Fault diagnosis method and system for power transmission line

Abstract

The invention provides a fault diagnosis method and system for a power transmission line, and relates to the technical field of power transmission line fault diagnosis. The fault line and the fault type can be quickly and accurately identified. The method comprises the following steps: decomposing and reconstructing an original fault signal by adopting wavelet packet operation to obtain a wavelet packet decomposition tree structure, and calculating wavelet packet energy of sub-band data of a last layer of the decomposition tree structure; constructing a wavelet packet energy characteristic matrix, performing normalization processing on the wavelet packet energy characteristic matrix, performing diversity processing on the normalized data, and dividing the data into a training set sample and a test set sample; performing model training according to the training set sample to obtain a fault line identification model and a fault type identification model; and carrying out power transmission line fault diagnosis based on the test set sample, the fault line identification model and the fault type identification model, and identifying a fault line and a fault type. The fault diagnosis system for the power transmission line is applied to the fault diagnosis method for the power transmission line.

Description

Technical Field

[0001] The present invention relates to the technical field of power transmission line fault diagnosis, and in particular to a fault diagnosis method and system for power transmission lines. Background Art

[0002] In the process of long-distance power transmission, the transmission lines generally experience a variety of terrains and landforms, and the climatic conditions are complex. Geographical conditions and environmental factors will affect the operation of the transmission lines. Once a short circuit fault occurs, it will seriously hinder the power transmission process. At this stage, relay protection methods have become a key technical means to ensure the stable operation of the transmission system, and can meet the requirements of rapidity, immediacy and reliability of fault removal. Fault diagnosis, as an important auxiliary tool for relay protection, has the main task of troubleshooting system faults based on massive and complex information such as protection action information and fault information when a fault occurs in the transmission system, and can eliminate the influence of uncertain factors on fault detection.

[0003] As an effective means of fault detection, domestic and foreign scholars have conducted a series of in-depth studies on fault diagnosis methods. The more common ones are expert systems, Bayesian networks, Petri nets, rough sets and artificial neural networks. However, in the face of today's increasingly complex power grids, the workload of maintaining and updating the knowledge base of expert systems has increased. At the same time, when facing fault conditions outside the knowledge base, a large amount of protection device action information is required to match and update the association rules in the knowledge base. The reasoning and calculation scale involved in this process has increased significantly, making it difficult to make accurate judgments. Therefore, experts have begun to work on developing other intelligent fault identification methods suitable for complex power grids. At present, Bayesian networks also have certain advantages in fault identification. Due to the strictness of their diagnostic and reasoning capabilities, this method has certain advantages in judging complex faults caused by uncertain factors. However, when facing complex power grids, the difficulty of network modeling increases, the diagnostic model becomes more complex, and the expansion performance is poor. Petri nets are also a common method for studying transmission line fault identification, but compared with Bayesian networks, this method has a significantly increased amount of calculation, and the fault tolerance is poor when the fault information is uncertain, so it is difficult to obtain good results. Rough set theory is another important tool for solving uncertainty problems after probability theory and fuzzy set theory. It can effectively deal with uncertainty factors in actual fault scenarios. This method can effectively simplify redundant data in the fault identification process and improve the efficiency of fault diagnosis. However, due to its high requirements for the redundancy of fault information, it is difficult to build a large-scale power grid fault model. Therefore, this method has certain limitations in the application of massive complex information fault systems. Artificial neural network is an intelligent network that imitates human brain neurons. Its powerful adaptive and self-learning capabilities make it stand out among artificial intelligence methods. With its unique advantages of high efficiency, self-organization and easy promotion, it has attracted the attention of researchers in the field of power grid fault diagnosis. However, this method still has obstacles such as slow convergence speed and difficulty in processing large sample data. Although conventional fault identification methods can achieve good identification results in small-scale transmission scenarios, it is difficult to adapt to complex large-scale transmission systems. There are problems such as insufficient expansion capabilities and difficulty in model construction. Therefore, it is of great practical significance to seek efficient and intelligent transmission line fault identification methods. Summary of the invention

[0004] In order to solve the above technical problems, the present invention provides a fault diagnosis method and system for power transmission lines, so as to quickly and accurately identify faulty lines and fault types during long-distance power transmission.

[0005] The present invention provides a fault diagnosis method for a power transmission line, the method comprising:

[0006] Step S1: Decompose and reconstruct the original fault signal using wavelet packet operation to obtain a wavelet packet decomposition tree structure, and calculate the wavelet packet energy of the sub-band data at the last layer of the decomposition tree structure;

[0007] Step S2: constructing a wavelet packet energy feature matrix, normalizing the wavelet packet energy feature matrix, and performing a diversity process on the normalized data to divide it into a training set sample and a test set sample;

[0008] Step S3: Perform model training based on training set samples to obtain a fault line identification model and a fault type identification model;

[0009] Step S4: Perform transmission line fault diagnosis based on the test set samples, the fault line identification model and the fault type identification model to identify the fault line and the fault type.

[0010] Preferably, the use of wavelet packet operation to decompose and reconstruct the original fault data to obtain a wavelet packet decomposition tree structure includes:

[0011] Formula (1) is used to orthogonally decompose the original fault signal into several frequency segments at different scales. Decomposed and

[0012]

[0013] Formula (2) is used to reconstruct the relatively important frequency bands in the decomposed sequence, and Back to

[0014]

[0015] in, is the original fault signal, h (2l-k) , g (2l-k) Respectively represent the low-pass and high-pass filter coefficients, k is the decomposition node number, l is the number of decomposition layers, They respectively represent the low-frequency detail signal and the high-frequency detail signal after decomposition.

[0016] Preferably, the calculation of the wavelet packet energy of the sub-band data at the last layer of the decomposition tree structure comprises: calculating the last layer 2 of the decomposition tree structure l The wavelet packet energy E1, E2, ..., E j ,

[0017] j represents the number of frequency bands contained in the last layer of wavelet packet decomposition, j = 1, 2, 3, ..., 2 l , the wavelet packet energy formula is:

[0018]

[0019] Among them, x j (k) represents the discrete data obtained by batch collection, k is the decomposition node number, n is the number of sampling points in the batch collection process of input data, S j (t) represents the original input electrical quantity data of the transmission line short circuit fault, E j Represents the wavelet packet energy of the jth node in the last layer of wavelet packet decomposition.

[0020] Preferably, the step of constructing a wavelet packet energy feature vector and normalizing the wavelet packet energy feature vector comprises:

[0021] Construct the wavelet packet energy eigenvector T, the formula is:

[0022]

[0023] in, It represents the energy value of the jth frequency band in the last layer of wavelet packet reconstruction, E j,max 、E j,min Respectively represent the minimum and maximum energy in each frequency band of the last layer, E j,m represents the normalized processing result of the energy value of the jth frequency band in the last layer, and T represents the energy feature vector of a single sample.

[0024] Preferably, the normalized data is subjected to diversity processing and divided into training set samples and test set samples according to a preset ratio of 7:3.

[0025] Preferably, the step S3: performing model training according to the training set samples to obtain a fault line identification model and a fault type identification model comprises:

[0026] Step S3.1: construct a convolutional neural network, the neural network includes an input layer, a convolutional layer, a pooling layer, a fully connected layer, and an output layer, and the weight and bias of the convolutional layer are set to {w, b}, where w represents the weight and b represents the bias;

[0027] Step S3.2: Update the weights and biases of the convolutional neural network RMSprop difference according to the back-propagation method;

[0028] Step S3.3: Input the training set samples into the convolutional neural network, perform network training based on the updated weights and biases, and obtain the fault line identification model and the fault type identification model.

[0029] Preferably, updating the weights and biases of the convolutional neural network RMSprop differential according to the back propagation method comprises:

[0030] The weights and biases of the convolutional neural network are initialized to {w(1), b(1)};

[0031] Set the number of initialization iterations h = 1;

[0032] The convolutional neural network is optimized using the expressions of weights and biases in the stochastic gradient descent method of the RMSprop differential convolutional neural network.

[0033] Update the number of iterations according to h=1+1;

[0034] The back-propagation method is used to update the weights and biases of the convolutional neural network RMSprop difference until the error gradient is satisfied.

[0035] Compared with the prior art, the fault diagnosis method for power transmission lines provided by the present invention has the following beneficial effects: by using wavelet packet operation to decompose and reconstruct the original fault signal, more accurate and detailed fault signal characteristics can be obtained, so as to better understand and grasp the characteristics and properties of the fault; by calculating the wavelet packet energy of the sub-band data of the last layer of the decomposition tree structure, the key features in the fault signal can be effectively extracted, avoiding the interference of redundant information, and improving the accuracy and efficiency of fault diagnosis; by using the wavelet packet energy spectrum to obtain the energy characteristic representation of the fault signal, the differences between different fault lines and fault types can be effectively reflected. Then, the CNN network is used to realize the rapid and accurate identification of the energy characteristics of the short-circuit fault of the power transmission line, which provides a new idea for assisting the fault detection and troubleshooting work of the power grid dispatching personnel.

[0036] The present invention also provides a fault diagnosis system for a power transmission line, the system comprising:

[0037] A decomposition module is used to decompose and reconstruct the original fault signal using wavelet packet operation, obtain the wavelet packet decomposition tree structure, and calculate the wavelet packet energy of the sub-band data at the last layer of the decomposition tree structure;

[0038] A diversity processing module is used to construct a wavelet packet energy feature matrix, normalize the wavelet packet energy feature matrix, and perform diversity processing on the normalized data to divide it into training set samples and test set samples;

[0039] A training module, used for performing model training according to training set samples to obtain a fault line identification model and a fault type identification model;

[0040] The fault diagnosis module is used to diagnose transmission line faults based on test set samples, fault line identification model and fault type identification model, and identify the fault line and fault type.

[0041] Compared with the prior art, the beneficial effects of the electronic device provided by the present invention are the same as the beneficial effects of the fault diagnosis method for a power transmission line described in the above technical solution, and will not be described in detail here.

[0042] The present invention also provides an electronic device, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, and when the computer program is executed by the processor, the steps of any one of the above-mentioned methods for diagnosing faults in power transmission lines are implemented.

[0043] Compared with the prior art, the beneficial effects of the electronic device provided by the present invention are the same as the beneficial effects of the fault diagnosis method for a power transmission line described in the above technical solution, and will not be described in detail here.

[0044] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the steps in any one of the above-mentioned methods for diagnosing faults in a power transmission line are implemented.

[0045] Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the present invention are the same as the beneficial effects of the fault diagnosis method for a power transmission line described in the above technical solution, and are not elaborated herein.

[0046] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0048] Figure 1 A flow chart of a fault diagnosis method for a power transmission line provided by an embodiment of the present invention is shown;

[0049] Figure 2 A schematic diagram of a fault diagnosis method for a power transmission line provided by an embodiment of the present invention is shown;

[0050] Figure 3 A schematic diagram of wavelet packet decomposition provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0051] In the description of the present invention, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0052] The "multiple" mentioned in this embodiment refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The words "exemplary" or "for example" are used to indicate examples, illustrations or explanations, and are intended to present related concepts in a specific way, and should not be interpreted as being more preferred or more advantageous than other embodiments or design solutions.

[0053] Figure 1 A flow chart of a method for diagnosing a fault in a power transmission line provided by an embodiment of the present invention is shown. Figure 2 FIG. 1 is a schematic diagram showing a method for diagnosing a fault in a power transmission line according to an embodiment of the present invention. Figure 1-Figure 2 As shown, the method includes:

[0054] Step S1: Figure 3 As shown, the original fault signal is decomposed and reconstructed by wavelet packet operation, the wavelet packet decomposition tree structure is obtained, and the wavelet packet energy of the sub-band data at the last layer of the decomposition tree structure is calculated.

[0055] It should be noted that formula (1) is used to orthogonally decompose the original fault signal into several frequency segments at different scales. Decomposed and

[0056]

[0057] The number of decomposition layers l is determined according to the complexity of the fault data. Wavelet packet decomposition can fully enhance the detail signal of the high-frequency part and improve the resolution of the high-frequency band.

[0058] Formula (2) is used to reconstruct the relatively important frequency bands in the decomposed sequence. The reconstructed signal frequency band is simpler than the original signal, but the signal length does not change. and Back to

[0059]

[0060] in, is the original fault signal, h (2l-k) , g (2l-k) Respectively represent the low-pass and high-pass filter coefficients, k is the decomposition node number, l is the number of decomposition layers, They represent the low-frequency detail signal and high-frequency detail signal after decomposition. Wavelet packet operation can adaptively adapt signal features to obtain frequency bands that match their frequency distribution, thereby achieving effective feature extraction.

[0061] Wavelet packet operation can divide the original signal into multiple independent frequency bands. When a short circuit occurs in a transmission line, the energy of the line will change significantly due to the impact of the short circuit current. At the same time, the energy characteristics of each frequency band will also be different. Therefore, the nature of the fault can be identified by the energy distribution characteristics of each frequency band.

[0062] Calculate the last layer of the decomposition tree structure 2 l The wavelet packet energy E1, E2, ..., E j ,

[0063] j represents the number of frequency bands contained in the last layer of wavelet packet decomposition, j = 1, 2, 3, ..., 2 l , the wavelet packet energy formula is:

[0064]

[0065] Among them, x j (k) represents the discrete data obtained by batch collection, k is the decomposition node number, n is the number of sampling points in the batch collection process of input data, S j (t) represents the original input electrical quantity data of the transmission line short circuit fault, E j Represents the wavelet packet energy of the jth node in the last layer of wavelet packet decomposition.

[0066] Using wavelet packet operation to decompose and reconstruct the original fault signal can obtain more accurate and detailed fault signal characteristics, improve the accuracy of fault identification and operation efficiency

[0067] Step S2: construct a wavelet packet energy feature matrix, normalize the wavelet packet energy feature matrix, and perform data segmentation on the normalized data to divide it into training set samples and test set samples.

[0068] It should be noted that the transmission line system is complex. When a short circuit fault occurs, the energy corresponding to each frequency band will change differently. Therefore, the nature of the fault can be represented by constructing an energy characteristic matrix T. The nature of the fault mainly involves the fault line and the fault type. The formula is:

[0069]

[0070] in, It represents the energy value of the jth frequency band in the last layer of wavelet packet reconstruction, E j,max 、E j,min Respectively represent the minimum and maximum energy in each frequency band of the last layer, E j,m represents the normalized processing result of the energy value of the jth frequency band in the last layer, and T represents the energy feature vector of a single sample.

[0071] The normalized data is divided into training set data and test set data according to a specific ratio. The optimal ratio is 7:3. The data is divided into training set samples and test set samples according to the ratio of 7:3 and input into the fault identification network.

[0072] Step S3: Perform model training based on the training set samples to obtain a fault line identification model and a fault type identification model.

[0073] It should be noted that the step S3: performing model training according to the training set samples to obtain the fault line identification model and the fault type identification model includes:

[0074] Step S3.1: Construct a convolutional neural network, which includes an output layer, a convolution layer, a pooling layer, a fully connected layer, and an output layer. The weight and bias of the convolution layer are set to {w, b}, where w represents the weight and b represents the bias.

[0075] Step S3.2: Update the weights and biases of the convolutional neural network RMSprop difference according to the back-propagation method;

[0076] Specifically, it is achieved through the following steps:

[0077] The weights and biases of the convolutional neural network are initialized to {w(1), b(1)};

[0078] Set the number of initialization iterations h = 1;

[0079] The convolutional neural network is optimized using the expressions of weights and biases in the stochastic gradient descent method of the RMSprop differential convolutional neural network. The expression formula is:

[0080]

[0081] in, represents the weight of the ith region of the jth feature map of the first layer at the hth iteration, represents the bias corresponding to the jth feature map of the first layer of the hth iteration, η represents the learning rate, J(h) represents the objective function of the hth iteration, and They respectively represent the momentum information of the weight and bias of the first layer at the h-th iteration.

[0082] Update the number of iterations according to h=1+1;

[0083] The back-propagation method is used to update the weights and biases of the convolutional neural network RMSprop difference until the error gradient is satisfied.

[0084] Step S3.3: Input the training set samples into the convolutional neural network, perform network training based on the updated weights and biases, and obtain the fault line identification model and the fault type identification model.

[0085] Step S4: Perform transmission line fault diagnosis based on the test set samples, the fault line identification model and the fault type identification model to identify the fault line and the fault type.

[0086] It should be noted that the present invention involves multiple fault lines, and the fault line needs to be determined first before the fault type of the line can be further identified. Therefore, the data is first input into the fault line identification model to identify the fault line, and then the test set sample data is input into the corresponding fault type identification model according to the identification result of the fault line.

[0087] Compared with the prior art, the fault diagnosis method for a power transmission line provided by the embodiment of the present invention has the following beneficial effects:

[0088] Firstly, the frequency domain information of the fault current signal is extracted by wavelet packet operation, and then its energy characteristic signal is obtained and input into the CNN fault identification network to identify the fault. By adopting the wavelet packet energy spectrum method, the rich frequency domain information in the transmission line fault data can be effectively extracted, and the difference in energy distribution of different lines and different fault types can be reflected. By preliminarily extracting fault characteristics, it is helpful to improve the fault identification accuracy. By optimizing the CNN fault identification network structure and parameters, the fault line and fault type can be accurately identified.

[0089] An embodiment of the present invention further provides a fault diagnosis system for a power transmission line, the system comprising:

[0090] A decomposition module is used to decompose and reconstruct the original fault signal using wavelet packet operation, obtain the wavelet packet decomposition tree structure, and calculate the wavelet packet energy of the sub-band data at the last layer of the decomposition tree structure;

[0091] A diversity processing module is used to construct a wavelet packet energy feature matrix, normalize the wavelet packet energy feature matrix, and perform diversity processing on the normalized data to divide it into training set samples and test set samples;

[0092] A training module, used for performing model training according to training set samples to obtain a fault line identification model and a fault type identification model;

[0093] The fault diagnosis module is used to diagnose transmission line faults based on test set samples, fault line identification model and fault type identification model, and identify the fault line and fault type.

[0094] Compared with the prior art, the beneficial effects of the electronic device provided by the present invention are the same as the beneficial effects of the fault diagnosis method for a power transmission line described in the above technical solution, and will not be described in detail here.

[0095] In addition, an embodiment of the present invention further provides an electronic device, including a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor. The transceiver, the memory, and the processor are respectively connected via a bus. When the computer program is executed by the processor, each process of the above-mentioned embodiment of a fault diagnosis method for a power transmission line is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be described here.

[0096] In addition, an embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the various processes of the above-mentioned embodiment of a fault diagnosis method for a transmission line are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0097] Computer readable storage media include: permanent and non-permanent, removable and non-removable media, which are tangible devices that can retain and store instructions for use by instruction execution devices. Computer readable storage media include: electronic storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, and any suitable combination of the above. Computer readable storage media include: phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), non-volatile random access memory (NVRAM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassette storage, magnetic tape disk storage or other magnetic storage devices, memory sticks, mechanical encoding devices (such as punched cards or raised structures in grooves with instructions recorded thereon) or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined in the embodiments of the present invention, computer-readable storage media do not include temporary signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (such as light pulses passing through fiber optic cables), or electrical signals transmitted through wires.

[0098] In the several embodiments provided in the present application, it should be understood that the disclosed devices, electronic devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, or it can be an electrical, mechanical or other form of connection.

[0099] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one location or distributed on multiple network units. Some or all of the units may be selected according to actual needs to solve the problem to be solved by the embodiments of the present invention.

[0100] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0101] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention is essentially or part of the contribution to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (including: a personal computer, a server, a data center or other network device) to perform all or part of the steps of the method described in each embodiment of the present invention. The above-mentioned storage medium includes various media that can store program codes as listed above.

[0102] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technical solution that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A fault diagnosis method for a power transmission line, characterized in that: include: Step S1: Decompose and reconstruct the original fault signal using wavelet packet operation to obtain a wavelet packet decomposition tree structure, and calculate the wavelet packet energy of the sub-band data at the last layer of the decomposition tree structure; Step S2: constructing a wavelet packet energy feature matrix, normalizing the wavelet packet energy feature matrix, and performing a diversity process on the normalized data to divide it into a training set sample and a test set sample; Step S3: Perform model training based on training set samples to obtain a fault line identification model and a fault type identification model; Step S4: Perform transmission line fault diagnosis based on the test set samples, the fault line identification model and the fault type identification model to identify the fault line and the fault type.

2. A method for fault diagnosis of a power transmission line according to claim 1, characterized in that: The wavelet packet operation is used to decompose and reconstruct the original fault data to obtain a wavelet packet decomposition tree structure, including: Formula (1) is used to orthogonally decompose the original fault signal into several frequency segments at different scales. Decomposed and Formula (2) is used to reconstruct the relatively important frequency bands in the decomposed sequence, and Back to in, is the original fault signal, h (2l-k) , g (2l-k) Respectively represent the low-pass and high-pass filter coefficients, k is the decomposition node number, l is the number of decomposition layers, They respectively represent the low-frequency detail signal and the high-frequency detail signal after decomposition.

3. A fault diagnosis method for a power transmission line according to claim 1, characterized in that: The step of calculating the wavelet packet energy of the sub-band data at the last layer of the decomposition tree structure comprises: Calculate the last layer of the decomposition tree structure 2 l The wavelet packet energy E1, E2, ..., E j , j represents the number of frequency bands contained in the last layer of wavelet packet decomposition, j = 1, 2, 3, ..., 2 l , the wavelet packet energy formula is: Among them, x j (k) represents the discrete data obtained by batch collection, k is the decomposition node number, n is the number of sampling points in the batch collection process of input data, S j (t) represents the original input electrical quantity data of the transmission line short circuit fault, E j Represents the wavelet packet energy of the jth node in the last layer of wavelet packet decomposition.

4. A method for fault diagnosis of a power transmission line according to claim 1, characterized in that: The step of constructing a wavelet packet energy feature vector and normalizing the wavelet packet energy feature vector comprises: Construct the wavelet packet energy eigenvector T, the formula is: in, It represents the energy value of the jth frequency band in the last layer of wavelet packet reconstruction, E j,max 、E j,min Respectively represent the minimum and maximum energy in each frequency band of the last layer, E j,m represents the normalized processing result of the energy value of the jth frequency band in the last layer, and T represents the energy feature vector of a single sample.

5. A fault diagnosis method for a power transmission line according to claim 1, characterized in that: The normalized data is divided into training set samples and test set samples according to the preset ratio of 7:

3.

6. A method for fault diagnosis of a power transmission line according to claim 1, characterized in that: The step S3: performing model training according to the training set samples to obtain a fault line identification model and a fault type identification model, including: Step S3.1: construct a convolutional neural network, the neural network includes an input layer, a convolutional layer, a pooling layer, a fully connected layer, and an output layer, and the weight and bias of the convolutional layer are set to {w, b}, where w represents the weight and b represents the bias; Step S3.2: Update the weights and biases of the convolutional neural network RMSprop difference according to the back-propagation method; Step S3.3: Input the training set samples into the convolutional neural network, perform network training based on the updated weights and biases, and obtain the fault line identification model and the fault type identification model.

7. A method for fault diagnosis of a power transmission line according to claim 6, characterized in that: The method of updating the weights and biases of the convolutional neural network RMSprop difference according to the back propagation method includes: The weights and biases of the convolutional neural network are initialized to {w(1), b(1)}; Set the number of initialization iterations h = 1; The convolutional neural network is optimized using the expressions of weights and biases in the stochastic gradient descent method of the RMSprop differential convolutional neural network. Update the number of iterations according to h=1+1; The back-propagation method is used to update the weights and biases of the convolutional neural network RMSprop difference until the error gradient is satisfied.

8. A fault diagnosis system for a power transmission line, characterized in that: include: A decomposition module is used to decompose and reconstruct the original fault signal using wavelet packet operation, obtain the wavelet packet decomposition tree structure, and calculate the wavelet packet energy of the sub-band data at the last layer of the decomposition tree structure; A diversity processing module is used to construct a wavelet packet energy feature matrix, normalize the wavelet packet energy feature matrix, and perform diversity processing on the normalized data to divide it into training set samples and test set samples; A training module, used for performing model training according to training set samples to obtain a fault line identification model and a fault type identification model; The fault diagnosis module is used to diagnose transmission line faults based on test set samples, fault line identification model and fault type identification model, and identify the fault line and fault type.

9. An electronic device, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, wherein: When the computer program is executed by the processor, the steps of a fault diagnosis method for a power transmission line as claimed in any one of claims 1 to 7 are implemented.

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 a fault diagnosis method for a power transmission line as claimed in any one of claims 1 to 7 are implemented.

Images