rate LT (millimeters per year), a remaining life (years) of the component, a remaining half life (also years), and an actual thickness (millimeters). Further, fig. 20B is a graph showing the correction of FSH (percent) value versus thickness value (millimeters or other units) for an acid battery facility. The figure also shows the thickness ranges of a gate 2004 and B gate 2006. Alternatively, the plot may be plotted with mV values instead of FSH. Further, in some examples, the map may be displayed as HF instead of corrected.

Fig. 21 shows a simplified example of an artificial neural network 2100 on which a machine learning algorithm may be executed. FIG. 21 is merely an example of a non-linear process employing an artificial neural network; other forms of nonlinear processing may be used to implement a machine learning algorithm in accordance with the features described herein.

In fig. 21, each of the input nodes is connected to a first set of processing nodes. The external source 2102 fed into the input node may be an evaluation index (metrics) from the results of steps through the methods disclosed herein. Each of the first set of processing nodes is connected to each of a second set of processing nodes. Each of the second set of processing nodes is connected to each of the output nodes. Although only two sets of processing nodes are shown, any number of processing nodes may be implemented. Also, while each group has only four input nodes, five processing nodes, and two output nodes in fig. 21, each group may implement any number of nodes. The data flow in fig. 21 is shown from left to right as: data may be input to an input node, may flow through one or more processing nodes, and may be output by an output node. The input to the input node may come from an external source 2102. The output 2104 may be sent to a feedback system 2106 and/or data store. The feedback system 2106 may send the output to the input node using the same or different input data for successive iterations of the process.

In one illustrative approach employing feedback system 2106, the system may employ machine learning to determine an output. The output may include a leak region boundary, a multi-sensor detection event, confidence value, and/or classification output. The system may employ a suitable machine learning model including xg-boosted (xg-boosted) decision trees, auto-encoders, perceptrons, decision trees, support vector machines, regression, and/or neural networks. The neural network may be a suitable type of neural network including a feed forward network, radial basis network, cyclic neural network, long/short term memory, gated cyclic unit, automatic encoder, variable automatic encoder, convolutional network, residual network, kohonen (r) Huo En network, and/or other types. In one example, the output data in the machine learning system may be represented as a multi-dimensional array, an extension (extension) of a two-dimensional table (such as a matrix) pair with higher dimensional data.

The neural network may include an input layer, a number of intermediate layers, and an output layer. Each layer may have its own weight. The input layer may be configured to receive as input more than one feature vector disclosed herein. The middle layer may be a convolutional layer, a pooled layer, a dense (fully-connected) layer, and/or other type. The input layer may pass input to the intermediate layer. In one example, each intermediate layer may process the output from a previous layer and then pass the output to the next intermediate layer. The output layer may be configured to output a classification or a real value. In one example, the layers in the neural network may use an activation function, such as an sigmoid function, a hyperbolic tangent function, a modified linear function, and/or other functions. In addition, the neural network may include a loss function. In some examples, a loss function may measure many missed positive examples; alternatively, the loss function may also measure many false positives. The loss function may be used to determine an error when comparing an output value to a target value. For example, when training a neural network, the output of the output layer may be used as a prediction and may be compared to a target value for a training example to determine an error. The errors may be used to update the weights of the layers of the neural network.

In one example, the neural network may include a technique for updating weights of one or more of the layers based on the error. The neural network may use gradient descent to update the weights. Alternatively, the neural network may use an optimizer to update the weights in the layers. For example, the optimizer may update the weights in the layers using various techniques or combinations of techniques. Where appropriate, the neural network may include a mechanism to prevent overfitting—regularization (such as L1 or L2), dropping, and/or other techniques. The neural network may also increase the amount of training data used to prevent overfitting.

In one example, FIG. 21 illustrates that a node may perform various types of processing, such as discrete calculations, computer programming, and/or mathematical functions implemented by a computing device. For example, an input node may comprise logical inputs of different data sources, such as more than one data server. The processing nodes may include parallel processing performed on multiple servers of a data center. And, the output node may be a logical output that is ultimately stored in a result data store (such as the same or a different data server as for the input node). In particular, the nodes need not be distinct. For example, two nodes in any two sets may perform exactly the same processing. The same node may repeat for the same or different sets.

Each node may be connected to more than one other node. The connection may connect the output of one node to the input of another node. A connection may be associated with a weight. For example, one connection may be weighted more important or important than another connection, thereby affecting the degree of further processing (degree) because the input traverses the artificial neural network. Such connections may be modified such that the artificial neural network 2100 is learning and/or dynamically reconfigurable. Although the nodes are shown as nodes having connections only to the connection of fig. 21, connections may be formed between any of the nodes. For example, a processing node may be configured to send output to a previous processing node.

The input received in the input node may be processed by processing nodes such as a first set of processing nodes and a second set of processing nodes. This processing may result in an output in the output node. As shown by the connections from the first set of processing nodes and the second set of processing nodes, the process may include multiple steps or sequences. For example, the first set of processing nodes may be a coarse data filter, while the second set of processing nodes may be a more detailed data filter.

The artificial neural network 2100 may be configured to perform decision making. As a simplified example for purposes of explanation, the artificial neural network 2100 may be configured to detect faces in photographs. The input node may be provided with a digital copy of a photograph. The first set of processing nodes may each be configured to perform specific steps to remove non-face content, such as large consecutive red portions. The second set of processing nodes may each be configured to find rough approximations of the face, such as face shape and skin tone. The processing may also be refined by multiple subsequent groups, each looking for further more specific tasks, where each node performs some form of processing that does not necessarily require operation in the promotion of that task. The artificial neural network 2100 may then make a prediction of the location on the face. The prediction may be correct or incorrect.

The feedback system 2106 may be configured to determine whether the artificial neural network 2100 makes a proper decision. The feedback may include an indication of a correct answer and/or an indication of an incorrect answer and/or a degree of accuracy (e.g., a percentage). For example, in the face recognition example provided above, the feedback system 2106 may be configured to determine whether a face was correctly identified and, if so, how much percentage of the face was correctly identified. The feedback system may be known to a correct answer so that the feedback system can train the artificial neural network 2100 by indicating whether it makes a correct decision. The feedback system may include manual inputs such as an administrator telling the artificial neural network 2100 whether it makes a correct decision. The feedback system may provide feedback (e.g., an indication of whether the previous output was correct or incorrect) via the input nodes to the artificial neural network 2100 or may transmit such information to one or more nodes. The feedback system may additionally or alternatively be coupled to a memory so that the output is stored. The feedback system may not have a correct answer at all, but rather have further processing as a basis for feedback: for example, the feedback system may include a system programmed to recognize a human face, such that the feedback allows the artificial neural network 2100 to compare its results with the results of an artificially programmed system.

The artificial neural network 2100 may be dynamically modified to learn and provide better input. For example, the artificial neural network 2100 may modify itself based on previous inputs and outputs and feedback from the feedback system 2106. For example, the processing in the node may be changed and/or the connections may be weighted differently. Continuing with the example provided above, face predictions may be incorrect because the photographs provided to the algorithm are colored in a way that makes all faces appear red. In this way, a node that excludes photographs containing a large number of red neighbors may be considered unreliable and connections to that node may be significantly reduced in weight. Additionally or alternatively, the node may be reconfigured to process the photograph in a different manner. The modification may be a prediction and/or estimation (guess) by the artificial neural network 2100, such that the artificial neural network 2100 may vary its nodes and connections to test hypotheses.

The artificial neural network 2100 need not have a collective number of processing nodes or a collective number of processing nodes, but may increase or decrease its complexity. For example, the artificial neural network 2100 may determine that more than one processing node is not necessary or should be reused (repurposed), and either discard or reconfigure the processing nodes on that basis. As another example, the artificial neural network 2100 may determine that further processing of all or part of the input is required and append further processing nodes and/or sets of processing nodes on that basis.

The feedback provided by the feedback system 2100 may be merely enhanced (e.g., provide an indication that the output is correct or incorrect, reward a machine learning algorithm for many points, etc.) or may be specified (e.g., provide a correct output). For example, a machine learning algorithm may be required to detect faces in photos. Based on an output, the feedback system may indicate a score (e.g., 75% accuracy, an indication that the estimate is accurate, etc.) or a specific response (e.g., specifically identifying where the face is located). In one example, a human operator/inspector may focus the inspection on a list of TMLs being screened. Once any localized corrosion is identified and repaired, the operator may indicate as much so that the model 1412 can be updated to reflect the new wall thickness value. Additionally, in some examples, a localized corrosion may be falsely identified in the system 1400 and the supervisor input into a machine learning or neural network executing in the digital analysis platform 112, the alarms and models of which may be refined accordingly.

The artificial neural network 2100 may be supported or replaced by other forms of machine learning. For example, one or more of the nodes of the artificial neural network 2100 may implement a decision tree, associated rule set, logic programming, regression model, cluster analysis mechanism, bayesian (Bayesian) network, proposition formula, generative model, and/or other algorithmic or formal decision making. The artificial neural network 2100 may perform deep learning.

Fig. 22 is a flowchart showing illustrative steps of a method 2200 performed in accordance with some embodiments disclosed herein. The method 2200 may be performed by a system 1400 when computer-executable instructions stored in a non-transitory computer-readable medium are executed by a processor. Method 2200 can screen from probe assemblies installed on a pipeline system in an industrial facility, among other things. As a result, systems and methods for optimizing asset health monitoring are improved because representative measurement sites are identified by screening and remaining probe assemblies can be disregarded when routine inspection of pipeline systems and other components in an industrial facility.

With respect to FIG. 22, in step 2202, in a computer memory 1406 communicatively coupled to the processor 1402, the system stores historical pipe wall thickness measurements acquired from probe assemblies 150A installed on a pipeline system in an industrial facility over a period of time. In step 2204, the data analysis platform 112 may set one or more superparameters, such as, but not limited to, a packet sensitivity superparameter, a threshold measurement superparameter, a packet size superparameter, and/or combinations thereof. Once these hyper-parameters are set, in step 2206, the system 1400 may begin training the model with at least historical pipe wall thickness measurements stored in computer memory 1406 and hyper-parameter values stored in computer memory 1404.

In step 2208, the model stored in the computer memory 1404 may group a first group of probe assemblies from among a plurality of probe assemblies installed on the pipeline system. As explained in this disclosure, such as with respect to fig. 15, several methods are provided by which groupings can be generated in the model. After grouping the probe assemblies, the data analysis platform 112 may assign a unique group identifier (groupID) to each group of probe assemblies. The unique group ID may be any identifier that the system 1400 can use to uniquely reference (refer to) the probe assembly group.

In step 2210, the data analysis platform 112 selects an optimization function for operation of the system 1400 based on at least the trained model. Many illustrative optimization functions are described in this disclosure, including, but not limited to, a media_tml_within_groupid optimization function, a minimum_average_tml_within_groupid optimization function, a minimum_variation_from_mean optimization function, and/or a tml_position optimization function. The decision to select a particular optimization function triggers the subsequent identification and measurement steps to take effect. For example, in steps 2212A, 2212B, 2212C, and 2212D (collectively, "step 2212"), the system 1400 identifies a probe assembly corresponding to each set of IDs for use in pipe wall thickness monitoring of the pipeline system based on the models stored in the computer memory 1404 and the selected optimization functions. In some examples, system 1400 may identify a single probe assembly for the entire group ID to represent the area being measured. In other examples, multiple probe assemblies may be identified that represent the group ID. The number of TMLs assigned to be screened from a group may be based on one or more rules. This is based on the maximum standard deviation of the group, which in one example: if the maximum standard deviation of any one of the groups is less than or equal to 0.25, then one TML is selected from that group. If the maximum standard deviation increases by 0.25, the number of TMLs selected increases by one (i.e., if it is between 0.25-0.5, two probe assemblies can be selected from the group, and so on). Furthermore, a step value of 0.25 may be modified for adjusting the sensitivity of the TML selection. Values below 0.25 will result in more TML being selected in each group (i.e., a conservative approach), while values above 0.25 will result in less TML being selected in each group (i.e., an aggressive approach).

In step 2214, during inspection, system 1400 may disregard (disregard) all remaining probe components in each group ID except for the probe components identified from each group ID. The system may measure the wall thickness of each of the identified probe assemblies in each group ID, but exclude other probe assemblies in the group ID. Thus, the system screens from multiple probe assemblies mounted on a pipeline system. At least one benefit of the use of screening the number of probe assemblies during an inspection is the time savings that it results. For example, a human inspector who previously inspected each probe assembly can now inspect the measurements on a reduced number of probe assemblies without greatly increasing the risk of missing dangerous localized corrosion. In one example, at step 2214, the system 1400 may output a human-readable report listing those probe assemblies that a human inspector should manually inspect the wall thickness measurements. The reports may be ranked in various ways, such as based on highest risk of localized corrosion, based on geographic convenience from a known starting location of a human inspector, or other order.

For example, as shown in fig. 16B and 16C, when the risk amount is plotted against the number of measurement sites, the risk amount is asymptotically below a threshold value of a minimum risk amount 1602, regardless of the increase in the number of measurement sites. Importantly, as the number of measurements decreases, as shown in graph 1604, the incremental change of risk (delta change) increases at an increasingly faster rate, in other words, decreasing the number of probe assemblies sampled increases the risk to an unsafe level. However, the system 1400 and method 2200 disclosed herein move graphics from an initial one-risk graphic 1606 to a more advantageous one-risk graphic 1608. Thus, by identifying those probe assemblies that are statistically most likely to cause general corrosion/degradation to the pipeline wall, the number of probe assemblies that need to be actively inspected during an inspection is screened, thereby reducing inspection time/cost while maintaining (or even reducing) risk conditions.

Finally, in step 2216 of fig. 22, the thickness monitoring controller 130 may receive and transmit the pipe wall thickness measurements from the probe assemblies of each group ID for verification. The thickness monitoring controller 130 may send the measurement data (and any other data) to the data store 1406 for historian retention and analysis, and to the data analysis platform 112 for analysis and visualization generation. For example, wall thickness measurements may show that a particular segment of a pipe in a pipeline system is experiencing degradation other than general corrosion, so that it rises to dangerous, localized corrosion levels and must be replaced within a particular period of time. In another example, the pipe wall thickness measurements may be made at one or more of a pipe, a tank, a vessel, and/or a pipeline at a facility.

Although specific embodiments are shown in the drawings and described with respect to the drawings, it is contemplated that various modifications may be devised by those skilled in the art without departing from the spirit and scope of the following claims. It will be appreciated, therefore, that the scope of the present disclosure and appended claims is not limited to the specific embodiments shown in and described with respect to the drawings and that modifications and other embodiments are intended to be included within the scope of the present disclosure and appended drawings. Furthermore, while the foregoing description and associated drawings describe example embodiments in the context of certain example combinations of elements and/or functions, it should be appreciated that different combinations of elements and/or functions may be provided by alternative embodiments without departing from the scope of the present disclosure and the appended claims. Furthermore, the foregoing description illustrates a method that describes the performance of a number of steps. Unless stated to the contrary, one or more steps in a method may not be required, one or more steps may be performed in a different order than illustrated, and one or more steps may be formed substantially simultaneously. The various aspects are capable of other embodiments and of being practiced or of being carried out in various ways. It is to be understood that the phraseology and terminology used herein is for the purpose of description and should not be regarded as limiting. Rather, the phraseology and terminology used herein is to be given its broadest interpretation and meaning. The use of "including" and "comprising" and variations thereof is meant to encompass the items listed thereafter and equivalents thereof as well as additional items and equivalents thereof.

Claims

1. A method for screening from a probe assembly mounted on a pipeline system, wherein the probe assembly is configured for pipeline wall thickness monitoring, the method comprising:

before training a model, setting a grouping sensitivity super-parameter, a threshold measurement super-parameter and a grouping size super-parameter for the model;

grouping a first set of probe assemblies based on at least historical pipe wall thickness measurements acquired from probe assemblies installed on the pipeline system over a period of time by executing the model on a processor;

assigning a unique group ID to each group of probe assemblies;

after training the model, selecting, by the model, an optimization function from a plurality of optimization functions for the model;

identifying a single probe assembly corresponding to each set of IDs by the model for use in pipe wall thickness monitoring of the pipeline system;

a tube wall thickness measurement from the individual probe assemblies of each set ID is sent for verification by a thickness monitoring controller associated with the tubing system.

2. The method of claim 1, wherein,

the probe assembly includes at least one resistance temperature detector configured to detect localized corrosion in the pipeline system, a thickness monitoring ultrasonic transducer, and an area monitoring ultrasonic transducer.

3. The method of claim 2, further comprising:

the single probe assembly pipe wall thickness measurement was verified to be general corrosion rather than localized corrosion by:

generating a probability map of all pipe wall thickness measurements associated with the pipeline system;

grouping drawn pipe wall thickness measurements by nominal thickness; and

a non-linear relationship was not identified on the probability map of the pipe wall thickness measurements grouped by nominal thickness to confirm general corrosion.

4. The method of claim 1, further comprising:

during inspection, all remaining probe assemblies in each group ID are not considered, except for a single probe assembly from each group ID, to reduce the number of inspection samples measured without compromising a risk condition of the pipeline system.

5. The method of claim 1, wherein,

the grouping of the first set of probe assemblies is further based at least on inspection information provided to the system and historical pipe wall thickness measurements acquired from probe assemblies installed on the pipeline system over a period of time.

6. The method of claim 1, wherein,

the plurality of optimization functions includes a median value of the TMLs within the group ID, a minimum average of the TMLs within the group ID, and a minimum deviation from the average.

7. The method of claim 6, wherein,

the plurality of optimization functions includes TML locations.

8. The method according to claim 1,

wherein the pipeline system may comprise a tank,

and wherein a first probe assembly of the probe assemblies is configured to measure a wall thickness of the tank.

9. The method according to claim 1,

wherein the conduit wall thickness monitoring comprises measuring the conduit wall thickness at a particular probe assembly,

wherein the conduit wall is at one or more of a conduit, tank, vessel and pipeline.

10. The method of claim 9, wherein,

the pipe wall thickness monitoring includes: the original wall thickness, loss of wall thickness over time, calibration errors, and measurement site repeatability errors are analyzed by the probe assembly.

11. The method of claim 1, further comprising:

storing in a computer memory communicatively coupled to the processor historical pipe wall thickness measurements acquired from the probe assembly mounted on the pipeline system over an extended period of time; and

the model is trained by the processor using at least the historical pipe wall thickness measurements stored in the computer memory.

12. The method of claim 11, wherein,

the model includes an artificial neural network.

13. A system for detecting localized corrosion of a plurality of components transporting material across a distance, the system comprising:

a plurality of probe assemblies attached to one or more of the components, wherein each of the plurality of probe assemblies corresponds to a unique identifier;

a data memory configured to store historical wall thickness measurements acquired over a period of time from measurements performed by the probe assembly;

a model trained on said historical wall thickness measurements in said data store, and wherein said superparameters may include at least one packet sensitivity superparameter; and

a monitoring device comprising a processor and a memory storing computer executable instructions that, when executed by the processor, cause the system to perform steps comprising:

grouping a first set of probe assemblies based on the model;

assigning a unique group ID to each group of probe assemblies;

selecting an optimization function from a plurality of optimization functions based on the model;

Identifying a probe assembly for wall thickness monitoring of the component based on the model and the selected optimization function for each set of IDs, wherein each set of IDs corresponds to the unique identifier corresponding to the identified probe assembly; and

a list corresponding to the unique identifier of any group ID is output.

14. The system according to claim 13,

wherein the plurality of probe assemblies comprises: at least one thickness monitoring ultrasonic transducer and one area monitoring ultrasonic transducer configured to detect localized corrosion of the component,

wherein the probe assemblies identified from each set of IDs include more than one probe assembly of the plurality of probe assemblies,

and wherein the memory of the monitoring device stores computer-executable instructions that, when executed by the processor, cause the system to perform steps that may include:

transmitting, by a thickness monitoring controller associated with the component, a wall thickness measurement of the probe assembly from each set of IDs for verification;

during inspection, not considering all remaining probe assemblies in each group ID except for the more than one probe assembly from each group ID; and

Verifying that wall thickness measurements of the more than one probe assembly for each set of IDs did not identify general corrosion.

15. The system of claim 13, wherein,

the wall thickness measurement of the probe assembly from a first set of IDs includes a wall thickness of a pipe component at the probe assembly.

16. The system of claim 13, wherein,

the wall thickness measurement of the probe assembly from a first set of IDs includes a thickness of a wall of a tank component at the probe assembly.

17. The system according to claim 13,

wherein the super-parameters include a packet sensitivity super-parameter, a threshold measurement super-parameter, and a packet size super-parameter

And wherein the plurality of optimization functions includes a median value of the TMLs within the group ID, a minimum average of the TMLs within the group ID, a minimum deviation from the average, and a TML location.

18. A non-transitory computer readable medium storing computer executable instructions that when executed by a processor cause a system to screen from probe assemblies installed on a pipeline system by performing steps comprising:

storing in a computer memory communicatively coupled to the processor historical pipe wall thickness measurements over a period of time acquired from the probe assembly mounted on the pipeline system;

Setting a super parameter for a model;

training, by the processor, the model using at least the historical pipe wall thickness measurements stored in the computer memory;

grouping a first set of probe assemblies based on the model executing on the processor;

assigning a unique group ID to each group of probe assemblies;

identifying a probe assembly corresponding to each set of IDs for use in pipe wall thickness monitoring of the pipeline system based on the model and the selected optimization function; and

a wall thickness measurement of the probe assembly from each set of IDs is sent for verification by a thickness monitoring controller associated with the part.

19. The non-transitory computer readable medium of claim 18,

wherein the super-parameters include at least one of a packet sensitivity super-parameter, a threshold measurement super-parameter, and a packet size super-parameter;

20. The non-transitory computer readable medium of claim 18, further storing computer executable instructions that, when executed by the processor, cause the system to perform steps comprising:

During inspection, all remaining probe assemblies in each group ID are not considered, except for the identified probe assembly from each group ID.