Fault diagnosis method and system based on Petri network and storage medium

Abstract

The invention discloses a Petri network-based fault diagnosis method and system and a storage medium, and the method comprises the steps: S1, obtaining a time sequence signal of substation equipment, the time sequence signal comprising alarm information and accident and displacement information; s2, on the basis of the time sequence signal of the substation equipment, constructing a fault judgment rule base and setting a front part weight, a rule starting threshold value and a back part credibility; s3, establishing a fuzzy function based on the preceding component weight, the rule starting threshold and the consequent component credibility; s4, on the basis of the Petri network and the fuzzy function, a network topology model of the substation equipment is established, and the network topology model of the substation equipment comprises a fault intelligent judgment rule base; and S5, based on the network topology model of the substation equipment, obtaining a current time sequence signal of the substation equipment, and judging a fault condition of the substation equipment. By establishing the network topology model of the substation equipment, the fault diagnosis analysis speed is increased, the fault diagnosis efficiency is improved, and the diagnosis effect of the power system is optimized.

Description

Technical Field

[0001] The present application relates to the technical field of power system protection, and in particular to a fault diagnosis method, system and storage medium based on Petri nets. Background Art

[0002] In recent years, high voltage direct current (HVDC) technology has achieved breakthrough development and has been widely used in the operation of power systems. As a very novel transmission technology, HVDC technology occupies an extremely important position in the new power system because of its unique advantages. Compared with traditional high voltage AC transmission technology, high voltage direct current transmission technology has many advantages such as low line cost, large transmission capacity, low loss, ability to complete asynchronous networking and limit short-circuit current of power system. Therefore, it is widely used in long-distance, high-power, asynchronous networking transmission occasions. The HVDC system is a very large and complex system. Various faults are inevitable during operation. When the system fails, it is necessary to identify the fault in time and organize personnel to repair it to avoid the bad situation of the entire system being shut down.

[0003] Expert systems are often used in the prior art for fault detection, and computer models of expert reasoning methods are used to judge faults. Expert system diagnostic methods can use rules to represent the action logic of protectors and circuit breakers and the diagnostic experience of operators, allowing the addition or deletion of rules to improve the real-time and effectiveness of the diagnostic system. They are suitable for fault diagnosis of small and medium-sized power systems and substations, but it is difficult to establish a knowledge base, the fault tolerance is poor, the knowledge base maintenance is difficult, and there will be problems such as combinatorial explosion and slow reasoning speed in fault diagnosis. It is necessary to improve and perfect it. Summary of the invention

[0004] In view of the defects existing in the prior art, the present application provides a fault diagnosis method, system and storage medium based on Petri nets to improve the existing power detection efficiency and optimize the diagnosis effect of power system faults.

[0005] To achieve the above object, the present invention provides a fault diagnosis method based on Petri net, comprising the following steps:

[0006] S1. Obtaining a timing signal of a substation device, wherein the timing signal includes alarm information, accident and displacement information;

[0007] S2. Based on the timing signal of the substation equipment, a fault discrimination rule base is constructed and antecedent weights, rule activation thresholds and consequent credibility are set;

[0008] S3, establishing a fuzzy function based on the antecedent weight, the rule activation threshold and the consequent credibility;

[0009] S4. Based on the Petri net and the fuzzy function, a network topology model of the substation equipment is established, wherein the network topology model of the substation equipment includes a fault intelligent judgment rule base;

[0010] S5. Based on the network topology model of the substation equipment, obtain the current timing signal of the substation equipment, and determine the fault condition of the substation equipment.

[0011] Furthermore, the antecedent weight, rule activation threshold and consequent credibility are all determined through technical specifications of the substation equipment and fault case analysis.

[0012] Furthermore, after the step of establishing a fuzzy function based on the antecedent weight, the rule activation threshold and the consequent credibility in S3, the Petri net-based fault diagnosis method further includes:

[0013] Determine an event token value of the substation equipment based on a timing signal of the substation equipment, wherein the event token value represents an abnormal condition of the substation equipment;

[0014] Determining a failure risk of the substation equipment based on an event token value of the substation equipment;

[0015] If the event token value is greater than a preset rule activation threshold, a fuzzy value is calculated based on the consequent credibility, and the event token value is adjusted.

[0016] Furthermore, the event token value is greater than 0 and less than 1.

[0017] Furthermore, the step of adjusting the event token value includes:

[0018] comparing the event token value with the rule activation threshold;

[0019] If the event token value is greater than the rule activation threshold, the difference between the event token value and the rule activation threshold is used as the adjusted event token value, and the product of the adjusted event token value and the consequent credibility is used as the fuzzy value.

[0020] Furthermore, the fuzzy function includes an input intensity function S(i,α), an output intensity function R(o,β), a state transfer control function F(t), a rated input quantity i and a rated output quantity o, and satisfies the relationship: S n =i n α n ; R n =o n β n ; i=M(p); o=F, where α n is the directed arc weight, β n is the credibility of the subsequent event, i, F, o are the token values ​​before, during and after the transition adjustment respectively, n is the serial number of the substation equipment, t is the time number of the substation equipment, and M(p) is the mark value of the event.

[0021] Further, after the step of S5, based on the network topology model of the substation equipment, acquiring the current timing signal of the substation equipment and determining the fault condition of the substation equipment, the Petri net-based fault diagnosis method further includes:

[0022] If the substation equipment fails, the fault handling method is determined according to the TR matrix, and the TR matrix satisfies the following relationship:

[0023]

[0024] Where K is the number of preset conditional propositions, 1 represents that the conditional proposition is true, 0 represents that the conditional proposition is false, -1 represents that the conditional proposition is in an uncertain state, R(i, j) is the value of each item in the TR matrix, P i , P j is a conditional proposition, i, j = 1, 2,…, K.

[0025] Furthermore, the Petri net-based fault diagnosis method further includes:

[0026] Determine a preset initial truth value vector, and operate the initial truth value vector with the TR matrix to determine an inference result that the substation equipment has a fault;

[0027] The present invention also provides a fault diagnosis system based on Petri net, the system comprising:

[0028] An acquisition module, used to acquire a timing signal of the substation equipment, wherein the timing signal includes alarm information, accident and displacement information;

[0029] A fault discrimination rule module is used to construct a fault discrimination rule base and set antecedent weights, rule activation thresholds and consequent credibility based on the timing signals of the substation equipment;

[0030] A model building module, used to build a network topology model of the substation equipment based on the Petri net and the antecedent weight, the rule activation threshold and the consequent credibility, wherein the network topology model of the substation equipment includes a fault intelligent judgment rule base;

[0031] The fault judgment module is used to obtain the current timing signal of the substation equipment based on the network topology model of the substation equipment and judge the fault condition of the substation equipment.

[0032] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the Petri net-based fault diagnosis method as described in any one of the above items is implemented.

[0033] Compared with the prior art, the advantages of the present invention are:

[0034] (1) The present invention uses Petri nets combined with computer simulation to perform rapid calculation and reasoning to draw conclusions, greatly improving the speed of fault diagnosis and analysis, and improving the efficiency of fault diagnosis.

[0035] (2) Since Petri nets have both strict mathematical expressions and intuitive graphical expressions, they have rich system description methods and system behavior analysis technologies, which can improve the work efficiency of staff and help them make reasonable judgments. They are more efficient in processing data with large amounts of information. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0037] Figure 1 A flowchart of a fault diagnosis method based on Petri nets according to an embodiment of the present application;

[0038] Figure 2 A network topology model diagram of a substation equipment in an embodiment of the present application; and

[0039] Figure 3 This is a system block diagram of a Petri net-based fault diagnosis system according to an embodiment of the present application.

[0040] Description of reference numerals:

[0041] 100, acquisition module; 200, fault identification rule module; 300, model building module; 400, fault identification module. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0043] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0044] In the description of the present application, the terms "first / second" involved are merely used to distinguish similar objects and do not represent a specific order for the objects. It can be understood that "first / second" can be interchanged with a specific order or sequence where permitted, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein.

[0045] It should be understood that the orientation or position relationship is based on the orientation or position relationship shown in the drawings. These orientation terms are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present application.

[0046] Power system fault judgment refers to the use of knowledge of the power system and its protection devices and information such as relay protection to accurately identify the location, type and malfunction of faulty components, and lay the foundation for taking measures to repair and improve existing deficiencies and improve the operation level of the power system. In the process of power system fault judgment, fault component identification is the focus and key content. At the same time, in order to improve the accuracy of fault judgment, effectively grasp the actual situation of the power system, grasp the key points of operation, and improve the level of fault judgment, it is necessary.

[0047] At present, with the continuous summary and development of technology, the fault judgment method of power system is gradually improved and perfected. The main methods and research status are reflected in the following aspects: expert system, artificial neural network, optimization technology, and fuzzy set theory. Among them, the expert system uses the computer model of the expert reasoning method to judge the fault. At present, this method is widely used and has obvious effects. It is a relatively successful method. According to the different fault diagnosis knowledge representation and the reasoning strategy used, the expert system can be divided into a heuristic rule-based reasoning system and a combined positive and negative reasoning system. At present, most fault judgments use the first method. The expert system diagnosis method can use rules to represent the action logic of the protector and circuit breaker and the diagnostic experience of the operator, allowing the addition or deletion of rules to improve the real-time and effectiveness of the diagnosis system. This method is suitable for small and medium-sized power systems and substation fault diagnosis. However, this method also has defects. For example, it is difficult to establish a knowledge base, the fault tolerance is poor, the knowledge base maintenance is difficult, and there will be problems such as combinatorial explosion and slow reasoning speed in fault diagnosis. It is necessary to improve and perfect it.

[0048] The fault judgment method based on artificial neural network has strong fault tolerance and learning ability. Common methods include forward neural network based on BP algorithm and neural network based on radial basis function. Through the application of these methods, diagnostic information can be effectively mastered and accurate diagnostic results can be obtained. This method can avoid the problem of difficult knowledge base construction and does not require the construction of inference engine. It is suitable for fault diagnosis of small and medium-sized power systems. The main problems faced in the application include: it is not easy to obtain complete sample data of the power system, the interaction function with symbolic data is weak, the heuristic knowledge processing is difficult, and the ability to explain its own behavior and output results is lacking. These defects restrict the promotion and application of this method and need to be further studied and analyzed.

[0049] The fault judgment method based on optimization technology is a solution method based on mathematical models. It describes the fault diagnosis problem as a 0-1 integer programming problem, constructs an analytical mathematical model, and uses optimization technology to find the optimal solution. This method is relatively strict in both diagnostic theory and diagnostic model, and has good results in specific applications. It does not require the introduction of heuristic knowledge, and conventional algorithms can accurately obtain detection results. It is suitable for power system fault diagnosis with relatively complete required information. The content that needs to be improved includes: establishing a reasonable mathematical model for power grid fault diagnosis, considering that multi-level backup protection is difficult, there are random factors, and the optimal solution may be lost, and the speed may be relatively slow. It is necessary to increase the running speed of the model.

[0050] The fault judgment method based on fuzzy set theory is to describe the possibility of fault existence with fuzzy membership, or to give fault diagnosis results by expert system or artificial neural network according to the system network topology and the action protection caused by the fault, the credibility of alarm information, etc. This method can handle the uncertainty of information and is usually used in combination with expert system and artificial neural network. It is difficult to use effective methods to describe uncertainty problems and establish large-scale fuzzy models. Therefore, it is necessary to propose an effective fault judgment method.

[0051] Figure 1 Schematic diagram of a Petri net-based fault diagnosis method according to an embodiment of the present application. Figure 2 This is a network topology model diagram of a substation equipment in an embodiment of the present application. Figure 1-Figure 2 As shown in FIG. 1 , the fault diagnosis method based on Petri net includes:

[0052] S1. Obtaining timing signals of substation equipment, wherein the timing signals include alarm information, accident and displacement information;

[0053] S2. Based on the timing signals of substation equipment, a fault discrimination rule base is constructed and the antecedent weight, rule activation threshold and consequent credibility are set;

[0054] S3, establishing a fuzzy function based on the antecedent weight, the rule activation threshold and the consequent credibility;

[0055] S4. Based on Petri net and fuzzy function, a network topology model of substation equipment is established, wherein the network topology model of substation equipment includes a fault intelligent judgment rule base;

[0056] S5. Based on the network topology model of the substation equipment, obtain the current timing signal of the substation equipment and determine the fault condition of the substation equipment.

[0057] Specifically, by monitoring the equipment operation status, fault records, etc., the timing signals of the substation equipment are obtained. These timing signals include alarm information, accident and displacement information. For example, the operation data of various equipment in the substation are collected in real time through the acquisition device, including alarm information, accident and displacement information. These data can come from protection relays, measurement and control devices, automation equipment, etc. By preprocessing these data, a unified timing signal is formed to facilitate subsequent fault diagnosis.

[0058] According to the acquired time series signals, a fault discrimination rule base is constructed. In the rule base, the antecedent weight, rule activation threshold and consequent credibility are set for each rule. These parameters can be optimized according to the operating characteristics of the equipment and historical fault data. Based on the antecedent weight, rule activation threshold and consequent credibility, a fuzzy function is established. The fuzzy function can convert the input time series signal into a fuzzy logic output to adapt to the uncertainty and ambiguity in the operation process of substation equipment. For example, based on the operating experience and expert knowledge of substation equipment, a fault discrimination rule base can be constructed. The rule base includes various fault types and their corresponding discrimination rules. At the same time, in order to improve the accuracy of fault diagnosis, a weight is set for the antecedent of each rule, a threshold is set for the activation of the rule, and a credibility is set for the consequent. Based on the antecedent weight, rule activation threshold and consequent credibility, a fuzzy function is established. The fuzzy function can map the input time series signal to a fault probability value for subsequent fault judgment.

[0059] The network topology model of substation equipment is established by using Petri nets and fuzzy functions. The network topology model includes a fault intelligent judgment rule base, which can be constructed according to the structure and operation characteristics of the equipment. Based on the network topology model of substation equipment, the current timing signal of the substation equipment is obtained, and the fault condition of the substation equipment is judged. The input timing signal is processed through fuzzy logic and fault judgment rule base, and the fault diagnosis result is output.

[0060] In some embodiments, the antecedent weight, the rule activation threshold and the consequent credibility are all determined through technical specifications of the substation equipment and fault case analysis.

[0061] In some embodiments, after the step of establishing a fuzzy function based on the antecedent weight, the rule activation threshold and the consequent credibility in S3, the fault diagnosis method based on the Petri net further includes:

[0062] Based on the timing signal of the substation equipment, an event token value of the substation equipment is determined, and the event token value represents an abnormal situation of the substation equipment;

[0063] Determine the failure risk of substation equipment based on the event token value of the substation equipment;

[0064] If the event token value is greater than the preset rule activation threshold, the fuzzy value is calculated based on the credibility of the consequent and the event token value is adjusted.

[0065] Specifically, the substation is an extremely critical part of the power system, and the safe operation of its equipment is crucial to the stability of the entire power system. Fault diagnosis technology can monitor the operating status of substation equipment in real time, timely discover and warn of potential fault risks, thereby ensuring the safe operation of the power system. A fault diagnosis model based on Petri nets, which includes the determination of antecedent weights, rule activation thresholds and consequent credibility, and the establishment of fuzzy functions. By analyzing the timing signals of substation equipment, the event token value is determined, the fault risk is evaluated, and the event token value is adjusted according to the consequent credibility. In particular, the event token value is greater than 0 and less than 1.

[0066] For example, according to the technical specifications of substation equipment and fault case analysis, the antecedent weight, rule activation threshold and consequent credibility are determined. Based on these parameters, a fuzzy function is established to characterize the uncertainty and fuzziness of the equipment status. Sensors and monitoring equipment are used to collect time series signals of substation equipment, such as voltage, current, temperature, etc. Based on these signals, the event token value of the substation equipment is determined, which characterizes the abnormal situation of the equipment. According to the event token value, the failure risk of the substation equipment is evaluated. The higher the event token value, the greater the possibility of abnormality of the equipment and the higher the failure risk. If the event token value is greater than the preset rule activation threshold, the fuzzy value is calculated based on the consequent credibility. The higher the consequent credibility, the more reliable the diagnosis result. According to the calculation results, the event token value is adjusted to more accurately reflect the actual status of the equipment. According to the adjusted event token value and fuzzy function, fault diagnosis is performed. If the diagnosis result shows that the equipment is faulty, an early warning is issued in time to notify the maintenance personnel to take corresponding maintenance measures. During the implementation process, the feasibility of the technical solution is verified through actual operation, and the details of the solution are adjusted in time according to feedback to ensure the efficiency, safety and reliability of substation equipment fault diagnosis.

[0067] In some embodiments, the step of adjusting the event token value includes:

[0068] Compare the event token value with the rule activation threshold;

[0069] If the event token value is greater than the rule activation threshold, the difference between the event token value and the rule activation threshold is used as the adjusted event token value, and the product of the adjusted event token value and the consequent credibility is used as the fuzzy value.

[0070] Specifically, the operation data of the substation equipment, including parameters such as voltage, current, and temperature, are collected, and the event token value is calculated using these data to characterize the abnormality of the equipment. The calculated event token value is compared with the preset rule start threshold. The rule start threshold is determined based on the technical specifications of the equipment and historical fault data, and is used to determine whether the equipment needs further fault diagnosis. If the event token value is greater than the rule start threshold, it indicates that the equipment may have a fault. At this time, the difference between the event token value and the rule start threshold is used as the adjusted event token value to eliminate the uncertainty caused by the event token value being higher than the threshold. The adjusted event token value is multiplied by the subsequent credibility to obtain a fuzzy value. The subsequent credibility is determined based on the technical specifications of the equipment and the fault case analysis, and is used to characterize the credibility of the fault diagnosis result. Fault diagnosis is performed based on the calculated fuzzy value. If the fuzzy value exceeds the preset fuzzy threshold, it is determined that the equipment has a fault, and an early warning signal is issued to notify the maintenance personnel to troubleshoot and handle the fault.

[0071] In some embodiments, the fuzzy function includes an input intensity function S(i, α), an output intensity function R(o, β), a state transition control function F(t), a rated input quantity i and a rated output quantity o, and satisfies the relationship: S n =i n α n ; R n =o n β n ; i=M(p); o=F, where α n is the directed arc weight, β n is the credibility of the subsequent event, i, F, o are the token values ​​before, during and after the transition adjustment, n is the serial number of the substation equipment, t is the time number of the substation equipment, and M(p) is the mark value of the event.

[0072] . When an accident occurs, a large amount of time series information will be generated. Through reasoning analysis, alarm information and accident information can be extracted from it to prompt the substation staff to check and handle it. However, if the amount of information is too large, the workload will increase exponentially, which is obviously not feasible. In some cases, the fault signals that the staff are preparing to handle can be completely ignored because their credibility is very low; or two faults with different credibility correspond to the same processing method. At this time, it is necessary to determine whether to choose this processing method; in some cases, there are multiple reasons for the generation of a fault signal, resulting in a variety of processing methods, and the specific choice also needs to be determined.

[0073] For example, p1, p2, p3...p8 represent the monitoring system sending out heavy gas protection action, differential protection action signal, fire extinguishing device alarm, fire general alarm information, transformer side circuit breaker trip signal, transformer three-phase current value 0, transformer smoke fire signal. The initial marker library M(Pi) takes the value of: if there is a proposition represented by the library, the token value is 1, otherwise it is 0. Here we also need to introduce a new concept token value. Each library, namely M(Pi), has an initial value not less than 0 called token. If the token value is greater than the weight in the directed arc, the transition is in the enabled state, also called the occurrence state. For each transition, the corresponding threshold is also set so that the transition in the occurrence state can be triggered and ignited. After the transition is ignited, the token in the library connected to the transition will be smoothly transferred to the next library until there is no transition in the Petri net to activate the transfer token. At this time, the Petri net enters the final stable state. In the model of the application embodiment, the token value will change in three situations. One is that before the transition, the token value becomes the product of the initial token and the weight of the directed arc; the other is that when judging whether the transition token value reaches the threshold, the token value is the sum of the arrived token values; and the other is that after the transition, the token value becomes the difference between the original token value and the threshold, and then the result of multiplying it with the credibility of the consequent is passed to the next place. Therefore, it can be seen that if the propositions of the places M(p7) and M(p8) are true, the token values ​​are 0.5 and 0.4 respectively, and the sum is 0.9, which is greater than the rule activation threshold of 0.7. Therefore, the transition is in the occurrence state, and the token value becomes 0.2. Multiplying it with the credibility of the consequent CF1 gives the fuzzy value of the conclusion of 0.45. The same reasoning can be used to infer other situations.

[0074] In some embodiments, after the step of S5, obtaining the current timing signal of the substation equipment based on the network topology model of the substation equipment and determining the fault condition of the substation equipment, the fault diagnosis method based on the Petri net further includes:

[0075] If a substation equipment fails, the fault handling method is determined according to the TR matrix, which satisfies the following relationship:

[0076] Where K is the number of preset conditional propositions, 1 represents that the conditional proposition is true, 0 represents that the conditional proposition is false, -1 represents that the conditional proposition is in an uncertain state, R(i, j) is the value of each item in the TR matrix, P i , P j is a conditional proposition, i, j = 1, 2,…, K.

[0077] Specifically, the TR matrix is ​​multiplied by the premise truth state vector to obtain the conclusion truth state vector, which is equivalent to one step of forward reasoning. The multiplication here is still the usual matrix multiplication, except that the multiplication is changed to AND (∧) and the addition is changed to OR (∨). Then the TR matrix is ​​multiplied by the truth state vector obtained in the previous step, which is equivalent to another reasoning. This cycle is repeated until the truth state vector remains unchanged, and the final reasoning result is obtained.

[0078] In some embodiments, the Petri net-based fault diagnosis method further includes:

[0079] A preset initial truth value vector is determined, and the initial truth value vector is operated with the TR matrix to determine the inference result of the substation equipment failure.

[0080] like Figure 3 As shown, the embodiment of the present application also provides a fault diagnosis system based on Petri nets, and the power line inspection system includes an acquisition module 100, a fault judgment rule module 200, a model building module 300 and a fault judgment module 400, wherein the acquisition module 100 is used to obtain the timing signal of the substation equipment, wherein the timing signal includes alarm information, accidents and displacement information; the fault judgment rule module 200 is used to build a fault judgment rule library based on the timing signal of the substation equipment and set the antecedent weight, the rule activation threshold and the consequent credibility; the model building module 300 is used to build a network topology model of the substation equipment based on the Petri net and the antecedent weight, the rule activation threshold and the consequent credibility, wherein the network topology model of the substation equipment includes a fault intelligent judgment rule library; the fault judgment module 400 is used to obtain the current timing signal of the substation equipment based on the network topology model of the substation equipment, and judge the fault condition of the substation equipment.

[0081] In order to better understand the fault diagnosis method based on Petri net in the embodiment of the present application, Figure 2 Provide a detailed description.

[0082] The total amount of various signals of substation equipment is very large. In the embodiment of the present application, the substation timing signals are divided into alarm information, accident and displacement information, and it is necessary to integrate and classify various types of original fault key information. The operating status of the equipment in the substation is complex, and the types and causes of defects are diverse. When a fault occurs, fault information from multiple sources is reported centrally, with a large amount of data and high repetitiveness, which is not conducive to the operation and maintenance personnel to quickly judge and locate the cause of the fault. Through the arrangement of typical faults of primary equipment, basic data is provided for establishing a rule base for fault judgment. A network topology model that conforms to the actual structure of the primary equipment of the power grid is established, and its simplification standard is: the bus, circuit breaker, and transformer are used as nodes, and the transmission line is equivalent to the edge in the network topology model. Add virtual connections between circuit breakers and busbars and lines and simplify them to edges in the network topology model. In this way, a connection matrix that characterizes the connection relationship of electrical equipment in the power grid can be obtained, and a state vector is established at the same time. After rule construction and rule pruning, a fuzzy function is established to complete the construction of the fuzzy Petri net.

[0083] This application takes the transformer as an example to build a fault judgment rule base. The rule base is stored in the database in the form of a table, including fields such as rule identification, rule activation threshold, rule description, rule credibility, rule reason, etc., and is associated with the antecedent table, which stores the fuzzy reasoning weight of the antecedent.

[0084] Next, we will use the example of transformer fire fuzzy function to illustrate the process of establishing fuzzy function:

[0085] The typical signals of transformer fire are shown in Table 1.

[0086] Table 1 Typical signals of transformer fire

[0087] Signal category Signal Source Related signals remote message Protective devices The monitoring system issues a heavy gas protection action remote message Protective devices Differential protection action signal remote message Measurement and control device Fire extinguishing device alarm remote message Fire protection system Fire General Alarm Information telemetry Protective devices Transformer circuit breaker tripping signal on each side telemetry Measurement and control device The three-phase current value of the transformer is 0 remote message Protective devices Transformer smoke fire signal remote message Protective devices Automatic sprinkler system start signal

[0088] Establish transformer fire fuzzy function: as shown in Table 2.

[0089] Table 2 Transformer fire fuzzy function

[0090]

[0091]

[0092] Among them, the weights of relevant signals, rule activation thresholds and consequent credibility CF all need to be determined through equipment failure case rule mining, technical specifications and operational experience to determine their specific values. These values ​​determine the final effect of fault identification.

[0093] The following describes the transformer fire cases in Tables 1 and 2. By using the correspondence between fuzzy Petri nets and fuzzy production rules, the fuzzy reasoning process can be simulated using Petri nets, such as Figure 2 As shown:

[0094] Among them, p1, p2, p3...p8 represent the monitoring system sending out heavy gas protection action, differential protection action signal, fire extinguishing device alarm, fire general alarm information, transformer side circuit breaker tripping signal, transformer three-phase current value 0, transformer smoke fire signal. The initial mark library M(Pi) takes the value: if there is a proposition represented by the library, the token value is 1, otherwise it is 0. Here we also need to introduce a new concept token value. Each library, namely M(Pi), has an initial value not less than 0 called token. If the token value is greater than the weight in the directed arc, the transition is in the enabled state, also called the occurrence state. For each transition, the corresponding threshold is also set so that the transition in the occurrence state can be triggered and ignited. After the transition is ignited, the token in the library connected to the transition will be smoothly transferred to the next library until there is no transition in the Petri net to activate the transfer token. At this time, the Petri net enters the final stable state. In the model of the embodiment of the present application, the token value will change in three situations. One is that before the transition, the token value becomes the product of the initial token and the weight of the directed arc; the other is that when judging whether the transition token value reaches the threshold, the token value is the sum of the arrived token values; and the other is that after the transition, the token value becomes the difference between the original token value and the threshold, and then the result after multiplying it with the credibility of the consequent is passed to the next library. Therefore, it can be seen that if the propositions of the library places M(p7) and M(p8) are true, the token values ​​are 0.5 and 0.4 respectively, and the sum is 0.9, which is greater than the rule activation threshold of 0.7. Therefore, the transition is in the occurrence state, and the token value becomes 0.2. Multiplying it with the credibility of the consequent CF1 to obtain the fuzzy value of the conclusion is 0.45. The same reasoning can be used to infer other situations.

[0095] It should be noted that the fault diagnosis method in the embodiment of the present application proposes an intelligent fault analysis technology based on Petri nets, compared with the expert system: the expert system has difficulty in establishing a knowledge base, poor fault tolerance, difficulty in maintaining the knowledge base, and problems such as combinatorial explosion and slow reasoning speed may occur in fault diagnosis; while the fault diagnosis method based on Petri nets proposed in the present application can quickly calculate and reason to draw conclusions through computer simulation through fuzzy Petri network structure, which greatly improves the reasoning speed compared with the expert system.

[0096] Compared with the fault judgment method based on artificial neural network, Petri net has both strict mathematical expression and intuitive graphical expression, rich system description means and system behavior analysis technology, and provides a solid conceptual foundation for computer science. However, the fault judgment method based on artificial neural network lacks the ability to explain its own behavior, and has weak interaction function with symbolic data, making it difficult to obtain complete sample data of the power system.

[0097] The fault judgment method based on optimization technology has randomness, so it may lose the optimal solution and run slowly. However, the Petri net structure is simple and intuitive, so the calculation speed is faster than the fault judgment method based on optimization technology and it is more stable.

[0098] Compared with other fault judgment methods based on fuzzy set theory, the intelligent fault judgment technology based on Petri net uses a Petri net model with a strict mathematical expression. It is faster than the method combining fuzzy set theory with expert system, and more efficient in processing large amounts of data; it is more interpretable than the method combining fuzzy set theory with artificial neural network.

[0099] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the above-mentioned Petri net-based fault diagnosis method is implemented.

[0100] It should be noted that the storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a ROM (Read-Only Memory), a RAM (Random Access Memory), a magnetic disk or an optical disk.

[0101] It will be appreciated by those skilled in the art that all or some of the steps, systems, and functional modules / units in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In a hardware implementation, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or may be implemented as hardware, or may be implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable storage medium, which may include a computer-readable storage medium (or a non-transitory medium) and a communication medium (or a temporary medium).

[0102] As is known to those of ordinary skill in the art, the term computer-readable storage media includes volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules or other data). Computer-readable storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disk (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is known to those of ordinary skill in the art that communication media typically contain computer-readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transmission mechanism, and may include any information delivery medium.

[0103] Exemplarily, the computer-readable storage medium may be an internal storage unit of the electronic device of the aforementioned embodiment, such as a hard disk or memory of the electronic device. The computer-readable storage medium may also be an external storage device of the electronic device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device.

[0104] The above is only a specific implementation of the present application, so that those skilled in the art can understand or implement the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest range consistent with the principles and novel features applied for herein.

Claims

1. A fault diagnosis method based on Petri net, characterized in that: The steps include: S1. Obtaining a timing signal of a substation device, wherein the timing signal includes alarm information, accident and displacement information; S2. Based on the timing signal of the substation equipment, a fault discrimination rule base is constructed and antecedent weights, rule activation thresholds and consequent credibility are set; S3, establishing a fuzzy function based on the antecedent weight, the rule activation threshold and the consequent credibility; S4. Based on the Petri net and the fuzzy function, a network topology model of the substation equipment is established, wherein the network topology model of the substation equipment includes a fault intelligent judgment rule base; S5. Based on the network topology model of the substation equipment, obtain the current timing signal of the substation equipment, and determine the fault condition of the substation equipment.

2. The Petri net-based fault diagnosis method according to claim 1, characterized in that: The antecedent weight, rule activation threshold and consequent credibility are all determined through technical specifications of the substation equipment and fault case analysis.

3. The Petri net-based fault diagnosis method according to claim 1, characterized in that: S3, after the step of establishing a fuzzy function based on the antecedent weight, the rule activation threshold and the consequent credibility, the Petri net-based fault diagnosis method further includes: Determine an event token value of the substation equipment based on a timing signal of the substation equipment, wherein the event token value represents an abnormal condition of the substation equipment; Determining a failure risk of the substation equipment based on an event token value of the substation equipment; If the event token value is greater than a preset rule activation threshold, a fuzzy value is calculated based on the consequent credibility, and the event token value is adjusted.

4. The Petri net-based fault diagnosis method according to claim 3, characterized in that: The event token value is greater than 0 and less than 1.

5. The Petri net-based fault diagnosis method according to claim 3, characterized in that: The step of adjusting the event token value comprises: comparing the event token value with the rule activation threshold; If the event token value is greater than the rule activation threshold, the difference between the event token value and the rule activation threshold is used as the adjusted event token value, and the product of the adjusted event token value and the consequent credibility is used as the fuzzy value.

6. The Petri net-based fault diagnosis method according to claim 5, characterized in that: The fuzzy function includes an input intensity function S(i, α), an output intensity function R(o, β), a state transfer control function F(t), a rated input quantity i and a rated output quantity o, and satisfies the relationship: S n =i n α n ; R n =o n β n ; i=M(p); o=F, where α n is the directed arc weight, β n is the credibility of the subsequent event, i, F, o are the token values ​​before, during and after the transition adjustment respectively, n is the serial number of the substation equipment, t is the time number of the substation equipment, and M(p) is the mark value of the event.

7. The Petri net-based fault diagnosis method according to claim 6, characterized in that: After the step of obtaining the current timing signal of the substation equipment based on the network topology model of the substation equipment and determining the fault condition of the substation equipment in step S5, the fault diagnosis method based on Petri nets further includes: If the substation equipment fails, the fault handling method is determined according to the TR matrix, and the TR matrix satisfies the following relationship: Where K is the number of preset conditional propositions, 1 represents that the conditional proposition is true, 0 represents that the conditional proposition is false, -1 represents that the conditional proposition is in an uncertain state, R(i, j) is the value of each item in the TR matrix, P i , P j is a conditional proposition, i, j = 1, 2,…, K.

8. The Petri net-based fault diagnosis method according to claim 7, characterized in that: The Petri net-based fault diagnosis method also includes: A preset initial truth value vector is determined, and the initial truth value vector is operated with the TR matrix to determine an inference result that the substation equipment has a fault.

9. A fault diagnosis system based on Petri net, characterized in that: The system comprises: An acquisition module, used to acquire a timing signal of the substation equipment, wherein the timing signal includes alarm information, accident and displacement information; A fault discrimination rule module is used to construct a fault discrimination rule base and set antecedent weights, rule activation thresholds and consequent credibility based on the timing signals of the substation equipment; A model building module, used to build a network topology model of the substation equipment based on the Petri net and the antecedent weight, the rule activation threshold and the consequent credibility, wherein the network topology model of the substation equipment includes a fault intelligent judgment rule base; The fault judgment module is used to obtain the current timing signal of the substation equipment based on the network topology model of the substation equipment and judge the fault condition of the substation equipment.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the Petri net-based fault diagnosis method according to any one of claims 1 to 8 is implemented.

Images