A method, system and device for health diagnosis of power transmission and transformation equipment

Abstract

The present invention proposes a method, system and equipment for health diagnosis of power transmission and transformation equipment, which belongs to the technical field of data processing of power transmission and transformation equipment. The method comprises: obtaining state parameters of historical power transmission and transformation equipment, analyzing the state parameters of historical power transmission and transformation equipment to first determine the single state quantity weight coefficient, and then using the single state quantity weight coefficient to determine the single state quantity coupling weight coefficient; using the single state quantity coupling weight coefficient as the input of the health diagnosis model and the equipment fault type as the output to construct an integrated health diagnosis model; training the integrated health diagnosis model to obtain a target integrated health diagnosis model; inputting the single state quantity coupling weight coefficient of the current power transmission and transformation equipment into the target integrated health diagnosis model, and predicting and outputting the fault type of the current power transmission and transformation equipment. Based on this method, a system and equipment for health diagnosis of power transmission and transformation equipment are also proposed. The present invention integrates the advantages of the decision tree being able to process irrelevant features and the high accuracy of the artificial neural network, thereby improving the overall accuracy of diagnosis.

Description

Technical Field

[0001] The present invention belongs to the technical field of data processing of power transmission and transformation equipment, and in particular relates to a health diagnosis method, system and equipment for power transmission and transformation equipment. Background Art

[0002] The stable operation of power transmission and transformation equipment is the core foundation for the reliable operation of the power grid. Equipment failure is the main source of regional power grid blackouts, reduced power supply reliability, and social instability. With the development of smart grids and the construction of new power systems, the requirements for safe, controllable, strong and intelligent power transmission and transformation equipment are increasing. Real-time grasp of the health status of equipment and accurate early warning of equipment abnormalities are the decisive factors for ensuring the safe operation of the power grid, and are also a major difficulty in the energy industry.

[0003] In recent years, the number of power transmission and transformation equipment has grown rapidly, and the equipment status parameters have grown exponentially. With the transition of extreme weather from sporadic to normal, the traditional equipment status assessment method is mainly based on a few monitoring parameters of a single time section, and the degree of intelligence is low. In terms of technical solutions, the current power transmission and transformation equipment diagnosis method does not consider the correlation between equipment status parameters, and mostly adopts a single fault diagnosis model, which leads to the problem of low accuracy of the diagnosis model. Therefore, the timeliness, accuracy, and adaptability of the process of health diagnosis of power transmission and transformation equipment in the existing technology are low, and it is difficult to effectively grasp the health status of the equipment. The safe and stable operation of power transmission and transformation equipment faces huge challenges. Summary of the invention

[0004] In order to solve the above technical problems, the present invention proposes a method, system and device for health diagnosis of power transmission and transformation equipment, which comprehensively utilizes the advantages of decision trees being able to process irrelevant features and artificial neural networks being highly accurate, thereby improving the accuracy of the diagnosis model as a whole.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] A method for health diagnosis of power transmission and transformation equipment comprises the following steps:

[0007] Acquire state parameters of historical power transmission and transformation equipment, analyze the state parameters of the historical power transmission and transformation equipment by using a correlation analysis method, first determine a single state quantity weight coefficient of the historical power transmission and transformation equipment, and then use the single state quantity weight coefficient to determine a single state quantity coupling weight coefficient of the state parameters of the historical power transmission and transformation equipment;

[0008] An integrated health diagnosis model for health diagnosis is constructed by taking the single state quantity coupling weight coefficient as the input of the health diagnosis model and the equipment fault type as the output of the health diagnosis model; the integrated health diagnosis model is a comprehensive model combining an integrated regression decision tree model and an artificial neural network;

[0009] The integrated health diagnosis model is trained to obtain a target integrated health diagnosis model; the single state quantity coupling weight coefficient of the current power transmission and transformation equipment state parameter is input into the target integrated health diagnosis model, and the current power transmission and transformation equipment fault type is output through prediction by the target integrated health diagnosis model.

[0010] Furthermore, the process of analyzing the state parameters of the historical power transmission and transformation equipment by using the correlation analysis method to first determine the single state quantity weight coefficient of the historical power transmission and transformation equipment includes:

[0011] Determine the weight coefficient of a single state quantity:

[0012] ; (1)

[0013] in, is the number of status parameters of power transmission and transformation equipment; For the The state parameters of power transmission and transformation equipment, For the Single state quantity weight coefficient of state parameters of power transmission and transformation equipment; For the A single scoring value for a power transmission and transformation equipment status parameter; For the Conventional weight coefficients of state parameters of power transmission and transformation equipment; For the A single scoring value for a power transmission and transformation equipment status parameter; For the Conventional weight coefficients of state parameters of power transmission and transformation equipment.

[0014] Furthermore, the process of determining the single state quantity coupling weight coefficient of the historical power transmission and transformation equipment state parameter by using the single state quantity weight coefficient is as follows:

[0015] ; (2)

[0016] in, is the single state quantity coupling weight coefficient of the state parameter of the power transmission and transformation equipment; is the coupling function; ; Representative is The coupling score value of the state parameters of the power transmission and transformation equipment, Representative is The coupling score value of the state parameters of the power transmission and transformation equipment; For the The single state quantity weight coefficient of the state parameter of the power transmission and transformation equipment.

[0017] Furthermore, the method further comprises:

[0018] When the coupling degree between each equipment status parameter and meteorological and environmental factors is less than the preset standard, ;

[0019] When the coupling degree between each equipment status parameter and meteorological and environmental factors is greater than the preset standard, ;

[0020] When the single state quantity coupling weight coefficient of the device state parameter is equal to the constant weight coefficient, .

[0021] Furthermore, the process of training the integrated health diagnosis model to obtain a target integrated health diagnosis model includes:

[0022] The data set of the single-state quantity coupling weight coefficient is divided into a first data set and a second data set according to whether or not there is a label; and the first data set and the second data set are used to train an integrated regression decision tree model;

[0023] The trained integrated regression decision tree model is used to output a third data set; and the first data set and the third data set are used to train an artificial neural network to obtain a target integrated health diagnosis model.

[0024] Furthermore, the first data set is a labeled sample set; the second data set is an unlabeled sample set; and the third data set is a labeled sample set.

[0025] Furthermore, the regression decision tree model uses the least squares method to find the optimal split point, and the objective function is defined as:

[0026] ; (3)

[0027] in, is the input and output sample set; represents the input of the sample, The output or label representing the sample; is the first sub-sample set; is the second sub-sample set; for The optimal split point pair with the smallest time; For the The output value of a node.

[0028] Furthermore, the artificial neural network includes an input layer, two hidden layers and an output layer; The input of the output neuron is ;in, For the The output layer weights, is the offset.

[0029] Furthermore, the power transmission and transformation equipment includes a transformer, a circuit breaker, a disconnector and a busbar.

[0030] The present invention also proposes a health diagnosis system for power transmission and transformation equipment, including a preprocessing module, a model training module and a prediction output module:

[0031] The preprocessing module is used to obtain the state parameters of the historical power transmission and transformation equipment, analyze the state parameters of the historical power transmission and transformation equipment by using the correlation analysis method, first determine the single state quantity weight coefficient of the historical power transmission and transformation equipment, and then use the single state quantity weight coefficient to determine the single state quantity coupling weight coefficient of the state parameters of the historical power transmission and transformation equipment;

[0032] The model training module is used to construct an integrated health diagnosis model for health diagnosis by using the single state quantity coupling weight coefficient as the input of the health diagnosis model and the equipment fault type as the output of the health diagnosis model; the integrated health diagnosis model is a comprehensive model combining an integrated regression decision tree model and an artificial neural network;

[0033] The prediction output module is used to train the integrated health diagnosis model to obtain a target integrated health diagnosis model; the single state quantity coupling weight coefficient of the current power transmission and transformation equipment state parameter is input into the target integrated health diagnosis model, and the current power transmission and transformation equipment fault type is output through the prediction of the target integrated health diagnosis model.

[0034] Furthermore, in the preprocessing module, the process of first determining the single state quantity weight coefficient of the historical power transmission and transformation equipment by using the correlation analysis method to analyze the state parameters of the historical power transmission and transformation equipment includes:

[0035] Determine the weight coefficient of a single state quantity:

[0036] ; (1)

[0037] in, is the number of status parameters of power transmission and transformation equipment; For the The state parameters of power transmission and transformation equipment, For the Single state quantity weight coefficient of state parameters of power transmission and transformation equipment; For the A single scoring value for a power transmission and transformation equipment status parameter; For the Conventional weight coefficients of state parameters of power transmission and transformation equipment; For the A single scoring value for a power transmission and transformation equipment status parameter; For the Conventional weight coefficients of state parameters of power transmission and transformation equipment.

[0038] Furthermore, in the preprocessing module, the process of determining the single state quantity coupling weight coefficient of the historical power transmission and transformation equipment state parameter by using the single state quantity weight coefficient is as follows:

[0039] ; (2)

[0040] in, is the single state quantity coupling weight coefficient of the state parameter of the power transmission and transformation equipment; is the coupling function; ; Representative is The coupling score value of the state parameters of the power transmission and transformation equipment, Representative is The coupling score value of the state parameters of the power transmission and transformation equipment; For the The single state quantity weight coefficient of the state parameter of the power transmission and transformation equipment.

[0041] Furthermore, in the model training module, the process of training the integrated health diagnosis model to obtain the target integrated health diagnosis model includes:

[0042] The data set of the single-state quantity coupling weight coefficient is divided into a first data set and a second data set according to whether or not there is a label; and the first data set and the second data set are used to train an integrated regression decision tree model;

[0043] The trained integrated regression decision tree model is used to output a third data set; and the first data set and the third data set are used to train an artificial neural network to obtain a target integrated health diagnosis model.

[0044] Furthermore, the first data set is a labeled sample set; the second data set is an unlabeled sample set; and the third data set is a labeled sample set.

[0045] Furthermore, in the model training module, the artificial neural network includes 1 input layer, 2 hidden layers and 1 output layer; The input of the output neuron is ;in, For the The output layer weights, is the offset.

[0046] The present invention also proposes a power transmission and transformation equipment health diagnosis device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor can implement the power transmission and transformation equipment health diagnosis method when executing the program.

[0047] The effects provided in the content of the invention are only the effects of the embodiments, not all the effects of the invention. One of the above technical solutions has the following advantages or beneficial effects:

[0048] The present invention proposes a method, system and device for health diagnosis of power transmission and transformation equipment, which belongs to the technical field of data processing of power transmission and transformation equipment. The method includes the following steps: obtaining the state parameters of historical power transmission and transformation equipment, analyzing the state parameters of the historical power transmission and transformation equipment by using the association analysis method, first determining the single state quantity weight coefficient of the historical power transmission and transformation equipment, and then determining the single state quantity coupling weight coefficient of the state parameters of the historical power transmission and transformation equipment by using the single state quantity weight coefficient; using the single state quantity coupling weight coefficient as the input of the health diagnosis model and the equipment fault type as the output of the health diagnosis model, constructing an integrated health diagnosis model for health diagnosis; the integrated health diagnosis model is a comprehensive model combining an integrated regression decision tree model and an artificial neural network; training the integrated health diagnosis model to obtain a target integrated health diagnosis model; inputting the single state quantity coupling weight coefficient of the current power transmission and transformation equipment state parameter into the target integrated health diagnosis model, and outputting the current power transmission and transformation equipment fault type through prediction by the target integrated health diagnosis model. Based on a method for health diagnosis of power transmission and transformation equipment, a system and device for health diagnosis of power transmission and transformation equipment are also proposed. The present invention firstly adopts the association analysis method to determine the coupling relationship between the various state parameters of the equipment, so as to determine the influence weight of each state parameter on the diagnosis result - the single state quantity coupling weight coefficient, and then adopts the comprehensive diagnosis algorithm of integrated decision tree and artificial neural network to construct the health diagnosis model of power transmission and transformation equipment, comprehensively utilizing the advantages of decision tree's ability to process irrelevant features and artificial neural network's high accuracy, so as to improve the accuracy of the diagnosis model as a whole. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 A flow chart of a method for health diagnosis of power transmission and transformation equipment proposed in Example 1 of the present invention;

[0050] Figure 2 A schematic diagram of a training process of an integrated health diagnosis model in a health diagnosis method for power transmission and transformation equipment proposed in Embodiment 1 of the present invention;

[0051] Figure 3 A schematic diagram of model input and output in a power transmission and transformation equipment health diagnosis method proposed in Example 1 of the present invention;

[0052] Figure 4 A comparison diagram of the output fault type number and the actual fault type number proposed in Embodiment 1 of the present invention;

[0053] Figure 5 This is a schematic diagram of a health diagnosis system for power transmission and transformation equipment proposed in Example 2 of the present invention;

[0054] Figure 6 This is a schematic diagram of a health diagnosis device for power transmission and transformation equipment proposed in Example 3 of the present invention. DETAILED DESCRIPTION

[0055] In order to clearly illustrate the technical features of the present solution, the present invention is described in detail below through specific implementation methods and in conjunction with the accompanying drawings. The disclosure below provides many different embodiments or examples for realizing different structures of the present invention. In order to simplify the disclosure of the present invention, the components and settings of specific examples are described below. In addition, the present invention may repeat reference numbers and / or letters in different examples. This repetition is for the purpose of simplification and clarity, and does not itself indicate the relationship between the various embodiments and / or settings discussed. It should be noted that the components illustrated in the accompanying drawings are not necessarily drawn to scale. The present invention omits the description of known components and processing techniques and processes to avoid unnecessary limitations on the present invention.

[0056] Example 1

[0057] Embodiment 1 of the present invention proposes a method for health diagnosis of power transmission and transformation equipment, which is used to solve the technical problems existing in health diagnosis of power transmission and transformation equipment in the prior art. Figure 1 A flow chart of a method for health diagnosis of power transmission and transformation equipment proposed in Example 1 of the present invention;

[0058] In step S1, the state parameters of historical power transmission and transformation equipment are obtained. In the present application, the power transmission and transformation equipment includes transformers, circuit breakers, disconnectors, busbars, etc.

[0059] Taking transformer equipment as an example, the equipment status parameters include oil temperature, oil level, load rate, oil dissolved gas, on-load voltage regulation gear, etc.

[0060] In step S2, the state parameters of the historical power transmission and transformation equipment are analyzed by using a correlation analysis method to first determine the single state quantity weight coefficient of the historical power transmission and transformation equipment, and then the single state quantity coupling weight coefficient of the state parameters of the historical power transmission and transformation equipment is determined using the single state quantity weight coefficient.

[0061] By calculating the single-state coupling weight coefficient, the subjectivity of the traditional weight coefficient can be improved to a certain extent. That is, if the weight of a certain state parameter of the equipment to the equipment fault state is very small, when the state parameter seriously deviates from the normal value of operation, the equipment has actually reached a fault state, while the equipment state diagnosed by the conventional weight method may still be a normal state, so the diagnostic accuracy of the conventional weight method is low. Therefore, it is proposed to calculate the single-state coupling weight coefficient of the equipment:

[0062] The single state quantity coupling weight coefficient can reflect the balanced weight of a single state parameter that affects the overall state of the device.

[0063] The process of analyzing the state parameters of the historical power transmission and transformation equipment by using the correlation analysis method to first determine the single state quantity weight coefficient of the historical power transmission and transformation equipment includes:

[0064] Determine the weight coefficient of a single state quantity:

[0065] ; (1)

[0066] in, is the number of status parameters of power transmission and transformation equipment; For the The state parameters of power transmission and transformation equipment, For the Single state quantity weight coefficient of state parameters of power transmission and transformation equipment; For the A single scoring value for a power transmission and transformation equipment status parameter; For the Conventional weight coefficients of state parameters of power transmission and transformation equipment; For the A single scoring value for a power transmission and transformation equipment status parameter; For the Conventional weight coefficients of state parameters of power transmission and transformation equipment.

[0067] The process of determining the single state quantity coupling weight coefficient of the historical power transmission and transformation equipment state parameters using the single state quantity weight coefficient is as follows:

[0068] ; (2)

[0069] in, is the single state quantity coupling weight coefficient of the state parameter of the power transmission and transformation equipment; is the coupling function; ; Representative is The coupling score value of the state parameters of the power transmission and transformation equipment, Representative is The coupling score value of the state parameters of the power transmission and transformation equipment; For the The single state quantity weight coefficient of the state parameter of the power transmission and transformation equipment.

[0070] When the coupling degree between each equipment status parameter and meteorological and environmental factors is less than the preset standard, , such as the transformer on-load voltage adjustment gear; when the degree of coupling between the status parameters of each device and meteorological and environmental factors is greater than the preset standard, , such as transformer oil level; when the single state quantity coupling weight coefficient of the equipment state parameter is equal to the constant weight coefficient, .

[0071] In step S3, an integrated health diagnosis model for health diagnosis is constructed with the single state quantity coupling weight coefficient as the input of the health diagnosis model and the equipment fault type as the output of the health diagnosis model; the integrated health diagnosis model is a comprehensive model combining an integrated regression decision tree model and an artificial neural network.

[0072] The regression decision tree model uses the least squares method to find the optimal split point, and the objective function is defined as:

[0073] ; (3)

[0074] in, is the input and output sample set; represents the input of the sample, The output or label representing the sample; is the first sub-sample set; is the second sub-sample set; for The optimal split point pair with the smallest time; For the The output value of a node.

[0075] In this application, the artificial neural network includes 1 input layer, 2 hidden layers and 1 output layer; The input of the output neuron is ;in, For the The output layer weights, is the offset.

[0076] In this application The single state quantity coupling weight coefficient representing the state parameters of the power transmission and transformation equipment is: . Represents the equipment fault type number. Taking the transformer as an example, 0 represents the normal equipment status, 1 represents low-energy discharge, 2 represents arc discharge and overheating, 3 represents high-energy discharge, 4 represents high-temperature overheating, and 5 represents medium-low temperature overheating. z is the data pair variable.

[0077] In step S4, the integrated health diagnosis model is trained to obtain a target integrated health diagnosis model;

[0078] The process of training the integrated health diagnosis model to obtain the target integrated health diagnosis model includes: dividing the data set of the single-state quantity coupling weight coefficient into a first data set and a second data set according to whether there is a label; using the first data set and the second data set to train the integrated regression decision tree model at the same time; using the trained integrated regression decision tree model to output a third data set; and then using the first data set and the third data set to train the artificial neural network to obtain the target integrated health diagnosis model.

[0079] The first data set is a labeled sample set; the second data set is an unlabeled sample set; and the third data set is a labeled sample set.

[0080] Figure 2 Schematic diagram of the training process of the integrated health diagnosis model in the health diagnosis method for power transmission and transformation equipment proposed in Example 1 of the present invention; comprehensively utilizing the ability of decision trees to process irrelevant features and the advantages of high accuracy of artificial neural networks, firstly using a portion of labeled A and unlabeled sample data sets B to train a regression decision tree model, using the trained regression decision tree model to calculate the labels of a portion of the unlabeled data set B to form a sample data set B', and then using the sample data sets A and B' to train the neural network, according to Figure 2 The target integrated health diagnosis model is obtained by the training method in .

[0081] In step S5, the single state quantity coupling weight coefficient of the current power transmission and transformation equipment state parameter is input into the target integrated health diagnosis model, and the current power transmission and transformation equipment fault type is output through prediction by the target integrated health diagnosis model. Figure 3 A schematic diagram of model input and output in a power transmission and transformation equipment health diagnosis method proposed in Example 1 of the present invention;

[0082] In order to fully illustrate the implementation process of a power transmission and transformation equipment health diagnosis method proposed in Example 1 of the present invention, the simulation analysis of the transformer fault diagnosis proposed in the present invention is carried out by taking the percentage of gas content in transformer oil as an example. A total of 200 sets of analysis data of dissolved gas in faulty transformer oil determined after maintenance are collected as labeled sample data set A, as shown in Table 1 below for transformer fault data samples.

[0083] Table 1: Transformer fault data sample

[0084]

[0085] Table 1 contains five gases, hydrogen, methane, ethane, ethylene, and acetylene, generated in transformer oil, and five determined fault types. 200 sets of dissolved gas analysis data in faulty transformer oil are collected as unlabeled sample data set B. The single state quantity coupling weight coefficient vector of the state parameters of the five gas parameters of hydrogen, methane, ethane, ethylene, and acetylene is calculated according to formula (1) and formula (2), and then according to Figure 2 The process trains the DT-ANN integrated health diagnosis model, and the output of the model is the fault type number, specifically: No. 1: low-energy discharge, No. 2: arc discharge and overheating, No. 3: high-energy discharge, No. 4: high-temperature overheating, and No. 5: medium and low temperature overheating.

[0086] 40 sets of transformer gas state parameters of hydrogen, methane, ethane, ethylene and acetylene were selected as verification samples and input into the trained DT-ANN integrated health diagnosis model to calculate the comparison between the output fault type number and the actual fault type number. Figure 4 As shown, Figure 4 : is a comparison diagram between the output fault type number and the actual fault type number proposed in Example 1 of the present invention; Figure 4 It can be seen that among the 40 verification samples, only sample No. 13 has an error, that is, the accuracy of the power transmission and transformation equipment health diagnosis method proposed in Example 1 of the present invention can reach 97.5%.

[0087] Embodiment 1 of the present invention proposes a method for health diagnosis of power transmission and transformation equipment. First, the coupling relationship between the various state parameters of the equipment is determined by the association analysis method, so as to determine the weight of each state parameter on the diagnosis result - the single state quantity coupling weight coefficient. Then, a comprehensive diagnosis algorithm integrating decision trees and artificial neural networks is adopted to construct a health diagnosis model for power transmission and transformation equipment. The decision tree can process irrelevant features and the advantages of the high accuracy of artificial neural networks are comprehensively utilized to improve the accuracy of the diagnosis model as a whole.

[0088] Example 2

[0089] Based on the power transmission and transformation equipment health diagnosis method proposed in Example 1 of the present invention, Example 2 of the present invention further proposes a power transmission and transformation equipment health diagnosis system. Figure 5 This is a schematic diagram of a power transmission and transformation equipment health diagnosis system proposed in Example 2 of the present invention; the system includes a preprocessing module, a model training module and a prediction output module:

[0090] The preprocessing module is used to obtain the state parameters of the historical power transmission and transformation equipment, analyze the state parameters of the historical power transmission and transformation equipment by using the correlation analysis method, first determine the single state quantity weight coefficient of the historical power transmission and transformation equipment, and then use the single state quantity weight coefficient to determine the single state quantity coupling weight coefficient of the state parameters of the historical power transmission and transformation equipment;

[0091] The model training module is used to construct an integrated health diagnosis model for health diagnosis by taking the single state quantity coupling weight coefficient as the input of the health diagnosis model and the equipment fault type as the output of the health diagnosis model; the integrated health diagnosis model is a comprehensive model combining an integrated regression decision tree model and an artificial neural network;

[0092] The prediction output module is used to train the integrated health diagnosis model to obtain the target integrated health diagnosis model; the single state quantity coupling weight coefficient of the current power transmission and transformation equipment state parameter is input into the target integrated health diagnosis model, and the current power transmission and transformation equipment fault type is output through the prediction of the target integrated health diagnosis model.

[0093] During the implementation of the preprocessing module: power transmission and transformation equipment includes transformers, circuit breakers, disconnectors, busbars, etc. Taking transformer equipment as an example, the equipment status parameters include oil temperature, oil level, load rate, oil dissolved gas, on-load voltage regulation gear, etc.

[0094] The process of analyzing the state parameters of the historical power transmission and transformation equipment by using the correlation analysis method to first determine the single state quantity weight coefficient of the historical power transmission and transformation equipment includes:

[0095] Determine the weight coefficient of a single state quantity:

[0096] ; (1)

[0097] in, is the number of status parameters of power transmission and transformation equipment; For the The state parameters of power transmission and transformation equipment, For the Single state quantity weight coefficient of state parameters of power transmission and transformation equipment; For the A single scoring value for a power transmission and transformation equipment status parameter; For the Conventional weight coefficients of state parameters of power transmission and transformation equipment; For the A single scoring value for a power transmission and transformation equipment status parameter; For the Conventional weight coefficients of state parameters of power transmission and transformation equipment.

[0098] The process of determining the single state quantity coupling weight coefficient of the historical power transmission and transformation equipment state parameters using the single state quantity weight coefficient is as follows:

[0099] ; (2)

[0100] in, is the single state quantity coupling weight coefficient of the state parameter of the power transmission and transformation equipment; is the coupling function; ; Representative is The coupling score value of the state parameters of the power transmission and transformation equipment, Representative is The coupling score value of the state parameters of the power transmission and transformation equipment; For the The single state quantity weight coefficient of the state parameter of the power transmission and transformation equipment.

[0101] When the coupling degree between each equipment status parameter and meteorological and environmental factors is less than the preset standard, ;

[0102] When the coupling degree between each equipment status parameter and meteorological and environmental factors is greater than the preset standard, ; When the single state quantity coupling weight coefficient of the device state parameter is equal to the constant weight coefficient, .

[0103] The process of implementing the model training module includes: dividing the data set of single-state quantity coupling weight coefficients into a first data set and a second data set according to whether or not there is a label; using the first data set and the second data set to train the integrated regression decision tree model at the same time; using the trained integrated regression decision tree model to output a third data set; and then using the first data set and the third data set to train the artificial neural network to obtain the target integrated health diagnosis model.

[0104] The first data set is a labeled sample set; the second data set is an unlabeled sample set; and the third data set is a labeled sample set.

[0105] The regression decision tree model uses the least squares method to find the optimal split point, and the objective function is defined as:

[0106] ; (3)

[0107] in, is the input and output sample set; represents the input of the sample, The output or label representing the sample; is the first sub-sample set; is the second sub-sample set; for The optimal split point pair with the smallest time; For the The output value of a node.

[0108] The artificial neural network consists of 1 input layer, 2 hidden layers and 1 output layer; The input of the output neuron is ;in, For the The output layer weights, is the offset.

[0109] The process of realizing the prediction output module includes: after inputting the single state quantity coupling weight coefficient vector of its state parameter, the equipment fault type can be output.

[0110] Embodiment 2 of the present invention proposes a health diagnosis system for power transmission and transformation equipment. First, the coupling relationship between the various state parameters of the equipment is determined by the association analysis method, so as to determine the weight of each state parameter on the diagnosis result - the single state quantity coupling weight coefficient. Then, a comprehensive diagnosis algorithm integrating decision trees and artificial neural networks is adopted to construct a health diagnosis model for power transmission and transformation equipment. The decision tree can handle irrelevant features and the advantages of the high accuracy of artificial neural networks are comprehensively utilized to improve the accuracy of the diagnosis model as a whole.

[0111] Example 3

[0112] The present invention 3 also proposes a health diagnosis device for power transmission and transformation equipment, Figure 6 A schematic diagram of a power transmission and transformation equipment health diagnosis device proposed in Embodiment 3 of the present invention includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, a power transmission and transformation equipment health diagnosis method can be implemented, and the specific process includes:

[0113] In step S1, the state parameters of the historical power transmission and transformation equipment are obtained;

[0114] In step S2, the state parameters of the historical power transmission and transformation equipment are analyzed by using a correlation analysis method to first determine the single state quantity weight coefficient of the historical power transmission and transformation equipment, and then the single state quantity coupling weight coefficient of the state parameters of the historical power transmission and transformation equipment is determined by using the single state quantity weight coefficient;

[0115] In step S3, an integrated health diagnosis model for health diagnosis is constructed by taking the single state quantity coupling weight coefficient as the input of the health diagnosis model and the equipment fault type as the output of the health diagnosis model; the integrated health diagnosis model is a comprehensive model combining an integrated regression decision tree model and an artificial neural network;

[0116] In step S4, the integrated health diagnosis model is trained to obtain a target integrated health diagnosis model;

[0117] In step S5, the single state quantity coupling weight coefficient of the current power transmission and transformation equipment state parameter is input into the target integrated health diagnosis model, and the current power transmission and transformation equipment fault type is output through prediction by the target integrated health diagnosis model.

[0118] Embodiment 3 of the present invention proposes a health diagnosis device for power transmission and transformation equipment. First, the coupling relationship between the various state parameters of the equipment is determined by the association analysis method, so as to determine the weight of each state parameter on the diagnosis result - the single state quantity coupling weight coefficient. Then, a comprehensive diagnosis algorithm integrating decision trees and artificial neural networks is adopted to construct a health diagnosis model for power transmission and transformation equipment. The decision tree can handle irrelevant features and the advantages of the high accuracy of artificial neural networks are comprehensively utilized to improve the accuracy of the diagnosis model as a whole.

[0119] The description of the relevant parts of a power transmission and transformation equipment health diagnosis device provided in Example 3 of the present application can be found in the detailed description of the corresponding parts of a power transmission and transformation equipment health diagnosis method provided in Example 1 of the present application, and will not be repeated here.

[0120] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is any such actual relationship or order between these entities or operations. Moreover, the term "include", "comprise" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or equipment that includes a series of elements are inherent to the elements. In the absence of more restrictions, the elements limited by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or equipment that includes the elements. In addition, the above-mentioned technical solution provided in the embodiment of the present application is consistent with the corresponding technical solution in the prior art in principle, and the part is not described in detail, so as not to repeat too much.

[0121] Although the above describes the specific implementation of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. For those skilled in the art, other different forms of modifications or deformations can be made on the basis of the above description. It is not necessary and impossible to list all the implementation methods here. On the basis of the technical solution of the present invention, various modifications or deformations that can be made by those skilled in the art without creative work are still within the scope of protection of the present invention.

Claims

1. A method for health diagnosis of power transmission and transformation equipment, characterized in that: The following steps are involved: Acquire state parameters of historical power transmission and transformation equipment, analyze the state parameters of the historical power transmission and transformation equipment by using a correlation analysis method, first determine a single state quantity weight coefficient of the historical power transmission and transformation equipment, and then use the single state quantity weight coefficient to determine a single state quantity coupling weight coefficient of the state parameters of the historical power transmission and transformation equipment; The process of analyzing the state parameters of the historical power transmission and transformation equipment by using the correlation analysis method to first determine the single state quantity weight coefficient of the historical power transmission and transformation equipment includes: Determine the weight coefficient of a single state quantity: ；（1） in, is the number of status parameters of power transmission and transformation equipment; For the The state parameters of power transmission and transformation equipment, For the Single state quantity weight coefficient of state parameters of power transmission and transformation equipment; For the A single scoring value for a power transmission and transformation equipment status parameter; For the Conventional weight coefficients of state parameters of power transmission and transformation equipment; For the A single scoring value for a power transmission and transformation equipment status parameter; For the Conventional weight coefficients of state parameters of power transmission and transformation equipment; The process of determining the single state quantity coupling weight coefficient of the historical power transmission and transformation equipment state parameter by using the single state quantity weight coefficient is as follows: ；（2） in, is the single state quantity coupling weight coefficient of the state parameter of the power transmission and transformation equipment; is the coupling function; ; Representative is The coupling score value of the state parameters of the power transmission and transformation equipment, Representative is The coupling score value of the state parameters of the power transmission and transformation equipment; For the Single state quantity weight coefficient of state parameters of power transmission and transformation equipment; An integrated health diagnosis model for health diagnosis is constructed by taking the single state quantity coupling weight coefficient as the input of the health diagnosis model and the equipment fault type as the output of the health diagnosis model; the integrated health diagnosis model is a comprehensive model combining an integrated regression decision tree model and an artificial neural network; The integrated health diagnosis model is trained to obtain a target integrated health diagnosis model; the single-state quantity coupling weight coefficient of the current power transmission and transformation equipment state parameter is input into the target integrated health diagnosis model, and the current power transmission and transformation equipment fault type is output through prediction by the target integrated health diagnosis model; the process of training the integrated health diagnosis model to obtain the target integrated health diagnosis model includes: dividing the data set of the single-state quantity coupling weight coefficient into a first data set and a second data set according to whether there is a label; simultaneously using the first data set and the second data set to train an integrated regression decision tree model; using the trained integrated regression decision tree model to output a third data set; and then using the first data set and the third data set to train an artificial neural network to obtain the target integrated health diagnosis model.

2. A method for health diagnosis of power transmission and transformation equipment according to claim 1, characterized in that: The method further comprises: When the coupling degree between each equipment status parameter and meteorological and environmental factors is less than the preset standard, ; When the coupling degree between each equipment status parameter and meteorological and environmental factors is greater than the preset standard, ; When the single state quantity weight coefficient of the device state parameter is equal to the constant weight coefficient, .

3. A method for health diagnosis of power transmission and transformation equipment according to claim 1, characterized in that: The first data set is a labeled sample set; the second data set is an unlabeled sample set; and the third data set is a labeled sample set.

4. A method for health diagnosis of power transmission and transformation equipment according to claim 1, characterized in that: The regression decision tree model uses the least squares method to find the optimal split point, and the objective function is defined as: ；（3） in, is the input and output sample set; represents the input of the sample, The output or label representing the sample; is the first sub-sample set; is the second sub-sample set; for The optimal split point pair with the smallest time; For the The output value of a node.

5. A method for health diagnosis of power transmission and transformation equipment according to claim 1, characterized in that: The artificial neural network includes an input layer, two hidden layers and an output layer; The input of the output neuron is ;in, For the The output layer weights, is the offset.

6. A method for health diagnosis of power transmission and transformation equipment according to claim 1, characterized in that: The power transmission and transformation equipment includes a transformer, a circuit breaker, a disconnector and a busbar.

7. A health diagnosis system for power transmission and transformation equipment, characterized in that: Including preprocessing module, model training module and prediction output module: The preprocessing module is used to obtain the state parameters of the historical power transmission and transformation equipment, analyze the state parameters of the historical power transmission and transformation equipment by using the correlation analysis method, first determine the single state quantity weight coefficient of the historical power transmission and transformation equipment, and then use the single state quantity weight coefficient to determine the single state quantity coupling weight coefficient of the state parameters of the historical power transmission and transformation equipment; The process of analyzing the state parameters of the historical power transmission and transformation equipment by using the correlation analysis method to first determine the single state quantity weight coefficient of the historical power transmission and transformation equipment includes: Determine the weight coefficient of a single state quantity: ；（1） in, is the number of status parameters of power transmission and transformation equipment; For the The state parameters of power transmission and transformation equipment, For the Single state quantity weight coefficient of state parameters of power transmission and transformation equipment; For the A single scoring value for a power transmission and transformation equipment status parameter; For the Conventional weight coefficients of state parameters of power transmission and transformation equipment; For the A single scoring value for a power transmission and transformation equipment status parameter; For the Conventional weight coefficients of state parameters of power transmission and transformation equipment; The process of determining the single state quantity coupling weight coefficient of the historical power transmission and transformation equipment state parameter by using the single state quantity weight coefficient is as follows: ；（2） in, is the single state quantity coupling weight coefficient of the state parameter of the power transmission and transformation equipment; is the coupling function; ; Representative is The coupling score value of the state parameters of the power transmission and transformation equipment, Representative is The coupling score value of the state parameters of the power transmission and transformation equipment; For the Single state quantity weight coefficient of state parameters of power transmission and transformation equipment; The model training module is used to construct an integrated health diagnosis model for health diagnosis by using the single state quantity coupling weight coefficient as the input of the health diagnosis model and the equipment fault type as the output of the health diagnosis model; the integrated health diagnosis model is a comprehensive model combining an integrated regression decision tree model and an artificial neural network; The prediction output module is used to train the integrated health diagnosis model to obtain a target integrated health diagnosis model; the single state quantity coupling weight coefficient of the current power transmission and transformation equipment state parameter is input into the target integrated health diagnosis model, and the current power transmission and transformation equipment fault type is output through the prediction of the target integrated health diagnosis model; the process of training the integrated health diagnosis model to obtain the target integrated health diagnosis model includes: The data set of the single-state quantity coupling weight coefficient is divided into a first data set and a second data set according to whether or not there is a label; and the first data set and the second data set are used to train an integrated regression decision tree model; The trained integrated regression decision tree model is used to output a third data set; and the first data set and the third data set are used to train an artificial neural network to obtain a target integrated health diagnosis model.

8. A power transmission and transformation equipment health diagnosis system according to claim 7, characterized in that: The first data set is a labeled sample set; the second data set is an unlabeled sample set; and the third data set is a labeled sample set.

9. A power transmission and transformation equipment health diagnosis system according to claim 7, characterized in that: In the model training module, the artificial neural network includes 1 input layer, 2 hidden layers and 1 output layer; The input of the output neuron is ;in, For the The output layer weights, is the offset.

10. A health diagnosis device for power transmission and transformation equipment, characterized in that: It comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, a method for health diagnosis of power transmission and transformation equipment as claimed in any one of claims 1 to 6 can be implemented.

Images