Petri net twinborn model fault diagnosis method for manufacturing system
By combining Petri net twin models and basic diagnostic tools, the difficulty of fault diagnosis in complex manufacturing systems using traditional Petri nets is solved, achieving efficient and accurate fault detection while reducing computational burden.
Patent Information
- Application Number
- CN202511219122.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-08-28
- Publication Date
- 2025-12-16
AI Technical Summary
Traditional Petri net-based state diagnostics cannot effectively diagnose deadlocks or long-term quiescent behaviors in complex manufacturing systems. They also have high computational complexity and cannot allocate resources reasonably in dynamic environments, leading to system bottlenecks and resource conflicts.
By employing a Petri net twin model, a basic diagnostic tool is constructed by adding control places and control transitions, and combining basic reachability techniques and fault propagation theory. This tool compresses the state space and performs fault diagnosis through transition sequence observation.
It reduces computational complexity, enables more accurate diagnosis of faults in complex manufacturing systems, avoids state explosion, and improves diagnostic efficiency and accuracy.
Smart Images

Figure CN121143219A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The application belongs to the field of fault diagnosis of manufacturing systems, and particularly relates to a Petri net twin model fault diagnosis method for a manufacturing system. BACKGROUND
[0002] Complex manufacturing systems are the core of modern industrial production, and their operation relies on the coordinated work of multiple key materials and advanced manufacturing equipment. In many manufacturing processes, reasonable resource allocation and efficient scheduling strategies are crucial to improving production efficiency and ensuring product quality. For example, in a highly precise production process, a manufacturing system typically consists of multiple processing units and intelligent logistics equipment such as robots and conveyor systems. During production, product workpieces need to be transferred between different processing modules and undergo a series of precise processing procedures. To ensure smooth production, logistics equipment such as robots and automated guided vehicles are responsible for efficiently transporting workpieces between modules. However, both processing modules and logistics equipment are limited resources and are subject to competition from multiple workpieces. The core problem of complex manufacturing system scheduling optimization is how to reasonably allocate these resources in a dynamic environment to ensure efficient operation of the system and avoid bottlenecks and resource conflicts.
[0003] With the continuous development of manufacturing technology, more and more enterprises choose to use combined equipment (integrated automation equipment with multiple processing units) to improve manufacturing flexibility and optimize production processes. In this context, modeling methods based on discrete event systems are widely used in production scheduling and optimization research to more accurately simulate the dynamic behavior of manufacturing systems and develop efficient scheduling strategies. Reasonable scheduling schemes not only improve system throughput but also reduce equipment idle time and processing delays, thereby further enhancing the overall competitiveness of the manufacturing system.
[0004] Petri nets are a modeling tool for designing and analyzing systems that can effectively describe behavior characteristics such as asynchrony, concurrency, and deadlock during the operation of combined equipment through digital and graphical representation. Therefore, they are widely used in related fields such as semiconductor wafer manufacturing. In partially observable Petri nets, calculating all possible states and state transition relationships of complex systems can lead to the problem of state explosion. The basic reachability technique introduces the concept of marking equivalence class, which can reduce the number of calculated and stored states, significantly reducing the required computational complexity. In addition, twin models can further reduce the required computational complexity, as the initial token number of all places is at most 1 and the topology of the net structure is more optimized than the original Petri net system. In the field of fault diagnosis of complex manufacturing systems, traditional Petri net-based state diagnosis tools cannot diagnose systems with deadlocks or long silent behaviors, and the required complexity is exponential. SUMMARY
[0005] In order to solve the above problems existing in the prior art, the application provides a Petri net twin model fault diagnosis method for a manufacturing system. In a first aspect, the application provides a Petri net twin model fault diagnosis method for a manufacturing system, comprising: establishing a Petri net model according to a processing flow of a combined device in the manufacturing system; adding a control place and a control transition to a topological structure of the Petri net model to construct a Petri net twin model; according to the Petri net twin model, compressing a state space to be calculated through a basic reachability technology, constructing a basic diagnostic device of the Petri net twin model based on classification of fault types; separating a state set based on observation of a transition sequence and fault propagation theory, using the basic diagnostic device to diagnose the manufacturing system and outputting a diagnosis result.
[0006] In an embodiment of the application, the combined device comprises: a robot TM, a loading and unloading site LP, a workpiece alignment module AL, a processing module PM and a buffer controller BC.
[0007] In an embodiment of the application, an expression of the Petri net model is as follows: ; wherein, represents a Petri net structure, , represents a place set, represents a transition set, and are directed arcs from to and from to , represents an initial marking, represents a preset alphabet used for representing an event set, represents a labeling function.
[0008] In an embodiment of the application, an expression of the Petri net twin model is as follows: ; wherein, represents a topological structure corresponding to a first transition of the Petri net after adding a control place and a control transition, , represents a second transition of the Petri net after adding the control place and the control transition, , a control library corresponding to the first transition, an initial identifier of a control library corresponding to the first transition, a control transition corresponding to the first transition, a credential of an initial identifier of a control library corresponding to the first transition, a preset alphabet for representing an event set, a marking function corresponding to the first transition. In an embodiment of the present application, according to the Petri net twin model, the state space to be calculated is compressed by using the basic reachability technique, and a basic diagnostic device of the Petri net twin model is constructed based on the classification of fault types, comprising: Based on the Petri net twin model, the state space to be calculated is compressed by using the basic reachability technique, and a basic state space is obtained;
[0009] According to the fault propagation rule, each basic reachable state in the basic state space is assigned a corresponding state token, and an assigned state space is obtained; Based on the classification of fault types and the assigned state space, a basic diagnostic device of the Petri net twin model is constructed. In an embodiment of the present application, the expression of the basic diagnostic device of the Petri net twin model is as follows:
[0010] ; ; Wherein, the expression of the basic diagnostic device, an event set, a state set of the basic diagnostic device, a transition function, an initial diagnostic device state.
[0011] In an embodiment of the present application, based on the observation of the transition sequence and the fault propagation theory, the state set is separated, the manufacturing system is diagnosed by using the basic diagnostic device, and the diagnostic result is output, comprising: Using the basic diagnostic device, the current state of the manufacturing system is calculated based on the observation of the transition sequence; the observation of the transition sequence includes the data output by the sensor of the manufacturing system, and the state of the manufacturing system includes: safe state, fault state and uncertain state; The current state of the manufacturing system is judged: If it is a safe state, the manufacturing system has not occurred a fault; If it is a fault state, the fault type is analyzed, and the identification number of the manufacturing system running and the number of observable events are output; If the state is uncertain, it is determined whether there is an uncertain path for the basic diagnoser: If not, the state in which the manufacturing system is located is calculated by waiting for the observation of the next transition sequence. If so, the manufacturing system cannot be diagnosed.
[0012] In a second aspect, the present application provides a Petri net twin model fault diagnosis device for a manufacturing system, comprising: A model construction module is configured to establish a Petri net model according to the processing flow of a combined device in the manufacturing system. A twin model construction module is configured to add control places and control transitions to the topological structure of the Petri net model to construct a Petri net twin model. A diagnoser construction module is configured to compress the state space that needs to be calculated by a basic reachability technique according to the Petri net twin model, and construct a basic diagnoser of the Petri net twin model based on the classification of fault types. A diagnosis module is configured to separate state sets based on the observation of the transition sequence and the fault propagation theory, diagnose the manufacturing system by using the basic diagnoser, and output a diagnosis result.
[0013] In a third aspect, the present application provides an electronic device, comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory complete communication with each other through the communication bus. The memory is configured to store a computer program. The processor is configured to execute the program stored on the memory to implement the steps of the Petri net twin model fault diagnosis method for a manufacturing system provided in the first aspect of the present application.
[0014] In a fourth aspect, the present application provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the steps of the Petri net twin model fault diagnosis method for a manufacturing system provided in the first aspect of the present application.
[0015] The present application has the following beneficial effects: In the scheme provided by the application, all transitions of the system are enabled at least once by introducing dynamic control places and transitions in the topological structure of the Petri net; in combination with the basic reachability technology, some basic representative identifiers can be more flexibly selected in the fault diagnosis process, and for a complex manufacturing system, the dynamic can more accurately reflect the actual production situation, so that the faults prone to occur can be more accurately diagnosed. Further, by means of the fault propagation theory, the basic diagnostic device is constructed, and the fault diagnosis problem of the complex system is converted into the problem of verifying whether the basic diagnostic device has an uncertain path, so that the required computing burden is greatly reduced. BRIEF DESCRIPTION OF DRAWINGS
[0016] Figure 1 A step schematic diagram of a Petri net twin model fault diagnosis method for a manufacturing system provided by an embodiment of the application; Figure 2 A structure schematic diagram of a Petri net twin model in a Petri net twin model fault diagnosis method for a manufacturing system provided by an embodiment of the application; Figure 3 A structure schematic diagram of an electronic device provided by an embodiment of the application. DETAILED DESCRIPTION
[0017] The application will be further described in detail below with specific embodiments, but the implementation of the application is not limited thereto.
[0018] An embodiment of the application provides a Petri net twin model fault diagnosis method, device, electronic equipment and storage medium for a manufacturing system.
[0019] It should be noted that the execution subject of the Petri net twin model fault diagnosis method for a manufacturing system provided by an embodiment of the application can be a device, and the device can run in an electronic equipment. The electronic equipment can be a server or a terminal device, but is not limited thereto.
[0020] Next, first, a Petri net twin model fault diagnosis method for a manufacturing system provided by an embodiment of the application is introduced.
[0021] As shown in the figure, the Petri net twin model fault diagnosis method for a manufacturing system provided by an embodiment of the application can include the following steps: Figure 1 S1, establishing a Petri net model according to the processing flow of a combined device in a manufacturing system.
[0022] Specifically, the combined device can include: A transfer module (TM), a load port (LP), an aligner (AL), a processing module (PM) and a buffer controller (BC). The TM is a transfer module, and the robot can perform detailed operations such as taking out and exchanging the workpiece, and can operate a single workpiece or multiple workpieces at a time.
[0023] It can be understood that in the combined device proposed in the embodiment of the application, the processing of the workpiece is as follows: after the workpiece is taken out from the LP by the TM, the workpiece is first placed in the AL module for preliminary correction, the corrected workpiece is transferred to the BC module (pre-cooling of the workpiece processing process), after being transferred to the buffer controller, the workpiece flows to the PM module to start processing, and after the workpiece is processed according to the set process in turn, the workpiece is again put into the buffer controller module, and is again placed in the LP in a certain order.
[0024] The processing of the combined device proposed in the embodiment of the application will process multiple workpieces at a time, and therefore, the Process-oriented Petri Net (POPN) is selected for the Petri net model of the combined device, that is, the process-oriented modeling method. The POPN is based on the modeling of the processing process of the manufacturing system, and the processing process of the workpiece is directly presented by constructing the model. This modeling method requires that the process of the workpiece processing be clear, and on this basis, each process and the resources in the system are modeled, and the established model is connected according to the processing order. In the model, the marks correspond to different states of the combined device in the running process, the places are used to represent each module or resource in the device, and the actions of the robot are represented by transitions. In the initial state, the tokens in the processing place represent the workpieces to be processed, and the tokens in the resource place represent the resources in the device, such as the occupation of the robot and the maximum number of workpieces that can be grasped.
[0025] When the Petri net is combined with the manufacturing system, in order to better consider the sensor data of the system, the Petri net with labels is selected. The labeled Petri net has two kinds of nodes, places and transitions, which are connected by directed arcs. Each mark corresponds to each state of the system, the place represents the state of a module of the system, and the transition can change the state. Therefore, the process of modeling the combined device is to use the place, the transition and the connecting arc to represent the state change of the system. Based on the complexity of the processing process of the combined device, the Process-oriented Petri Net (POPN) is used when running.
[0026] In the tag Petri net model constructed in the embodiment of the application, different states in the motion process of the corresponding combined device are marked, the places are divided into processing places and resource places, in the initial state, the processing places represent the processing modules, and the tokens in the processing places represent the workpieces to be processed. The resource places represent the resource control places with control capacity, and the tokens in the resource places represent the resources in the device.
[0027] The structure diagram of the Petri net twin model in the Petri net twin model fault diagnosis method for a manufacturing system provided in the embodiment of the application is shown in Figure 2 Figure 2 The solid line part in the figure is the tag Petri net model corresponding to the combined device proposed in the embodiment of the application, the combined device includes two double-arm manipulators, VTR1 and VTR2, which are responsible for picking and placing the workpieces in the two production lines. AL1 and AL2 are preliminary correction modules of the two production lines, and the processing process of the workpieces needs strict shape, quality, purity and other requirements, and the module can preliminarily process the workpieces into workpieces to be processed. The transition represents the operation of the manipulator, and the square represents pick, place and swap actions. Figure 2 The red part in the figure represents a fault. The number of tokens in the figure represents the number of workpieces to be processed, and the initial number is 10. The number of tokens in the resource place represents the ability of the corresponding device to process workpieces at the same time, for example, the token number of VTR1 is 3, which represents the number of workpieces that can be picked and placed by the manipulator at the same time. The approximate device processing flow is as follows: the workpieces to be processed are transferred from the VTR1 and VTR2 to the AL1 and AL2 for preliminary processing, and then transferred to the buffer controller BC by the VTR1 and VTR2 for cooling after the preliminary processing is completed. When the workpieces to be processed reach the standard, the workpieces are transferred to the PM processing module for processing, and the workpieces of the two production lines are combined into the required finished products. Finally, the workpieces are transferred to the LP for packaging and disassembling, and new workpieces to be processed are unloaded.
[0028] In the process of constructing the tag Petri net model in the embodiment of the application, the transition firing rule of the model is also determined. The firing rule defines the state transition rule. In the tag Petri net, the firing condition and sensor data of each transition need to be determined. To fire a certain transition, all input places connected to the transition must meet certain token conditions, specifically, there must be a sufficient number of tokens in the input place. The sensor data refers to the observation of the tag by the firing transition, and the premise is that the sensor is associated before the transition is enabled.
[0029] The expression of the Petri net model established in step S1 is as follows: ; wherein, denotes the Petri net structure, , denotes the set of places, denotes the set of transitions, and are directed arcs from to and from to respectively, denotes the initial marking, denotes a preset alphabet, which can be used to represent the set of events, denotes the labeling function, which assigns a label in to the observable transition or does not assign a label to the unobservable transition.
[0030] In the labeled Petri net model, the POPN (Process-oriented Petri Net) is selected to identify the different states in the motion process of the corresponding combined device. The places are divided into processing places and resource places. In the initial state , the processing places represent the processing modules, and the tokens inside represent the workpieces to be processed. The resource places represent the resource control places with control capacity, and the tokens inside represent the resources in the device.
[0031] S2, add control places and control transitions to the topological structure of the Petri net model to construct the Petri net twin model.
[0032] The expression of the Petri net twin model is as follows: ; wherein, denotes the topological structure of the Petri net after adding control places and control transitions , , , denotes the control place corresponding to the th transition, denotes the control transition corresponding to the th transition, denotes the initial marking of the th transition, denotes the preset alphabet, denotes the labeling function corresponding to the th transition. Specifically, , ,for formula and Establishment, for If there are non-emittable transitions , . It is the initial identifier assigned to all places. A token, The scaling function does not assign labels to the added control transitions. In the twin model... In the middle, all changes Each can be launched at least once.
[0033] This invention uses a Petri net twin model to diagnose faults in a complex manufacturing system. First, a scalarized Petri net model is established for the complex manufacturing system. Based on this, construct The topology. For each non-reflectable transition in this topology. By adding control libraries and control change and the corresponding directed arcs Make It can be launched at least once.
[0034] First, construct the topology of the Petri net, find the unenabled transitions in the topology, and add control places and transitions to enable them. Figure 2 The dashed lines represent the added control places and control transitions; the system formed by the dashed lines and the solid lines is a twin system. In the Petri net twin model, transition enabling is crucial. By observing the transition sequence, the original manufacturing system is diagnosed for faults, especially unobservable ones. The established twin model is as follows: , in It is a Petri net with added control library set and control transition set The subsequent topology. Specifically... It is a collection of warehouses; It is a set of transitions; for all places and transitions ,formula and Established. For all control locations and control transitions. If there are unenable transitions Add an arc .make Let be the initial set of places containing tokens, i.e., for all , Established.
[0035] To ensure that the introduction of control places and transitions does not cause the twin system to become unbounded or introduce new deadlocks, this embodiment of the invention selects all places containing tokens in their initial states. Add directed arcs At the same time, for all unenable transitions... For all its pre-requisite warehouses Add enable arc Meanwhile, for those overloaded transitions For all its pre-requisite warehouses Add two control transitions and the corresponding suppression arc . Is the initial identifier given to the first Each transition corresponds to a control voucher for all vouchers. A token, It is the first The scaling function corresponding to each transition: If , ;if , In other words, for the original Petri net transitions, their assigned labels remain unchanged, and no labels are assigned to any added control transitions. This results in a property: in the Siamese model... In the middle, all changes Each can be launched at least once.
[0036] S3, based on the Petri net twin model, compresses the state space that needs to be calculated through basic reachability techniques, and constructs a basic diagnostic tool for the Petri net twin model based on fault type classification, which may include: Based on the Petri net twin model, the basic reachability technique is used to compress the state space that needs to be calculated, thus obtaining the basic state space. ; According to the fault propagation rules, a corresponding state label is assigned to each basic reachable state in the basic state space. or Thus, the allocated state space is obtained; Based on the classification and assignment of fault types and the resulting state space, a basic diagnostic tool for Petri net twin models is constructed.
[0037] According to the fault propagation rules, a corresponding state label is assigned to each basic reachable state. or This indicates whether the state is reached without fault transition or with fault transition. Based on the classification of fault types, the basic diagnostic tool for constructing the twin model is a deterministic finite state automaton.
[0038] The expression for the basic diagnostic tool of the Petri net twin model is as follows: ; in, The expression representing the basic diagnostic tool. This represents the set of states of the basic diagnostic tool. Describe the transfer function, This indicates the initial diagnostic state. ,in It is the set of states of the basic diagnostic tool. Represents a set of events. .
[0039] Specifically, the fundamental reachability graph, as a method for compressing the state space, focuses on calculating the minimum interpretation vector. Specifically, let... For identification Downward transition The set of explanations; let For identification Downward transition The set of explanatory vectors; let As an identifier Downward transition The minimum set of explanations; let As an identifier Downward transition The set of minimum explanatory vectors. For a Petri net, Basic identifier set Satisfy: (1) (2) Given For all changes and all minimum interpretation vectors logo belong Right now, , .
[0040] Using the recursive method described above, we can obtain the twin system. Basic reachability graph: Deterministic finite state automata ,in It is the previous set of basic identifiers; It is a collection of events. It is a logical event used to describe the propagation of faults; It is a transfer function; It is the initial identifier.
[0041] S4, based on the observation of transition sequences and the theory of fault propagation, separates the state set and uses a basic diagnostic tool to diagnose the manufacturing system, outputting diagnostic results, which may include: Using a basic diagnostic tool, the current state of the manufacturing system is calculated based on observations of the transition sequence. The observations of the transition sequence include data output from the sensors of the manufacturing system, and the state of the manufacturing system includes: safe state, fault state, and uncertain state. Determine the current state of the manufacturing system: If the system is in a safe state, then no malfunction has occurred in the manufacturing system; If the system is in a fault state, the fault type is analyzed, and the number of identifiers and observable events of the manufacturing system operation are output. If the state is uncertain, then determine whether there is an uncertain path in the basic diagnostic tool: If it does not exist, wait for the next observation of the transition sequence to calculate the state of the manufacturing system; If it exists, the manufacturing system cannot diagnose it.
[0042] Specifically, embodiments of the present invention address the basic diagnostic device. The state set in the data is based on fault labels. Categorize: If all labels are identified , Established, Status This is recorded as a safe state; if all labels indicate... , Established, Status This is recorded as a faulty state; if a label is present... , , Established, Status This is denoted as an uncertain state. A cycle consisting entirely of uncertain states is called an uncertain cycle. If no event can occur in this state... Enabled, this nondeterministic state This is called a dead state. If the path traverses an indeterminate loop, a path starting from the initial diagnostic state and ending at any dead indeterminate state is called an indeterminate path. A manufacturing system is diagnosable if and only if the basic fault diagnostic tool of the twin system of the manufacturing system has no indeterminate paths.
[0043] In S4, when calculating the current state of the manufacturing system, the basic diagnostic tool can be used to abstract the observations of the transition sequence into symbolic data in order to calculate the current state of the manufacturing system.
[0044] When the current state of the manufacturing system is determined to be a fault state, the fault type is further analyzed, the number of steps from the occurrence of the fault to the fault state is calculated, and the number of identifiers and observable events of the manufacturing system operation are output.
[0045] The final output data results of this invention embodiment include: the number of states required by the diagnostic system based on different numbers of workpieces being processed and different equipment processing capabilities. As shown in the bolded part of the index comparison results in Table 1, in order to compare with other methods for fault diagnosis of complex manufacturing systems based on Petri nets, this embodiment of the invention also calculates the number of identifiers and the number of basic identifiers required by the traditional method.
[0046] Table 1. Comparison Results of Indicators
[0047] Table 1 compares the performance indicators of traditional Petri net-based fault diagnosis methods and the Petri net twin model-based fault diagnosis method proposed in this invention. As can be seen from Table 1, the Petri net twin model-based fault diagnosis method designed in this embodiment has a significant advantage in calculating the state space of complex manufacturing systems, and the number of basic identifiers in the twin model is significantly lower. Reducing the number of basic identifiers significantly reduces the state space of the underlying diagnostic tool, greatly decreasing the computational burden of searching uncertain paths. In complex systems, the state space can grow exponentially, especially in systems with concurrency and resource contention. Reducing the number of basic identifiers helps control the size of the state space, avoids the state explosion problem, and improves diagnostic efficiency. During the state space expansion process, each identifier needs to be stored in memory. If the identifier expansion is uncontrolled, the system may run out of memory. Reducing the number of expanded identifiers can also significantly reduce memory consumption, allowing the analysis process to run with limited memory resources, which is more suitable for large-scale Petri net models.
[0048] Understandably, this embodiment of the invention constructs a Petri net topology and adds control places and control transitions to obtain a twin model where all transitions are enabled at least once. This model construction avoids calculating the original system's state space, thus preventing the state explosion problem. Simultaneously, the basic reachability graph, by calculating the minimum interpretation vector, can retain some representative identifiers and discard those that have no impact on fault diagnosis, thereby further compressing the required computational state space. The feasibility and applicability of the reported method stem from the use of twin systems from complex systems, making the computational cost of real-world systems acceptable. The graphical representation of Petri nets and twin systems make this method easy to use and operate; by introducing labels into the Petri net model, characterizing sensor readings, the cost of data analysis and storage is reduced. Combined with the basic reachability graph technique, it allows for more flexible selection of some basic representative identifiers during fault diagnosis. For complex manufacturing systems, this dynamic can more accurately reflect the actual production situation, thus more accurately diagnosing potential faults. Finally, through fault propagation theory, a basic diagnostic tool is constructed, transforming the fault diagnosis problem of complex systems into the problem of verifying whether the basic diagnostic tool has uncertain paths, thereby greatly reducing the required computational burden.
[0049] Secondly, corresponding to the above method embodiments, this invention also provides a Petri net twin model fault diagnosis device for a manufacturing system, the device comprising: The model building module is used to create Petri net models based on the processing flow of combined equipment in a manufacturing system. The twin model building module is used to add control places and control transitions to the topology of the Petri net model to build a Petri net twin model. The diagnostic tool building module is used to compress the state space that needs to be calculated based on the Petri net twin model through basic reachability techniques, and build a basic diagnostic tool for the Petri net twin model based on the classification of fault types. The diagnostic module is used to separate the state set based on the observation of the transition sequence and the fault propagation theory. Using the basic diagnostic tool, it diagnoses the manufacturing system and outputs the diagnostic results.
[0050] Understandably, the working principle of the model building module can be found in step S1 of the Petri net twin model fault diagnosis method for manufacturing systems proposed in the first aspect; the working principle of the twin model building module can be found in step S2 of the Petri net twin model fault diagnosis method for manufacturing systems proposed in the first aspect; the working principle of the diagnostic tool building module can be found in step S3 of the Petri net twin model fault diagnosis method for manufacturing systems proposed in the first aspect; and the working principle of the diagnostic module can be found in step S4 of the Petri net twin model fault diagnosis method for manufacturing systems proposed in the first aspect. Further details will not be provided here.
[0051] Thirdly, embodiments of the present invention also provide an electronic device, such as... Figure 3 As shown, it includes a processor 001, a communication interface 002, a memory 003, and a communication bus 004, wherein the processor 001, the communication interface 002, and the memory 003 communicate with each other through the communication bus 004. The memory is used to store computer programs; When the processor executes the program stored in the memory, it implements the steps of any of the Petri net twin model fault diagnosis methods for manufacturing systems provided in the first aspect of the present invention.
[0052] The communication bus mentioned in the above electronic devices can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in the diagram, but this does not mean that there is only one bus or one type of bus.
[0053] The communication interface is used for communication between the aforementioned electronic devices and other devices.
[0054] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.
[0055] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.
[0056] The method provided in this invention can be applied to electronic devices. Specifically, the electronic device can be a desktop computer, a portable computer, a smart mobile terminal, a server, etc. No limitation is made herein; any electronic device that can implement this invention falls within the protection scope of this invention.
[0057] Fourthly, corresponding to the Petri net twin model fault diagnosis method for manufacturing systems provided in the first aspect, this embodiment of the invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of any of the Petri net twin model fault diagnosis methods for manufacturing systems provided in the first aspect of this invention.
[0058] For the embodiments of the device / electronic device / storage medium, since they are basically similar to the method embodiments, the description is relatively simple, and relevant parts can be referred to in the description of the method embodiments.
[0059] It should be noted that the device, electronic device, and storage medium in the embodiments of the present invention are respectively the device, electronic device, and storage medium for applying the above-described Petri net twin model fault diagnosis method for manufacturing systems. Therefore, all embodiments of the above-described Petri net twin model fault diagnosis method for manufacturing systems are applicable to the device, electronic device, and storage medium, and can achieve the same or similar beneficial effects.
[0060] It should be noted that, in the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.
[0061] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the apparatus embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.
[0062] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention are included within the scope of protection of the present invention.
Claims
1. A method for fault diagnosis using a Petri net twin model of a manufacturing system, characterized in that, include: Establish a Petri net model based on the processing flow of combined equipment in the manufacturing system; To construct a Petri net twin model, control places and control transitions are added to the topology of the Petri net model. Based on the Petri net twin model, the state space that needs to be calculated is compressed through basic reachability techniques, and a basic diagnostic tool for the Petri net twin model is constructed based on the classification of fault types. Based on the observation of transition sequences and the theory of fault propagation, the state set is separated, and the basic diagnostic tool is used to diagnose the manufacturing system and output the diagnostic results.
2. The method for fault diagnosis of Petri net twin models for manufacturing systems according to claim 1, characterized in that, The combined device includes: Robotic arm™, loading / unloading position LP, workpiece alignment module AL, machining module PM, and buffer controller BC.
3. The method for fault diagnosis of Petri net twin models for manufacturing systems according to claim 1, characterized in that, The expression for the Petri net model is as follows: ; in, This represents a Petri net structure. , Represents the set of places. Represents a set of transitions. and They are from arrive and from arrive The directed arc, Indicates the initial identifier. This represents a predefined alphabet used to represent event sets. This indicates a scalar function.
4. The method for fault diagnosis of Petri net twin model for manufacturing systems according to claim 1, characterized in that, The expression for the Petri net twin model is as follows: ; in, This indicates that after adding control locations and control transitions to the Petri net, the 1st... The topology corresponding to each transition. , Indicates the first The control library corresponding to each transition. Indicates the first Each change corresponds to a control change. Indicates the initial identifier assigned to the first The voucher in the control library corresponding to each transition. This represents a predefined alphabet used to represent event sets. Indicates the first The scaling function corresponding to each transition.
5. The method for fault diagnosis of Petri net twin models for manufacturing systems according to claim 1, characterized in that, Based on the Petri net twin model, the state space to be computed is compressed using basic reachability techniques. A basic diagnostic tool for the Petri net twin model is constructed based on fault type classification, including: Based on the Petri net twin model, the basic reachability technique is used to compress the state space that needs to be calculated, and the basic state space is obtained. According to the fault propagation rules, a corresponding state label is assigned to each basic reachable state in the basic state space to obtain the assigned state space. Based on the classification of fault types and the assigned state space, a basic diagnostic tool for Petri net twin models is constructed.
6. The method for fault diagnosis of Petri net twin models for manufacturing systems according to claim 1, characterized in that, The expression for the basic diagnostic tool of the Petri net twin model is as follows: ; in, The expression representing the basic diagnostic tool. Represents a set of events. This represents the set of states of the basic diagnostic tool. Describe the transfer function, This indicates the initial diagnostic state.
7. The method for fault diagnosis of Petri net twin models for manufacturing systems according to claim 1, characterized in that, Based on the observation of transition sequences and the theory of fault propagation, the state set is separated. Using the aforementioned basic diagnostic tool, the manufacturing system is diagnosed, and the diagnostic results are output, including: Using the aforementioned basic diagnostic tool, the current state of the manufacturing system is calculated based on the observation of the transition sequence; the observation of the transition sequence includes data output by the sensors of the manufacturing system, and the state of the manufacturing system includes: safe state, fault state, and uncertain state; Determine the current state of the manufacturing system: If the system is in a safe state, then no malfunction has occurred in the manufacturing system; If the system is in a fault state, the fault type is analyzed, and the number of identifiers and observable events of the manufacturing system operation are output. If the state is uncertain, then determine whether the basic diagnostic tool has an uncertain path: If it does not exist, wait for the next observation of the transition sequence to calculate the state of the manufacturing system; If present, the manufacturing system cannot diagnose it.
8. A Petri net twin model fault diagnosis device for manufacturing systems, characterized in that, include: The model building module is used to create Petri net models based on the processing flow of combined equipment in a manufacturing system. The twin model construction module is used to add control places and control transitions to the topology of the Petri net model to construct a Petri net twin model. The diagnostic tool building module is used to compress the state space that needs to be calculated based on the Petri net twin model by means of basic reachability techniques, and to build a basic diagnostic tool for the Petri net twin model based on the classification of fault types. The diagnostic module is used to separate the state set based on the observation of the transition sequence and the fault propagation theory. Using the basic diagnostic tool, it diagnoses the manufacturing system and outputs the diagnostic results.
9. An electronic device, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; The memory is used to store computer programs; When the processor executes the program stored in the memory, it implements the steps of the Petri net twin model fault diagnosis method for manufacturing systems according to any one of claims 1-7.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the Petri net twin model fault diagnosis method for manufacturing systems as described in any one of claims 1-7.
