High-speed railway transformer abnormality monitoring method and system, electronic device and medium

Abstract

The application provides a high-speed railway transformer abnormality monitoring method, system, electronic equipment and medium, and belongs to the technical field of power system monitoring. The method comprises the following steps: constructing a high-speed railway transformer abnormality monitoring model, collecting input and output data of the transformer under no load and different loads to form a data set, and training the model to realize real-time operation state abnormality monitoring of the transformer. The system can effectively utilize a large amount of normal operation monitoring data to establish a model, and judge the operation state of the transformer through the probability of actual data in the probability distribution given by the model, thereby being independent of fault data. The application combines the advantages of mechanism models and data-driven methods, and through a time-varying state space model and a recurrent neural network, the operation mechanism of the transformer is met, and the monitoring data can be fully utilized, so that the accuracy and interpretability of monitoring are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power system monitoring, in particular to a high-speed railway transformer abnormality monitoring method and system, an electronic device and a medium. BACKGROUND

[0002] With the increase of high-speed railway transport capacity and the expansion of line scale, the operation safety problem of high-speed railway becomes more and more important. The high-speed railway system is complex and has many devices. The transformer is one of the important devices in the high-speed railway system, which plays an important role in power conversion and power transmission. The operation failure of the transformer will cause serious problems such as damage to the equipment of the high-speed railway and train shutdown, so the abnormality monitoring of the high-speed railway transformer is particularly important.

[0003] The existing high-speed railway transformer abnormality monitoring methods can be roughly divided into the following two types: (1) using sensors to monitor the temperature, vibration and other variables inside the transformer, and comparing them with the set threshold to determine whether the operation is normal; (2) establishing a mechanism model of the transformer, comparing the model output with the actual data to determine whether the operation is normal; (3) training a neural network model with a large amount of data to directly give the transformer operation result; in the above methods, the first type has low intelligence and is prone to false positives and false negatives, the second type has good interpretability but modeling is often difficult and many parameters are difficult to calculate, and the third type makes full use of a large amount of operation monitoring data of the transformer device, but its interpretability is poor. Combining data and mechanism for abnormality monitoring of the device can make full use of existing mechanism modeling information and a large amount of operation data, and at present there is still a lack of a method for combining the operation mechanism and operation data of the high-speed railway transformer for abnormality monitoring. SUMMARY

[0004] The purpose of the present application is to provide a high-speed railway transformer abnormality monitoring method, system, electronic device and medium, which combines the advantages of mechanism model and data-driven method, uses time-varying state space model and recurrent neural network, conforms to the operation mechanism of the transformer, makes full use of monitoring data, and improves the accuracy and interpretability of monitoring. In addition, the design of the present application enables non-professionals to perform transformer abnormality monitoring, and the monitoring accuracy can be continuously improved with the increase of data volume and optimization of the model.

[0005] To achieve the above purpose, the present application provides a high-speed railway transformer abnormality monitoring method, comprising the following steps:

[0006] Step S1, collecting monitoring data of the high-speed railway transformer, including input and output data under no load and different loads, to form a data set;

[0007] Step S2, constructing a high-speed railway transformer abnormality monitoring model;

[0008] Step S3, training the high-speed railway transformer abnormality monitoring model;

[0009] Step S4, obtaining real-time operation data of the high-speed railway transformer to perform abnormality monitoring on the operation state.

[0010] Preferably, in step S1, the data set is represents the input of the kth monitoring sequence at time t, represents the output of the kth monitoring sequence at time t, T k represents the length of the kth monitoring sequence, and L represents that there are L monitoring sequences in total.

[0011] Preferably, in step S2, the high-speed railway transformer abnormality monitoring model comprises a time-varying state space model of a high-speed railway transformer model, a parameter prior matrix set, a parameter mask matrix set, and a recurrent neural network, and the construction method of the high-speed railway transformer abnormality monitoring model is as follows:

[0012] Step S21, establishing a time-varying state space model of a high-speed railway transformer model according to the operation mechanism of the transformer;

[0013]

[0014] wherein x t is an n-dimensional vector representing the operation state of the transformer at time t; u t is a d-dimensional vector representing the input vector of the transformer at time t; z t is an m-dimensional vector representing the output vector of the transformer at time t; Φ t , B t , and H t represent the state transition matrix, the input matrix, and the output matrix of the transformer at time t, respectively; ω t and v t are zero-mean Gaussian noise, and their covariance matrices are Q t and R t , respectively; the symbol represents Gaussian distribution;

[0015] The symbol Θ t represents the parameter set in the time-varying state space model, Θ t ={Φ t , B t , H t , Q t , R t}, and Θ t is a time-varying parameter;

[0016] Step S22, constructing a parameter prior matrix set Θ pand a parameter mask matrix set Θ m ;

[0017] Step S23, combining the time-varying state space model in step S21, a recurrent neural network is constructed;

[0018]

[0019] Let denote the output set of the recurrent neural network, where and are the state transition matrix prediction value, input matrix prediction value, output matrix prediction value, ω t covariance matrix prediction value and v t covariance matrix prediction value given by the recurrent neural network, respectively, the matrix dimensions of Φ t , B t , H t , Q t , R t are the same.

[0020] Preferably, in step S22, Θ t corresponding parameter prior matrix set Θ p ={Φ p , B p , H p , Q p , R p}, Φ p , B p , H p , Q p and R p are the corresponding prior matrices of Φ t , B t , H t , Q t and R t , Θ p and Θ t have the following relationship:

[0021]

[0022]

[0023]

[0024]

[0025]

[0026] where, the symbol [] ijrepresents the element in the i-th row and j-th column of the matrix in [].

[0027] Preferably, in step S22, Θ t The corresponding parameter prior matrix set Θ m = {Φ m , B m , H m , Q m , R m}, Φ m , B m , H m , Q m and R m are the corresponding mask matrices of Φ t , B t , H t , Q t and R t , respectively, and Θ m and Θ t have the following relationship:

[0028]

[0029]

[0030]

[0031]

[0032]

[0033] Preferably, in step S3, the high-speed railway transformer abnormality monitoring model is trained, including the following steps:

[0034] Step S31, input the input data in the data set into the recurrent neural network, and calculate the network output

[0035] Step S32, mask calculation is performed on to obtain The specific calculation formula is:

[0036]

[0037]

[0038]

[0039]

[0040]

[0041] wherein, and are the state transition matrix, input matrix, output matrix, and ω t are the covariance matrix of v t , respectively, and symbol denotes the multiplication of corresponding elements of matrices;

[0042] Step S33, using the Kalman filter, calculate the training loss of the network;

[0043]

[0044] wherein, and denote the optimal state estimation value at time t and time t-1, respectively, denotes the prior estimation of the state at time t, denotes the prior estimation of the state covariance matrix at time t, and superscript T denotes the matrix transpose operation, denotes the prior estimation of the output vector at time t, and S t denotes the prior estimation of the covariance matrix of the output vector at time t, and K t denotes the Kalman gain at time t, and ∑ t and ∑ t-1 denote the optimal state covariance matrix estimation at time t and time t-1, respectively, and in particular, when t = 0, ∑0 = I;

[0045] The training loss of the network is obtained as:

[0046]

[0047] wherein, log denotes the logarithmic function;

[0048] Step S34, using the training loss to train the recurrent neural network.

[0049] Preferably, in step S4, the real-time operation data of the high-speed railway transformer is obtained to perform abnormal monitoring of the operation state, including the following steps:

[0050] Step S41, inputting the transformer input vector u t into the trained recurrent neural network to calculate

[0051] Step S42, according to step S32, performing mask calculation on

[0052]

[0053] Step S43, given a threshold value δ, according to step S33, calculating the state estimation value at time t​ and S t , obtaining a Gaussian probability distribution When , it indicates that the transformer is working normally, otherwise, it indicates that the transformer is working abnormally.

[0054] The application further provides a high-speed railway transformer abnormality monitoring system, comprising:

[0055] A data acquisition module is configured to acquire input and output data of the high-speed railway transformer under no load and different loads.

[0056] A model construction module is configured to construct a high-speed railway transformer abnormality monitoring model, which comprises a time-varying state space model, a parameter prior matrix, a parameter mask matrix and a recurrent neural network.

[0057] A training module is configured to train the abnormality monitoring model, calculate the output of the recurrent neural network through the input data, perform mask calculation and Kalman filter processing, calculate the training loss of the network, and train the recurrent neural network using the training loss.

[0058] A real-time monitoring module is configured to acquire real-time operation data of the high-speed railway transformer and perform abnormality monitoring on the operation state using the trained recurrent neural network.

[0059] The application further provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the high-speed railway transformer abnormality monitoring method when executing the computer program.

[0060] The application further provides a computer readable storage medium, which stores a computer program, and the computer program implements the steps of the high-speed railway transformer abnormality monitoring method when executed by a processor.

[0061] Therefore, the application adopts the high-speed railway transformer abnormality monitoring method, system, electronic device and medium, and the technical effects are as follows:

[0062] (1) Although the abnormality monitoring of the high-speed railway transformer device is very important, the fault data is relatively less, the normal operation monitoring data is more, and the sample quantity is unbalanced.

[0063] (2) In abnormality monitoring, each model has its own characteristics and focus direction, and it is often difficult to establish an accurate mechanism model, while the data-driven method has poor interpretability. The present application combines the advantages of both, and the time-varying state space model makes the present application basically consistent with the operation mechanism of the transformer, and the recurrent neural network enables the present application to make full use of a large amount of monitoring data.

[0064] (3) Although there are many references for modeling high-speed railway transformers, it is still difficult for non-professionals. The present application combines the mechanism model of the transformer, but as long as the parameters in the mechanism model are not fixed, the present application is essentially a data-driven method. This enables non-professionals to use the present application for transformer abnormality monitoring. Of course, the larger the amount of data, the more accurate the time-varying state space model, the higher the accuracy of the present application, so theoretically the monitoring accuracy of the present application can be continuously improved. BRIEF DESCRIPTION OF DRAWINGS

[0065] Figure 1 A flowchart of the present application, a high-speed railway transformer abnormality monitoring method. DETAILED DESCRIPTION

[0066] The technical solutions of the present application are further described below through the drawings and examples.

[0067] Unless otherwise defined, the technical terms or scientific terms used in the present application shall have the usual meaning understood by those skilled in the art to which the present application belongs.

[0068] Example 1

[0069] The present application provides a high-speed railway transformer abnormality monitoring method, comprising the following steps:

[0070] Step S1, collecting monitoring data of the high-speed railway transformer, including input and output data under no load and different loads, to form a data set.

[0071] The data set is represents the input of the kth monitoring sequence at time t, represents the output of the kth monitoring sequence at time t, T k represents the length of the kth monitoring sequence, and L represents a total of L monitoring sequences.

[0072] Step S2, constructing a high-speed railway transformer abnormality monitoring model;

[0073] Step S21, establishing a time-varying state space model of the high-speed railway transformer model according to the operation mechanism of the transformer;

[0074]

[0075] wherein x t is an n-dimensional vector representing the operating state of the transformer at time t; u t is a d-dimensional vector representing the input vector of the transformer at time t; z t is an m-dimensional vector representing the output vector of the transformer at time t, the input, state and output of the transformer being different under no-load and different loads; Φ t , B t and H t represent the state transition matrix, input matrix and output matrix of the transformer at time t, respectively; ω t and v t are Gaussian noises with zero mean, and the covariance matrices thereof are Q t and R t , respectively; symbol represents Gaussian distribution;

[0076] The parameter set in the time-varying state space model is represented by symbol Θ t , Θ t ={Φ t , B t , H t , Q t , R t}, and Θ t is a time-varying parameter;

[0077] Step S22, constructing a parameter prior matrix set Θ p and a parameter mask matrix set Θ m ;

[0078] The parameter prior matrix set Θ t corresponding to Θ p ={Φ p , B p , H p , Q p , R p}, Φ p , B p , H p , Q p and R p are prior matrices corresponding to Φ t , B t , H t , Q t and R t , respectively, and Θ p and Θ t have the following relationship:

[0079]

[0080]

[0081]

[0082]

[0083]

[0084] where the symbol [] ij represents the element in the i-th row and j-th column of the matrix in [].

[0085] Constructing Θ t the corresponding parameter prior matrix set Θ m = {Φ m , B m , H m , Q m , R m}, Φ m , B m , H m , Q m and R m are the corresponding mask matrices of Φ t , B t , H t , Q t and R t , respectively, and Θ m and Θ t have the following relationship:

[0086]

[0087]

[0088]

[0089]

[0090]

[0091] Step S23, in combination with the time-varying state space model in step S21, constructs a recurrent neural network;

[0092]

[0093] Let denote the output set of the recurrent neural network, where and are the predicted values of the state transition matrix, the input matrix, the output matrix, the covariance matrix of ω t and the covariance matrix of v t given by the recurrent neural network, respectively. The matrix dimensions of Φ t , B t , H t , Q t , R t are the same.

[0094] Step S3, training the high-speed railway transformer abnormality monitoring model, comprising the following steps:

[0095] Step S31, inputting the input data in the data set into the recurrent neural network, and calculating the network output

[0096] Step S32, performing mask calculation on to obtain The specific calculation formula is:

[0097]

[0098]

[0099]

[0100]

[0101]

[0102] Wherein, and are the state transition matrix, the input matrix, the output matrix, the covariance matrix of ω t and the covariance matrix of v t after mask calculation, and the symbol represents the multiplication of the corresponding elements of the matrix;

[0103] Step S33, using the Kalman filter to calculate the training loss of the network;

[0104]

[0105] Wherein, and represent the best estimation value of the state at time t and time t-1, represents the prior estimation of the state at time t, represents the prior estimation of the state covariance matrix at time t, and the superscript T represents the matrix transpose operation, represents the prior estimation of the output vector at time t, and S t represents the prior estimation of the covariance matrix of the output vector at time t, and K t represents the Kalman gain at time t, and ∑ t and ∑ t-1respectively represent the optimal state covariance matrix estimation at time t and time t-1, in particular, when t=0, ∑0=I;

[0106] The training loss of the network is obtained as:

[0107]

[0108] wherein, log represents a logarithmic function;

[0109] Step S34, training the recurrent neural network using the training loss.

[0110] Step S4, obtaining real-time operation data of the high-speed railway transformer, and performing abnormal monitoring of the operation state, including the following steps:

[0111] Step S41, inputting the transformer input vector u t into the trained recurrent neural network to calculate

[0112] Step S42, according to step S32, calculating mask calculation:

[0113]

[0114] Step S43, given a threshold value δ, according to step S33, calculating the t time and S t , obtaining the Gaussian probability distribution When , it indicates that the transformer is working normally; otherwise, it indicates that the transformer is working abnormally.

[0115] Embodiment two

[0116] A high-speed railway transformer abnormal monitoring system, comprising:

[0117] A data acquisition module for acquiring input and output data of the high-speed railway transformer under no load and different loads;

[0118] A model construction module for constructing a high-speed railway transformer abnormal monitoring model, which includes a time-varying state space model, a parameter prior matrix, a parameter mask matrix and a recurrent neural network;

[0119] A training module for training the abnormal monitoring model, calculating the output of the recurrent neural network through the input data, performing mask calculation and Kalman filter processing to calculate the training loss of the network, and training the recurrent neural network using the training loss;

[0120] Real-time monitoring module, for obtaining real-time operation data of the high-speed railway transformer, and performing abnormal monitoring on the operation state by using the trained recurrent neural network.

[0121] If the above functions are realized in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the parts of the prior art that essentially contribute or the parts of the technical solutions can be embodied in the form of a software product, which is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0122] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered a list of executable instructions for implementing logical functions, and can be specifically embodied in any computer-readable medium for use by an instruction execution system, apparatus or device, such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, apparatus or device and execute the instructions, or in conjunction with these instructions execution system, apparatus or device. For the purpose of this specification, the "computer-readable medium" can be any device that can contain, store, communicate, propagate or transport programs for use by an instruction execution system, apparatus or device or in conjunction with these instruction execution system, apparatus or device.

[0123] More specific examples (non-exhaustive list) of computer-readable medium include the following: electrical connections having one or more wires (electronic devices), portable computer disks (magnetic devices), random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memories), fiber optic devices, and portable compact disk read-only memories (CDROMs). In addition, the computer-readable medium can even be paper or other suitable medium on which the program can be printed, as the program can be electronically obtained, for example, by optical scanning of the paper or other medium, followed by editing, interpreting or otherwise processing the program as necessary, and then storing it in a computer memory.

[0124] Therefore, the application adopts the above high-speed railway transformer abnormality monitoring method, system, electronic device and medium, combines the advantages of mechanism models and data-driven methods, and through a time-varying state space model and a recurrent neural network, both the operation mechanism of the transformer and the monitoring data can be fully utilized, and the accuracy and interpretability of monitoring are improved. In addition, the design of the application enables non-professionals to also perform transformer abnormality monitoring, and as the amount of data increases and the model is optimized, the monitoring accuracy can be continuously improved.

[0125] It should be noted that the contents not elaborated in the application are all prior art and are well known to those skilled in the art.

[0126] Finally, it should be noted that: the above examples are only used to illustrate the technical solutions of the application and not to limit them, although the application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that: the technical solutions of the application can still be modified or replaced by equivalents, and these modifications or equivalent replacements cannot make the modified technical solutions deviate from the spirit and scope of the technical solutions of the application.

Claims

1. A method for monitoring abnormality of a high-speed railway transformer, characterized in that: The following steps are involved: Step S1: collecting monitoring data of high-speed railway transformers, including input and output data under no-load and different loads, to form a data set; Step S2: constructing a high-speed railway transformer abnormality monitoring model; Step S3: training a high-speed railway transformer abnormality monitoring model; Step S4: obtaining real-time operating data of the high-speed railway transformer and performing abnormal monitoring of the operating status; In step S2, the high-speed railway transformer anomaly monitoring model includes a time-varying state space model of the high-speed railway transformer model, a parameter prior matrix set, a parameter mask matrix set, and a recurrent neural network. The method for constructing the high-speed railway transformer anomaly monitoring model is as follows: Step S21: establishing a time-varying state space model of the high-speed railway transformer model according to the operating mechanism of the transformer; ； in, is a dimensional vector, indicating that the transformer The operating status at the moment; is a dimensional vector, representing The input vector of the transformer at time instant; yes dimensional vector, representing Output vector of the transformer at moment t; , and Respectively The state transfer matrix, input matrix and output matrix of the moment transformer; and is zero-mean Gaussian noise, and its covariance matrix is and ;symbol represents Gaussian distribution; Use symbols represents the set of parameters in the time-varying state-space model, , is a time-varying parameter; Step S22: Construct a parameter prior matrix set and the parameter mask matrix set ; Step S23: construct a recurrent neural network by combining the time-varying state space model in step S21; ； make represents the output set of the recurrent neural network, where 、 、 、 and They are the state transfer matrix prediction value, input matrix prediction value, output matrix prediction value given by the recurrent neural network, The covariance matrix of the predicted value and The covariance matrix of the predicted value, 、 、 、 、 The matrix dimensions are 、 、 、 、 same.

2. The method for monitoring abnormality of a high-speed railway transformer according to claim 1, characterized in that: In step S1, the data set is , Indicates the Monitoring sequence Input at the moment, Indicates the Monitoring sequence Output at any moment, Indicates the The length of the monitoring sequence, Indicates a total of A monitoring sequence.

3. The method for monitoring abnormality of a high-speed railway transformer according to claim 2, characterized in that: In step S22, construct The corresponding parameter prior matrix set , 、 、 、 and They are 、 、 、 and The corresponding prior matrix, and With the following relationship: ； ； ； ； ； Among them, the symbol express The matrix Rank Column element.

4. The method for monitoring abnormality of a high-speed railway transformer according to claim 3, characterized in that: In step S22, construct The corresponding parameter prior matrix set , 、 、 、 and They are 、 、 、 and The corresponding mask matrix, and With the following relationship: ； ； ； ； 。 5. The method for monitoring abnormality of a high-speed railway transformer according to claim 4, characterized in that: In step S3, training a high-speed railway transformer anomaly monitoring model includes the following steps: Step S31: Input the input data in the data set into the recurrent neural network and calculate the network output ; Step S32: Do mask calculation and get , the specific calculation formula is: ； ； ； ； ； in, 、 、 、 and They are the state transfer matrix after mask calculation, input matrix, output matrix, The covariance matrix and The covariance matrix of Indicates the multiplication of corresponding elements of the matrix; Step S33: Use the Kalman filter to calculate the training loss of the network; ； in, and express Moment and The best estimate of the state at time t, express The prior estimate of the state at time t, express The prior estimate of the state covariance matrix at time , the superscript symbol represents the matrix transpose operation, express A priori estimate of the output vector at time , express The prior estimate of the covariance matrix of the output vector at time t, express The Kalman gain at time t, and Respectively Moment and The best state covariance matrix estimation at time t; The training loss of the network is obtained as: ； in, represents the logarithmic function; Step S34: Use the training loss to train the recurrent neural network.

6. A high-speed railway transformer abnormality monitoring method according to claim 5, characterized in that: In step S4, real-time operation data of the high-speed railway transformer is obtained to monitor abnormal operation status, including the following steps: Step S41: Transformer input vector Input into the trained recurrent neural network and calculate ; Step S42: According to step S32, Do mask calculation: ； Step S43: Given a threshold , according to step S33, calculate Moment and , we get the Gaussian probability distribution ;when If the voltage is 0, it indicates that the transformer is working normally; otherwise, it indicates that the transformer is working abnormally.

7. A high-speed railway transformer abnormality monitoring system, characterized in that: The method for monitoring abnormality of a high-speed railway transformer according to any one of claims 1 to 6 comprises: Data acquisition module, used to collect input and output data of high-speed railway transformers under no-load and different loads; A model building module is used to build a high-speed railway transformer anomaly monitoring model, which includes a time-varying state space model, a parameter prior matrix, a parameter mask matrix, and a recurrent neural network; A training module, configured to train the anomaly monitoring model, calculate the output of the recurrent neural network through input data, perform mask calculation and Kalman filter processing to calculate the training loss of the network, and use the training loss to train the recurrent neural network; The real-time monitoring module is used to obtain real-time operating data of high-speed railway transformers and use the trained recurrent neural network to monitor abnormal operating status.

8. A computer device comprising: memory and processor; The memory stores a computer program, characterized in that when the processor executes the computer program, the steps of the high-speed railway transformer abnormality monitoring method according to any one of claims 1 to 6 are implemented.

9. 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 high-speed railway transformer abnormality monitoring method according to any one of claims 1 to 6 are implemented.

Images