CN116907772A - Self-diagnosis and fault source identification method and system of bridge structure monitoring sensor - Google Patents

Self-diagnosis and fault source identification method and system of bridge structure monitoring sensor Download PDF

Info

Publication number
CN116907772A
CN116907772A CN202310878884.XA CN202310878884A CN116907772A CN 116907772 A CN116907772 A CN 116907772A CN 202310878884 A CN202310878884 A CN 202310878884A CN 116907772 A CN116907772 A CN 116907772A
Authority
CN
China
Prior art keywords
sensor
data
abnormal
anomaly
fault
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202310878884.XA
Other languages
Chinese (zh)
Inventor
龚加兴
韩坤林
刘大洋
石永燕
宋刚
桑晓玉
宋纯冰
斯新华
邢春超
刘文韬
刘鹏
杨超华
李磊
杨小庆
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China Merchants Chongqing Highway Engineering Testing Center Co ltd
Original Assignee
China Merchants Chongqing Highway Engineering Testing Center Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by China Merchants Chongqing Highway Engineering Testing Center Co ltd filed Critical China Merchants Chongqing Highway Engineering Testing Center Co ltd
Priority to CN202310878884.XA priority Critical patent/CN116907772A/en
Publication of CN116907772A publication Critical patent/CN116907772A/en
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01MTESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M5/00Investigating the elasticity of structures, e.g. deflection of bridges or air-craft wings
    • G01M5/0008Investigating the elasticity of structures, e.g. deflection of bridges or air-craft wings of bridges
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01MTESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M5/00Investigating the elasticity of structures, e.g. deflection of bridges or air-craft wings
    • G01M5/0066Investigating the elasticity of structures, e.g. deflection of bridges or air-craft wings by exciting or detecting vibration or acceleration
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01MTESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M5/00Investigating the elasticity of structures, e.g. deflection of bridges or air-craft wings
    • G01M5/0075Investigating the elasticity of structures, e.g. deflection of bridges or air-craft wings by means of external apparatus, e.g. test benches or portable test systems
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks
    • G06N3/04Architecture, e.g. interconnection topology
    • G06N3/044Recurrent networks, e.g. Hopfield networks
    • G06N3/0442Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks
    • G06N3/04Architecture, e.g. interconnection topology
    • G06N3/048Activation functions
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00Computing arrangements using knowledge-based models
    • G06N5/04Inference or reasoning models
    • G06N5/048Fuzzy inferencing

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Computing Systems (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Mathematical Physics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Molecular Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biophysics (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Automation & Control Theory (AREA)
  • Fuzzy Systems (AREA)
  • Testing Or Calibration Of Command Recording Devices (AREA)

Abstract

The application provides a self-diagnosis and fault source identification method and system of a bridge structure monitoring sensor.

Description

Self-diagnosis and fault source identification method and system of bridge structure monitoring sensor
Technical Field
The application relates to the technical field of bridges, in particular to a self-diagnosis and fault source identification method and system of a bridge structure monitoring sensor.
Background
At present, the bridge structure monitoring system is widely applied to real-time monitoring of the health condition of a bridge structure. Each structural health monitoring sensor is mainly used for monitoring the environment, action, structural response and structural change of a bridge. The sensor is used as an important component of a bridge structure monitoring system and is responsible for collecting and transmitting monitoring data of a structure. However, with the increasing complexity and large scale of monitoring systems, relying solely on manual analysis and diagnosis of large amounts of sensor data has become increasingly difficult. In addition, with the development of sensor technology, the accuracy and stability of the sensor have reached a very high level, but it is still unavoidable that malfunctions and anomalies occur. Once sensor data is abnormal, the current mode is to diagnose whether the sensor is faulty or not, and under the condition that the sensor is not abnormal, whether the bridge structure is abnormal or not is judged manually. The sensor cannot realize self-diagnosis in the mode, mainly relies on manual investigation, and has relatively low determination efficiency of abnormal sources. Therefore, a method capable of rapidly realizing self-diagnosis and fault source identification of the bridge structure monitoring sensor is needed.
Disclosure of Invention
Aiming at the defects existing in the prior art, the application provides a self-diagnosis and fault source identification method and system of a bridge structure monitoring sensor, which are used for solving the technical problems that faults and abnormal sources are difficult to determine in the prior art.
A self-diagnosis and fault source identification method of a bridge structure monitoring sensor comprises the following steps: periodically collecting transmission data of each sensor as initial data; constructing a long-term and short-term memory model to perform exception analysis on the initial data to obtain exception data carrying a sensor mark; analyzing the abnormal data by adopting a fuzzy logic algorithm, and determining a fault mode of a corresponding sensor; based on the corresponding sensor, an anomaly source is determined from the failure mode.
In one embodiment, the step of periodically collecting the transmission data of each sensor as initial data includes: setting data acquisition frequency according to the initial condition; and based on the data acquisition frequency, periodically acquiring the transmission data of the sensor as initial data.
In one embodiment, after the step of periodically collecting the transmission data of each sensor as initial data, the method further includes: preprocessing the initial data; and analyzing the preprocessed initial data according to a preset strategy, and determining an abnormal mode.
In one embodiment, after the step of constructing the long-term and short-term memory model to perform the anomaly analysis on the initial data, the method further includes: if the abnormal data carrying the sensor marks is not output, the initial data is recorded, and the transmission data of each sensor is repeatedly and periodically acquired as the initial data.
In one embodiment, the construction is long-short termThe memory model carries out an abnormality analysis step on the initial data, which comprises the following steps: let the input data be x t The hidden state is h t Forgetting door f t The input gate is i t The output gate is o t The unit state is C t The following steps are:
f t =σ(W f ·[h t-1 ,x t ]+b f )
i t =σ(W i ·[h t-1 ,x t ]+b i )
o t =σ(W o ·[h t-1 ,x t ]+b o )
h t =0 t *tanh(C t )
where W and b are model parameters, σ is Sigmoid activation function, which represents element multiplication, [ h ] t-1 ,x t ]Indicating that the last hidden state and current input was connected.
In one embodiment, the step of analyzing the abnormal data by using a fuzzy logic algorithm to determine the fault mode of the corresponding sensor includes: setting an abnormal state vector a= [ a ] 1 ,a 2 ,...a n ] T Wherein a is i Is the confidence of the ith abnormal state, and the value range is [0,1]]Wherein 0 indicates that the anomaly is not believed at all, and 1 indicates that the anomaly is believed to be fully; setting a fault mode vector b= [ B ] 1 ,b 2 ,...b m ] T Wherein b j Is the confidence of the j-th fault mode, and the value range is also [0,1]The method comprises the steps of carrying out a first treatment on the surface of the Setting a fuzzy relation matrix R, wherein R is an n multiplied by m matrix ij Indicating the confidence of the jth fault mode in the ith abnormal state; in performing fuzzy reasoning, the following equation is used to calculate the failure mode vector b= [ B ] 1 ,b 2 ,...b m ] T
B=A T .R
Confidence b of a certain failure mode j And determining that the label after defuzzification is a corresponding fault mode, and determining that the label is a fault mode of a corresponding sensor.
In one embodiment, the step of determining the source of the anomaly from the failure mode based on the corresponding sensor includes: weighted average of sensors triggering abnormal states, the weight being determined by the weight factor of each sensor, the weighted average abnormal statesThe calculation can be made by the following formula:
wherein q represents the number of sensors, W l Representing the weight factor of each sensor, C l Is the reported abnormal state.
In one embodiment, the step of determining the source of the abnormality from the failure mode based on the corresponding sensor further includes: preset upper and lower threshold value theta' 1 And θ' 2 If (if)Or->The anomaly originates from the bridge structure; if it isThe anomaly originates from the sensor.
In one embodiment, after the step of determining the source of the abnormality according to the failure mode based on the corresponding sensor, the method further includes: when the abnormality is derived from a sensor, performing self-checking on the corresponding sensor; when the abnormality comes from the bridge structure, a preset alarm system is triggered to perform abnormality early warning.
The self-diagnosis and fault source identification system of the bridge structure monitoring sensor comprises a data acquisition module, a data analysis module and a fault determination module, wherein: the data acquisition module is used for periodically acquiring the transmission data of each sensor as initial data; the data analysis module is used for constructing a long-term and short-term memory model to perform exception analysis on the initial data to obtain exception data carrying a sensor mark; the data analysis module is also used for analyzing the abnormal data by adopting a fuzzy logic algorithm and determining a fault mode of the corresponding sensor; the fault determination module is used for determining an abnormal source according to the fault mode based on the corresponding sensor.
According to the technical scheme, the beneficial technical effects of the application are as follows:
1. the method comprises the steps of periodically collecting transmission data of the sensor, analyzing the collected sensor data in real time by adopting a long-short-term memory (LSTM) model, finding abnormal conditions timely, and simultaneously, predicting and self-diagnosing a fault mode by applying a fuzzy logic algorithm to further determine the source of the fault abnormality, so that the efficiency and the accuracy of monitoring the bridge abnormality source are greatly improved.
2. The source of the abnormal state is compared with a preset threshold value by calculating the weighted average abnormal state value of the sensor, so as to distinguish whether the problem is caused by the sensor fault or the bridge structure problem.
3. And the continuous monitoring of the abnormal state is carried out at regular intervals, so that the continuity of data acquisition is ensured, and the maintenance efficiency is improved.
4. When the problem that self-repairing cannot be achieved is encountered, namely, the condition that the abnormality is derived from a bridge structure, the method can also inform maintenance personnel of intervention in time by triggering an alarm system so as to ensure the safety of the bridge.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below. Like elements or portions are generally identified by like reference numerals throughout the several figures. In the drawings, elements or portions thereof are not necessarily drawn to scale.
FIG. 1 is a schematic flow chart of a method for self-diagnosis and fault source identification of a bridge structure monitoring sensor according to one embodiment;
FIG. 2 is a block diagram of a device for self-diagnosis and fault source identification of a bridge structure monitoring sensor in one embodiment.
Detailed Description
Embodiments of the technical scheme of the present application will be described in detail below with reference to the accompanying drawings. The following examples are only for more clearly illustrating the technical aspects of the present application, and thus are merely examples, and are not intended to limit the scope of the present application.
It is noted that unless otherwise indicated, technical or scientific terms used herein should be given the ordinary meaning as understood by one of ordinary skill in the art to which this application belongs. The terms first, second and the like in the description and in the claims of the embodiments of the disclosure and in the above-described figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate in order to describe embodiments of the present disclosure. Furthermore, the terms "comprise" and "have," as well as any variations thereof, are intended to cover a non-exclusive inclusion. The term "plurality" means two or more, unless otherwise indicated. In the embodiment of the present disclosure, the character "/" indicates that the front and rear objects are an or relationship. For example, A/B represents: a or B. The term "and/or" is an associative relationship that describes an object, meaning that there may be three relationships. For example, a and/or B, represent: a or B, or, A and B. The term "corresponding" may refer to an association or binding relationship, and the correspondence between a and B refers to an association or binding relationship between a and B.
In one embodiment, as shown in fig. 1, there is provided a self-diagnosis and fault source identification method of a bridge structure monitoring sensor, including:
s110, periodically collecting transmission data of each sensor as initial data.
In particular, this step aims at forming a data stream by periodically reading and recording the output data of the sensor.
In one embodiment, step S110 includes: setting data acquisition frequency according to the initial condition; based on the data acquisition frequency, the transmission data of the sensor is periodically acquired as initial data.
Specifically, the initial conditions are factors such as the type of the bridge, the geographic position, the climate conditions, the traffic conditions, the sensor type and the like. According to factors such as the type of the bridge, the geographic position, the climate condition, the traffic condition, the sensor type and the like, and relevant industry specifications, a proper data acquisition frequency is set. For example, for a large suspension bridge, a relatively high data acquisition frequency, such as 50hz, may be required to acquire acceleration due to its complex structure. For a municipal overpass, only a low data acquisition frequency, such as 1 hour, is required. And at the set frequency, the field industrial personal computer sends a data request to each sensor, and the sensors respond to the request and return the current measurement data. Such data may include various information of bridge displacement, vibration, temperature, humidity, wind speed, traffic flow, etc.
In one embodiment, after step S110, the method further includes: preprocessing initial data; and analyzing the preprocessed initial data according to a preset strategy, and determining an abnormal mode.
Specifically, pre-processing operations such as denoising, standardization, filling in missing values, data smoothing and the like are performed on the monitored data before the data is analyzed, and the operations can help to improve the quality of the data and improve the accuracy of analysis. The preset strategy is as follows: whether the data exceeds the limit is determined by setting threshold values, and when the data exceeds or falls below the threshold values, the data is marked as abnormal. The threshold value is obtained by taking 0.7 times of the limit value through finite element software simulation, and relevant abnormal data are input into the field industrial personal computer for real-time analysis, so that a possible abnormal mode is searched.
Possible abnormal failure modes include: anomalies in the sensor (power down, failure, drift, offset, etc.); the bridge structure is abnormal.
S120, constructing a long-term and short-term memory model, and carrying out anomaly analysis on the initial data to obtain anomaly data carrying sensor marks.
Specifically, in this step, the data stream is analyzed and evaluated using a Long Short Term Memory (LSTM) model. The returned abnormal data carries a sensor mark, so that the sensor of the abnormal data source can be clearly determined.
In one embodiment, the model construction and anomaly analysis in step S120 includes: let the input data be x t The hidden state is h t Forgetting door f t The input gate is i t The output gate is o t The unit state is C t The following steps are:
f t =σ(W f ·[h t-1 ,x t ]+b f )
i t =σ(W i ·[h t-1 ,x t ]+b i )
o t =σ(W o ·[h t - 1 ,x t ]+b o )
h t =o t *tanh(C t )
where W and b are model parameters, σ is Sigmoid activation function, which represents element multiplication, [ h ] t-1 ,x t ]Indicating that the last hidden state and current input was connected.
In particular, sensor data x is acquired in real time t On-line discrimination is performed to confirm whether the state is abnormal. When the output data deviates significantly from the expected model, an abnormal state will be returned.
In one embodiment, after step S120, the method further includes: if the abnormal data carrying the sensor marks is not output, recording initial data, and repeatedly and periodically acquiring the transmission data of each sensor as the initial data.
Specifically, when the analysis result of the established model on the initial data is normal, the batch of initial data is recorded, then the step S110 is iterated and repeated to continue to repeat the method, so that the continuous monitoring of the abnormal state is carried out regularly, the continuity of data acquisition is ensured, and the maintenance efficiency is improved. In the event that the previous steps are completed without returning to an abnormal state, the method will continue with conventional data acquisition and analysis to maintain continuous monitoring of the bridge structure. At the same time, all sensor readings, self-test results, and possible anomalies are recorded in the maintenance log. When a problem occurs in the future, the possible fault modes can be found according to the history record, so that the problem can be better positioned, and the problem can be used as reference data for optimizing the performance of a monitoring system, adjusting the arrangement of the sensors and improving a fault detection algorithm.
S130, analyzing the abnormal data by adopting a fuzzy logic algorithm to determine a fault mode of the corresponding sensor.
Specifically, once an abnormal state is returned in the data analysis, the abnormal data is analyzed by adopting a fuzzy logic algorithm, and a possible fault mode is judged. For each possible failure mode, there is an associated confidence level that indicates the likelihood of that failure mode. In fuzzy logic, confidence is typically represented by a membership function that maps a given input to a confidence value between 0, 1. This confidence value represents the degree or likelihood that the input belongs to a certain fuzzy set.
The principle and method of setting the confidence level are as follows:
membership function: first, a membership function needs to be defined that describes the relationship between the input value and the fuzzy set. The membership function uses a gaussian function:
where x is the input value, c is the center position of the gaussian, σ is the standard deviation of the gaussian, and the width and steepness of the curve are controlled.
Blurring: the input values are obfuscated into a set of confidence values by mapping the input values into a range of values for the membership function. Each confidence value represents the likelihood that the input belongs to a certain fuzzy set. For example, for an input value x, its blurring result can be expressed as μ A (x)。
Reasoning and rules: and matching and reasoning the input value after blurring with a predefined blurring rule by utilizing a reasoning mechanism of the blurring logic. Each rule may contain a plurality of conditions and conclusions, and each condition and conclusion has a corresponding confidence level. In the reasoning process, fuzzy rules are used for matching and reasoning the input values after blurring. Rules can be expressed as: if conditions 1 and 2, conclusion 1.
Polymerization: in the reasoning process, the confidence of each conclusion is calculated by an aggregation method (such as maximum value, weighted average and the like) according to the confidence of the rule and the confidence of the condition.
Defuzzification: according to the confidence coefficient obtained by reasoning and aggregation, defuzzification operation can be carried out, and the fuzzy confidence coefficient is converted into a specific label.
By verifying each possible fault mode, the accuracy of alarm is ensured, false alarm is avoided, and the reliability of the system is improved. If the label after the confidence degree of a certain fault mode is defuzzified is the corresponding fault mode label, returning to the fault problem existing in the sensor. And further fault localization and analysis is performed at step S140.
In one embodiment, step S130 includes: setting an abnormal state vector a= [ a ] 1 ,a 2 ,...a n ] T Wherein a is i Is the confidence of the ith abnormal state, and the value range is [0,1]]Wherein 0 indicates that the anomaly is not believed at all, and 1 indicates that the anomaly is believed to be fully; setting a fault mode vector b= [ B ] 1 ,b 2 ,...b m ] T Wherein b j Is the confidence of the j-th fault mode, and the value range is also [0,1]The method comprises the steps of carrying out a first treatment on the surface of the Setting a fuzzy relationThe matrix R, R is an n m matrix, where R ij Indicating the confidence of the jth fault mode in the ith abnormal state; in performing fuzzy reasoning, the following equation is used to calculate the failure mode vector b= [ B ] 1 ,b 2 ,...b m ] T
B=A T ·R
Confidence b of a certain failure mode j And determining that the label after defuzzification is a corresponding fault mode, and determining that the label is a fault mode of a corresponding sensor.
Specifically, the fuzzy logic algorithm is adopted to analyze the abnormal data, and a specific mode can be judged by setting a fuzzy matrix and utilizing fuzzy reasoning. Fuzzy logic is a powerful tool for dealing with uncertainty and can make inferences and decisions in the presence of uncertainty or fuzzy information.
S140 determines the source of the anomaly from the failure mode based on the corresponding sensor.
Specifically, after the corresponding sensor is first determined, the same bridge structure is verified according to the peripheral sensors of the corresponding sensor or other sensors belonging to the same type. If all sensors report the same anomaly, the problem may appear on the bridge structure, requiring analysis and resolution of the structural problem. If only one or a few sensors report anomalies, then a problem may arise at the sensors themselves, requiring sensor self-tests.
In one embodiment, step S140 includes: weighted average of sensors triggering abnormal states, the weight being determined by the weight factor of each sensor, the weighted average abnormal statesThe calculation can be made by the following formula:
wherein q represents the number of sensors, W l Representing the weight factor of each sensor, C l Is the reported abnormal state.
Specifically, the abnormal state reported by all sensors is weighted and the weight is determined by the weight factor of each sensor. Let us assume that we have q sensors, the weight factor of each sensor being denoted W l The reported abnormal state is denoted as C l
In one embodiment, step S140 further includes: preset upper and lower threshold value theta' 1 And θ' 2 If (if) Or (b)The anomaly originates from the bridge structure; if->The anomaly originates from the sensor.
Specifically, by setting a threshold upper and lower threshold value θ' 1 And θ' 2 . If it isExceeding the threshold value theta' 1 And less than theta' 2 The problem may occur in bridge structures where analysis and resolution of structural problems is required. Otherwise, the problem may arise on a single or a small number of sensors, requiring sensor self-test.
Wherein the preset upper and lower threshold value theta 'can be determined by analyzing the known normal operation data' 1 And θ' 2 Wherein θ' 1 At the upper limit, θ' 2 Is the lower limit. These normal operating data may cover the reporting of a plurality of sensors and should be representative of the normal operating conditions of the bridge structure or system. By statistically analyzing these data, it can be determined that the sensor is normally sensedThe weighted average of the abnormal state reported by the device can determine the upper and lower threshold value theta 'by setting the preset threshold value as the average value plus or minus 3 times of standard deviation' 1 And θ' 2 An abnormal state exceeding this limit may be considered a bridge structure abnormality.
In one embodiment, after step S140, the method further includes: when the abnormality is from the sensor, performing self-detection on the corresponding sensor; when the abnormality comes from the bridge structure, a preset alarm system is triggered to perform abnormality early warning.
Specifically, when the problem that self-repairing cannot be achieved is encountered, the method can also timely inform maintenance personnel of intervention by triggering an alarm system so as to ensure the safety of the bridge.
In one embodiment, as shown in fig. 2, a self-diagnosis and fault source discrimination system of a bridge structure monitoring sensor is provided, comprising a data acquisition module 210, a data analysis module 220, and a fault determination module 230, wherein:
the data acquisition module 210 is configured to periodically acquire transmission data of each sensor as initial data;
the data analysis module 220 is configured to construct a long-term and short-term memory model to perform anomaly analysis on the initial data, so as to obtain anomaly data carrying sensor marks;
the data analysis module 220 is further configured to analyze the abnormal data by using a fuzzy logic algorithm, and determine a fault mode of the corresponding sensor;
the fault determination module 230 is configured to determine a source of the anomaly from the fault mode based on the corresponding sensor.
It will be appreciated by those skilled in the art that the modules or steps of the application described above may be implemented in a general purpose computing device, they may be centralized on a single computing device, or distributed across a network of computing devices, or they may alternatively be implemented in program code executable by computing devices, such that they may be stored on a computer storage medium (ROM/RAM, magnetic or optical disk) for execution by computing devices, and in some cases, the steps shown or described may be performed in a different order than what is shown or described herein, or they may be individually manufactured as individual integrated circuit modules, or a plurality of modules or steps in them may be manufactured as a single integrated circuit module. Therefore, the present application is not limited to any specific combination of hardware and software.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present application, and are not limiting; although the application has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the application, and are intended to be included within the scope of the appended claims and description.

Claims (10)

1. The self-diagnosis and fault source identification method of the bridge structure monitoring sensor is characterized by comprising the following steps of:
periodically collecting transmission data of each sensor as initial data;
constructing a long-term and short-term memory model to perform exception analysis on the initial data to obtain exception data carrying a sensor mark;
analyzing the abnormal data by adopting a fuzzy logic algorithm, and determining a fault mode of a corresponding sensor;
based on the corresponding sensor, an anomaly source is determined from the failure mode.
2. The method of claim 1, wherein the step of periodically collecting the transmission data of each sensor as initial data comprises:
setting data acquisition frequency according to the initial condition;
and based on the data acquisition frequency, periodically acquiring the transmission data of the sensor as initial data.
3. The method of claim 2, wherein after the step of periodically collecting the transmission data of each sensor as initial data, further comprising:
preprocessing the initial data;
and analyzing the preprocessed initial data according to a preset strategy, and determining an abnormal mode.
4. The method of claim 1, wherein after the step of constructing a long-short term memory model for anomaly analysis of the initial data, further comprising:
if the abnormal data carrying the sensor marks is not output, the initial data is recorded, and the transmission data of each sensor is repeatedly and periodically acquired as the initial data.
5. The method of claim 1, wherein said constructing a long-short term memory model performs an anomaly analysis step on said initial data, comprising:
let the input data be x t The hidden state is h t Forgetting door f t The input gate is i t The output gate is o t The unit state is C t The following steps are:
f t =σ(W f ·[h t-1 ,x t ]+b f )
i t =σ(W i ·[h t-1 ,x t ]+b i )
o t =σ(W o ·[h t-1 ,x t ]+b o )
h t =o t *tanh(C t )
where W and b are model parameters, σ is a Sigmoid activation function, * representing element multiplication, [ h ] t-1 ,x t ]Indicating that the last hidden state and current input was connected.
6. The method of claim 1, wherein the step of analyzing the anomaly data using a fuzzy logic algorithm to determine a failure mode of the corresponding sensor comprises:
setting an abnormal state vector a= [ a ] 1 ,a 2 ,…a n ] T Wherein a is i Is the confidence of the ith abnormal state, and the value range is [0,1]]Wherein 0 indicates that the anomaly is not believed at all, and 1 indicates that the anomaly is believed to be fully;
setting a fault mode vector b= [ B ] 1 ,b 2 ,…b m ] T Wherein b j Is the confidence of the j-th fault mode, and the value range is also [0,1];
Setting a fuzzy relation matrix R, wherein R is an n multiplied by m matrix ij Indicating the confidence of the jth fault mode in the ith abnormal state;
in performing fuzzy reasoning, the following equation is used to calculate the failure mode vector b= [ B ] 1 ,b 2 ,…b m ] T
B=A T ·R
Confidence b of a certain failure mode j And determining that the label after defuzzification is a corresponding fault mode, and determining that the label is a fault mode of a corresponding sensor.
7. The method of claim 6, wherein determining a source of anomalies from the failure mode based on the corresponding sensor comprises:
weighted average of sensors triggering abnormal states, the weight being determined by the weight factor of each sensor, the weighted average abnormal statesThe calculation can be made by the following formula:
wherein q represents the number of sensors, W l Representing the weight factor of each sensor, C l Is the reported abnormal state.
8. The method of claim 7, wherein determining a source of anomaly from the failure mode based on the corresponding sensor further comprises:
presetting an upper and lower limit threshold value theta 1 And theta 2 If (if)Or->The anomaly originates from the bridge structure;
if it isThe anomaly originates from the sensor.
9. The method of claim 8, wherein after the step of determining the source of the anomaly from the failure mode based on the corresponding sensor, further comprising:
when the abnormality is derived from a sensor, performing self-checking on the corresponding sensor;
when the abnormality comes from the bridge structure, a preset alarm system is triggered to perform abnormality early warning.
10. The self-diagnosis and fault source identification system of the bridge structure monitoring sensor is characterized by comprising a data acquisition module, a data analysis module and a fault determination module, wherein:
the data acquisition module is used for periodically acquiring the transmission data of each sensor as initial data;
the data analysis module is used for constructing a long-term and short-term memory model to perform exception analysis on the initial data to obtain exception data carrying a sensor mark;
the data analysis module is also used for analyzing the abnormal data by adopting a fuzzy logic algorithm and determining a fault mode of the corresponding sensor;
the fault determination module is used for determining an abnormal source according to the fault mode based on the corresponding sensor.
CN202310878884.XA 2023-07-17 2023-07-17 Self-diagnosis and fault source identification method and system of bridge structure monitoring sensor Pending CN116907772A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310878884.XA CN116907772A (en) 2023-07-17 2023-07-17 Self-diagnosis and fault source identification method and system of bridge structure monitoring sensor

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310878884.XA CN116907772A (en) 2023-07-17 2023-07-17 Self-diagnosis and fault source identification method and system of bridge structure monitoring sensor

Publications (1)

Publication Number Publication Date
CN116907772A true CN116907772A (en) 2023-10-20

Family

ID=88357708

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310878884.XA Pending CN116907772A (en) 2023-07-17 2023-07-17 Self-diagnosis and fault source identification method and system of bridge structure monitoring sensor

Country Status (1)

Country Link
CN (1) CN116907772A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117723782A (en) * 2024-02-07 2024-03-19 山东大学 Sensor fault identification positioning method and system for bridge structure health monitoring

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117723782A (en) * 2024-02-07 2024-03-19 山东大学 Sensor fault identification positioning method and system for bridge structure health monitoring
CN117723782B (en) * 2024-02-07 2024-05-03 山东大学 Sensor fault identification positioning method and system for bridge structure health monitoring

Similar Documents

Publication Publication Date Title
JP7017861B2 (en) Anomaly detection system and anomaly detection method
US11314242B2 (en) Methods and systems for fault detection and identification
KR102618023B1 (en) Failure prediction diagnosis system and method through pattern analysis according to failure type
Worden Structural fault detection using a novelty measure
US7496798B2 (en) Data-centric monitoring method
JP4528335B2 (en) Sensor performance verification apparatus and method
JP5016519B2 (en) Diagnostic system and method for predictive condition monitoring
KR101713985B1 (en) Method and apparatus for prediction maintenance
CN110909822B (en) Satellite anomaly detection method based on improved Gaussian process regression model
CN109186813A (en) A kind of temperature sensor self-checking unit and method
CN110334740A (en) The electrical equipment fault of artificial intelligence reasoning fusion detects localization method
CN118154174B (en) Intelligent operation and maintenance cloud platform for industrial equipment
CN112284440B (en) Sensor data deviation self-adaptive correction method
CN109522948A (en) A kind of fault detection method based on orthogonal locality preserving projections
US8560279B2 (en) Method of determining the influence of a variable in a phenomenon
CN108427400B (en) Aircraft airspeed head fault diagnosis method based on neural network analytic redundancy
KR20170127430A (en) Method and system for detecting, classifying and / or mitigating sensor error
CN108650139A (en) A kind of powerline network monitoring system
CN113574480B (en) Device for predicting equipment damage
CN108257365B (en) Industrial alarm design method based on global uncertainty evidence dynamic fusion
CN114323671A (en) Method and device for determining the remaining service life by means of an artificial intelligence method on the basis of a predictive diagnosis of a component of an electric drive system
CN117454221A (en) Abnormality identification method for multi-sensor fusion of high-speed rail traction braking system
CN116907772A (en) Self-diagnosis and fault source identification method and system of bridge structure monitoring sensor
CN114003422A (en) Host anomaly detection method, computer device, and storage medium
US8271421B1 (en) Nonparametric fuzzy inference system and method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination