A method for evaluating the state of an oil-immersed transformer

Abstract

The present application discloses a method for evaluating the state of an oil-immersed transformer, which relates to the technical field of power transformers. The method includes: inputting the test data of the fast-response items into a preset state factor calculation model for calculation to obtain the state factors of the fast-response items of the oil-immersed transformer, and determining the first operating state of the oil-immersed transformer; performing weighted summation on the same type of test data in the test data of the slow-response items and / or the test data of the maintenance items to obtain the state factors of the slow-response items of the oil-immersed transformer, and determining the second operating state of the oil-immersed transformer; simultaneously using a preset fault detection network to determine the fault diagnosis result of the oil-immersed transformer; based on the second operating state and the fault diagnosis result, verifying the first operating state, and determining the target operating state of the oil-immersed transformer. The present application can improve the accuracy of evaluating the operating state of the oil-immersed transformer.

Description

Technical Field

[0001] This application relates to the technical field of power transformers, and particularly to a method for evaluating the state of an oil-immersed transformer. Background Art

[0002] Oil-immersed power transformers can effectively transmit large-capacity electrical energy and provide stable and reliable power supply for industrial production and commercial activities. The operating state of power transformers is very important for the safe and efficient transmission of electrical energy. Therefore, it is necessary to obtain the operating state of transformers through operation and maintenance projects.

[0003] Currently, when evaluating the operating state of an oil-immersed transformer, it is usually evaluated by an evaluator based on the operating parameters of the oil-immersed transformer. However, this method is greatly affected by the subjective factors of the evaluator and it is difficult to accurately evaluate the operating state of the oil-immersed transformer. Summary of the Invention

[0004] In view of this, this application provides a method for evaluating the state of an oil-immersed transformer, and the main purpose is to improve the accuracy of evaluating the operating state of the oil-immersed transformer.

[0005] According to the first aspect of this application, a method for evaluating the state of an oil-immersed transformer is provided, and the method includes:

[0006] After testing the fast-response items involved in the operation and maintenance detection process of the oil-immersed transformer, input the test data of the fast-response items into a preset state factor calculation model for calculation to obtain the fast-response item state factor of the oil-immersed transformer;

[0007] Based on the fast-response item state factor, determine the first operating state of the oil-immersed transformer;

[0008] After testing the slow-response items and / or maintenance items involved in the operation and maintenance detection process of the oil-immersed transformer, perform weighted summation on the same type of test data in the test data of the slow-response items and / or the test data of the maintenance items to obtain the slow-response item state factor of the oil-immersed transformer;

[0009] Based on the slow-response item state factor, determine the second operating state of the oil-immersed transformer;

[0010] At the same time, input the test data of the slow-response items and / or the test data of the maintenance items into a preset fault detection network for fault detection to obtain the fault diagnosis result of the oil-immersed transformer and the fault location information when a fault exists. Among them, when the fault diagnosis result is that a fault exists, update the weights of the slow-response items and / or maintenance items involved in the fault diagnosis result;

[0011] Based on the second operating state and the fault diagnosis result, verify the first operating state, and determine the target operating state of the oil-immersed transformer according to the verification result.

[0012] According to a second aspect of the present application, there is provided an oil-immersed transformer state evaluation device, which includes:

[0013] A first calculation unit, which is used to input the test data of the fast response items involved in the operation and maintenance detection process of the oil-immersed transformer into a preset state factor calculation model for calculation after testing,

[0014] to obtain the fast response item state factor of the oil-immersed transformer;

[0015] A determination unit, which is used to determine the first operating state of the oil-immersed transformer based on the fast response item state factor;

[0016] A second calculation unit, which is used to perform weighted summation on the same type of test data in the test data of the slow response items and / or the overhaul items involved in the operation and maintenance detection process of the oil-immersed transformer after testing, to obtain the slow response item state factor of the oil-immersed transformer;

[0017] The determination unit is further used to determine the second operating state of the oil-immersed transformer based on the slow response item state factor;

[0018] A detection unit, which is used to input the test data of the slow response items and / or the overhaul items into a preset fault detection network for fault detection at the same time, to obtain the fault diagnosis result of the oil-immersed transformer and the fault location information when a fault exists, wherein when the fault diagnosis result is that a fault exists, update the weights of the slow response items and / or the overhaul items involved in the fault diagnosis result;

[0019] A verification unit, which is used to verify the first operating state based on the second operating state and the fault diagnosis result, and determine the target operating state of the oil-immersed transformer according to the verification result.

[0020] According to a third aspect of the present application, there is provided a storage medium, on which a computer program is stored, and when the program is executed by a processor, the above-mentioned oil-immersed transformer state evaluation method is implemented.

[0021] According to the fourth aspect of the present application, an electronic device is provided, including a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor. When the processor executes the program, the above-mentioned oil-immersed transformer state evaluation method is implemented.

[0022] By means of the above technical solution, an oil-immersed transformer state evaluation method provided by the present application calculates the rapid response item state factor by using a preset state factor calculation model, and determines the first operating state of the oil-immersed transformer, which can avoid human participation in the operation state evaluation of the oil-immersed transformer, thereby improving the accuracy of the operation state evaluation of the oil-immersed transformer. In addition, by calculating the slow response item state factor and determining the second operating state of the oil-immersed transformer, the present application can correct the first operating state of the oil-immersed transformer, thereby further improving the accuracy of the operation state evaluation of the oil-immersed transformer. Furthermore, by using a preset fault detection network, the present application determines whether there is a fault in the oil-immersed transformer and the fault location information when there is a fault, and can correct the first operating state of the oil-immersed transformer based on the fault diagnosis result, thereby ensuring the accuracy of the operation state evaluation result of the oil-immersed transformer. Further, the present application can also update the weights of the slow response items and / or maintenance items involved in the fault diagnosis result when there is a fault, thereby improving the calculation accuracy of the slow response item state factor and further ensuring the accuracy of the operation state evaluation result of the oil-immersed transformer.

[0023] The above description is only an overview of the technical solution of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of the present application more obvious and understandable, the specific embodiments of the present application are specifically given below. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The drawings described herein are used to provide a further understanding of the present application, and constitute a part of the present application. The schematic embodiments and descriptions thereof of the present application are used to explain the present application, and do not constitute an improper limitation of the present application. In the drawings:

[0025] Figure 1 A flowchart showing an oil-immersed transformer state evaluation method provided by an embodiment of the present application is shown;

[0026] Figure 2 A flowchart showing another oil-immersed transformer state evaluation method provided by an embodiment of the present application is shown;

[0027] Figure 3 A flowchart showing a model establishment process provided by an embodiment of the present application is shown;

[0028] Figure 4Shows a schematic diagram of the overall evaluation process provided by the embodiments of the present application;

[0029] Figure 5 Shows a structural schematic diagram of an oil-immersed transformer status evaluation device provided by the embodiments of the present application. Detailed implementation manners

[0030] The present application will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments. It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments can be combined with each other.

[0031] The prior art is greatly affected by subjective factors of the evaluators and it is difficult to accurately evaluate the operating status of oil-immersed transformers.

[0032] To solve the above problems, the embodiments of the present invention provide an oil-immersed transformer status evaluation method, as Figure 1 shown, the method includes:

[0033] Step 101, after testing the rapid response items involved in the operation and maintenance detection process of the oil-immersed transformer, input the test data of the rapid response items into a preset state factor calculation model for calculation to obtain the rapid response item status factor of the oil-immersed transformer.

[0034] Among them, the operation and maintenance detection items of the oil-immersed transformer include insulation oil dielectric loss, water content in oil, oil breakdown voltage, acid value detection in oil, particulate matter detection in oil, furfural content determination in oil, dissolved gas detection in oil (including H 2 、CH 4 、C 2 H 6 、C 2 H 4 、C 2 H 2 、CO、CO 2 and O 2Content detection), insulation resistance test, partial discharge test, dielectric loss test, core grounding current, core insulation resistance, winding DC resistance difference, oil-paper insulation polymerization degree, isothermal relaxation current detection, load rate, etc. According to the length of the test time interval of the operation and maintenance project, the project is divided into quick response projects, slow response projects and maintenance projects. Among them, according to the implementation cycle of the power company's operation and maintenance inspection projects, the detection interval of dissolved gas in oil is relatively short, which belongs to the quick response project; the insulation oil dielectric loss, oil acid value detection, oil particle detection, furfural content determination in oil, insulation resistance test, partial discharge test, dielectric loss test, core grounding current, core insulation resistance, winding DC resistance difference and other inspection items have a longer test interval, which belongs to the slow response project; because the oil-paper insulation polymerization degree and isothermal relaxation current detection items cannot be carried out when the transformer is in operation, they can only be carried out after the transformer is out of operation, which belongs to the maintenance project. The test data of the quick response project includes H 2 , CH 4 , C 2 H 6 , C 2 H 4 , C 2 H 2 ,CO,CO 2 and O 2 In addition, the preset state factor calculation model can specifically be a neural network model.

[0035] The embodiment of the present invention is mainly applicable to the scenario of evaluating the operating status of an oil-immersed transformer. The execution subject of the embodiment of the present invention is a device or equipment capable of evaluating the operating status of an oil-immersed transformer, which can be specifically set on the server side.

[0036] Since the detection cycle of the quick response project is short, the detection frequency of the quick response project is high in the daily operation of the oil-immersed transformer, such as once every hour. In order to avoid human participation in the operation status evaluation of the oil-immersed transformer, thereby affecting the accuracy of the operation status evaluation of the oil-immersed transformer, the embodiment of the present invention calculates the quick response project state factor of the oil-immersed transformer based on the test data of the quick response project and the preset state factor calculation model, so as to determine the first operation state of the oil-immersed transformer based on the quick response project state factor. For this process, step 101 specifically includes: inputting the dissolved gas content in the oil into the preset neural network model for calculation to obtain the quick response project state factor of the oil-immersed transformer. Among them, the preset neural network model includes an input layer, a first hidden layer, a second hidden layer and an output layer.

[0037] When calculating the rapid response project status factor of an oil-immersed transformer, the dissolved gas content in the oil is input into the first hidden layer through the input layer for calculation to obtain the output vector of the first hidden layer; the output vector of the first hidden layer is input into the second hidden layer for calculation to obtain the output vector of the second hidden layer; the output vector of the second hidden layer is input into the output layer for calculation to obtain the rapid response project status factor of the oil-immersed transformer.

[0038] Specifically, the dissolved gas content in the oil is used as the input vector x = [x 1 , x 2 , x 3 , x 4 , …, x i , where x i represents the content of a certain gas. For example, the concentration of H 2 , the concentration of CH 4 , the concentration of C 2 H 6 , the concentration of C 2 H 4 , etc. Taking the sigmoid function as an example of the activation function, the input vector is input into the first hidden layer through the input layer for calculation, and the output vector of the first hidden layer is: h 1 = sigmoid(W 1 x + b 1 ), where x is the input vector, h 1 is the output vector of the first hidden layer, W 1 is the weight matrix of the first hidden layer, b 1 is the bias vector of the first hidden layer, and sigmoid() is the activation function.

[0039] After that, the output vector of the first hidden layer is input into the second hidden layer for calculation, and the output vector of the second hidden layer is: h 2 = sigmoid(W 2 h 1 + b 2 ), where h 1 is the output vector of the first hidden layer, h 2 is the output vector of the second hidden layer, W 2 is the weight matrix of the second hidden layer, b 2 is the bias vector of the second hidden layer, and sigmoid() is the activation function.

[0040] Then, the output vector of the second hidden layer is calculated through the output layer, and the final output is y = sigmoid(W 3 h 2 + b 3), where h 2 is the output vector of the second hidden layer, W 3 is the weight matrix of the third hidden layer, y is the quick response project status factor, and b 3 is the bias vector of the output layer, and sigmoid() is the activation function.

[0041] It should be noted that the dimensions of the weight matrices and bias vectors of the first hidden layer, the second hidden layer, and the output layer are determined according to the number of elements in each layer.

[0042] Step 102: Determine the first operating state of the oil-immersed transformer based on the quick response project status factor.

[0043] Among them, the first operating state of the oil-immersed transformer includes a relatively good operating state, aging, severe aging, etc.

[0044] For the embodiments of the present invention, in order to determine the first operating state of the oil-immersed transformer, step 102 specifically includes: determining the first evaluation level of the oil-immersed transformer based on the state interval where the quick response project status factor is located; and determining the first operating state of the oil-immersed transformer according to the first evaluation level.

[0045] Specifically, the value range of the quick response project status factor can be divided into multiple state intervals in advance. Different state intervals correspond to different evaluation levels, and different evaluation levels correspond to different operating states. For example, the value range of the quick response project status factor is 0-100 points. The range of 0-100 points is divided into multiple state intervals, which are above 80 points, from 40 to 80 points, and below 40 points. If the calculated quick response project status factor is in the state interval above 80 points, it is determined that the corresponding evaluation level is A level, indicating that the operating state of the oil-immersed transformer is good and there is no aging problem; if the calculated quick response project status factor is in the state interval from 40 to 80 points, it is determined that the corresponding evaluation level is B level, indicating that the oil-immersed transformer has aging and requires attention from maintenance personnel; if the calculated quick response project status factor is in the state interval above 80 points, it is determined that the corresponding evaluation level is C level, indicating that the oil-immersed transformer is severely aged and needs to be immediately stopped for maintenance.

[0046] It should be noted that the setting of the state interval, the evaluation level, and the operating state can be set according to actual business requirements, and the embodiments of the present invention do not make specific limitations here.

[0047] Step 103: After testing the slow response items and / or maintenance items involved in the operation and maintenance detection of the oil-immersed transformer, perform weighted summation on the same type of test data in the test data of the slow response items and / or the test data of the maintenance items to obtain the slow response item status factor of the oil-immersed transformer.

[0048] For the embodiments of the present invention, since the detection cycles of the slow response items and / or maintenance items are relatively long, the detection frequencies of the slow response items and / or maintenance items are relatively low during the daily operation of the oil-immersed transformer, such as once every 3 months, half a year, or one year. In addition, each item in the slow response items and / or maintenance items is usually not detected on the same day. Therefore, weighted summation can be performed on the test data of the same batch of detections, that is, the same type of test data, to obtain the slow response item status factor.

[0049] Taking the insulation resistance test as an example, the weighted summation of the measured insulation resistance values is performed to obtain the slow response item status factor. The specific formula is as follows:

[0050]

[0051] Among them, F slowreact is the slow response item status factor, W i,old is the weight corresponding to the i-th insulation resistance detection item, and R i is the resistance value of the i-th insulation resistance detected.

[0052] Step 104: Based on the slow response item status factor, determine the second operating state of the oil-immersed transformer.

[0053] Among them, the first operating states of the oil-immersed transformer include good operating state, aging, severe aging, etc.

[0054] For the embodiments of the present invention, in order to determine the first operating state of the oil-immersed transformer, step 102 specifically includes: determining the second evaluation grade of the oil-immersed transformer based on the state interval where the slow response item status factor is located; and determining the second operating state of the oil-immersed transformer according to the second evaluation grade.

[0055] Specifically, the value range of the slow reaction project status factor can be divided into multiple status intervals in advance. Different status intervals correspond to different evaluation levels, and different evaluation levels correspond to different operating states. For example, the value range of the slow reaction project status factor is 0-100 points. The range of 0-100 points is divided into multiple status intervals, namely above 80 points, 40-80 points, and below 40 points. If the calculated slow reaction project status factor is in the status interval above 80 points, it is determined that the corresponding evaluation level is A level, indicating that the operation state of the oil-immersed transformer is good and there is no aging problem. If the calculated slow reaction project status factor is in the status interval of 40-80 points, it is determined that the corresponding evaluation level is B level, indicating that the oil-immersed transformer has aging and requires attention from the operation and maintenance personnel. If the calculated slow reaction project status factor is in the status interval above 80 points, it is determined that the corresponding evaluation level is C level, indicating that the oil-immersed transformer is severely aged and needs to be stopped for maintenance immediately.

[0056] It should be noted that the setting of the status interval, evaluation level, and operating state can be set according to actual business requirements, and the embodiments of the present invention do not make specific limitations here.

[0057] Step 105: At the same time, input the test data of the slow reaction project and / or the test data of the maintenance project into a preset fault detection network for fault detection, to obtain the fault diagnosis result of the oil-immersed transformer and the fault location information when a fault occurs.

[0058] Among them, each node in the preset fault detection network is composed of a determination condition. In addition, when the fault diagnosis result is that a fault exists, update the weights of the slow reaction project and / or the maintenance project involved in the fault diagnosis result.

[0059] For the slow reaction project and / or the maintenance project, the embodiments of the present invention can also determine whether there is a fault in the oil-immersed transformer and where the fault occurs through the test data of the slow reaction project and / or the test data of the maintenance project, and a preset fault detection network.

[0060] Specifically, the test data of the slow reaction project and / or the test data of the maintenance project can be input into a preset fault detection network, and sequentially determine whether the test data of the slow reaction project and / or the test data of the maintenance project meet the determination conditions of each node in the network. If the determination condition of a certain node is not met, it indicates that there is a fault in the oil-immersed transformer. At the same time, since the manifestation forms of the oil-immersed transformer are different for different faults, when there is a fault in the oil-immersed transformer, the cause and location of the fault of the oil-immersed transformer can be determined according to the determination conditions that are not met in the preset fault detection network.

[0061] Step 106: Based on the second operating state and the fault diagnosis result, verify the first operating state, and determine the target operating state of the oil-immersed transformer according to the verification result.

[0062] For the embodiments of the present invention, since there may be errors in the monitoring of the content of each gas dissolved in oil, resulting in inaccurate results of the first operating state, it is necessary to verify the first operating state based on the second operating state and the fault diagnosis result obtained from the test data of the slow reaction items and / or the test data of the maintenance items.

[0063] Specifically, if the first operating state is consistent with the second operating state or the fault diagnosis result, it indicates that the evaluated first operating state is accurate, and directly determine the first operating state as the target operating state of the oil-immersed transformer; if the first operating state is inconsistent with the second operating state and the fault diagnosis result, it indicates that the evaluated first operating state is inaccurate, and determine the second operating state as the target operating state of the oil-immersed transformer.

[0064] An oil-immersed transformer state evaluation method provided by the embodiments of the present invention can avoid human participation in the evaluation of the operating state of the oil-immersed transformer by calculating the state factors of the fast reaction items using a preset state factor calculation model and determining the first operating state of the oil-immersed transformer, thereby improving the accuracy of the evaluation of the operating state of the oil-immersed transformer. In addition, the embodiments of the present invention can correct the first operating state of the oil-immersed transformer by calculating the state factors of the slow reaction items and determining the second operating state of the oil-immersed transformer, thereby further improving the accuracy of the evaluation of the operating state of the oil-immersed transformer. Moreover, the embodiments of the present invention can determine whether there is a fault in the oil-immersed transformer and the fault location information when there is a fault by using a preset fault detection network, and can correct the first operating state of the oil-immersed transformer based on the fault diagnosis result, thereby ensuring the accuracy of the evaluation result of the operating state of the oil-immersed transformer. Further, the embodiments of the present invention can also update the weights of the slow reaction items and / or the maintenance items involved in the fault diagnosis result when there is a fault, thereby improving the calculation accuracy of the state factors of the slow reaction items and further ensuring the accuracy of the evaluation result of the operating state of the oil-immersed transformer.

[0065] Further, as a refinement and extension of the specific implementation manner of the above embodiment, in order to fully illustrate the implementation manner of this embodiment, this embodiment also provides another oil-immersed transformer state evaluation method, as Figure 2 shown, this method includes:

[0066] Step 201: After testing the fast - response items involved in the operation and maintenance detection of the oil - immersed transformer, input the test data of the fast - response items into a preset state factor calculation model for calculation to obtain the state factor of the fast - response items of the oil - immersed transformer.

[0067] Among them, the preset state factor calculation model is specifically a neural network model.

[0068] For the embodiments of the present invention, before calculating the state factor of the fast - response items of the oil - immersed transformer using the preset state factor calculation model, it is necessary to train the preset state factor calculation model, as Figure 3 shown. Specifically, it is necessary to collect the historical test data of the fast - response items, construct a sample training set, and train the neural network model based on this sample training set to obtain the preset state factor calculation model. Among them, the loss function during the training process is the mean - square error function, which is specifically as follows:

[0069]

[0070] Among them, L is the mean - square error, y i represents the state factor of the fast - response items output by the model corresponding to the i - th sample data, and y true,i represents the true state factor of the fast - response items corresponding to the i - th sample data.

[0071] When training the neural network model according to the mean - square error loss function, the gradient descent method or its variant (such as the Adam optimizer) is used to update the weights and biases to minimize the loss function. Among them, the process of updating the weights and biases can be expressed as: Among them, η is the learning rate. The training process continues until the loss drops below a specified value. At this time, the obtained and are the structural parameters of the model based on the state factor of the fast - response items.

[0072] In actual calculation, first normalize the test data of the fast - response items, and then input the normalized test data into the preset state factor calculation model for calculation to obtain the state factor of the fast - response items Among them, and are the weight and bias parameters of the preset state factor calculation model, and x is the content of dissolved gases in oil.

[0073] Step 202: Based on the state factor of the fast - response items, determine the first operating state of the oil - immersed transformer.

[0074] In an embodiment of the present invention, after calculating the fast response project status factor, based on the fast response project status factor, the first operating state of the oil-immersed transformer is determined. The specific process is exactly the same as that of step 102 and will not be elaborated here.

[0075] Further, in an embodiment of the present invention, an early warning range and an early warning value can also be set according to the fast response project status factor obtained from historical operation and maintenance data. If during subsequent operation and maintenance, the fast response project status factor F fastreact enters the early warning range from the safe range, the system will send an early warning message. If it remains in the early warning range for a long time, the operation and maintenance personnel can be reminded to check for possible faults or perform optimization operations. If during subsequent operation and maintenance, F fastreact reaches above the early warning value, power outage for maintenance is required.

[0076] Step 203: After testing the slow response items and / or maintenance items involved in the operation and maintenance detection of the oil-immersed transformer, based on the weight corresponding to the slow response item and / or the weight corresponding to the maintenance item, the same type of test data in the test data of the slow response item and / or the test data of the maintenance item is weighted and summed to obtain the slow response project status factor of the oil-immersed transformer.

[0077] Taking the insulation resistance test as an example, the measured insulation resistance values are weighted and summed to obtain the slow response project status factor. The specific formula is as follows:

[0078]

[0079] Among them, F slowreact is the slow response project status factor, W i,old is the weight corresponding to the i-th insulation resistance detection item, and R i is the resistance value of the i-th insulation resistance detected.

[0080] Step 204: Based on the slow response project status factor, determine the second operating state of the oil-immersed transformer.

[0081] In an embodiment of the present invention, after calculating the slow response project status factor, based on the slow response project status factor, the second operating state of the oil-immersed transformer is determined. The specific process is exactly the same as that of step 104 and will not be elaborated here.

[0082] Step 205: At the same time, input the test data of the slow response item and / or the test data of the maintenance item into a preset fault detection network for fault detection to obtain the fault diagnosis result of the oil-immersed transformer and the fault location information when a fault exists.

[0083] Wherein, when the fault diagnosis result indicates the existence of a fault, the weights of the slow response items and / or maintenance items involved in the fault diagnosis result are updated. In addition, the preset fault detection network includes a first detection network and a second detection network. Specifically, the first detection network may be an Alpha network, and the second detection network may be a Beta network.

[0084] For the embodiments of the present invention, expert knowledge, experience, and decision-making logic can be collected for the slow response items and / or maintenance items and used as original rules. Then, the original rules are transformed into a form that can be processed by a computer, such as rule representation, semantic network, frame representation, or logical representation, to form a knowledge base, as Figure 3 shown. Then, an inference engine (preset fault detection network) and an interpreter are constructed based on the knowledge base. The interpreter is used to analyze and parse the corresponding test data and decision conditions in the event of a fault, so as to locate the fault.

[0085] When performing fault diagnosis using the preset fault detection network, step 205 specifically includes: inputting the test data of the slow response items and the test data of the maintenance items into the first detection network to obtain the first conditional determination results corresponding to each node in the first detection network, where each node in the first detection network corresponds to a single determination condition; inputting the first conditional determination results into the second detection network to obtain the second conditional determination results corresponding to each node in the second detection network, where each node in the second detection network corresponds to at least two determination conditions; if any of the conditional determination results corresponding to the nodes in the second conditional determination results is abnormal, it is determined that the oil-immersed transformer has a fault, and the test data of the slow response items and / or the test data of the maintenance items involved in the abnormal conditional determination results, as well as the determination conditions, are added to the conflict set; based on the test data of the slow response items and / or the test data of the maintenance items in the conflict set, as well as the determination conditions, the fault location information of the oil-immersed transformer is analyzed.

[0086] Specifically, the test data of the slow response items and the test data of the maintenance items are input into the Alpha network, and each node is traversed in turn to perform a single-condition determination, obtaining the first conditional determination results corresponding to each node in the Alpha network and storing them. Then, the first conditional determination results are input into the Beta network, and each node is traversed in turn to perform a multi-condition combination determination. When performing the combination determination, the Beta network nodes can use the intermediate results passed down from the Alpha network and directly perform the condition combination determination based on each intermediate result, thereby reducing the amount of calculation and improving the efficiency of fault diagnosis determination.

[0087] If the test data of a slow - response item and / or the test data of an overhaul item evaluated by any node in the Beta network do not meet a certain determination condition, it is determined that the oil - immersed transformer has a fault, and the determination condition of this node, as well as the relevant test data of the slow - response item and / or the test data of the overhaul item, are added to the conflict set, as Figure 4 shown. Then, the interpreter analyzes the test data of the slow - response item and / or the test data of the overhaul item in the conflict set, as well as the determination condition, so as to determine the fault location information of the oil - immersed transformer, that is, specifically what the fault is and where the fault occurs. Since for different faults, the forms shown by the oil - immersed transformer are different. For example, for fault A, the test data of slow - response item a and the test data of slow - response item b do not meet the determination condition; for fault B, the test data of slow - response item b, the test data of slow - response item c, and the test data of slow - response item d do not meet the determination condition. Therefore, the interpreter can locate where the oil - immersed transformer has a fault and the type of the fault according to the abnormal test data and the determination condition in the conflict set, which can play a role in assisting decision - making and guiding decision - making.

[0088] In some embodiments, the weights of the slow - response items and / or the overhaul items in the conflict set can also be updated. The method for this process includes: updating the weights of the slow - response items and / or the overhaul items in the conflict set to obtain the updated weights; based on the updated weights, performing a weighted sum on the same type of test data in the test data of the slow - response items and / or the test data of the overhaul items to obtain a recalculated state factor of the slow - response items. Taking the insulation resistance test as an example, the specific formula is as follows:

[0089]

[0090] where, F slowreact is the recalculated state factor of the slow - response items, W i,new is the updated weight of the i - th insulation resistance detection item, and R i is the resistance value of the i - th insulation resistance detected.

[0091] By updating the weights of the slow - response items and / or the overhaul items in the conflict set in the embodiments of the present invention, the calculation accuracy of the state factor of the slow - response items can be improved, thereby improving the accuracy of the operation state evaluation of the oil - immersed transformer.

[0092] Step 206: If the first operating state is different from the second operating state and the fault diagnosis result, determine that the first operating state is inaccurate. Based on the second operating state and the fault diagnosis result, determine the target operating state of the oil-immersed transformer, and update the preset state factor calculation model.

[0093] For the embodiments of the present invention, if the first operating state is inconsistent with the second operating state and the fault diagnosis result, it indicates that the evaluated first operating state is inaccurate. At this time, the target operating state of the oil-immersed transformer can be comprehensively determined according to the second operating state and the fault diagnosis result. For example, if the second operating state is that the oil-immersed transformer is severely aged and the fault detection result is that there is a fault in the oil-immersed transformer, and the two detection results are consistent at this time, the second operating state can be determined as the target operating state of the oil-immersed transformer. In addition, the preset state factor calculation model can be continuously updated and trained to improve the calculation accuracy of the fast-response project state factor.

[0094] Another oil-immersed transformer state evaluation method provided by the embodiments of the present invention can calculate the fast-response project state factor by using a preset state factor calculation model and determine the first operating state of the oil-immersed transformer, which can avoid human participation in the operating state evaluation of the oil-immersed transformer, thereby improving the accuracy of the operating state evaluation of the oil-immersed transformer. In addition, by calculating the slow-response project state factor and determining the second operating state of the oil-immersed transformer, the embodiments of the present invention can correct the first operating state of the oil-immersed transformer, thereby further improving the accuracy of the operating state evaluation of the oil-immersed transformer. Moreover, by using a preset fault detection network to determine whether there is a fault in the oil-immersed transformer and the fault location information when there is a fault, the embodiments of the present invention can correct the first operating state of the oil-immersed transformer based on the fault diagnosis result, thereby ensuring the accuracy of the operating state evaluation result of the oil-immersed transformer. Further, the embodiments of the present invention can also update the weights of the slow-response projects and / or maintenance projects involved in the fault diagnosis result when there is a fault, thereby improving the calculation accuracy of the slow-response project state factor and further ensuring the accuracy of the operating state evaluation result of the oil-immersed transformer.

[0095] Further, as Figure 1 and Figure 2 a specific implementation of the method shown, this embodiment provides an oil-immersed transformer state evaluation device, as Figure 5 shown. The device includes: a first calculation unit 31, a determination unit 32, a second calculation unit 33, a detection unit 34, and a verification unit 35.

[0096] The first calculation unit 31 can be used to input the test data of the fast response items involved in the operation and maintenance detection process of the oil-immersed transformer into a preset state factor calculation model for calculation, so as to obtain the state factor of the fast response items of the oil-immersed transformer.

[0097] The determination unit 32 can be used to determine the first operating state of the oil-immersed transformer based on the state factor of the fast response items.

[0098] The second calculation unit 33 can be used to perform weighted summation on the same type of test data in the test data of the slow response items and / or the overhaul items involved in the operation and maintenance detection process of the oil-immersed transformer after testing, so as to obtain the state factor of the slow response items of the oil-immersed transformer.

[0099] The determination unit 32 can also be used to determine the second operating state of the oil-immersed transformer based on the state factor of the slow response items.

[0100] The detection unit 34 can be used to input the test data of the slow response items and / or the test data of the overhaul items into a preset fault detection network for fault detection at the same time, so as to obtain the fault diagnosis result of the oil-immersed transformer and the fault location information when a fault exists. Among them, when the fault diagnosis result is that a fault exists, the weights of the slow response items and / or the overhaul items involved in the fault diagnosis result are updated.

[0101] The verification unit 35 can be used to verify the first operating state based on the second operating state and the fault diagnosis result, and determine the target operating state of the oil-immersed transformer according to the verification result.

[0102] In some embodiments, the first calculation unit 31 can specifically be used to input the dissolved gas content in the oil into the preset neural network model for calculation, so as to obtain the state factor of the fast response items of the oil-immersed transformer.

[0103] In some embodiments, when the preset neural network model includes an input layer, a first hidden layer, a second hidden layer, and an output layer, the first calculation unit 31 can also specifically be used to input the dissolved gas content in the oil into the first hidden layer through the input layer for calculation, so as to obtain the output vector of the first hidden layer; input the output vector of the first hidden layer into the second hidden layer for calculation, so as to obtain the output vector of the second hidden layer; input the output vector of the second hidden layer into the output layer for calculation, so as to obtain the state factor of the fast response items of the oil-immersed transformer.

[0104] In some embodiments, the determining unit 32 may be specifically configured to determine a first evaluation level of the oil-immersed transformer based on a state interval in which the fast-response item status factor is located; and determine a first operating state of the oil-immersed transformer according to the first evaluation level.

[0105] In some embodiments, the determining unit 32 may also be specifically configured to determine a second evaluation level of the oil-immersed transformer based on a state interval in which the slow-response item status factor is located; and determine a second operating state of the oil-immersed transformer according to the second evaluation level.

[0106] In some embodiments, the second calculating unit 33 may be specifically configured to, based on the weight corresponding to the slow-response item and / or the weight corresponding to the overhaul item, perform weighted summation on the same type of test data in the test data of the slow-response item and / or the test data of the overhaul item, to obtain a slow-response item status factor of the oil-immersed transformer.

[0107] In some embodiments, when the preset fault detection network includes a first detection network and a second detection network, the detection unit 34 includes: a detection module, a determination module, and an analysis module.

[0108] The detection module may be configured to input the test data of the slow-response item and the test data of the overhaul item into the first detection network, to obtain a first conditional determination result corresponding to each node in the first detection network, where each node in the first detection network corresponds to a single determination condition.

[0109] The detection module may also be configured to input the first conditional determination result into the second detection network, to obtain a second conditional determination result corresponding to each node in the second detection network, where each node in the second detection network corresponds to at least two determination conditions.

[0110] The determination module may be configured to, if a conditional determination result corresponding to any node in the second conditional determination result is abnormal, determine that the oil-immersed transformer has a fault, and add the test data of the slow-response item and / or the test data of the overhaul item involved in the abnormal conditional determination result, as well as the determination condition, to a conflict set.

[0111] The analysis module may be configured to analyze fault location information of the oil-immersed transformer based on the test data of the slow-response item and / or the test data of the overhaul item in the conflict set, as well as the determination condition.

[0112] In some embodiments, the verification unit 35 may be specifically configured to determine that the first operating state is inaccurate if the first operating state is different from the second operating state and the fault diagnosis result, determine the target operating state of the oil-immersed transformer based on the second operating state and the fault diagnosis result, and update the preset state factor calculation model.

[0113] In some embodiments, the detection unit 34 further includes an update module.

[0114] The update module may be configured to update the weights of the slow response items and / or the maintenance items in the conflict set to obtain updated weights; based on the updated weights, perform weighted summation on the same type of test data in the test data of the slow response items and / or the test data of the maintenance items to obtain a recalculated state factor of the slow response items.

[0115] It should be noted that for other corresponding descriptions of each functional unit involved in the oil-immersed transformer state evaluation device provided in this embodiment, reference may be made to Figure 1 and Figure 2 the corresponding descriptions therein, which will not be elaborated here.

[0116] Based on the above methods as shown in Figure 1 and Figure 2 correspondingly, this embodiment further provides a storage medium on which a computer program is stored, and when the program is executed by a processor, it implements the oil-immersed transformer state evaluation method as shown in Figure 1 and Figure 2 above.

[0117] Based on such an understanding, the technical solution of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), and includes several instructions for causing an electronic device (which can be a personal computer, a server, or a network device, etc.) to execute the methods in various implementation scenarios of the present application.

[0118] Based on the above methods as shown in Figure 1 and Figure 2 and the virtual device embodiment as shown in Figure 5 in order to achieve the above object, an electronic device is further provided in an embodiment of the present application, which may specifically be a personal computer, a tablet computer, a server, or other network devices, etc. The device includes a storage medium and a processor; the storage medium is used to store a computer program; the processor is used to execute the computer program to implement the oil-immersed transformer state evaluation method as shown in Figure 1 and Figure 2 above.

[0119] Optionally, the above-mentioned physical device may further include a user interface, a network interface, a camera, a Radio Frequency (RF) circuit, sensors, an audio circuit, a WI-FI module, and so on. The user interface may include a display, an input unit such as a keyboard, etc. Optionally, the user interface may further include a USB interface, a card reader interface, etc. The network interface may optionally include a standard wired interface, a wireless interface (such as a WI-FI interface), etc.

[0120] Those skilled in the art can understand that the above-mentioned physical device structure provided in this embodiment does not constitute a limitation on the physical device, and may include more or fewer components, or combine some components, or have different component arrangements.

[0121] The storage medium may further include an operating system and a network communication module. The operating system is a program for managing the hardware and software resources of the above-mentioned physical device, and supports the operation of information processing programs and other software and / or programs. The network communication module is used to implement communication between components inside the storage medium, and communication between other hardware and software in the information processing physical device.

[0122] Through the description of the above embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus a necessary general hardware platform, or can be implemented by hardware.

[0123] In the embodiment of the present invention, by using a preset state factor calculation model, calculating a fast-response project state factor, and determining a first operating state of the oil-immersed transformer, it is possible to avoid human participation in the assessment of the operating state of the oil-immersed transformer, thereby improving the accuracy of the assessment of the operating state of the oil-immersed transformer. In addition, in the embodiment of the present invention, by calculating a slow-response project state factor and determining a second operating state of the oil-immersed transformer, it is possible to correct the first operating state of the oil-immersed transformer, thereby further improving the accuracy of the assessment of the operating state of the oil-immersed transformer. Furthermore, in the embodiment of the present invention, by using a preset fault detection network, determining whether there is a fault in the oil-immersed transformer and the fault location information when there is a fault, it is possible to correct the first operating state of the oil-immersed transformer based on the fault diagnosis result, thereby ensuring the accuracy of the assessment result of the operating state of the oil-immersed transformer. Further, the embodiment of the present invention can also update the weights of the slow-response projects and / or maintenance projects involved in the fault diagnosis result when there is a fault, thereby improving the calculation accuracy of the slow-response project state factor and further ensuring the accuracy of the assessment result of the operating state of the oil-immersed transformer.

[0124] Those skilled in the art can understand that the attached drawings are only schematic diagrams of a preferred implementation scenario, and the modules or processes in the attached drawings are not necessarily essential for implementing the present application. Those skilled in the art can understand that the modules in the devices in the implementation scenario can be distributed in the devices in the implementation scenario according to the description of the implementation scenario, or can be correspondingly changed and located in one or more devices different from this implementation scenario. The modules in the above implementation scenario can be combined into one module, or can be further split into multiple sub-modules.

[0125] The above serial numbers of the present application are only for description and do not represent the advantages or disadvantages of the implementation scenario. The above-disclosed are only several specific implementation scenarios of the present application. However, the present application is not limited thereto, and any changes that can be conceived by those skilled in the art should fall within the protection scope of the present application.

Claims

1. A method for evaluating the state of an oil-immersed transformer, characterized in that: include: After testing the quick response items involved in the operation and maintenance detection process of the oil-immersed transformer, the test data of the quick response items are input into a preset state factor calculation model for calculation to obtain the quick response item state factor of the oil-immersed transformer, wherein the quick response items include dissolved gas detection in oil; Based on the quick response item state factor, determining a first operating state of the oil-immersed transformer; After testing the slow-reaction items and / or maintenance items involved in the operation and maintenance detection process of the oil-immersed transformer, weighted summation is performed on the test data of the slow-reaction items and / or the test data of the maintenance items of the same type to obtain the slow-reaction item state factor of the oil-immersed transformer, wherein the slow-reaction items include insulating oil dielectric loss detection, oil acid value detection, oil particle detection, oil furfural content determination, insulation resistance test, partial discharge test, dielectric loss test, core grounding current detection, core insulation resistance detection and winding DC resistance mutual difference detection; Determining a second operating state of the oil-immersed transformer based on the slow-reaction item state factor; At the same time, the test data of the slow-response items and / or the test data of the maintenance items are input into a preset fault detection network for fault detection, so as to obtain a fault diagnosis result of the oil-immersed transformer and fault location information when a fault exists, wherein when the fault diagnosis result indicates that a fault exists, the weights of the slow-response items and / or maintenance items involved in the fault diagnosis result are updated; Based on the second operating state and the fault diagnosis result, verifying the first operating state, and determining a target operating state of the oil-immersed transformer according to the verification result; The test data of the quick response project includes the content of dissolved gas in oil, and the preset state factor calculation model is a preset neural network model. The test data of the quick response project is input into the preset state factor calculation model for calculation to obtain the quick response project state factor of the oil-immersed transformer, including: The dissolved gas content in the oil is input into the preset neural network model for calculation to obtain the rapid response project state factor of the oil-immersed transformer.

2. The method according to claim 1, characterized in that When the preset neural network model includes an input layer, a first hidden layer, a second hidden layer and an output layer, the dissolved gas content in the oil is input into the preset neural network model for calculation to obtain the rapid response item state factor of the oil-immersed transformer, including: Inputting the dissolved gas content in the oil into the first hidden layer through the input layer for calculation, and obtaining an output vector of the first hidden layer; Inputting the output vector of the first hidden layer into the second hidden layer for calculation to obtain the output vector of the second hidden layer; The output vector of the second hidden layer is input into the output layer for calculation to obtain the fast response item state factor of the oil-immersed transformer.

3. The method according to claim 1, characterized in that Determining a first operating state of the oil-immersed transformer based on the quick response item state factor includes: Determining a first assessment level of the oil-immersed transformer based on the state interval of the quick response item state factor; determining a first operating state of the oil-immersed transformer according to the first evaluation level; and Determining a second operating state of the oil-immersed transformer based on the slow-reaction item state factor includes: Determining a second evaluation level of the oil-immersed transformer based on the state interval of the slow-reaction item state factor; A second operating state of the oil-immersed transformer is determined according to the second evaluation level.

4. The method according to claim 1, characterized in that The test data of the slow-reaction item and / or the test data of the maintenance item of the same type are weighted summed to obtain the slow-reaction item state factor of the oil-immersed transformer, including: Based on the weight corresponding to the slow response item and / or the weight corresponding to the maintenance item, the test data of the slow response item and / or the test data of the maintenance item are weighted and summed to obtain the slow response item status factor of the oil-immersed transformer.

5. The method according to claim 1, characterized in that: When the preset fault detection network includes a first detection network and a second detection network, the test data of the slow response item and / or the test data of the maintenance item are input into the preset fault detection network for fault detection to obtain the fault diagnosis result of the oil-immersed transformer and the fault location information when a fault exists, including: Input the test data of the slow-reaction project and the test data of the maintenance project into the first detection network to obtain the first condition determination result corresponding to each node in the first detection network, wherein each node in the first detection network corresponds to a single determination condition; Inputting the first condition determination result into the second detection network to obtain the second condition determination result corresponding to each node in the second detection network, wherein each node in the second detection network corresponds to at least two determination conditions; If the condition determination result corresponding to any node in the second condition determination result is abnormal, it is determined that the oil-immersed transformer has a fault, and the test data of the slow response item and / or the test data of the maintenance item involved in the abnormal condition determination result and the determination condition are added to the conflict set; Analyzing the fault location information of the oil-immersed transformer based on the test data of the slow-response items and / or the test data of the maintenance items in the conflict set and the determination conditions; and Based on the second operating state and the fault diagnosis result, verifying the first operating state, and determining a target operating state of the oil-immersed transformer according to the verification result, including: If the first operating state is different from the second operating state and the fault diagnosis result, it is determined that the first operating state is inaccurate, and based on the second operating state and the fault diagnosis result, the target operating state of the oil-immersed transformer is determined, and the preset state factor calculation model is updated.

6. The method according to claim 5, characterized in that The method further comprises: Updating the weights of the slow-response items and / or maintenance items in the conflict set to obtain updated weights; Based on the updated weights, weighted summation is performed on the test data of the slow-response item and / or the test data of the maintenance item of the same type to obtain a recalculated slow-response item state factor.

7. An oil-immersed transformer condition assessment device, characterized in that: include: A first calculation unit is used for, after testing the quick response items involved in the operation and maintenance detection process of the oil-immersed transformer, inputting the test data of the quick response items into a preset state factor calculation model for calculation to obtain the quick response item state factor of the oil-immersed transformer, wherein the quick response items include dissolved gas detection in oil, the test data of the quick response items include the content of dissolved gas in oil, and the preset state factor calculation model is a preset neural network model; A determination unit, configured to determine a first operating state of the oil-immersed transformer based on the quick response item state factor; A second calculation unit is used for, after testing the slow-reaction items and / or maintenance items involved in the operation and maintenance detection process of the oil-immersed transformer, weighted summing the test data of the slow-reaction items and / or the test data of the maintenance items of the same type to obtain the slow-reaction item state factor of the oil-immersed transformer, wherein the slow-reaction items include insulating oil dielectric loss detection, oil acid value detection, oil particle detection, oil furfural content determination, insulation resistance test, partial discharge test, dielectric loss test, core grounding current detection, core insulation resistance detection and winding DC resistance mutual difference detection; The determining unit is further used to determine the second operating state of the oil-immersed transformer based on the slow-reaction item state factor; A detection unit, used for simultaneously inputting the test data of the slow-response items and / or the test data of the maintenance items into a preset fault detection network for fault detection, and obtaining a fault diagnosis result of the oil-immersed transformer, and fault location information when a fault exists, wherein when the fault diagnosis result is that a fault exists, the weights of the slow-response items and / or maintenance items involved in the fault diagnosis result are updated; a verification unit, configured to verify the first operating state based on the second operating state and the fault diagnosis result, and determine a target operating state of the oil-immersed transformer according to the verification result; The first calculation unit is specifically used to input the dissolved gas content in the oil into the preset neural network model for calculation, so as to obtain the rapid response item state factor of the oil-immersed transformer.

8. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

9. An electronic device comprising a storage medium, a processor, and a computer program stored in the storage medium and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 6 is implemented.

Images