Intelligent fault diagnosis method and system for circuit breaker motor driving mechanism

Abstract

The invention discloses an intelligent fault diagnosis method and system for a motor driving mechanism of a circuit breaker, and the method comprises the steps: collecting parameter data of the motor driving mechanism in the circuit breaker, and building a fuzzy rule between fault features such as parameter changes of the motor driving mechanism and a fault mode, fuzzy logic such as fuzzy AND and fuzzy OR is used for fuzzy rule reasoning to obtain the membership degree after operation, and finally fuzzy synthesis operation is used for processing all fuzzy rule operation results contained in a certain fault mode to achieve comprehensive decision making, and the membership degree of fault features to a main fault mode is obtained. And meanwhile, the membership degree of the fault mode subjected to potential influence can be obtained. In order to overcome the problem that the accuracy of fuzzy rule reasoning fault diagnosis is limited by personnel experience and level, the method proposes to construct a fuzzy neural network consistent with fuzzy rule reasoning to intelligently diagnose the fault of the motor driving mechanism.

Description

Technical Field

[0001] The present invention relates to the technical field of high-voltage switchgear, and in particular to an intelligent fault diagnosis method and system for a circuit breaker motor drive mechanism. Background Art

[0002] Failure of the motor drive mechanism of a high-voltage circuit breaker will not only affect the accurate opening and closing of the moving contact of the high-voltage circuit breaker, but also may cause the high-voltage circuit breaker to trip. The failure of the motor drive mechanism of a high-voltage circuit breaker is likely not to occur in isolation. The occurrence of a certain fault event will have a potential impact on the operation of other components of the system. For example, faults such as motor bearing wear and shaft misalignment may cause body vibration. At this time, it is necessary to determine whether the component itself has failed or is affected by other fault modes. Therefore, it is necessary to find an intelligent fault diagnosis method suitable for the motor drive mechanism of a high-voltage circuit breaker, which can not only be used to describe the abnormality of fault parameters and the cause of component failure, but also to describe the potential impact of a certain component abnormality on other components of the system. The fault diagnosis methods based on analytical models such as state estimation, residual generation, and parameter identification can detect and identify the signs of fault occurrence, and then use the classic "If-Then" reasoning rules to diagnose the fault phenomenon corresponding to the fault signs. The classic "If-Then" reasoning rules usually only use two values ​​1 ("true") and 0 ("false"), and if the antecedent of the rule is "true", then the latter is also "true". In order to correspond the existing fault parameters to a more accurate description of the fault severity, it is necessary to propose a more intelligent fault diagnosis method to further process the detected fault feature data. In order to describe the essence of things more objectively, fuzzy rule reasoning is introduced into fault diagnosis and other reasoning. Using fuzzy rule reasoning, reasoning can be carried out according to the natural thinking process of human beings, and all real numbers between 0 and 1 are used as logical values ​​to describe the uncertainty of the fault boundary. Fuzzy rule reasoning can simulate the vague or ambiguous language that may be used by experts in decision-making. The rule generally judges the severity of the fault (for example, divided into: no fault, initial fault, slight, serious, severe, etc.) based on the quantified fault signs (for example, divided into: small, medium, large, etc.). Using fuzzy variables to deal with the uncertainty problem in the system simulates the reasoning process of humans in solving similar problems, reducing the difficulty of fault severity analysis for this large and complex hydraulic system. Considering the complexity and uncertainty of fault propagation between various components or subsystems in the high-voltage circuit breaker motor drive mechanism system, it is not easy to accurately infer the fault status of each component of the high-voltage circuit breaker motor drive mechanism system based on the fault data obtained by the sensor. The present invention proposes an intelligent fault diagnosis method for a circuit breaker motor drive mechanism, which utilizes the changes in fault characteristic parameters, as well as expert advice, existing empirical data and other knowledge to infer the fault phenomenon corresponding to the quantified fault characteristic parameters; uses fuzzy logic values ​​to accurately describe the degree of fault membership, and can provide fault diagnosis results within a larger range between 0 and 1, objectively describing the potential impact of the occurrence of a certain fault in the high-voltage circuit breaker motor drive mechanism on other fault modes. Summary of the invention

[0003] In view of the above-mentioned problems, the present invention is proposed.

[0004] Therefore, the present invention provides an intelligent fault diagnosis method for a circuit breaker motor drive mechanism, which uses fuzzy synthesis operations to process all fuzzy rule operation results contained in a certain fault mode to achieve comprehensive decision-making, obtain the membership of the fault characteristics to the main fault mode, and also obtain the membership of the potentially affected fault mode.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: an intelligent fault diagnosis method for a circuit breaker motor drive mechanism, comprising:

[0006] Collecting first object parameter data and establishing a fault identification rule for a change in the first object parameter;

[0007] Utilize the identification logic in the fault identification rules to calculate the rule measurement indicators;

[0008] By synthesizing the rule measurement indicators contained in the fault mode, comprehensive decision-making is achieved and fault diagnosis is completed.

[0009] As a preferred solution of the intelligent fault diagnosis method for the circuit breaker motor drive mechanism of the present invention, wherein: the fault identification rule includes selecting parameters containing fault characteristics in the first object and preliminarily establishing a relationship between the selected parameters and the partial fault;

[0010] After determining the fault mode and fault characteristics, the fuzzy relationship is used to describe the fault characteristic variables for each fault mode diagnosis, forming a fuzzy rule reasoning method for all fault modes.

[0011] As a preferred solution of the intelligent fault diagnosis method for the circuit breaker motor drive mechanism described in the present invention, wherein: the synthesis of the rule measurement indicators contained in the fault mode includes selecting a minimum operation, a maximum operation, a product operation, a probability or operation, and a fuzzy synthesis operation formed by a combination of the four operations according to the fault mode to form a comprehensive decision-making algorithm based on fuzzy rules.

[0012] As a preferred solution of the intelligent fault diagnosis method for a circuit breaker motor drive mechanism described in the present invention, the fuzzy synthesis operation also includes introducing a corresponding weight vector into the fuzzy synthesis operation to correct the fault diagnosis result.

[0013] As a preferred solution of the intelligent fault diagnosis method for a circuit breaker motor drive mechanism described in the present invention, the weight vector includes constructing a fuzzy neural network consistent with fuzzy rule reasoning to train the weights.

[0014] As a preferred solution of the intelligent fault diagnosis method for a circuit breaker motor drive mechanism of the present invention, wherein: the first object includes but is not limited to a motor drive mechanism in a circuit breaker;

[0015] The parameter data includes, but is not limited to, obtaining, through sensors, the supply voltage and current of the circuit breaker motor drive mechanism, as well as the motor speed, winding inductance and resistance, inertia and damping of the motor drive mechanism as fault characteristics;

[0016] The calculation of rule measurement indicators by using the identification logic in the fault identification rule includes, but is not limited to, using fuzzy logic of fuzzy AND and fuzzy OR to perform fuzzy rule reasoning to obtain the membership degree after operation.

[0017] As a preferred solution of the intelligent fault diagnosis method for the circuit breaker motor drive mechanism described in the present invention, wherein: the relationship between the selected parameters and partial faults is initially established, including transmission mechanism wear, motor winding phase-to-phase short circuit, motor winding turn-to-turn short circuit, transmission mechanism blockage, abnormal data measured by the current sensor, abnormal data measured by the voltage sensor, and abnormal data measured by the speed sensor.

[0018] An intelligent fault diagnosis system for a circuit breaker motor drive mechanism using the method of the present invention is characterized in that:

[0019] A data collection module, a data storage module, a calculation module and an observer module;

[0020] The data collection module is used to collect the power supply voltage, power supply current and motor speed of the motor drive mechanism in the circuit breaker;

[0021] The data storage module is used to store and upload information during the control process;

[0022] The calculation module is used to process the calculation corresponding to the parameters;

[0023] The observer module is used to carry the state observer for actual use.

[0024] A computer device comprises: a memory and a processor; the memory stores a computer program, wherein: the processor implements the steps of any one of the methods of the present invention when executing the computer program.

[0025] A computer-readable storage medium stores a computer program, wherein: when the computer program is executed by a processor, the steps of any one of the methods of the present invention are implemented.

[0026] Beneficial effects of the present invention: The intelligent fault diagnosis method for the circuit breaker motor drive mechanism provided by the present invention, in view of the complexity and uncertainty of fault propagation between various components or subsystems in the circuit breaker motor drive mechanism, uses fuzzy rules to establish a fuzzy relationship between fault characteristics and one or more fault modes to deal with this uncertainty, and finally obtains the degree to which a group of fault characteristics belong to the main fault mode and the degree to which they belong to other affected fault modes through fuzzy synthesis operation. In view of the problem that the diagnosis results obtained by using only fuzzy rule reasoning are sometimes not satisfactory, the diagnosis results are corrected by introducing weights in the fuzzy synthesis operation, and a fuzzy neural network consistent with fuzzy rule reasoning is constructed to train the weights, thereby overcoming the problem that manually setting weights is limited by experience and level. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0028] Figure 1 An overall flow chart of an intelligent fault diagnosis method for a circuit breaker motor drive mechanism provided in the first embodiment of the present invention.

[0029] Figure 2 A fault diagnosis principle framework diagram based on fuzzy rule reasoning for an intelligent fault diagnosis method for a circuit breaker motor drive mechanism provided in the second embodiment of the present invention.

[0030] Figure 3 A fuzzy neural network structure diagram of an intelligent fault diagnosis method for a circuit breaker motor drive mechanism provided in accordance with the second embodiment of the present invention. DETAILED DESCRIPTION

[0031] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0032] Example 1, reference Figure 1 , as an embodiment of the present invention, provides an intelligent fault diagnosis method for a circuit breaker motor drive mechanism, comprising:

[0033] S1: Collecting parameter data of a first object and establishing a fault identification rule for a change in the parameter of the first object.

[0034] Furthermore, in the present application, the first object is the motor drive mechanism in the circuit breaker. Specifically, the parameter data of the motor drive mechanism in the circuit breaker is collected, and fuzzy rules between fault characteristics such as changes in motor drive mechanism parameters and fault modes are established; the parameter data of the circuit breaker motor drive mechanism includes the power supply voltage and power supply current of the circuit breaker motor drive mechanism, as well as the motor speed, winding inductance and resistance, inertia and damping of the motor drive mechanism, obtained through sensors. These parameters are selected as fault characteristics.

[0035] It is important to know that when the supply voltage and current change, it may indicate the load or fault of the motor; the motor speed can reflect its working efficiency; the winding inductance and resistance help analyze the electrical characteristics of the motor; and inertia and damping affect the mechanical response. Through these data, we can deeply explore the relationship between fault characteristics and fault modes, establish more effective fuzzy rules, and ultimately improve the reliability and safety of circuit breakers.

[0036] Using the selected parameters containing fault characteristics, a preliminary relationship is established with the following faults, including: transmission mechanism wear, motor winding phase short circuit, motor winding turn-to-turn short circuit, transmission mechanism blockage, abnormal data measured by the current sensor, abnormal data measured by the voltage sensor, and abnormal data measured by the speed sensor.

[0037] It is important to know that these types of faults are common in motor drive mechanisms and occur frequently, so they require priority attention to ensure the reliability and safety of the system. Each fault will produce obvious characteristics on specific parameters. For example:

[0038] Wear of transmission mechanism: will cause the motor speed to decrease and the current to increase.

[0039] Motor winding interphase short circuit: will cause abnormal increase in current and may also cause voltage drop.

[0040] Short circuit between motor winding turns: It will cause local heating and abnormal current waveform.

[0041] Transmission mechanism blockage: will cause the motor speed to drop significantly and the current to increase.

[0042] By identifying these characteristics, early fault warning can be achieved, maintenance measures can be taken in time, and potential losses can be reduced. Building these relationships can provide a basis for subsequent data analysis and model building, and enhance the accuracy and effectiveness of fault diagnosis.

[0043] After determining the fault mode and fault characteristics, the fuzzy relationship is used to describe the fault characteristic variables that can diagnose each fault mode, forming a fuzzy rule reasoning method for all fault modes. The relationship between fault characteristics, fault modes and fuzzy rules is clearly described.

[0044] Among them, fault characteristics refer to parameter data monitored by sensors in the motor drive mechanism. These characteristics can reflect the operating status of the system and show specific changes when a fault occurs. Fault mode refers to a specific type of fault that occurs in the system due to certain reasons. For example, wear of the transmission mechanism, short circuit of the motor winding, etc. Each fault mode corresponds to a specific fault characteristic. Fuzzy rules are inference rules established based on the logical relationship between fault characteristics and fault modes. These rules use fuzzy logic methods to associate fuzzy inputs (fault characteristics) with clear outputs (fault modes) to achieve fault diagnosis.

[0045] The following example describes the relationship between fault characteristics, fault modes and fuzzy rules:

[0046] Each failure mode corresponds to a specific failure characteristic. For example:

[0047] Transmission mechanism wear → current increases and speed decreases.

[0048] The motor winding phase is short-circuited → the current increases abnormally and the voltage drops.

[0049] Establish fuzzy rules based on the changes in fault characteristics. For example:

[0050] If the current is "high" and the speed is "low", it may be "gear mechanism wear".

[0051] If the current is "very high" and the voltage is "low", it may be "a short circuit between phases of the motor winding".

[0052] Example:

[0053] Fault symptoms:

[0054] Current (I): High, Medium, Low.

[0055] Speed ​​(N): High, Medium, Low.

[0056] Voltage(V): High, Medium, Low.

[0057] Failure Mode:

[0058] A: The transmission mechanism is worn.

[0059] B: The motor winding phases are short-circuited.

[0060] Fuzzy rules:

[0061] If I is "high" and N is "low", the failure mode is A (transmission wear).

[0062] If I is "very high" and V is "low", the fault mode is B (motor winding phase short circuit).

[0063] In this way, the fuzzy logic reasoning method can be used to effectively diagnose the system fault. This method can handle uncertainty and fuzziness and improve the accuracy and reliability of fault diagnosis.

[0064] In an optional embodiment, the first object may also be a "motor drive system". Such a system is commonly used in industrial automation and mechanical equipment. The key parameters of the motor drive system, such as the power supply voltage, power supply current, and speed, are acquired through sensors to establish real-time data monitoring. Fuzzy logic is used to analyze the collected parameters to identify potential fault characteristics, such as motor overload, overheating, or transmission system jamming. Fuzzy rules for the fault modes related to the motor drive system are formed, and fuzzy logic is used for reasoning to determine the current state of the system and the risk of failure. According to the reasoning results, the power supply parameters of the motor are automatically adjusted.

[0065] S2: Calculate the rule measurement index by using the identification logic in the fault identification rule.

[0066] Specifically, fuzzy logic such as fuzzy AND and fuzzy OR is used to perform fuzzy rule reasoning to obtain the membership after operation. Among them, the rule measurement index in the present invention is the membership. In many actual fault diagnosis scenarios, the system state and characteristic values ​​are often not absolute, but fuzzy. Fuzzy logic allows us to reason under uncertain and incomplete information, thereby providing more flexible decision support. Through fuzzy AND and fuzzy OR operations, the membership of multiple fault features can be combined to obtain a comprehensive evaluation of a certain fault mode. This comprehensive evaluation can more comprehensively reflect the system state and avoid misjudgment that may be caused by a single feature. Through the membership after operation, the applicability of each fuzzy rule can be quantified. This helps to identify which fault modes are most consistent with the actual measured characteristic values, thereby providing a basis for fault diagnosis. The calculation result of the membership can be used as the basis for fault decision-making. For example, a threshold can be set, and when the membership of a certain fault mode exceeds the threshold, the fault mode can be determined to be a major fault. This method can improve the accuracy of diagnosis. By continuously acquiring new data and updating the membership, the judgment of the fault mode can be dynamically adjusted. This enables the system to maintain a high level of fault identification capability under different operating conditions.

[0067] S3: By synthesizing the rule measurement indicators contained in the fault mode, comprehensive decision-making is achieved and fault diagnosis is completed.

[0068] Specifically, the corresponding weight vector is introduced into the fuzzy synthesis operation to correct the fault diagnosis result, and the minimum value (min) operation, maximum value (max) operation, product (prod) operation, probability or (probor) operation, and the fuzzy synthesis operation formed by their combination are selected according to the fault mode to form a comprehensive decision-making algorithm based on fuzzy rules.

[0069] In the embodiment of the present application, a fuzzy neural network (FNN) consistent with fuzzy rule reasoning is constructed to train weights, and the training algorithm steps of the FNN network are as follows:

[0070] 1: Randomly select samples from the training samples and input them into the FNN network to calculate the actual output.

[0071] 2: Combine the actual output and calculate the error with the expected output.

[0072] 3: Adjust the network weights and thresholds along the negative gradient direction.

[0073] 4: Repeat steps 1 to 3 until the target error is reached and the iteration ends.

[0074] In an optional embodiment, principal component analysis can be used to reduce the dimension of the fault mode data and extract the main features, thereby calculating the importance of each feature to fault diagnosis and assigning weights accordingly. Alternatively, a linear regression model can be used to fit the relationship between the fault feature and the fault mode by the least squares method to obtain the regression coefficient of the feature, which is then used as the weight. Alternatively, a genetic algorithm can be used to optimize the weight, and the accuracy of fault diagnosis can be evaluated by the fitness function to find the optimal weight combination.

[0075] Example 2, reference Figure 2 , 3 , which is the first embodiment of the present invention, provides an intelligent fault diagnosis method for a circuit breaker motor drive mechanism, comprising:

[0076] Assume that F is the set of all fault modes of the object to be diagnosed. The fault modes in F can be divided into m categories, namely F1, F2, …, F m , each fuzzy subset F i The specific definition of (i=1,…,m) is expressed as:

[0077]

[0078] Where f i represents the i-th fuzzy subset F i The corresponding fuzzy variables.

[0079] Assume that a set S is used to describe all the fault characteristics of the object to be diagnosed. These fault characteristics are divided into n categories, denoted as S1, S2, L, S n , each fuzzy subset S i The specific definition of (i=1,…,n) is expressed as:

[0080]

[0081] Where s i is the i-th fuzzy subset S i The corresponding fuzzy variables.

[0082] In the verification system, the circuit breaker motor drive mechanism supply voltage U can be obtained through the sensor d , supply current I0, and motor speed ω m and winding inductance L a and resistor R a , Inertia of the motor drive mechanism J m and damping B m . These parameters are selected as fault characteristics. Note:

[0083] S=(B m ,J m ,R a ,L a ,I0,U d ,ω m ) T

[0084] According to the fault characteristics S = (B m ,J m ,R a ,L a ,I0,U d ,ω m ) T The fault modes that can be diagnosed are:

[0085] F=(F1,F2,F3,F4,F5,F6,F7) T

[0086] The description of each failure mode is shown in Table 1 below.

[0087] Table 1 Fault mode and fault characteristic correspondence table

[0088]

[0089]

[0090] Note: In the table, “+” indicates that the fault characteristics increase; “-” indicates that the fault characteristics decrease; “±” indicates that the fault characteristics increase or decrease.

[0091] Assume that based on the fault characteristic variables s1, s2, L, s n Each failure mode can be diagnosed i (i=1,L,m). If there is a fuzzy rule reasoning method to diagnose these fault modes, then each fault mode f1,f2,L,f m The number of fuzzy rules is q1, q2, L, q m The i-th failure mode f i The corresponding fuzzy rules can be described as:

[0092] R ij :IF<…>THEN<…>,j=1,…,q i

[0093] In order to more clearly describe the fault characteristics s1, s2, L, s n 、Failure mode f1,f2,L,f m and fuzzy rule R ij The relationship between the three. The fault diagnosis principle based on fuzzy rule reasoning can be used Figure 1 Description. Theoretically, each fault feature s1,s2,…,s n can be used to diagnose a certain fault mode; each fault mode f1, f2, ..., f m The number of fuzzy rules required may be different. After determining the fault mode and fault characteristics, the problem to be solved is: using fuzzy rule reasoning to obtain a fault diagnosis result close to the actual situation.

[0094] The comprehensive decision-making method based on fuzzy rules is given below.

[0095] In the fuzzy rule comprehensive decision-making process, fault features s1, s2, L, s n As the fuzzy input vector, it is recorded as:

[0096] S=[s1,s2,L,s n ] T

[0097] Failure mode f1,f2,L,f m As the fuzzy output vector, it is recorded as:

[0098] F=[f1,f2,L,f m ] T

[0099] The i-th failure mode f i Corresponding fuzzy rule R ij The result of the operation (j=1,L,q1) is:

[0100]

[0101] In the formula, r ij is the fuzzy rule R ij The membership degree after operation.

[0102] Failure mode f1,f2,L,f m The calculation results of the corresponding fuzzy rules are:

[0103]

[0104] In engineering practice, only Find f i , the results obtained are sometimes not satisfactory. An effective method is to introduce the R i The corresponding weight vector W i To correct the diagnosis result i , weight vector W i Expressed as:

[0105]

[0106] In the formula, w ij For ij The corresponding weights satisfy

[0107] The result of fuzzy rule operation R i and weight vector W i , the fuzzy output f can be obtained through fuzzy synthesis operation i ,Right now:

[0108] f i =R i oW i ,i=1,L,m

[0109] The symbol "o" in the formula represents a fuzzy synthesis operation, which should be selected and designed according to actual needs. The fuzzy synthesis operation "o" used in the present application includes: minimum value (min) operation, maximum value (max) operation, product (prod) operation, probability or (probor) operation, and the fuzzy synthesis operation formed by their combination, namely:

[0110]

[0111] Set the weight w according to the actual situation and prior knowledge ij After that, any of the above fuzzy synthesis operations can be used to obtain the fault mode f i The diagnostic result of (i=1,L,m). The schematic diagram of this operation is as follows Figure 2 shown.

[0112] In order to solve the problem that the diagnosis results obtained by using fuzzy rule reasoning alone are sometimes not satisfactory, weights are introduced in the fuzzy synthesis operation to correct the diagnosis results, and a fuzzy neural network (FNN) consistent with fuzzy rule reasoning is constructed to train the weights, overcoming the problem that manual setting of weights is limited by experience and level.

[0113] like Figure 3 As shown in Figure 1, the designed FNN consists of four layers: input layer, fuzzy language layer, fuzzy rule layer and output layer. The input layer inputs the fault features s1, s2, L, s n The clear value of the fuzzy language layer is the fault feature s1, s2, L, s n Output after fuzzification; the fuzzy rule layer is the calculation result R describing the fuzzy rules ij (j=1,L,q1); the operation result of the output layer R ij (j=1,L,q1), that is, each failure mode f i The diagnostic result of (i=1,L,m).

[0114] The FNN input layer to the fuzzy language layer is used to calculate the fault features s1, s2, L, s n After fuzzification, the membership of each fuzzy rule is obtained by fuzzy rule reasoning. This process does not consider weight adjustment. The output layer of FNN is f i (i=1,L,m), the i-th output f of the output layer i The output of the corresponding fuzzy rule layer (hidden layer) is R ij (j=1,L,q1), the weight of the network line connecting the fuzzy rule layer and the output layer is W ij (j=1,L,q1), if the fuzzy synthesis operation is realized, then the weight of the network connecting the fuzzy rule layer and the output layer is exactly the same as R in the fuzzy synthesis operation. ij The weights corresponding to (j=1,L,q1) are consistent, that is, they all use W ij (j=1,L,q1) represents.

[0115] Assume that a sample pair (S, F) is S = [s1, s2, ..., s n ] T ,Z=[z1,z2,…,z m ] T Among them, Z can be regarded as F=[f1,f2,…,f m ] T The expected output. If the threshold of the output layer neuron is θ=[θ1,θ2,…,θ m ] T .

[0116] Then the i-th output f of the output layer iThe output of the corresponding fuzzy rule layer neuron is:

[0117]

[0118] The output of the output layer neurons is:

[0119]

[0120] In the formula, g(·) is the transfer function of the output layer.

[0121] The i-th output f of the output layer i and the expected output z i The error is:

[0122]

[0123] Error E i The weight w between the neurons in the fuzzy rule layer and the output layer ij The partial derivative of is:

[0124]

[0125] In the formula, δ i =(f i -z i )g′(net i ).

[0126] From the above, we can get the weight w ij The adjustment formula is:

[0127]

[0128] Where η is the learning step size.

[0129] Similarly, the error E i The threshold θ for the output layer neurons i The partial derivative of is:

[0130]

[0131] From the above, we can get the threshold value θ i The adjustment formula is:

[0132]

[0133] The learning algorithm steps of the FNN network are as follows:

[0134] Step 1: Randomly select samples from the training samples and input them into the FNN network to calculate the actual output;

[0135] Step 2: Combine the actual output and calculate the error with the expected output;

[0136] Step 3: Adjust the network weights and thresholds along the negative gradient direction;

[0137] Step4: Repeat steps Step to Step3 until the target error is reached and the iteration ends.

[0138] In addition, although the neural network is used to train the weights w ij ,However, the method proposed in this paper mainly relies on fuzzy rule reasoning for fault ,decision-making, therefore, the decision-making ability of the method proposed in this ,paper is mainly realized by setting reasonable fuzzy rules.

[0139] Embodiment 3 is the first embodiment of the present invention, which provides an intelligent fault diagnosis system for a circuit breaker motor drive mechanism, comprising:

[0140] Data collection module, data storage module, calculation module and observer module; the data collection module is used to collect the power supply voltage, power supply current and motor speed of the motor drive mechanism in the circuit breaker; the data storage module is used to store and upload the information in the control process; the calculation module is used to process the calculation corresponding to the parameters; the observer module is used to carry the state observer for actual use.

[0141] If the above functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0142] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0143] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0144] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0145] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. An intelligent fault diagnosis method for a circuit breaker motor drive mechanism, characterized in that: include: Collecting first object parameter data and establishing a fault identification rule for a change in the first object parameter; Utilize the identification logic in the fault identification rules to calculate the rule measurement indicators; By synthesizing the rule measurement indicators contained in the fault mode, comprehensive decision-making is achieved and fault diagnosis is completed.

2. The intelligent fault diagnosis method for a circuit breaker motor drive mechanism according to claim 1, characterized in that: The fault identification rule includes selecting parameters containing fault characteristics in the first object and preliminarily establishing a relationship between the selected parameters and the partial fault; After determining the fault mode and fault characteristics, the fuzzy relationship is used to describe the fault characteristic variables for each fault mode diagnosis, forming a fuzzy rule reasoning method for all fault modes.

3. The intelligent fault diagnosis method for a circuit breaker motor drive mechanism according to claim 2, characterized in that: The synthesis of the rule measurement indicators contained in the fault mode includes selecting a minimum operation, a maximum operation, a product operation, a probability or operation, and a fuzzy synthesis operation formed by a combination of the four operations according to the fault mode to form a comprehensive decision-making algorithm based on fuzzy rules.

4. The intelligent fault diagnosis method for a circuit breaker motor drive mechanism according to claim 3, characterized in that: The fuzzy synthesis operation also includes introducing a corresponding weight vector into the fuzzy synthesis operation to correct the fault diagnosis result.

5. The intelligent fault diagnosis method for a circuit breaker motor drive mechanism according to claim 4, characterized in that: The weight vector includes constructing a fuzzy neural network consistent with fuzzy rule reasoning to train the weights.

6. The intelligent fault diagnosis method for a circuit breaker motor drive mechanism according to claim 1, characterized in that: The first object includes, but is not limited to, a motor drive mechanism in a circuit breaker; The parameter data includes, but is not limited to, obtaining, through sensors, the supply voltage and current of the circuit breaker motor drive mechanism, as well as the motor speed, winding inductance and resistance, inertia and damping of the motor drive mechanism as fault characteristics; The calculation of rule measurement indicators by using the identification logic in the fault identification rule includes, but is not limited to, using fuzzy logic of fuzzy AND and fuzzy OR to perform fuzzy rule reasoning to obtain the membership degree after operation.

7. The intelligent fault diagnosis method for a circuit breaker motor drive mechanism according to claim 6, characterized in that: The relationship between the selected parameters and some faults is initially established, including transmission mechanism wear, motor winding phase-to-phase short circuit, motor winding turn-to-turn short circuit, transmission mechanism blockage, abnormal data measured by the current sensor, abnormal data measured by the voltage sensor, and abnormal data measured by the speed sensor.

8. An intelligent fault diagnosis system for a circuit breaker motor drive mechanism using the method according to any one of claims 1 to 7, characterized in that: A data collection module, a data storage module, a calculation module and an observer module; The data collection module is used to collect the power supply voltage, power supply current and motor speed of the motor drive mechanism in the circuit breaker; The data storage module is used to store and upload information during the control process; The calculation module is used to process the calculation corresponding to the parameters; The observer module is used to carry the state observer for actual use.

9. A computer device comprising: A memory and a processor; the memory stores a computer program, wherein the processor implements the steps of any method as claimed in claim 1 when executing the computer program.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images