CN117270482A - Automobile factory control system based on digital twin - Google Patents
Automobile factory control system based on digital twin Download PDFInfo
- Publication number
- CN117270482A CN117270482A CN202311559210.XA CN202311559210A CN117270482A CN 117270482 A CN117270482 A CN 117270482A CN 202311559210 A CN202311559210 A CN 202311559210A CN 117270482 A CN117270482 A CN 117270482A
- Authority
- CN
- China
- Prior art keywords
- equipment
- data
- plant
- digital twin
- maintenance
- 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
Links
- 238000004519 manufacturing process Methods 0.000 claims abstract description 132
- 238000013507 mapping Methods 0.000 claims abstract description 32
- 238000012544 monitoring process Methods 0.000 claims abstract description 21
- 230000036541 health Effects 0.000 claims abstract description 13
- 230000002159 abnormal effect Effects 0.000 claims abstract description 8
- 238000004891 communication Methods 0.000 claims abstract description 5
- 238000012423 maintenance Methods 0.000 claims description 82
- 238000000034 method Methods 0.000 claims description 75
- 230000008569 process Effects 0.000 claims description 39
- 230000004927 fusion Effects 0.000 claims description 14
- 230000003993 interaction Effects 0.000 claims description 12
- 238000012549 training Methods 0.000 claims description 6
- 238000013527 convolutional neural network Methods 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 5
- 230000001360 synchronised effect Effects 0.000 claims description 5
- 238000013386 optimize process Methods 0.000 claims description 4
- 230000005856 abnormality Effects 0.000 claims description 3
- 238000010801 machine learning Methods 0.000 claims description 3
- 238000012394 real-time manufacturing Methods 0.000 claims description 2
- 210000002569 neuron Anatomy 0.000 description 12
- 238000005516 engineering process Methods 0.000 description 10
- 238000011176 pooling Methods 0.000 description 10
- 239000000047 product Substances 0.000 description 10
- 239000000463 material Substances 0.000 description 9
- 238000007726 management method Methods 0.000 description 8
- 239000000243 solution Substances 0.000 description 8
- 230000006870 function Effects 0.000 description 6
- 230000000007 visual effect Effects 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 238000013145 classification model Methods 0.000 description 3
- 238000010276 construction Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 230000010354 integration Effects 0.000 description 3
- 238000004088 simulation Methods 0.000 description 3
- 239000013598 vector Substances 0.000 description 3
- 230000004913 activation Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000006835 compression Effects 0.000 description 2
- 238000007906 compression Methods 0.000 description 2
- 238000013480 data collection Methods 0.000 description 2
- 238000013523 data management Methods 0.000 description 2
- 230000003862 health status Effects 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000005299 abrasion Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 239000006227 byproduct Substances 0.000 description 1
- 210000004027 cell Anatomy 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 238000013079 data visualisation Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012905 input function Methods 0.000 description 1
- 238000003780 insertion Methods 0.000 description 1
- 230000037431 insertion Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 230000013016 learning Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000007639 printing Methods 0.000 description 1
- 210000001747 pupil Anatomy 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000009877 rendering Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 239000012086 standard solution Substances 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 238000012384 transportation and delivery Methods 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 230000031836 visual learning Effects 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
- G05B19/41865—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM] characterised by job scheduling, process planning, material flow
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/30—Nc systems
- G05B2219/32—Operator till task planning
- G05B2219/32252—Scheduling production, machining, job shop
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Manufacturing & Machinery (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
A digital twinning-based automotive plant control system comprising: a digital twin platform configured for communication connection with the plant to enable real-time data exchange between the digital twin platform and the plant, enabling real-time monitoring of the state of the plant and execution of control of the plant by the digital twin platform. The digital twin platform comprises: a digital twin model, the digital twin model constituting a virtual space of the mapping plant; the digital twin model comprises a mapping of physical entities in a factory and a mapping of manufacturing process parameters in the factory, wherein the mapping of the physical entities in the factory comprises a three-dimensional model of manufacturing equipment; and a warning module including a fault prediction model therein configured to determine a health condition of the equipment based on the status of the equipment in the plant and the manufacturing process parameters, and issue a warning when the health condition of the equipment is abnormal.
Description
Technical Field
The application relates to an automobile factory control system constructed based on a digital twin technology.
Background
With the rapid development of artificial intelligence, industrial internet and other technologies, manufacturing industry is facing increasing challenges. Information exchange technology has covered various levels of field devices, personnel and management, formed a powerful information system based on data and supported by network integration, greatly improving the production automation level and work efficiency.
Digital twinning is a currently newly developed technology. By means of digital twinning techniques, a digital representation of the physical system, i.e. digital twinning, may be provided, and various elements in digital twinning may establish a connection with corresponding elements in the physical system through a network, e.g. internet of things techniques, e.g. as described in US2017/286572 A1. Through this connection, digital twinning can receive data related to physical system state, such as sensor readings, etc. Based on this data, digital twinning may analyze or interpret the state history of the physical system through simulation, as well as predict the actual or future state of the physical system. In some manufacturing industries, digital twinning techniques have been employed, for example, to predict the life of components in a system in order to replace the components in time.
Compared with other manufacturing industries, automobile manufacturing, particularly the industries of automobile part production, assembly and the like, has various unique characteristics, such as complex processes, mass production batches, various automation devices, on-time delivery, zero quality defects and the like. Therefore, it is also desirable for automobile manufacturers, particularly automobile parts manufacturers, to increase operating efficiency by employing digital means. Although digital twinning technology has been applied in some other areas, it has not been substantially developed in the field of highly automated automotive manufacturing.
Disclosure of Invention
The digital twin technology is introduced into the automobile manufacturing industry, particularly the automobile part production, assembly and other industries.
According to one aspect of the present application, there is provided a digital twinning-based automotive plant control system comprising:
a digital twin platform configured to be in communication connection with the plant to enable real-time data exchange between the digital twin platform and the plant, such that the state of the plant can be monitored and control of the plant performed in real time by the digital twin platform;
wherein, the digital twin platform comprises:
a digital twin model, the digital twin model constituting a virtual space of the mapping plant; the digital twin model comprises a mapping of physical entities in a factory and a mapping of manufacturing process parameters in the factory, wherein the mapping of the physical entities in the factory comprises a three-dimensional model of manufacturing equipment; and
a warning module including a fault prediction model therein configured to determine a health condition of the equipment based on a state of the equipment in the plant and manufacturing process parameters and to issue a warning when the health condition of the equipment is abnormal.
In one embodiment, the fault prediction model is constructed by machine learning, wherein raw data from manufacturing equipment is classified, and for each class of raw data, some or all of the following values are calculated as equipment status monitor indicators, respectively: k-means, maximum, minimum, standard deviation, skewness, kurtosis, JB statistics, and p-value.
In one embodiment, each category of data is partitioned into two sets of data sets using expert experience: an acceptable range data set and an operational failure data set.
In one embodiment, a convolutional neural network is utilized to process the equipment state monitoring indexes to obtain probability values of abnormality of each monitoring index of the equipment, so that a fault prediction model is established; and training and validating the fault prediction model using data in the two sets of data sets.
In one embodiment, the digital twin platform further comprises a monitoring module, a video module and a decision module; the monitoring module is configured to detect equipment and manufacturing processes in a factory in real time to achieve synchronous mapping of manufacturing process states in the digital twin platform; the video module is configured to display manufacturing equipment and its operating state, and manufacturing process parameters; the decision module is configured to process the real-time manufacturing process parameters and update the manufacturing process parameters in the digital twin model and can generate an optimized process scheme.
In one embodiment, the three-dimensional model of the manufacturing apparatus is constructed using a holographic apparatus by means of a QR code positioning method.
In one embodiment, the mapping of the plant manufacturing process parameters in the digital twin model is achieved by fusion of manufacturing process parameters, including data fusion and physical fusion.
In one embodiment, the digital twinning platform further includes a maintenance module configured to provide a maintenance intelligent guide of the manufacturing apparatus, the maintenance intelligent guide including: and guiding staff to complete maintenance operation based on the three-dimensional model of the manufacturing equipment, and updating current maintenance operation data into an equipment history maintenance database.
In one embodiment, the dimension protection module is further configured to provide dimension protection intelligent interactions of the manufacturing device, the dimension protection intelligent interactions comprising: based on the equipment history maintenance operation data, the health state of the equipment is visually displayed, a solution is provided for the equipment maintenance, and the maintenance operation data is updated into an equipment history maintenance database.
In one embodiment, the automotive plant control system further includes a user interface communicatively coupled to the digital twinning platform and configured to enable an operator to monitor and perform control of the plant in real-time through the digital twinning platform and to be directed by the digital twinning platform to perform operations within the plant.
According to the automobile factory control system, the production management system, the enterprise resource system and the Internet of things equipment are connected through the digital twin platform, so that real-time display of production resources is realized in the background, and the production state is transparent. Other technical effects of the aspects of the present application will be described in detail in the following description of the specific embodiments.
Drawings
The foregoing and other aspects of the present application will be more fully understood and appreciated from the following detailed description taken with reference to the accompanying drawings, in which:
FIG. 1 is a schematic illustration of a layout of an automotive plant control system constructed in accordance with the digital twinning technique of the present application;
FIG. 2 is a schematic diagram of one exemplary method of constructing a fault prediction module in an automotive plant control system of the present application.
Detailed Description
The present application relates generally to the introduction of digital twinning technology in the automotive industry, particularly in the automotive parts production, assembly, and the like.
It is first noted that in conventional automotive manufacturing, and particularly in automotive parts factories, production management systems (MES) focus on manufacturing process control, while enterprise resource systems (ERP/SAP) focus on financial, logistical, etc. Optimization of production performance indicators is difficult to achieve in conventional ways.
According to the method, in the digital twin technology automobile manufacturing industry, particularly in an automobile part factory, a production management system, an enterprise resource system and an internet of things (IoT) device are connected through a digital twin platform, so that production resources (personnel, devices, materials, methods, environments and the like) can be displayed in real time, and a transparent production state is realized.
To this end, according to one possible embodiment of the present application, a control system for an automotive plant (such as an automotive component production, assembly plant, whole vehicle assembly plant, etc.) is constructed based on digital twinning technology, as schematically illustrated in fig. 1.
The automobile factory control system comprises a digital twin platform DT, and can interact with an actual factory (or workshop) PS to realize real-time data exchange between the digital twin platform DT and the actual factory (or workshop) PS. The digital twin platform DT may be a cloud platform, or a background computer system, which may be communicatively connected to the factory PS by wired or wireless means. The schematically represented background user interfaces U1, U2, U3. are communicatively connected to the digital twinning platform DT by wired or wireless means. Further, in the factory PS, a mixed reality device MR or a smart mobile device (such as a cell phone, tablet computer, etc.) worn by an operator, etc. may interact with the digital twin platform DT. Therefore, these mixed reality devices MR or smart mobile devices also constitute the user interface. The operator can monitor the state in the plant PS and perform control of the plant PS in real time through the digital twin platform DT by means of a user interface, and perform operations within the plant PS as directed by the digital twin platform DT.
The factory PS includes therein mechanical equipment (such as a processing equipment or a production line) EQP, control devices EC (such as a PLC or an edge computer) associated with the mechanical equipment EQP, sensing devices SEN (such as cameras disposed in the factory, various sensors carried by or associated with the mechanical equipment EQP, etc.), ioT hardware devices, and the like.
The digital twin platform DT contains a digital twin model MDT, which constitutes a digital space, i.e. a virtual space, mapping an actual plant (shop) PS, i.e. a physical space, constituting digital twin and physical twin. The digital twin platform DT further comprises various functional modules constructed for monitoring the plant PS, such as a monitoring module M1, a video module M2, a decision module M3, a warning module M4, a maintenance module M5, etc., which are capable of interacting with the digital twin model MDT, as well. These modules may be provided independently of the digital twin model MDT or may be integrated in the digital twin model MDT in the form of functional modules. The video module M2 may be just a software module that displays various information through a display connected via a user interface; alternatively, the video module M2 may be in the form of a software module plus a display, itself having a display function.
In the construction of the digital twin model MDT, mapping of physical entities and process parameters needs to be implemented. In order to accurately describe the manufacturing process control flow in the digital space, it is necessary to describe it from multiple dimensions and multiple proportions. The multi-dimension mainly comprises a visual mode, a calculation mode and a data mode, and realizes synchronous mapping, prediction analysis and data management of the state of the manufacturing process respectively. The multiple ratios are meant to include mainly the product manufacturing process, the production plant and other different levels of particle size. By establishing associations between various scale patterns, and integrating visual, computational, and data patterns, a digital twin model for automotive manufacturing can be formed. The mapping of the digital twin model in the production plant can be displayed by the video module M2.
As an example, in the construction of the digital twin model MDT, the mapping of the physical entities may be implemented by means of a mixed reality device MR, as described below.
In the factory PS, the operator activates the mixed reality device MR, and the camera reads the surrounding environment to generate a three-dimensional model of the equipment entity from a preset recognition object by understanding the surrounding scene.
In particular, a three-dimensional full process monitoring of the running state of the plant can be created in the digital twin platform DT, wherein a digital mirror image of the plant and its operating process, i.e. a three-dimensional model, is formed, mainly based on a built production site digital twin vision model, which virtually and in real time synchronously maps the logistics of the production site, the product processing state changes, the plant operation and the personnel position.
An exemplary method of creating a three-dimensional model of a device comprises the specific steps of: (1) By installing UWB positioning sensors on each logistics vehicle in the production place, realizing workshop material flow mapping, UWB continuously transmitting position information to a digital twin platform, and simulating the vehicle position in a virtual space based on the actual position by the digital twin platform; (2) The product processing state change map, i.e., the product model is dynamically presented in real time in different manufacturing process stages.
Another exemplary method of creating a three-dimensional model of a device includes image acquisition of the device using a holographic device. Wherein, it relates to locating the device using QR code locating method, thereby spatially synchronizing the device three-dimensional model with the real device. For this purpose, anchor points must be used in the creation of the three-dimensional model. The position of the device in the three-dimensional model in the real world environment can be determined using a holographic device. An exemplary method of creating an anchor point includes the steps of:
selecting a QR code positioning point by using a positioning point guide;
printing a locating point from a PDF file created by the locating point guide;
attaching the anchor point to the physical object (device) in the real environment;
the person gazes at the anchor point, capturing pupil angles by a camera to locate the physical object (device) in the three-dimensional model.
Other methods of creating a three-dimensional model of a production facility may also be used herein, provided that a three-dimensional model of the production facility can be created in the digital twin model MDT.
The mapping of process parameters in the digital twin model MDT may comprise the following three steps:
(1) And (3) collecting: various data collection methods may be employed to collect equipment used in the manufacture of automobiles (such as component production, assembly, etc.), as well as dynamic data generated, including methods through system interfaces, human-machine interactions, hardware collection terminals, bar codes, sensors, radio frequency identification, etc., and the collected data may include inspection data related to quality of production, consumable materials data, process (procedure) completion data, equipment data, etc.
(2) Fusion: real-time multi-source heterogeneous data (particularly multi-source sensor data) collected during the production process is preprocessed and analyzed by data cleaning, data unified modeling, data out-of-control alignment (alignment) and other methods, providing reliable data and information for subsequent state synchronization mapping, prediction, analysis and feedback control.
(3) Mapping: the digital twin model MDT is visualized and synchronously operated with a physical process by using methods of three-dimensional model reconstruction, two-dimensional data display, process simulation and the like, so that comprehensive real-time video monitoring of the automobile manufacturing (such as component production, assembly and the like) process is realized.
It should be noted that in automotive manufacturing, a large number of manufacturing process parameters are generated. These data describe various information in the manufacturing process, and manufacturing process control is focused primarily on the collection of manufacturing process parameters. The manufacturing process parameters mainly comprise manufacturing process parameters, production personnel data, generating equipment data, working hour data, manufacturing process parameters and production quality data. Based on this data, the manufacturing process can be simulated in a virtual platform and anomaly problems can be predicted and tracked quickly. Thus, real-time data collection is a prerequisite for manufacturing process control.
Automobile manufacturing processes are very numerous in number, types, and collected in a variety of different ways. Currently, most automotive manufacturing process parameters are collected by hardware sensor devices and uploaded to a database of a manufacturing execution system. Thus, data interaction and collection between systems typically needs to be accomplished through a data interface. The method and the system realize the collection of manufacturing process parameters through interfaces and hardware sensors, and store real-time data combined manufacturing process flows and production nodes in a structured mode through data clustering so as to comprehensively and accurately store the relation between the data. This provides data support for establishing subsequent production video modes and computation modes.
The fusion of manufacturing process parameters includes data fusion and physical fusion. Data fusion refers to the integration of data and is a process in which data is generated by a single data source. Real-time data (raw data) collected from the physical generation line is typically multi-sourced and heterogeneous, i.e., sparse, not suitable for direct use with and learning models thereof. Thus, the raw data needs to be processed according to the following three characteristics: 1) Identity: the manufacturing process parameters include three-dimensional model data and structured data stored in a relational database, among other types of data. These data often exhibit significant multi-source heterogeneity and therefore require definition of a common engineering data format; 2) Normalization: manufacturing data stored on different platforms is typically accessed (acquired) in different ways. In order to realize real-time interaction between the data, a corresponding interface and a standard communication protocol need to be established in each system, and meanwhile, the requirements of communication quality, physical security and information security are met; 3) Traceability: in the process of fusing data using various algorithms and tools, it is necessary to ensure that the logical relationship between the data is unchanged, i.e., at any network device, at any node, any other data related to the digital twin model can be accessed (acquired) through the logical relationship.
Physical fusion refers to the interconnection between manufacturing data and physical entities, and mainly includes two aspects: 1) The physical fabrication process runs in synchronization with its digital twin model. To ensure that virtual lines, devices can be mapped ideally, synchronously and completely with physical lines, devices, models and real objects must maintain dynamic consistency on physical, behavioral and rule-based data; 2) Fusion of manufacturing process parameters and manufacturing process. Starting with a workflow-based data management concept, each manufacturing process node is associated one-to-one with a generated manufacturing process parameter. For dynamic and real-time multi-source and heterogeneous data collected during complex product manufacturing, multi-layer data filters are constructed by defining different detection and processing rules, and multi-stage filter combinations that minimize the loss of raw data information are selected to filter the data. Next, based on the ontology construction, a multidimensional context and corresponding measured values of the defined data are established and converted into a global data model in the data integration middleware, so that unified modeling of structured, semi-structured and unstructured production data is realized.
Next, each constituent element of the production site is three-dimensionally modeled using a simulated graphics rendering engine, for example Unity, and the scene is arranged. The mapping relation between the fusion data and the digital twin model is established, so that the operation of the production place is reflected from the three-dimensional whole process and the two-dimensional whole element state, and the visual synchronous operation of the physical production place is realized. For three-dimensional whole process monitoring, virtual-real synchronous mapping is respectively carried out on the material circulation, the process state change of the product, the equipment operation and the figure position of the production site based on the constructed production site digital twin body visualization model, so as to form a digital mirror image of the workshop operation process.
An exemplary method for implementing digitized mirroring includes the specific steps of: (1) The method comprises the steps of (1) mapping workshop material flow, namely, installing UWB positioning sensors on each logistics transportation vehicle, enabling UWB to continuously send position information to a digital twin platform, enabling the digital twin platform to simulate the vehicle position in a virtual space according to the actual position, and (2) mapping product process state change, namely, displaying a product model in real time and dynamically in different assembly process stages. In one aspect, the ongoing process (procedure) procedure and progress is determined based on the process flow, process (procedure) completion, and material usage data, and after the process (procedure) is completed, the three-dimensional model of the product is converted to a corresponding process stage model. On the other hand, based on the collected detected data or process parameters (such as insertion pressure, deformation, displacement, etc.), the process state is displayed in real time through the video module M2 by product model correction or reconstruction, state update and data visualization, and the product line operation state can be updated in real time. (3) The device running state mapping, i.e. the user interface accessing the digital twin platform DT in real time, such as a unified API interface, receives interface data and reverts to virtual device running in combination with the simulation model. User interface data is received with the test tool and stored in a virtual device with the digital twin model.
When the digital twin platform DT is put into use, the mapping of the devices in the plant PS in the digital twin model MDT is revealed to the user by the video module M2. The video module M2 can present the current operation state of all devices, warning information, output information, the locations of interior personnel and logistics vehicles, and other information in three dimensions in real time. By means of the monitoring module M1, real-time process parameters of the production line or equipment in the factory PS can be obtained. The decision module M3 processes the acquired real-time process parameters and updates them in the digital twin model MDT, and can generate an optimized process scheme. The digital twin platform DT may send the optimized process recipe to the factory PS for manual or automatic adjustment of the production process.
The warning module M4 in the digital twin model MDT includes a failure prediction model. Based on the equipment status and real-time process parameters acquired by the monitoring module M1, the health status of the equipment is determined using a fault prediction model, and a warning is given when an abnormality in the equipment or process is determined. For example, the warning module M4 may determine anomalies in personnel, equipment, materials, methods, environments, measurements, etc. in the plant PS based on the refined data provided by the monitoring module M1 using the digital twin model MDT and track the specific cause of the anomalies. Alternatively, the sensing device SEN in the factory PS may include an AI visual device tailored to be able to identify the device operational status, and the alert module M4 is able to determine the health of the device based on the identification information of the AI visual device.
A method of constructing the failure prediction model included in the warning module M4 by machine learning is described below with reference to fig. 2.
First, in step S1, the raw data is classified according to the material model using a k-means clustering algorithm. Because equipment production can be according to customer individuation demand for equipment model is different, consequently need show the difference to different model production operation cycle and classify the raw data that the sensor gathered according to the model in the equipment production process through classification model. Therefore, in step S1, the raw data (the actual operation value of the device) from the device in the factory PS is classified by using the classification model, the raw data is classified into category 1, category 2..category M, and the raw data sequence of the state detection is constructed based on the raw data collected by the sensor in the device production process, respectively, and the raw data trend lines under different model types are drawn.
Next, in step S2, a device status monitor index is constructed. For each type of data divided in step S1, the following eight indices are calculated as the device status monitor indices, respectively: k-means, maximum, minimum, standard deviation, skewness, kurtosis, JB statistics, and p-value. Data of each category is divided into two sets of data sets using expert experience: an acceptable range data set and an operational failure data set. The acceptable range means that the equipment can still keep running, and the running fault means that the equipment has a certain degree of fault and needs maintenance.
Next, in step S3, an acceptable operation database is constructed based on the data in the acceptable range data set, and a failure database is constructed based on the data in the operation failure data set. Further, the device state partitioning result data is re-coarsely clustered, i.e., the data in the acceptable operations database is partitioned into normal, mild anomaly, moderate anomaly, and severe anomaly subsets using a classification model, such as a CNN model; data in the operational failure database is partitioned into light failure, moderate failure and heavy failure subsets.
Then, for each category of data divided in step S1, a probability value of its abnormal pattern is determined as described below.
According to one specific implementation, the input data "V1", "V2", "V3", "V4", "V5", "V6", "V7", "V8" employed in the input layer of the convolutional neural network are k-means, maximum, minimum, standard deviation, skewness, kurtosis, JB statistics, and p-values, respectively. It should be noted that some of the above eight values may be selected and/or other values may be added as input data in the input layer of the convolutional neural network. The result output by the convolution layer is composed of a plurality of characteristic surfaces, each value of the characteristic surfaces represents a neuron, and each neuron is connected with the characteristic surface of the upper layerIs connected to the region of the first layer. The characteristic surfaces are in one-to-one correspondence with the convolution kernels, and the value of each neuron in the characteristic surfaces is obtained through calculation of the corresponding convolution kernels. Taking the first convolution layer as an example, the operation formula is as follows:
（1）
wherein,the output value of the jth neuron which is the kth feature plane in the first convolution layer; f is a ReLU activation function; />Is the weight of the ith row and the ith column in the kth convolution kernel; />Is an input function; />Is the offset value corresponding to the kth convolution kernel.
And each convolution layer is connected with a pooling layer, and the feature plane output by the convolution layer is further subjected to dimension reduction on the premise of not increasing training parameters, so that parameters of a network are reduced, and the robustness of the model is improved. Dividing the input feature surface into size in the pooling layerAnd (3) carrying out pooling operation on each area, wherein each compressed characteristic surface output by the pooling layer corresponds to one characteristic surface output by the convolution layer of the upper layer. The method selects common maximum value pooling for operation, namely extracting the maximum value of elements in the pooling domain to form a compression characteristic surface.
After repeated convolution and pooling, the full connection layer connects all the compressed feature planes output by the last pooling layer in series to form feature vectors and inputs the feature vectors into the full connection layer. Each neuron in the full-connection layer is fully connected with the neuron in the previous layer, and the features extracted by the convolution layer and the pooling layer can be integrated, so that the feature with more differentiation is obtained. Assuming that K compression characteristic surfaces are output by the upper pooling layer in total, the value of the s-th neuron of the full-connection layerCan be calculated by the following formula:
（2）
wherein,for the eigenvectors obtained in series, < > and>for the connection weight between the s-th neuron of the full connection layer and the i-th element of the input vector, is>And f is a ReLU activation function, and h is the number of neurons of the fully connected layer. The output layer is fully connected with the fully connected layer, so that the output value of the j-th output layer node is +.>Can be calculated by the following formula:
（3）
wherein,to output the connection weight between the jth neuron of the layer and the s-th neuron of the full connection layer,a bias layer that is the jth output layer neuron; t is the number of output categories, t=m. For subsequent pattern recognition and classification, the result of equation (3) needs to be converted into a probability form, and then the probability value of the input data belonging to the j-th class quality pattern can be calculated by a normalized exponential function softmax:
（4）
wherein,e is a natural constant, which is a probability value that the input data belongs to the j-th class abnormal mode. Based on->The respective data may be partitioned into respective subsets of the acceptable operational database and the failure database.
Next, in step S4, a failure prediction model is constructed, and a device next-stage operation state value (probability value including an abnormal pattern) is acquired. And collecting part of data (70% -80% for example) in each subset of the acceptable operation database and the fault database according to time sequence to form an original sequence X, constructing a prediction model, and collecting real-time data in fine granularity according to a prediction value. The specific steps may include:
(a) Decomposing the original sequence X into a smooth trend item sequence X_t and a multicycle item sequence X_p by using an EMD method;
(b) Fitting and predicting trend terms x_t using NGM (1, k) models;
(c) Using fourier series fitting and predicting a periodic term sequence x_p;
(d) Obtaining a comprehensive prediction equation from the results of step (b) and step (c) and making predictions;
(e) The LVSVM is used to correct the prediction error of the failure prediction model.
Next, in step S5, a model reliability analysis is performed. The remaining (e.g., 20% -30%) data in the acceptable operation database and the fault database is used for prediction and verification to determine the prediction accuracy and fitting capability of the fault prediction model. And adopting a posterior difference ratio and a small error probability to evaluate the prediction registration and the precision of the model. And (5) carrying out model fitting and prediction accuracy inspection on the prediction result by adopting absolute errors, root mean square errors and the like.
Thereby, a failure prediction model suitable for the warning module M4 is established.
It can be seen that based on the mapping, prediction and analysis results of the digital space, the warning module M4 is able to warn the abnormal (faulty) condition in the production process and inform the relevant personnel to adjust the execution parameters of the relevant process to avoid the abnormal condition or to repair the equipment fault.
With respect to the maintenance module M5, it may guide the maintenance operation of the field device, and provide a scientific maintenance basis for the operation state of the target device.
In this connection, it should be pointed out that maintenance work of an automobile apparatus (for example, a component apparatus) involves various small components and a large number of operation steps, and thus, the apparatus maintenance and management process is complicated. The requirements of the devices on the running environment, which are affected by the performance of the devices, are not completely the same, and the maintenance methods of the devices are different. Meanwhile, due to abrasion and service life consumption of equipment in the running production process, each equipment maintenance needs to give a new solution conforming to the current situation based on the past maintenance data.
In addition, for staff just touching the maintenance work of the automobile manufacturing equipment, the maintenance tasks of the equipment with different functions of a plurality of production lines are easy to generate the conditions of missing steps and not-in-place part installation, and the safety of the equipment and the benefit of enterprises can be reduced. The traditional maintenance training carries out visual learning through methods such as pictures, videos and the like, and can understand the maintenance flow and the specific operation method of the equipment quickly, but has weaker spatial third dimension and experience sense, higher comprehensive quality requirements on maintenance staff, and certain difficulty in completing practice by hands in person for staff with weak spatial imagination and manual ability. In addition, the daily maintenance data of the equipment needs to be recorded and recorded in detail by maintenance staff, and the work efficiency of the maintenance staff is greatly reduced by manual paper registration or manual input to the system. Therefore, by means of the mixed reality technology, workers can be effectively helped to feel and experience the manufacturing process in an immersive manner, and accordingly maintenance quality and working efficiency are improved. However, the conventional VR device can only bring the vision and consciousness of the person into a completely virtual world, and cannot provide the surrounding real environment information.
The maintenance module M5 provided in the digital twin platform DT can provide two core functions of maintenance intelligent guidance and intelligent interaction of the automobile manufacturing equipment.
The maintenance intelligent guide of the automobile manufacturing equipment comprises the following steps: based on the three-dimensional model of the virtual current equipment in the digital twin platform DT, the entity and the model are combined, so that staff can understand the working principle of the equipment deeply, guide the staff to finish maintenance operation, and update the current maintenance operation data into the equipment history maintenance database.
The maintenance intelligent interaction of the automobile manufacturing equipment comprises the following steps: based on the equipment history maintenance operation data, the health state of the equipment is visually displayed, a solution is provided for the equipment maintenance, and the maintenance operation data is updated into an equipment history maintenance database. In the state detection process, the equipment history maintenance data is displayed on a three-dimensional model of the equipment history maintenance data in a concise and visual manner through a virtual interaction means, so that an operator can control the running state of complex equipment on the whole. The multi-element data of the equipment maintenance is added into the equipment maintenance database by means of a holographic image (such as hollens) device through voice and gesture operation.
As described above, the warning module M4 including the failure prediction model is built in the digital twin platform DT, whereby the warning module M4 can give the health status of the current device. If the health state is normal, the maintenance module M5 can carry out on-site maintenance training and guidance on staff through voice interaction of a user interface, and update current maintenance operation data into the equipment history maintenance database, wherein the maintenance data comprises task information, starting time, ending time, operator number, maintenance period, maintenance result and picture uploading. If the health state is not normal, the maintenance module M5 invokes the equipment maintenance history database through voice interaction of the user interface, the fault type is matched, a guiding flow is added, training guidance of on-site maintenance is carried out, and the current maintenance data is updated to the equipment history maintenance database. The maintenance data comprises task information, starting time, ending time, operator number, maintenance period, maintenance result and picture uploading.
By means of the maintenance module M5, the threshold operated by the maintenance personnel can be lowered. Meanwhile, the diagnosis result of the health state of the equipment is displayed visually through the history maintenance data, a solution is provided for the maintenance and the service of the equipment, and scientific and fine management of the equipment maintenance and the service is assisted.
According to the method and the device, the production management system, the enterprise resource system and the IoT device are connected through the digital twin platform, so that production resources (personnel, equipment, materials, methods and environments) are displayed in real time, transparent production states are displayed, and when deviation occurs, the problem solving requirements are triggered in real time and pushed to related personnel. The digital twin platform can provide the information needed to solve the problem (three dimensional device operation in real time, process parameters, initial statistical analysis of the problem, including cause and validity tracking of previous problem solutions) to help quickly solve the problem.
The method can establish a standardized fault code database, establish a fault knowledge graph, utilize various modes to carry out statistics tracking, automatically transfer an effective solution to a problem solution standard solution knowledge base and interconnect with a full transparent management system.
The intelligent mobile device is connected to the digital twin platform, various operation indexes are collected in real time according to business requirements and displayed in a graphical mode, such as productivity, equipment efficiency, inventory level and the like, and maintenance personnel can receive all types of warning and exception information in real time. The present invention is based on a data distribution intersection system and platform at the current level in production and operation, and utilizes a digital twinning platform to connect various systems and IoT devices to achieve intelligent decisions in production and operation.
In the basic scheme (state monitoring) of the digital twin platform DT, the operation state of production equipment is monitored, and production cycle data of each equipment is decomposed. The actual operating state of each device (such as time, temperature, voltage, current, and other parameters) is monitored. By means of state monitoring, a large number of data samples are obtained to optimize the model and to select the most suitable model. This enables proactive equipment failure prediction. In the digital twin platform DT extension scheme (fault warning and maintenance guidance), equipment faults are avoided in time, field equipment operation and maintenance operation are guided, and scientific maintenance basis is provided for the operation state of target equipment. The invention provides a new intelligent solution to equipment fault monitoring and warning problems in factory production.
Although the present application is described herein with reference to specific embodiments, the scope of the application is not intended to be limited to the details shown. Various modifications may be made to these details without departing from the underlying principles of the present application.
Claims (10)
1. A digital twinning-based automotive plant control system, comprising:
a digital twin platform configured to be in communication connection with the plant to enable real-time data exchange between the digital twin platform and the plant, such that the state of the plant can be monitored and control of the plant performed in real time by the digital twin platform;
the digital twin platform is characterized by comprising the following components:
a digital twin model, the digital twin model constituting a virtual space of the mapping plant; the digital twin model comprises a mapping of physical entities in a factory and a mapping of manufacturing process parameters in the factory, wherein the mapping of the physical entities in the factory comprises a three-dimensional model of manufacturing equipment; and
a warning module including a fault prediction model therein configured to determine a health condition of the equipment based on a state of the equipment in the plant and manufacturing process parameters and to issue a warning when the health condition of the equipment is abnormal.
2. The automotive plant control system of claim 1, wherein the fault prediction model is constructed by machine learning, wherein raw data from manufacturing equipment is classified, and for each class of raw data, some or all of the following values are calculated as equipment condition monitoring indicators, respectively: k-means, maximum, minimum, standard deviation, skewness, kurtosis, JB statistics, and p-value.
3. The automotive factory control system of claim 2, wherein each category of data is partitioned into two sets of data using expert experience: an acceptable range data set and an operational failure data set.
4. The automobile factory control system according to claim 3, wherein the fault prediction model is established by processing the equipment state monitoring indexes by using a convolutional neural network to obtain probability values of abnormality of each monitoring index of the equipment; and training and validating the fault prediction model using data in the two sets of data sets.
5. The automotive plant control system of any one of claims 1-4, further comprising a monitoring module, a video module, a decision module; the monitoring module is configured to detect equipment and manufacturing processes in a factory in real time to achieve synchronous mapping of manufacturing process states in the digital twin platform; the video module is configured to display manufacturing equipment and its operating state, and manufacturing process parameters; the decision module is configured to process the real-time manufacturing process parameters and update the manufacturing process parameters in the digital twin model and can generate an optimized process scheme.
6. The automotive factory control system according to any one of claims 1 to 4, wherein the three-dimensional model of the manufacturing apparatus is constructed by means of a QR code positioning method using a hologram apparatus.
7. The automotive plant control system of any one of claims 1-4, wherein mapping of plant manufacturing process parameters in the digital twin model is accomplished by fusion of manufacturing process parameters, including data fusion and physical fusion.
8. The automotive plant control system of any one of claims 1-4, further comprising a maintenance module in the digital twinning platform, the maintenance module configured to provide a maintenance intelligent boot of a manufacturing facility, the maintenance intelligent boot comprising: and guiding staff to complete maintenance operation based on the three-dimensional model of the manufacturing equipment, and updating current maintenance operation data into an equipment history maintenance database.
9. The automotive factory control system of claim 8, wherein the maintenance module is further configured to provide a maintenance intelligence interaction for a manufacturing device, the maintenance intelligence interaction comprising: based on the equipment history maintenance operation data, the health state of the equipment is visually displayed, a solution is provided for the equipment maintenance, and the maintenance operation data is updated into an equipment history maintenance database.
10. The automotive plant control system of any one of claims 1-4, further comprising a user interface communicatively coupled to the digital twinning platform and configured to enable an operator to monitor and perform control of the plant in real-time through the digital twinning platform and to be directed by the digital twinning platform to perform operations within the plant.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311559210.XA CN117270482A (en) | 2023-11-22 | 2023-11-22 | Automobile factory control system based on digital twin |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311559210.XA CN117270482A (en) | 2023-11-22 | 2023-11-22 | Automobile factory control system based on digital twin |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117270482A true CN117270482A (en) | 2023-12-22 |
Family
ID=89204932
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311559210.XA Pending CN117270482A (en) | 2023-11-22 | 2023-11-22 | Automobile factory control system based on digital twin |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117270482A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117518957A (en) * | 2023-12-29 | 2024-02-06 | 深圳市玩视科技股份有限公司 | Digital twinning-based central control system control method, device, equipment and medium |
| CN117787099A (en) * | 2023-12-28 | 2024-03-29 | 南京拓控信息科技股份有限公司 | Digital twin method for self-diagnosis of state of vehicle in-out warehouse detection shed equipment |
| CN117991731A (en) * | 2023-12-28 | 2024-05-07 | 安徽同新源科技有限公司 | Intelligent factory management and control method and system based on digital twinning |
| CN118192479A (en) * | 2024-05-17 | 2024-06-14 | 南京微思软件有限公司 | Factory workshop digital twin model correction method and system based on data acquisition |
| CN118211415A (en) * | 2024-04-08 | 2024-06-18 | 北京国信动力源科技发展有限公司 | Industrial digital twin platform configuration method, system and readable storage medium |
| CN118350228A (en) * | 2024-06-17 | 2024-07-16 | 山东捷瑞数字科技股份有限公司 | Dynamic equipment fault prediction method and system based on digital twinning |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110263846A (en) * | 2019-06-18 | 2019-09-20 | 华北电力大学 | The method for diagnosing faults for being excavated and being learnt based on fault data depth |
| CN110929918A (en) * | 2019-10-29 | 2020-03-27 | 国网重庆市电力公司南岸供电分公司 | 10kV feeder line fault prediction method based on CNN and LightGBM |
| CN111523659A (en) * | 2020-04-14 | 2020-08-11 | 北京华控智加科技有限公司 | Machine fault prediction diagnosis method based on three-level neural network modeling |
| CN111767398A (en) * | 2020-06-30 | 2020-10-13 | 国网新疆电力有限公司电力科学研究院 | Secondary equipment fault short text data classification method based on convolutional neural network |
| CN112163798A (en) * | 2020-10-29 | 2021-01-01 | 黑龙江东方学院 | Big data driven self-regulating automobile processing assembly line digital twin system |
| CN112818446A (en) * | 2021-01-26 | 2021-05-18 | 西安交通大学 | Construction method of intelligent workshop digital twin system |
| CN113409213A (en) * | 2021-06-22 | 2021-09-17 | 中铁工程装备集团有限公司 | Plunger pump fault signal time-frequency graph noise reduction enhancement method and system |
| KR20220050235A (en) * | 2013-03-11 | 2022-04-22 | 매직 립, 인코포레이티드 | System and method for augmented and virtual reality |
| CN114881162A (en) * | 2022-05-19 | 2022-08-09 | 广东电网有限责任公司 | Method, apparatus, device and medium for predicting failure of metering automation master station |
| CN115857439A (en) * | 2022-09-05 | 2023-03-28 | 中国船舶集团有限公司第七一六研究所 | Production line data mapping and fusing system and method based on digital twinning |
| CN115907700A (en) * | 2022-09-26 | 2023-04-04 | 北京卫星制造厂有限公司 | Real-time mapping and visual monitoring method for digital twin model of workshop |
| US20230186201A1 (en) * | 2016-05-09 | 2023-06-15 | Strong Force Iot Portfolio 2016, Llc | Industrial digital twin systems providing neural net-based adjustment recommendation with data relevant to role taxonomy |
| CN116680867A (en) * | 2023-04-27 | 2023-09-01 | 贵州电网有限责任公司 | Transformer visual fault diagnosis method and system based on refined three-dimensional model |
| CN116703368A (en) * | 2023-08-08 | 2023-09-05 | 国网信通亿力科技有限责任公司 | Synchronous line loss intelligent closed-loop monitoring method based on data mining |
| CN116756857A (en) * | 2023-08-16 | 2023-09-15 | 中汽研汽车工业工程(天津)有限公司 | Method and system for constructing digital twin body of automobile factory |
-
2023
- 2023-11-22 CN CN202311559210.XA patent/CN117270482A/en active Pending
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20220050235A (en) * | 2013-03-11 | 2022-04-22 | 매직 립, 인코포레이티드 | System and method for augmented and virtual reality |
| US20230186201A1 (en) * | 2016-05-09 | 2023-06-15 | Strong Force Iot Portfolio 2016, Llc | Industrial digital twin systems providing neural net-based adjustment recommendation with data relevant to role taxonomy |
| CN110263846A (en) * | 2019-06-18 | 2019-09-20 | 华北电力大学 | The method for diagnosing faults for being excavated and being learnt based on fault data depth |
| CN110929918A (en) * | 2019-10-29 | 2020-03-27 | 国网重庆市电力公司南岸供电分公司 | 10kV feeder line fault prediction method based on CNN and LightGBM |
| CN111523659A (en) * | 2020-04-14 | 2020-08-11 | 北京华控智加科技有限公司 | Machine fault prediction diagnosis method based on three-level neural network modeling |
| CN111767398A (en) * | 2020-06-30 | 2020-10-13 | 国网新疆电力有限公司电力科学研究院 | Secondary equipment fault short text data classification method based on convolutional neural network |
| CN112163798A (en) * | 2020-10-29 | 2021-01-01 | 黑龙江东方学院 | Big data driven self-regulating automobile processing assembly line digital twin system |
| CN112818446A (en) * | 2021-01-26 | 2021-05-18 | 西安交通大学 | Construction method of intelligent workshop digital twin system |
| CN113409213A (en) * | 2021-06-22 | 2021-09-17 | 中铁工程装备集团有限公司 | Plunger pump fault signal time-frequency graph noise reduction enhancement method and system |
| CN114881162A (en) * | 2022-05-19 | 2022-08-09 | 广东电网有限责任公司 | Method, apparatus, device and medium for predicting failure of metering automation master station |
| CN115857439A (en) * | 2022-09-05 | 2023-03-28 | 中国船舶集团有限公司第七一六研究所 | Production line data mapping and fusing system and method based on digital twinning |
| CN115907700A (en) * | 2022-09-26 | 2023-04-04 | 北京卫星制造厂有限公司 | Real-time mapping and visual monitoring method for digital twin model of workshop |
| CN116680867A (en) * | 2023-04-27 | 2023-09-01 | 贵州电网有限责任公司 | Transformer visual fault diagnosis method and system based on refined three-dimensional model |
| CN116703368A (en) * | 2023-08-08 | 2023-09-05 | 国网信通亿力科技有限责任公司 | Synchronous line loss intelligent closed-loop monitoring method based on data mining |
| CN116756857A (en) * | 2023-08-16 | 2023-09-15 | 中汽研汽车工业工程(天津)有限公司 | Method and system for constructing digital twin body of automobile factory |
Non-Patent Citations (6)
| Title |
|---|
| 刘鹏;李洪儒;王卫国;许葆华;: "基于补偿距离评估技术与灰色关联分析的滚动轴承故障程度识别", 机械传动, no. 10 * |
| 张国政 , 申君歌: "基于多周期时间序列的灰色预测模型及其应用", 统计与决策, pages 14 - 19 * |
| 李训铭;周志权;: "基于BP神经网络和灰色理论的示功图故障诊断", 电子设计工程, no. 17 * |
| 杨迪: "基于特征组合优化与卷积神经网络的滚动轴承故障诊断", 《中国优秀硕士学位论文全文库》, pages 32 - 36 * |
| 林晓清: "基于数字孪生理念的智能工厂与案例分析", 数字制造科学, pages 314 - 318 * |
| 王时龙，刘检华，杨波，王四宝，黄涛编: "中国战略性新兴产业研究与发展 智慧工业", 机械工业出版社, pages: 282 - 350 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117787099A (en) * | 2023-12-28 | 2024-03-29 | 南京拓控信息科技股份有限公司 | Digital twin method for self-diagnosis of state of vehicle in-out warehouse detection shed equipment |
| CN117991731A (en) * | 2023-12-28 | 2024-05-07 | 安徽同新源科技有限公司 | Intelligent factory management and control method and system based on digital twinning |
| CN117518957A (en) * | 2023-12-29 | 2024-02-06 | 深圳市玩视科技股份有限公司 | Digital twinning-based central control system control method, device, equipment and medium |
| CN118211415A (en) * | 2024-04-08 | 2024-06-18 | 北京国信动力源科技发展有限公司 | Industrial digital twin platform configuration method, system and readable storage medium |
| CN118192479A (en) * | 2024-05-17 | 2024-06-14 | 南京微思软件有限公司 | Factory workshop digital twin model correction method and system based on data acquisition |
| CN118350228A (en) * | 2024-06-17 | 2024-07-16 | 山东捷瑞数字科技股份有限公司 | Dynamic equipment fault prediction method and system based on digital twinning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116129366B (en) | Digital twinning-based park monitoring method and related device | |
| CN117270482A (en) | Automobile factory control system based on digital twin | |
| Wang | Towards zero-defect manufacturing (ZDM)—a data mining approach | |
| CN112085261B (en) | Enterprise production status diagnosis method based on cloud fusion and digital twin technology | |
| Massaro | Electronics in advanced research industries: Industry 4.0 to Industry 5.0 Advances | |
| WO2021160260A1 (en) | Digital platform using cyber-physical twin structures providing an evolving digital representation of a risk-related real world asset for quantifying risk measurements, and method thereof | |
| CN105372581A (en) | Flexible circuit board manufacturing process automatic monitoring and intelligent analysis system and method | |
| Cohen et al. | A smart process controller framework for Industry 4.0 settings | |
| CN114118673A (en) | Workshop intelligent fault diagnosis early warning method based on digital twin technology | |
| CN112327780A (en) | Digital twin system construction method and architecture of electronic equipment test production line | |
| CN112799369A (en) | Product assembly process control method and device | |
| CN117494905B (en) | Construction method and system for digital counterweight balance arch bridge | |
| CN112184007B (en) | Workshop equipment remote diagnosis method based on digital twin | |
| CN117078163A (en) | Logistics storage management and control system and method based on digital twinning | |
| CN116894211A (en) | System for generating human perceptible interpretive output, method and computer program for monitoring anomaly identification | |
| Kumari et al. | Advanced automobile manufacturing: an industry 4.0 | |
| Ebni et al. | Digital twin based smart manufacturing; from design to simulation and optimization schema | |
| CA3155163A1 (en) | Development of a product using a process control plan digital twin | |
| Zheng et al. | Visual analytics for digital twins: a conceptual framework and case study | |
| CN117807431A (en) | Visual management system of intelligent mill | |
| CN117993613A (en) | Artificial intelligence analysis method and system for operation data of digital factory | |
| Ren et al. | Research on digital twin framework for customized product manual assembly systems | |
| CN116551466B (en) | Intelligent monitoring system and method in CNC (computerized numerical control) machining process | |
| Xiao et al. | Digital twin-driven prognostics and health management for industrial assets | |
| KR102437917B1 (en) | Equipment operation system |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20231222 |
|
| RJ01 | Rejection of invention patent application after publication |