Method and system for predicting reliable operation time of relay protection device and electronic equipment

Abstract

The invention relates to a relay protection device reliable operation time prediction method and system and electronic equipment, and the method comprises the steps: obtaining fault statistical data, carrying out the training of a Transform model according to the fault statistical data, and obtaining a fault rate prediction model, the fault statistical data comprises environment temperature, operation time of the relay protection device and average fault rate; obtaining expansion data according to the obtained random operation time, the random environment temperature and the fault rate prediction model; determining a reliability model and a joint failure rate model according to an Arrhenius model and Weibull distribution, fitting the joint failure rate model based on the fault statistical data and the expansion data, and determining parameters of the joint failure rate model; and determining the reliable operation time of the relay protection device through the reliability model based on the parameters, the target environment temperature and a preset reliability threshold.

Description

Technical Field

[0001] The present application relates to the field of relay protection device operation time prediction, and in particular to a method, system and electronic equipment for predicting the reliable operation time of a relay protection device. Background Art

[0002] At present, the scale of the power system is constantly expanding, and the structural complexity is increasing. Under actual working conditions, environmental factors such as temperature, humidity, vibration, and dust will affect the reliability of the equipment. With the increase of operating years, the failure rate of a large number of electronic components in the device will gradually increase, thereby consuming the remaining life of the device. As the first line of defense of the power system, the high reliability of relay protection itself is the basic guarantee for the normal operation of the power system. Accurate and rapid evaluation of the reliability of relay protection devices is of great significance to the safe and stable operation of substations, and it also helps to improve the utilization rate of the device and reduce cost losses.

[0003] Existing device life assessment models usually require a large amount of failure data to train the model. However, high-reliability relay protection devices and systems have less operating failure data, which is insufficient to meet the data requirements of the model.

[0004] The existing technology is difficult to achieve accurate prediction of the reliable operation time of relay protection devices under small sample fault data. Summary of the invention

[0005] The embodiments of the present application provide a method, system and electronic device for predicting the reliable operation time of a relay protection device, so as to at least solve the problem in the related art that it is difficult to accurately predict the reliable operation time of a relay protection device under small sample fault data.

[0006] In a first aspect, an embodiment of the present application provides a method for predicting the reliable operation time of a relay protection device, comprising:

[0007] Obtaining fault statistical data, and training a Transformer model according to the fault statistical data to obtain a fault rate prediction model, wherein the fault statistical data includes ambient temperature, operating time of the relay protection device, and average failure rate;

[0008] Obtaining expanded data according to the acquired random operating time, random ambient temperature and the failure rate prediction model;

[0009] Determine a reliability model and a joint failure rate model according to the Arrhenius model and the Weibull distribution, fit the joint failure rate model based on the fault statistical data and the expanded data, and determine parameters of the joint failure rate model;

[0010] Based on the parameters, the target ambient temperature and a preset reliability threshold, the reliable operation time of the relay protection device is determined by the reliability model.

[0011] In one embodiment, determining the reliability model and the joint failure rate model according to the Arrhenius model and the Weibull distribution includes:

[0012] Determine a distribution failure function according to the Weibull distribution, determine a reliability function according to the distribution failure function, and determine a failure rate function according to the distribution failure function and the reliability function;

[0013] Determining the correlation between the scale parameters in the reliability function and the failure rate function and the ambient temperature according to the Arrhenius model;

[0014] A joint failure rate model is determined according to the failure rate function and the association relationship, and a reliability model is determined according to the association relationship and the reliability function.

[0015] In one embodiment, determining a joint failure rate model according to the Arrhenius model and the Weibull distribution includes:

[0016] The distribution failure function is determined according to the Weibull distribution. The specific formula of the distribution failure function is:

[0017]

[0018] Where F(t) represents the distribution failure function, α represents the scale parameter, β represents the shape parameter, e is the base of the natural logarithm, and t represents the running time.

[0019] The reliability function R(t) is expressed as:

[0020]

[0021] The failure rate function λ(t) is expressed as:

[0022]

[0023] The correlation between the scale parameter and the ambient temperature is determined according to the Arrhenius model, which is specifically expressed as follows:

[0024]

[0025] Wherein, α represents a scale parameter, A represents a proportional constant, E represents the product activation energy corresponding to the failure rate, K represents the Boltzmann constant, and T represents the ambient temperature;

[0026] A joint failure rate model is determined according to the failure rate function and the association relationship, and the joint failure rate model is expressed as:

[0027]

[0028] In one embodiment, fitting the joint failure rate model based on the fault statistical data and the expanded data to determine the parameters of the joint failure rate model includes:

[0029] Iterating the joint failure rate model based on the least square method according to the fault statistical data and the expanded data;

[0030] In response to the residual sum of squares between the predicted value of the joint failure rate model and the average failure rate being less than a preset threshold, the iteration is stopped, and the parameters of the joint failure rate model are determined, the parameters including the proportional constant and the shape parameter.

[0031] In one embodiment, the determining the reliable operation time of the relay protection device through the reliability model based on the parameter, the target ambient temperature and the preset reliability threshold value includes:

[0032] Determining a target reliability model according to the parameters and the reliability model;

[0033] Obtaining a reliability variation trend with operating time according to the target ambient temperature and the target reliability model;

[0034] The reliable operation time of the relay protection device is determined according to the change trend and the preset reliability threshold.

[0035] In one embodiment, the extended data is obtained according to the acquired random operating time, random ambient temperature and the failure rate prediction model, including:

[0036] According to the acquired random operating time and random ambient temperature, prediction is performed using the failure rate prediction model to obtain a corresponding predicted failure rate;

[0037] The random operating time, random ambient temperature and the predicted failure rate are used as extended data.

[0038] In one embodiment, the Transformer model includes: an input module, an encoding module, a decoding module and an output module.

[0039] The input module includes a data embedding layer and a position encoder, which is used to perform a word embedding operation on each data in the input sequence;

[0040] The encoding module and the decoding module include a self-attention module, which is used to map the input sequence into question-key-value and calculate an attention score.

[0041] In a second aspect, an embodiment of the present application provides a system for predicting the reliable operation time of a relay protection device, comprising:

[0042] Acquisition module: used to acquire fault statistical data, and train the Transformer model according to the fault statistical data to obtain a fault rate prediction model, wherein the fault statistical data includes ambient temperature, operating time of the relay protection device and average failure rate;

[0043] Data expansion module: used to obtain expanded data according to the acquired random operating time, random ambient temperature and the failure rate prediction model;

[0044] A fitting module: used for determining a reliability model and a joint failure rate model according to the Arrhenius model and the Weibull distribution, fitting the joint failure rate model based on the fault statistical data and the extended data, and determining parameters of the joint failure rate model;

[0045] Result module: used to determine the reliable operation time of the relay protection device through the reliability model based on the parameters, the target ambient temperature and the preset reliability threshold.

[0046] In a third aspect, an embodiment of the present application provides a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for predicting the reliable operation time of a relay protection device as described in the first aspect above is implemented.

[0047] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for predicting the reliable operation time of a relay protection device as described in the first aspect above.

[0048] The embodiments of the present application provide a method, system and electronic device for predicting the reliable operation time of a relay protection device, which have at least the following technical effects.

[0049] This application uses the Transformer model's ability to process small sample data, trains the Transformer model with small sample data, and performs data expansion through the trained model, which to a certain extent overcomes the problem of the subsequent prediction effect being affected by too few original data samples. A reliability model and a joint failure rate model are constructed based on the Arrhenius model and Weibull distribution, and the joint failure rate model is iteratively solved based on the original data and expanded data to improve the prediction accuracy of the prediction model. At the same time, the reliable operation time of the relay protection device products is evaluated, which is beneficial to the safe and stable operation of the substation. At the same time, it helps to improve the utilization rate of several protection devices and reduce cost losses.

[0050] Details of one or more embodiments of the present application are set forth in the following drawings and description to make other features, objects, and advantages of the present application more readily apparent. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0052] Figure 1 It is a flow chart of a method for predicting the reliable operation time of a relay protection device according to an embodiment of the present application;

[0053] Figure 2 is a schematic diagram showing a Transformer model structure according to an exemplary embodiment;

[0054] Figure 3 It is a structural block diagram of a system for predicting the reliable operation time of a relay protection device according to an embodiment of the present application;

[0055] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0056] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is described and illustrated below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. Based on the embodiments provided in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present application.

[0057] Obviously, the drawings described below are only some examples or embodiments of the present application. For ordinary technicians in this field, the present application can also be applied to other similar scenarios based on these drawings without creative work. In addition, it can also be understood that although the efforts made in this development process may be complicated and lengthy, for ordinary technicians in this field related to the content disclosed in this application, some changes in design, manufacturing or production based on the technical content disclosed in this application are just conventional technical means, and should not be understood as insufficient content disclosed in this application.

[0058] Reference to "embodiments" in this application means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those of ordinary skill in the art that the embodiments described in this application may be combined with other embodiments without conflict.

[0059] Unless otherwise defined, the technical terms or scientific terms involved in this application should be understood by people with ordinary skills in the technical field to which this application belongs. The words "one", "a", "a", "the" and the like involved in this application do not indicate a quantitative limitation, and may represent the singular or plural. The terms "include", "comprise", "have" and any of their variations involved in this application are intended to cover non-exclusive inclusions; for example, a process, method, system, product or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units that are not listed, or may also include other steps or units inherent to these processes, methods, products or devices. The words "connect", "connected", "coupled" and the like involved in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The "multiple" involved in this application refers to two or more. "And / or" describes the association relationship of associated objects, indicating that there may be three relationships, for example, "A and / or B" can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the objects before and after are in an "or" relationship. The terms "first", "second", "third", etc. involved in this application are only used to distinguish similar objects and do not represent a specific ordering of the objects.

[0060] In a first aspect, an embodiment of the present application provides a method for predicting the reliable operation time of a relay protection device. Figure 1 is a flow chart of a method for predicting the reliable operation time of a relay protection device according to an embodiment of the present application, such as Figure 1 As shown, the method includes:

[0061] Step S101, acquiring fault statistical data, and training a Transformer model according to the fault statistical data to obtain a fault rate prediction model, wherein the fault statistical data includes ambient temperature, operating time of the relay protection device, and average failure rate.

[0062] Optionally, the fault statistical data is the average failure rate of the relay protection devices in operation in the substation under different operating years and temperature conditions, and the fault statistical data is small sample data. The average failure rate is calculated based on the statistical average of the number of failures of each device per year.

[0063] The Transformer model is trained with the two-dimensional data of operating time and ambient temperature as input and the corresponding average failure rate as output. The specific training process includes: using the cross-validation method, dividing the training set and the test set according to the preset ratio, using the training set data to train the Transformer, and after the training, applying the Transformer model to the test set to verify the prediction accuracy until the test results reach the preset prediction accuracy.

[0064] In one example, Figure 2 is a schematic diagram showing the structure of a Transformer model according to an exemplary embodiment. Figure 2 As shown, the Transformer model in step S101 includes: an input module, an encoding module, a decoding module and an output module. The output module includes a linear layer and a softmax layer.

[0065] The input module includes a data embedding layer and a position encoder, which is used to perform word embedding operations on each data in the input sequence.

[0066] Optionally, a data input encoding strategy is used to implement word embedding operations, where each value in the input sequence is mapped to a 512-dimensional feature row vector, and then the input sequence is encoded through sine and cosine functions to generate a fixed absolute position representation, namely positional encoding (PE), which is then added to the previously completed word embedding sequence. The formula for positional encoding is as follows:

[0067]

[0068] Among them, pos represents the position index of the input sequence where the data of a certain time step is located, d model represents the dimension of the input sequence word embedding, and m represents the dimension of the vector.

[0069] The encoding module and the decoding module include a self-attention module, which is used to map the input sequence into question-key-value and calculate an attention score.

[0070] Alternatively, if Figure 2As shown in the figure, the encoding module and the decoding module also include a feedforward fully connected layer and a residual connection module. Among them, the self-attention module maps the input sequence to a query-key-value and calculates the attention score. It calculates the dot product of a question and all keys to obtain the weight, thereby learning the relative importance of each data and all other data in the sequence. The specific calculation formula is as follows:

[0071]

[0072] Among them, Q stands for question, K stands for key, V stands for value, and d model represents the word embedding dimension, Q represents the question matrix, K represents the key matrix, V represents the value matrix, and W Q , W K , W V Represents the weight matrix.

[0073] Furthermore, the encoding module and the decoding module include a multi-head self-attention layer. The multi-head attention mechanism splits a single attention into 8 groups and performs attention operations separately within the group. Finally, the output of each group is reassembled to the original size as the output of the multi-head attention layer.

[0074] In this way, by processing multiple attention heads in parallel through the multi-head attention layer, each head can learn different attention patterns and capture different relationships and features in the input data, so that the model can capture richer information and features. This is conducive to fully focusing on and utilizing the features in small sample data, reducing the model's demand for data volume, and improving the model's recognition and prediction accuracy.

[0075] Continue to refer to Figure 1 , after step S101, step S102 is performed, as follows.

[0076] Step S102, obtaining extended data according to the acquired random operating time, random ambient temperature and failure rate prediction model.

[0077] Optionally, through the trained failure rate prediction model, a set of random operating time and ambient temperature matrices within a limited time range are input to generate the average failure rate prediction results of the relay protection device under the operating time and ambient temperature, and obtain the expanded data of the relay protection device that conforms to the failure distribution model.

[0078] In one example, step S102 includes:

[0079] According to the acquired random operating time and random ambient temperature, the fault rate prediction model is used to predict and obtain the corresponding predicted fault rate. The random operating time, random ambient temperature and predicted fault rate are used as the extended data. Optionally, the data is expanded through the trained fault rate prediction model to obtain the extended data of the relay protection device that conforms to the law of the failure distribution model. In this way, the small sample data is effectively expanded, which overcomes the problem of affecting the subsequent prediction effect due to too few original data samples to a certain extent.

[0080] Step S103, determining a reliability model and a joint failure rate model according to the Arrhenius model and the Weibull distribution, fitting the joint failure rate model based on the fault statistical data and the extended data, and determining parameters of the joint failure rate model.

[0081] Optionally, considering that the failure rate model of the relay protection device conforms to the two-parameter Weibull distribution, and at the same time, the scale parameter α of the Weibull distribution and the ambient temperature satisfy the Arrhenius model. Therefore, the joint failure rate model is determined by the Arrhenius model and the Weibull distribution, and the parameters of the joint failure rate model are iteratively obtained using the original data and the expanded data.

[0082] In one example, step S103 includes:

[0083] Step S301, determining a distribution failure function according to Weibull distribution, determining a reliability function according to the distribution failure function, and determining a failure rate function according to the distribution failure function and the reliability function.

[0084] Alternatively, experiments show that the failure rate of relay protection devices conforms to the reliability "bathtub curve" characteristics, and Weibull distribution, as the theoretical basis for reliability analysis and device life prediction, can describe the bathtub curve as a whole. The specific formula of the distribution failure function is:

[0085]

[0086] Among them, F(t) represents the distribution failure function, α represents the scale parameter, β represents the shape parameter, e is the base of the natural logarithm, and t represents the running time.

[0087] The reliability function is determined according to the distribution failure function. The reliability function R(t) is expressed as:

[0088]

[0089] The failure rate function is determined according to the distribution failure function and the reliability function. The failure rate function λ(t) is expressed as:

[0090]

[0091] Among them, f(t) is the reciprocal of F(t), F(t) represents the distribution failure function, α represents the scale parameter, β represents the shape parameter, e is the base of the natural logarithm, and t represents the running time.

[0092] Step S302: determining the correlation between the scale parameters in the reliability function and the failure rate function and the ambient temperature according to the Arrhenius model.

[0093] Optionally, the Arrhenius model represents the relationship between the degradation rate of the components of the protection device and parameters such as temperature. The scale parameters in the reliability function and the failure rate function and the ambient temperature satisfy the Arrhenius model, that is, the correlation between the scale parameters and the ambient temperature is specifically expressed as:

[0094]

[0095] Among them, α represents the scale parameter, A represents the proportional constant, E represents the product activation energy corresponding to the failure rate, K represents the Boltzmann constant, K=8.617*10-5eV / K, T represents the ambient temperature, and e is the base of the natural logarithm.

[0096] Step S303, determining a joint failure rate model according to the failure rate function and the correlation relationship, and determining a reliability model according to the correlation relationship and the reliability function.

[0097] Optionally, a joint failure rate model is obtained by combining the failure rate function and the correlation relationship. The joint failure rate model is expressed as:

[0098]

[0099] Among them, λ(t) represents the failure rate function, A represents the proportionality constant, β represents the shape parameter, E represents the product activation energy corresponding to the failure rate, K represents the Boltzmann constant, K=8.617*10-5eV / K, T represents the ambient temperature, e is the base of the natural logarithm, and t represents the operating time.

[0100] Combining the reliability function and the correlation relationship, the reliability model is obtained. The reliability model R(t) is expressed as:

[0101]

[0102] Among them, F(t) represents the distributed failure function, t represents the operating time, α represents the scale parameter, β represents the shape parameter, A represents the proportionality constant, E represents the product activation energy corresponding to the failure rate, K represents the Boltzmann constant, K=8.617*10-5eV / K, T represents the ambient temperature, and e is the base of the natural logarithm.

[0103] In one example, step S103 includes:

[0104] Step S304: iterate the joint failure rate model based on the least square method according to the fault statistical data and the expanded data.

[0105] Optionally, partial derivatives of β and A are obtained respectively through the joint failure rate model, an initial iteration value is selected, and based on the original fault statistics data and the augmented data, the LM method is used to iterate the joint failure rate model through the least squares method to solve the values ​​of each parameter and calculate the residual sum of squares.

[0106] Step S305 , in response to the residual sum of squares between the predicted value of the joint failure rate model and the average failure rate being less than a preset threshold, the iteration is stopped, and the parameters of the joint failure rate model are determined, the parameters including a proportional constant and a shape parameter.

[0107] Optionally, the new parameter value and the new residual sum of squares are iteratively calculated until the residual sum of squares is less than a preset threshold, and the iteration is stopped to obtain the proportional constant and the shape parameter. It should be noted that although the proportional constant and the shape parameter are solved based on the joint failure rate model, the proportional constant and the shape parameter in the reliability model are the same as those in the joint failure rate model, and since the reliability model is part of the calculation process of the joint failure rate model, the optimal solution obtained by the joint failure rate model can be used in the reliability model.

[0108] In this way, by considering the temperature factor in step S103, a joint failure rate model with better accuracy and applicability is constructed, and the joint failure rate model is iteratively solved through the original data and the expanded data, thereby improving the prediction accuracy of the prediction model. At the same time, a fast and reliable life estimation method is provided for relay protection device products, saving time and cost for operation and maintenance, and providing guarantee for the rapid and extensive application of secondary equipment.

[0109] Step S104, based on the parameters of the joint failure rate model, the target ambient temperature and the preset reliability threshold, the reliable operation time of the relay protection device is determined through the reliability model.

[0110] Optionally, the parameter values ​​of the joint failure rate model obtained in step S103 are used as parameter values ​​of the reliability model, and the reliable operation time of the relay protection device is determined according to the target ambient temperature and a preset reliability threshold, wherein the reliability threshold is generally determined according to expert experience or test results.

[0111] In one example, step S104 includes:

[0112] Step S401, determining a target reliability model according to the parameters of the joint failure rate model and the reliability model. Optionally, the parameters of the joint failure rate model are substituted into the reliability model to obtain the target reliability model.

[0113] Step S402, obtaining a reliability variation trend with operating time according to the target ambient temperature and the target reliability model. Substituting the target ambient temperature into the target reliability model can obtain a reliability variation function with operating time.

[0114] Step S403, determine the reliable operation time of the relay protection device according to the change trend and the preset reliability threshold. Optionally, determine the operation time when the reliability in the change trend is equal to the reliability threshold, and use this time as the reliable operation time. The reliable operation time is generally in years, and the recommended exit operation time of the relay protection device can be obtained according to the reliable operation time.

[0115] In this way, the reliable operation time of the relay protection device product is evaluated through step S104, which is beneficial to the safe and stable operation of the substation. At the same time, it helps to improve the utilization rate of the protection device and reduce cost loss.

[0116] In summary, this application utilizes the characteristic of the Transformer model that can process small sample data, trains the Transformer model through small sample data, and expands the data through the trained model, which to a certain extent overcomes the problem of affecting the subsequent prediction effect due to too few original data samples. By constructing a joint failure rate model with better accuracy and applicability, the joint failure rate model is iteratively solved based on the original data and expanded data to improve the prediction accuracy of the prediction model. At the same time, the reliable operation time of the relay protection device products is evaluated, which is beneficial to the safe and stable operation of the substation. At the same time, it helps to improve the utilization rate of several protection devices and reduce cost losses.

[0117] In a second aspect, the embodiments of the present application provide a system for predicting the reliable operation time of a relay protection device. Figure 3 is a structural block diagram of a system for predicting the reliable operation time of a relay protection device according to an embodiment of the present application, such as Figure 3 As shown, the system includes:

[0118] Acquisition module 100: used to acquire fault statistical data, and train the Transformer model according to the fault statistical data to obtain a fault rate prediction model. The fault statistical data includes ambient temperature, operating time of the relay protection device and average failure rate.

[0119] Data expansion module 200: used to obtain expanded data according to the acquired random operating time, random ambient temperature and failure rate prediction model.

[0120] Fitting module 300: used to determine the reliability model and the joint failure rate model according to the Arrhenius model and the Weibull distribution, fit the joint failure rate model based on the fault statistical data and the extended data, and determine the parameters of the joint failure rate model.

[0121] Result module 400 is used to determine the reliable operation time of the relay protection device through a reliability model based on the parameters, the target ambient temperature and the preset reliability threshold.

[0122] In one example, the fitting module 300 includes:

[0123] It is used to determine the distribution failure function according to the Weibull distribution, determine the reliability function according to the distribution failure function, and determine the failure rate function according to the distribution failure function and the reliability function.

[0124] The correlation between the scale parameters in the reliability function and the failure rate function and the ambient temperature is determined based on the Arrhenius model.

[0125] The joint failure rate model is determined based on the failure rate function and the correlation relationship, and the reliability model is determined based on the correlation relationship and the reliability function.

[0126] In one example, the fitting module 300 includes:

[0127] It is used to determine the distribution failure function according to the Weibull distribution. The specific formula of the distribution failure function is:

[0128]

[0129] Among them, F(t) represents the distribution failure function, α represents the scale parameter, β represents the shape parameter, e is the base of the natural logarithm, and t represents the running time.

[0130] The reliability function R(t) is expressed as:

[0131]

[0132] The failure rate function λ(t) is expressed as:

[0133]

[0134] The relationship between the scale parameter and the ambient temperature is determined according to the Arrhenius model, which is specifically expressed as:

[0135]

[0136] Among them, α represents the scale parameter, A represents the proportional constant, E represents the product activation energy corresponding to the failure rate, K represents the Boltzmann constant, and T represents the ambient temperature;

[0137] The joint failure rate model is determined based on the failure rate function and the correlation relationship. The joint failure rate model is expressed as:

[0138]

[0139] In one example, the fitting module 300 includes:

[0140] The joint failure rate model is iterated based on the least squares method according to the fault statistics data and the augmented data.

[0141] In response to the residual sum of squares between the predicted value of the joint failure rate model and the average failure rate being less than a preset threshold, the iteration is stopped, and the parameters of the joint failure rate model are determined, the parameters including a proportional constant and a shape parameter.

[0142] In one example, the results module 400 includes:

[0143] The target reliability model is determined based on the parameters and the reliability model.

[0144] According to the target ambient temperature and the target reliability model, the reliability variation trend with the operating time is obtained.

[0145] The reliable operation time of the relay protection device is determined based on the change trend and the preset reliability threshold.

[0146] In one example, the data expansion module 200 includes:

[0147] It is used to make predictions through a failure rate prediction model according to the acquired random operating time and random ambient temperature to obtain a corresponding predicted failure rate.

[0148] Random running time, random ambient temperature and predicted failure rate are used as augmented data.

[0149] In one example, the Transformer model in the acquisition module 100 includes: an input module, an encoding module, a decoding module, and an output module.

[0150] The input module includes a data embedding layer and a position encoder, which is used to perform word embedding operations on each data in the input sequence.

[0151] The encoding module and the decoding module include a self-attention module, which is used to map the input sequence into question-key-value and calculate the attention score.

[0152] In summary, this application utilizes the characteristic of the Transformer model that can process small sample data, trains the Transformer model through small sample data, and expands the data through the trained model, which to a certain extent overcomes the problem of affecting the subsequent prediction effect due to too few original data samples. By constructing a joint failure rate model with better accuracy and applicability, the joint failure rate model is iteratively solved based on the original data and expanded data to improve the prediction accuracy of the prediction model. At the same time, the reliable operation time of the relay protection device products is evaluated, which is beneficial to the safe and stable operation of the substation. At the same time, it helps to improve the utilization rate of several protection devices and reduce cost losses.

[0153] In a third aspect, an embodiment of the present application provides an electronic device, Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the method for predicting the reliable operation time of the relay protection device provided in the first aspect is implemented. Figure 4 The electronic device 60 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0154] The electronic device 60 may be in the form of a general-purpose computing device, for example, it may be a server device. The components of the electronic device 60 may include but are not limited to: at least one processor 61, at least one memory 62, and a bus 63 connecting different system components (including the memory 62 and the processor 61).

[0155] The bus 63 includes a data bus, an address bus, and a control bus.

[0156] The memory 62 may include a volatile memory, such as a random access memory (RAM) 621 and / or a cache memory 622 , and may further include a read-only memory (ROM) 623 .

[0157] The memory 62 may also include a program / utility 625 having a set (at least one) of program modules 624, such program modules 624 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0158] The processor 61 executes various functional applications and data processing by running the computer program stored in the memory 62, such as the method for predicting the reliable operation time of the relay protection device provided in the first aspect of the present application.

[0159] The electronic device 60 may also communicate with one or more external devices 64 (e.g., keyboards, pointing devices, etc.). Such communication may be performed via an input / output (I / O) interface 65. Furthermore, the model-generated device 60 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 66. As shown, the network adapter 66 communicates with other modules of the model-generated device 60 via a bus 63. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the model-generated device 60, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (RAID) systems, tape drives, and data backup storage systems.

[0160] It should be noted that although several units / modules or sub-units / modules of the electronic device are mentioned in the above detailed description, this division is merely exemplary and not mandatory. In fact, according to an embodiment of the present invention, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided into multiple units / modules to be embodied.

[0161] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having a program stored thereon, and when the program is executed by a processor, the method for predicting the reliable operation time of a relay protection device provided in the first aspect is implemented.

[0162] The readable storage medium may include but is not limited to: a portable disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical storage device, a magnetic storage device or any suitable combination of the above.

[0163] In a possible implementation, the present invention can also be implemented in the form of a program product, which includes a program code. When the program product is run on a terminal device, the program code is used to enable the terminal device to execute the steps of the method for predicting the reliable operating time of the relay protection device provided in the first aspect.

[0164] The program code for executing the present invention may be written in any combination of one or more programming languages, and may be executed entirely on a user device, partially on a user device, as an independent software package, partially on a user device and partially on a remote device, or entirely on a remote device.

[0165] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0166] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A method for predicting the reliable operation time of a relay protection device, characterized in that: include: Obtaining fault statistical data, and training a Transformer model according to the fault statistical data to obtain a fault rate prediction model, wherein the fault statistical data includes ambient temperature, operating time of the relay protection device, and average failure rate; Obtaining expanded data according to the acquired random operating time, random ambient temperature and the failure rate prediction model; Determine a reliability model and a joint failure rate model according to the Arrhenius model and the Weibull distribution, fit the joint failure rate model based on the fault statistical data and the expanded data, and determine parameters of the joint failure rate model; Based on the parameters, the target ambient temperature and a preset reliability threshold, the reliable operation time of the relay protection device is determined by the reliability model.

2. A method for predicting the reliable operation time of a relay protection device according to claim 1, characterized in that: Determining the reliability model and the joint failure rate model according to the Arrhenius model and the Weibull distribution includes: Determine a distribution failure function according to the Weibull distribution, determine a reliability function according to the distribution failure function, and determine a failure rate function according to the distribution failure function and the reliability function; Determining the correlation between the scale parameters in the reliability function and the failure rate function and the ambient temperature according to the Arrhenius model; A joint failure rate model is determined according to the failure rate function and the association relationship, and a reliability model is determined according to the association relationship and the reliability function.

3. A method for predicting the reliable operation time of a relay protection device according to claim 2, characterized in that: The joint failure rate model is determined based on the Arrhenius model and Weibull distribution, including: The distribution failure function is determined according to the Weibull distribution. The specific formula of the distribution failure function is: Where F(t) represents the distribution failure function, α represents the scale parameter, β represents the shape parameter, e is the base of the natural logarithm, and t represents the running time. The reliability function R(t) is expressed as: The failure rate function λ(t) is expressed as: The correlation between the scale parameter and the ambient temperature is determined according to the Arrhenius model, which is specifically expressed as follows: Wherein, α represents a scale parameter, A represents a proportional constant, E represents the product activation energy corresponding to the failure rate, K represents the Boltzmann constant, and T represents the ambient temperature; A joint failure rate model is determined according to the failure rate function and the association relationship, and the joint failure rate model is expressed as:

4. A method for predicting the reliable operation time of a relay protection device according to claim 3, characterized in that: The fitting of the joint failure rate model based on the fault statistical data and the expanded data to determine the parameters of the joint failure rate model includes: Iterating the joint failure rate model based on the least square method according to the fault statistical data and the expanded data; In response to the residual sum of squares between the predicted value of the joint failure rate model and the average failure rate being less than a preset threshold, the iteration is stopped, and the parameters of the joint failure rate model are determined, the parameters including the proportional constant and the shape parameter.

5. A method for predicting the reliable operation time of a relay protection device according to claim 4, characterized in that: The step of determining the reliable operation time of the relay protection device by the reliability model based on the parameter, the target ambient temperature and the preset reliability threshold value includes: Determining a target reliability model according to the parameters and the reliability model; Obtaining a reliability variation trend with operating time according to the target ambient temperature and the target reliability model; The reliable operation time of the relay protection device is determined according to the change trend and the preset reliability threshold.

6. A method for predicting the reliable operation time of a relay protection device according to claim 1, characterized in that: The extended data is obtained according to the acquired random operating time, random ambient temperature and the failure rate prediction model, including: According to the acquired random operating time and random ambient temperature, prediction is performed using the failure rate prediction model to obtain a corresponding predicted failure rate; The random operating time, random ambient temperature and the predicted failure rate are used as extended data.

7. A method for predicting the reliable operation time of a relay protection device according to claim 1, characterized in that: The Transformer model includes: an input module, an encoding module, a decoding module and an output module. The input module includes a data embedding layer and a position encoder, which is used to perform a word embedding operation on each data in the input sequence; The encoding module and the decoding module include a self-attention module, which is used to map the input sequence into question-key-value and calculate an attention score.

8. A prediction system for the reliable operation time of a relay protection device, characterized in that: include: Acquisition module: used to acquire fault statistical data, and train the Transformer model according to the fault statistical data to obtain a fault rate prediction model, wherein the fault statistical data includes ambient temperature, operating time of the relay protection device and average failure rate; Data expansion module: used to obtain expanded data according to the acquired random operating time, random ambient temperature and the failure rate prediction model; A fitting module: used for determining a reliability model and a joint failure rate model according to the Arrhenius model and the Weibull distribution, fitting the joint failure rate model based on the fault statistical data and the extended data, and determining parameters of the joint failure rate model; Result module: used to determine the reliable operation time of the relay protection device through the reliability model based on the parameters, the target ambient temperature and the preset reliability threshold.

9. An electronic device, characterized in that: include Memory, processor, and A computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for predicting the reliable operation time of a relay protection device according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method for predicting the reliable operation time of a relay protection device according to any one of claims 1 to 7 is implemented.

Images