Method and system for evaluating operation state of key equipment of main power grid

Abstract

The invention relates to the technical field of power management, in particular to a main power grid key equipment operation state evaluation method and system. Multi-source data are collected, the running state of current key equipment is preliminarily determined through the distribution state of the multi-source data, features of the multi-source data are extracted and fused based on preliminary judgment, deep state information is extracted through a neural network, and therefore the running state of the current key equipment is determined. Compared with the prior art, the embodiment of the invention has the advantages that various data are associated and fused as a whole to be processed, so that the running state can be comprehensively evaluated.

Description

Technical Field

[0001] The present invention relates to the technical field of power management, and is a method and system for evaluating the operating status of key equipment in a main power grid. Background Art

[0002] As a critical component of the power system, transformers play a vital role in ensuring its safe and stable operation. Reducing the probability of failure is key to improving operational safety. As transformers operate longer, their failure rate increases, and the types of failures are often diverse and complex, making accurate assessment of the cause of transformer failures challenging. Furthermore, as grid voltage levels increase, transformer voltage levels also increase, increasing the risk of failures and posing a potential threat to the safe and stable operation of the grid. This is especially true for large, technically complex 500kV, 360MVA power transformers. Once a failure occurs, the maintenance costs are extremely high, increasing not only the operating costs of power companies but also the stability of the power system.

[0003] In order to avoid major accidents caused by failures of electrical equipment such as transformers, it is necessary to strengthen on-site monitoring and diagnosis of electrical equipment in substations and transformer stations to detect potential failures as early as possible. Summary of the Invention

[0004] In order to solve the above technical problems, the present application provides a method and system for evaluating the operating status of key equipment in the main power grid. By obtaining operation monitoring data corresponding to multiple indicators of the power transformer and performing feature processing and fusion on the data, the operating status of the power transformer can be comprehensively evaluated.

[0005] In order to achieve the above objectives, the technical solutions adopted in the embodiments of the present application are as follows: In a first aspect, a method for evaluating the operating status of key equipment in a main power grid is provided, which is applied to the status evaluation of a power transformer. The method comprises: obtaining a multi-index data sequence corresponding to a subsystem in the power transformer, and performing data cleaning on the data to remove abnormal data; determining a continuous distribution state of the data, and determining whether the operating status meets an initial preset condition based on the continuous distribution state; when the operating status meets the initial preset condition, extracting a feature sequence of the data sequence corresponding to each indicator, and fusing the features corresponding to multiple indicators to obtain a fused feature sequence; inputting the fused feature sequence into a prediction model to obtain a status evaluation result for the power transformer; the prediction model is constructed by a gated recurrent network.

[0006] In some specific implementations, the subsystem includes a bushing, a winding, an iron core and a tap, and multi-index data is used to characterize the system status corresponding to multiple subsystems; the system status includes the bushing insulation performance, the bushing electrical performance, the winding insulation performance, the winding electrical performance, the winding mechanical performance, the iron core insulation performance to ground and the tap electrical performance; the data includes the bushing dielectric loss value, the bushing insulation resistance value, the bushing DC resistance value, the winding insulation resistance value, the winding DC leakage current value, the winding capacitance, the winding short-circuit impedance value, the iron core grounding current value, the iron core insulation resistance value, and the tap voltage ratio.

[0007] In some specific implementations, determining whether the operating state meets the initial preset conditions based on the continuous distribution state includes: determining the associated indicators corresponding to the data and the continuous data distribution states corresponding to the associated indicators, and establishing a data distribution state relationship between the indicator and the associated indicator, determining the deviation between the data distribution state relationship and the standard data distribution state relationship, and determining whether the current operating state meets the preset conditions based on whether the deviation meets the standard deviation.

[0008] In some specific implementations, the method also includes determining the weights of the indicator and the associated indicator, and obtaining a joint weight based on the weights of the two, and updating the deviation based on the joint weight to obtain a deviation update value, and determining whether the current operating status meets the preset conditions based on whether the deviation update value meets the standard deviation.

[0009] In some specific implementations, determining the weight value of each of the indicators includes: converting multiple data corresponding to each of the indicators into a probability matrix, and determining the contribution of each data to the indicator, determining the information entropy and information entropy redundancy corresponding to each of the indicators based on the contribution of each data, and obtaining the weight of the indicator based on the information entropy redundancy.

[0010] In some specific implementations, the method also includes: updating the weight to obtain a target weight; specifically including: retrieving a second weight corresponding to the indicator, the second weight being obtained based on expert decision-making; integrating the second weight with the calculated weight value based on a combination coefficient to obtain the target weight.

[0011] In some specific implementations, the data is cleaned to remove abnormal data, including: obtaining multiple data corresponding to each type of data sequence, and calculating the path length corresponding to each data in the isolation tree and the average path length corresponding to the multiple data, and removing data with a path length less than the average path length as abnormal data.

[0012] In some specific implementations, the features of the data sequence corresponding to each indicator are extracted through a one-dimensional convolutional neural network configured with an attention mechanism; the features corresponding to multiple indicators are fused to obtain a fused feature sequence, including: fusing the weight values corresponding to each indicator based on a fully connected layer to obtain a fused feature.

[0013] In the second aspect, a main power grid key equipment operation status assessment system is provided, which is used for status assessment of power transformers, including: a data acquisition unit, used to obtain multi-index data corresponding to each subsystem of the power transformer; a data storage unit, used to store the multi-index data correspondingly to form multiple data sequences; a data processing unit, used to retrieve data from the data storage unit and execute any of the above methods.

[0014] In some specific implementations, the data processing unit includes: a first evaluation module, used to obtain a multi-indicator data sequence corresponding to the subsystem of the power transformer, and perform data cleaning on the data to eliminate abnormal data; determine the continuous distribution state of the data, and determine whether the current operating state meets the preset conditions based on the continuous distribution state; a data processing module, used to extract the feature sequence of the data sequence corresponding to each indicator when the current operating state meets the preset conditions, and fuse multiple features to obtain a fused feature sequence; a second evaluation module, used to input the fused feature sequence into a prediction model to obtain a state evaluation result of the power transformer.

[0015] The technical solution provided in the embodiments of this application collects multi-source data and preliminarily determines the current operating status of key equipment based on the distribution of the multi-source data. Based on this preliminary judgment, the features of the multi-source data are extracted and integrated through a neural network to extract deep-level status information to determine the current operating status of key equipment. Compared with the existing technology, the embodiments of this application associate and integrate multiple data and process them as a whole, which can comprehensively evaluate the operating status. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0017] The methods, systems, and / or programs in the accompanying drawings will be further described according to exemplary embodiments. These exemplary embodiments will be described in detail with reference to the drawings. These exemplary embodiments are non-limiting exemplary embodiments, wherein example numerals represent similar structures in the various views of the drawings.

[0018] Figure 1 It is a schematic diagram of the system structure provided in an embodiment of the present application.

[0019] Figure 2 This is a flow chart of the main power grid key equipment operating status assessment method provided in an embodiment of the present application.

[0020] Figure 3 This is a structural diagram of a data processing unit provided in an embodiment of the present application.

[0021] Figure 4 This is a schematic diagram of the terminal device structure provided in an embodiment of the present application. DETAILED DESCRIPTION

[0022] In order to better understand the above technical solution, the technical solution of the present application is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present application and the specific features in the embodiments are detailed descriptions of the technical solution of the present application, rather than limitations on the technical solution of the present application. In the absence of conflict, the embodiments of the present application and the technical features in the embodiments can be combined with each other.

[0023] In the following detailed description, numerous specific details are set forth by way of example in order to provide a thorough understanding of the relevant teachings. However, it will be apparent to one skilled in the art that the present application can be practiced without these details. In other instances, well-known methods, procedures, systems, compositions, and / or circuits have been described at a relatively high level, without detail, to avoid unnecessarily obscuring aspects of the present application.

[0024] Flowcharts are used in this application to illustrate the execution processes performed by the system according to the embodiments of the present application. It should be clearly understood that the execution processes of the flowcharts may not be executed in sequence. Instead, these execution processes may be executed in reverse order or simultaneously. In addition, at least one additional execution process may be added to the flowchart. One or more execution processes may be deleted from the flowchart.

[0025] Before further explaining the embodiments of the present invention in detail, the nouns and terms involved in the embodiments of the present invention are explained. The nouns and terms involved in the embodiments of the present invention are subject to the following interpretations.

[0026] (1) In response to, it is used to indicate the conditions or states on which the executed operations depend. When the dependent conditions or states are met, one or more operations executed may be in real time or with a set delay. Unless otherwise specified, there is no restriction on the order in which the multiple operations executed are executed.

[0027] (2) Based on, used to indicate the conditions or states on which the executed operations depend. When the dependent conditions or states are met, one or more operations executed may be in real time or with a set delay. Unless otherwise specified, there is no restriction on the order in which the multiple operations executed are executed.

[0028] The present invention provides a method and system for assessing the operating status of key equipment in a main power grid, particularly a wet-type power transformer. Prior art methods for assessing the operating status of a main power grid, particularly a power transformer, primarily include oil chromatography analysis, electrical characteristics testing, and infrared thermal imaging. While all three methods can assess the operating status of power transformers to a certain extent, they each present their own challenges in practical application. For example, electrical characteristics testing can be affected by environmental factors, leading to measurement deviations. Furthermore, different transformer models may have varying electrical characteristics, necessitating the development of appropriate testing protocols and assessment criteria tailored to specific circumstances. Infrared thermal imaging may affect measurement results when exposed to varying ambient temperatures, humidity, and wind speeds. Furthermore, different transformer models may have varying surface structures and material properties. Therefore, employing any of these methods alone for condition assessment may be subject to influences caused by environmental and object variations. Furthermore, single-factor assessments of single parameters can also result in inaccurate assessment results.

[0029] Therefore, based on the aforementioned state of the art, the main grid key equipment operating status assessment system provided in this embodiment collects multi-type, multi-source data, fuses it, and then performs inference. This increases data diversity and considers the correlation between different types of data in their status representation, resulting in more accurate assessment results. Furthermore, this system incorporates a neural network model that extracts potential information from the data, thereby enhancing the foresight and depth of the assessment.

[0030] For details about the structure of this system, please refer to Figure 1 The embodiment of the present application provides a main power grid key equipment operating status assessment system 10, which includes three main units: a data acquisition unit 11, a data storage unit 12, and a data processing unit 13. The data acquisition unit is used to obtain multi-index data corresponding to each subsystem of the power transformer, the data storage unit is used to store the multi-index data accordingly, and the data processing unit is used to retrieve data from the data storage unit and obtain the current operating status of the key equipment by processing the data.

[0031] Among them, the subsystems in this embodiment are the key components of the power transformer, namely the bushing, winding, core and tap. The multi-indicator data refers to the data corresponding to one or more indicators in the above subsystems, which is used to characterize the system status corresponding to multiple subsystems.

[0032] In this embodiment, the system status is determined by subsystem, including bushing insulation performance, bushing electrical performance, winding insulation performance, winding electrical performance, winding mechanical performance, core-to-ground insulation performance, and tap electrical performance. The corresponding data are bushing dielectric loss, bushing insulation resistance, bushing DC resistance, winding insulation resistance, winding DC leakage current, winding capacitance, winding short-circuit impedance, core-to-ground current, core insulation resistance, and tap voltage ratio.

[0033] Furthermore, the above data are all obtained through a data acquisition unit, wherein the data acquisition unit is a data acquisition circuit or an electronic component such as a sensor configured at a corresponding position in the above subsystem, which obtains data corresponding to multiple indicators in the above subsystem through the electronic components and transmits the data to the data storage unit.

[0034] It is worth noting that the data acquisition unit in this embodiment should collect the above data based on time intervals. In other embodiments, time interval collection and event-triggered collection can also be used. Specifically, the above data is first collected based on time intervals, and when the current collection value of any of the above data exceeds a preset threshold, the event collection mode is triggered. Event-triggered collection means that the circuit of the data acquisition unit is connected and real-time data is collected until no event of continuously exceeding the preset threshold is generated in the agreed time interval or the number of events exceeding the preset threshold within the preset time is less than a preset number.

[0035] It is worth noting that when event-triggered acquisition is performed, all data acquisition devices in the data acquisition unit are in a connected state.

[0036] The data storage unit stores the collected data in a corresponding manner according to the storage space and stores the data in a sequence.

[0037] The data processing unit is the main unit for data processing in this embodiment. A method for evaluating the operating status of key grid equipment is configured in this unit to retrieve data from the data storage unit and process the data to obtain the operating status of key grid equipment.

[0038] Among them, see Figure 2 , the method for evaluating the operating status of key power grid equipment in the data processing unit of this embodiment includes the following steps: Step S21. Acquire a multi-index data sequence corresponding to the subsystem of the power transformer, and perform data cleaning on the data to remove abnormal data; determine the continuous distribution state of the data, and determine whether the operating state meets the initial preset conditions based on the continuous distribution state.

[0039] In this embodiment, the power transformer is a wet-type transformer, and the subsystems in this power transformer include bushings, windings, cores, and taps. Multi-index data is used to characterize the system states corresponding to the multiple subsystems, where the corresponding system states include bushing insulation performance, bushing electrical performance, winding insulation performance, winding electrical performance, winding mechanical performance, core-to-ground insulation performance, and tap electrical performance; the data includes bushing dielectric loss value, bushing insulation resistance value, bushing DC resistance value, winding insulation resistance value, winding DC leakage current value, winding capacitance, winding short-circuit impedance value, core grounding current value, core insulation resistance value, and tap voltage ratio.

[0040] In this embodiment, the collected data is generally sequence data, meaning it is not data corresponding to an isolated time point but rather a sequence data built up from multiple time points. The state assessment method involves integrating the sequence data corresponding to the aforementioned indicators and determining the state information contained in the integrated sequence data through an algorithm.

[0041] In this embodiment, the evaluation of the state includes a preliminary evaluation and an in-depth evaluation. The preliminary evaluation is used to statistically calculate the statistical distribution of multiple data and determine the shallow state based on the distribution state. When the power transformer is not undergoing task adjustment, its operation should be steady-state under ideal conditions, and the data should be uniform in time series. Therefore, when the data fluctuates, it means that the current operation of the power transformer has changed. However, it is worth noting that the above reasoning is only applicable under ideal conditions. The operation of the power transformer also changes accordingly with changes in the external environment. Therefore, in this embodiment, it is not possible to directly apply the data performance under ideal conditions to map the operating state.

[0042] Therefore, in this technical context, it is also possible to obtain the continuous distribution state of time series data, that is, the distribution of data over time, and determine the operating state based on the continuous distribution state, thereby reducing the error problem caused by judging isolated data. Specifically, the data in the time series is obtained, and the standard deviation, mean, and variance corresponding to each data point in the time series are determined. The degree of deviation of the variance from the ideal variance is determined, and a weight coefficient is determined based on the degree of deviation; the discrete value of the data in the current time series is determined based on the ratio of the standard deviation to the mean, and the degree of discreteness is updated using the weight coefficient to obtain the final discrete value, which is used to indicate the continuous distribution state of the current data.

[0043] While this method solves the problem of errors caused by isolated data, power transformers are complex systems, and their status cannot be fully evaluated using the statistical results corresponding to a single indicator. Therefore, to address this technical issue, this embodiment combines indicators during the preliminary assessment. Specifically, a data relationship is established between the indicator and the associated indicator, and the status assessment is performed based on this data relationship.

[0044] Specifically, the associated indicator corresponding to any indicator is determined, and the continuous data distribution state corresponding to the associated indicator is determined separately. This continuous data distribution state is also characterized by discrete values, that is, the discrete values corresponding to the time series data corresponding to the indicator and the associated indicator are determined separately. A data distribution state relationship between the indicator and the associated indicator is established, where this relationship is the ratio between the two. Based on this data distribution state relationship, the deviation from the standard data distribution state relationship is determined. Whether the current motion state meets the preset conditions is determined based on whether this deviation meets the standard deviation.

[0045] In this embodiment, the standard deviation is determined by pre-simulating the standard discrete value ranges corresponding to the two indicators and then determining the standard deviation based on these ranges. In this embodiment, the relationship between indicators and associated indicators is prioritized over indicators within the same subsystem. For example, this could be the relationship between the bushing dielectric loss value and either the bushing insulation resistance value or the bushing DC resistance value. In this embodiment, the optimal indicator is selected by determining the indicator with the largest discrete value of the two associated indicators. Furthermore, for the tap in this embodiment, which has only one indicator, an indicator-associated indicator relationship can be established with any of the aforementioned indicators.

[0046] In this embodiment, the overall distribution of the related indicators is determined, so that the evaluation is more complete and in-depth compared with a single indicator.

[0047] However, it is worth noting that each indicator of the power transformer system has a different weight for the system's operation, and the expression of data with different weights produces different results for the evaluation of the operating status. Therefore, in order to more accurately perform a preliminary evaluation in this embodiment, the weight corresponding to each indicator in the entire system should be considered, and the data should be updated based on the weight. Specifically, the weight corresponding to the indicator and the associated indicator is obtained separately, and a joint weight is obtained based on the weights of the two. The above deviation is updated using this joint weight, and the updated deviation is used to determine whether the current operating status meets the preset conditions.

[0048] In this embodiment, the weight of each indicator is determined by the entropy weight method to determine the objective weight of the characterization quantity. First, the data corresponding to each indicator is normalized, and then the normalized data is transformed into a probability matrix to determine the proportion of each sample under each indicator. This process is used to reflect the contribution of each data to the indicator, and based on the proportion corresponding to each data, the information entropy and information redundancy corresponding to each indicator are determined based on the information entropy formula. The calculation of information entropy is based on the following formula: , where k = 1 / ln (n) is the adjustment coefficient, pij is the weight of the i-th data of the j-th indicator, and ej represents the information entropy corresponding to the j-th indicator; the information redundancy is calculated based on the following formula: , and then the weight of this indicator is obtained based on the information entropy redundancy, which is expressed based on the following formula: .

[0049] The above method can determine the weight corresponding to each indicator, and use the average value of the sum of the weights of the two indicators as the joint weight and update the deviation through this joint weight to obtain the deviation update value, and use the updated deviation to determine whether the current operating status meets the preset conditions.

[0050] This embodiment further calculates the deviation through the above method to make the status evaluation more accurate.

[0051] In this embodiment, the weight determination method described above is an objective weighting method based on the degree of data dispersion. This method is highly objective and can rely on the inherent characteristics of the data to avoid subjective bias. However, this method is completely dependent on objectivity and is sensitive to data quality. When the data quality is low or the data sample is small, it cannot reflect the actual importance of the indicator. Therefore, in order to further obtain accurate indicator weights, this embodiment uses a combination of subjective and objective weights to obtain a comprehensive weight that can better reflect the characteristics of the data.

[0052] Specifically, in this embodiment, the second weight corresponding to the indicator is retrieved. This weight is obtained based on expert decision-making and can be understood as a subjective scoring weight. The details of how this weight is obtained will not be described in detail. The second weight is integrated with the weight calculated in the above manner based on the combination coefficient to obtain the final target weight.

[0053] Among them, first, the weight vector sequence of the above two weights is obtained, and the above two weight vector sequences are combined through a linear method to construct a possible weight set, and an optimization objective function is established, where the goal of this function is to minimize the difference between the comprehensive weight and the above two weights, and solve the combination coefficient according to the first-order derivative condition of the matrix differential, and finally combine the above two weights based on the combination coefficient to obtain the final target weight.

[0054] Based on the obtained target weight, the corresponding weight of each of the above indicators is obtained, and the average value of the sum of the weights of the two indicators is used as the joint weight. The deviation is updated by this joint weight to obtain the deviation update value, and the updated deviation is used to determine whether the current operating status meets the preset conditions.

[0055] This embodiment uses the above-described method to process data corresponding to multiple indicators and preliminarily determine the current operating status. If the updated deviation determines that the current operating status does not meet the preset conditions, it indicates that the current system operating status is abnormal. Based on the current indicators, the specific subsystem with the abnormality can be determined. Because this process distributes the status through data distribution, it is suitable for determining the shallow operating status of the system and is suitable for preliminary judgment scenarios. However, the above-described method cannot determine the deeper operating status of the power transformer. In this embodiment, further processing of the data that meets the preset conditions is required.

[0056] It is worth noting that when performing the above processing, because the amount of collected data is large, and the collected data may contain abnormal data and noise data due to reasons in the collection unit, the collected data needs to be cleaned before processing.

[0057] The data cleaning process in this embodiment is implemented using the isolation forest algorithm. Specifically, the collected data is organized into a table format, and the data table is pre-processed based on the isolation forest algorithm. The expected path length from the data to the isolation tree is obtained based on the amount of data, and the anomaly score of each data is calculated. The data is sorted according to the anomaly score, and the corresponding data is judged to be abnormal based on the anomaly score. The calculation of abnormal data is based on the expected path length and the average path length, which is expressed based on the following formula: ,in is the expected path length, is the average path length; the calculation of the average path length is based on the following formula: ,in For harmonious numbers Indicates that is Euler's constant.

[0058] Step S22: When the operating state meets the initial preset conditions, extract the feature sequence of the data sequence corresponding to each indicator, and fuse the features corresponding to multiple indicators to obtain a fused feature sequence.

[0059] Step S21 determines the current preliminary state of the power transformer. If the preliminary state meets the requirements, further determination of the power transformer's advanced state is required. In this embodiment, this process characterizes the data corresponding to multiple indicators and fuses the feature sequences corresponding to each indicator to obtain a fused feature that comprehensively reflects the power transformer. This fused feature, which is the result of fusing the data corresponding to multiple indicators, can comprehensively reflect the characteristics of the power transformer.

[0060] It is worth noting that the data collected in step S21 may not have the same time sequence because they involve different indicators and different types, which may result in missing data in the time series. Therefore, before starting step S22, it is necessary to time-align the data in step S21.

[0061] Specifically, by determining the data with the least number of time points among the multiple indicator data as a reference object, the remaining indicator data are subjected to redundant data elimination based on the reference object, and only the data with the same time point is retained.

[0062] In this embodiment, the feature extraction of the data sequence corresponding to each indicator is achieved by configuring a one-dimensional convolutional neural network with an attention mechanism. The one-dimensional convolutional neural network includes an input layer, a one-dimensional convolutional layer, an activation layer, and a pooling layer, wherein the input layer is used to receive multiple data sequences corresponding to multiple indicators after data cleaning, and the one-dimensional convolutional layer is the core structure of this network, which is used for the convolution kernel to perform a regional sliding operation on the multiple input data sequences to achieve the extraction of features corresponding to each indicator, and obtain features corresponding to multiple indicators; wherein the activation layer is configured with a loss function, and in this embodiment, the ReLu function is used for this loss function to solve the problem of link gradient disappearance; the pooling layer is used to reduce the sampling dimension of the data, and a maximum pooling strategy is adopted.

[0063] The above-mentioned network structure can realize feature extraction in the data sequence corresponding to the indicator and obtain multiple features. Among them, the extracted features are fused through the fully connected layer. Specifically, the weight value corresponding to the indicator obtained by step S21 in the fusion mechanism of the fully connected layer in this embodiment is fused. Among them, the target weight after weight integration is preferably selected for the weight value. Specifically, the updated corresponding features are obtained by weighted multiplication of the corresponding features by the weight value or target weight corresponding to the indicator, and the updated features are fused based on the fusion mechanism of the fully connected layer. Among them, the feature fusion of the fully connected layer can be directly performed using the method in the prior art, which will not be repeated in this embodiment.

[0064] In this embodiment, because the input data is continuous data, in order to better extract the potential connections between the data, the ability to capture cross-channel information is achieved by adding an attention mechanism to the one-dimensional convolutional layer.

[0065] Among them, the attention mechanism in this embodiment adopts a separable attention module, which is placed after the final convolution layer of each bottleneck structure in the one-dimensional convolution layer. This module generates channel attention for the extracted features by means of one-dimensional convolution. Specifically, the input features are sequentially processed by convolution layers of different levels and then average pooling operations are performed, and the corresponding feature vectors are generated by the one-dimensional convolution layer. Then, the feature vectors are processed by the loss function and normalization, and the generated weight coefficients are multiplied by the input features to obtain weighted features, and the weighted features are element-wise added to the input features to obtain the final feature matrix. After processing by the above-mentioned multiple modules and the corresponding convolution layers, the final features are obtained, which are the features to be fused.

[0066] Step S23: Input the fused feature sequence into a prediction model to obtain a status assessment result of the power transformer.

[0067] Step S22 can implement features for multiple indicators corresponding to multiple subsystems of the current power transformer, and in this embodiment, a gated recurrent network is used to implement state assessment. The gated recurrent network includes multiple gated recurrent units connected in series, which are used to predict long time series. Each gated recurrent unit in this network has the same structure, with each neuron having an update gate and a reset gate. The output of the reset gate at the current time step is multiplied by the hidden state at the previous time step. Then, in the fully connected layer of the activation function, the current candidate hidden state is calculated by connecting the output of the element-wise multiplication of the current time step input and the hidden state at the previous time step. The update gate is then used to combine the hidden state of the previous time step with the candidate hidden state of the current time step to calculate the final hidden state of each gated recurrent unit. The input to the series of gated recurrent units is the final hidden state of the output of the previous gated recurrent unit, which is used as the hidden state of the previous time step. The final hidden state of the last gated recurrent unit is then determined again using the above processing method as the final hidden state.

[0068] The final hidden state is passed through the output layer to obtain the prediction result, where the prediction result is the probability corresponding to each running state.

[0069] The effectiveness of the above prediction results depends on the determination of the hyperparameters of the above network, especially the gated recurrent unit network, such as the number of hidden layer neurons and the learning rate. Usually, these hyperparameters are determined through multiple experiments. However, this model inevitably brings about the problem of increased training costs. Therefore, in order to solve this technical problem, the embodiments of the present application determine the hyperparameters through an optimization method when training the above network, especially the gated recurrent unit.

[0070] Among them, the Harris Hawk optimization method is adopted for the hyperparameter optimization method in this embodiment. Regarding this method, parameters such as the population size, the number of iterations, and the initial position of the population are first initialized; the above initial position is continuously updated and the prey position is determined, and the escape energy of the prey is initialized; and the hunting strategy is updated according to the size of the escape energy. If the escape energy value is less than 1, the attack is started, and the fitness values of all individuals in the current iteration number are calculated and the optimal position is determined, and the next hunting strategy is determined based on whether the optimal position is reached; it is judged whether the maximum number of iterations has been reached. If the iteration time has been reached and the optimal value has been obtained, the update is stopped to obtain the global optimal solution, that is, the optimal hyperparameter. If the optimal value has not been obtained, the operation is returned and repeated until the global optimal solution is obtained.

[0071] The present invention provides a method and system for evaluating the operating status of key equipment in a main power grid. This method collects multi-source data and preliminarily determines the current operating status of key equipment based on the distribution of the multi-source data. Based on this preliminary judgment, the system extracts and fuses the features of the multi-source data, extracting deep-level status information through a neural network to determine the current operating status of the key equipment. Compared to existing technologies, the present invention integrates and fuses multiple data sets, processing them as a whole, enabling a comprehensive evaluation of the operating status.

[0072] In this embodiment, see Figure 3 The data processing unit 13 is used to execute the processing from step S21 to step S23. This unit includes the following modules: The first evaluation module 131 is used to obtain a multi-index data sequence corresponding to the subsystem of the power transformer, clean the data and remove abnormal data; determine the continuous distribution state of the data, and determine whether the operating state meets the initial preset conditions based on the continuous distribution state; The data processing module 132 is used to extract the feature sequence of the data sequence corresponding to each indicator when the operating state meets the initial preset conditions, and fuse the features corresponding to multiple indicators to obtain a fused feature sequence; The second evaluation module 233 is configured to input the fused feature sequence into a prediction model to obtain a status evaluation result of the power transformer.

[0073] See Figure 4 In other embodiments, the above method can also be integrated into the provided terminal device 40. In view of the fact that the device may have relatively large differences due to different configurations or performance, it can include one or more processors 401 and memory 402. The memory 402 can store one or more application programs or data. Among them, the memory 402 can be a temporary storage or a persistent storage. The application stored in the memory 402 can include one or more modules (not shown in the figure), and each module can include a series of computer-executable instructions in the terminal device. Furthermore, the processor 401 can be configured to communicate with the memory 402, and the terminal device executes the series of computer-executable instructions in the memory 402. The terminal device can also include one or more power supplies 403, one or more wired or wireless network interfaces 404, one or more input / output interfaces 405, one or more keyboards 406, etc.

[0074] In a specific embodiment, the terminal device includes a memory and one or more programs, wherein the one or more programs are stored in the memory, and the one or more programs may include one or more modules, and each module may include a series of computer-executable instructions in the terminal device, and the one or more programs are configured to be executed by one or more processors, including computer-executable instructions for performing the following: Acquire a multi-index data sequence corresponding to the subsystem of the power transformer, and perform data cleaning on the data to remove abnormal data; determine a continuous distribution state of the data, and determine whether the current operating state meets a preset condition based on the continuous distribution state; When the current operating state meets the preset conditions, the feature sequence of the data sequence corresponding to each indicator is extracted respectively, and multiple features are fused to obtain a fused feature sequence; The fused feature sequence is input into a prediction model to obtain a state assessment result of the power transformer; the prediction model is constructed by a gated recurrent network.

[0075] The following is a detailed introduction to the various components of the processor: In this embodiment, the processor is an application specific integrated circuit (ASIC), or is configured to implement one or more integrated circuits of the embodiments of the present application, such as one or more microprocessors (digital signal processors, DSPs), or one or more field programmable gate arrays (FPGAs).

[0076] Optionally, the processor can execute various functions by running or executing the software program stored in the memory and calling the data stored in the memory, such as executing the above Figure 1 The method shown.

[0077] In a specific implementation, as an embodiment, the processor may include one or more microprocessors.

[0078] The memory is used to store the software program for executing the solution of the present application, and the execution is controlled by the processor. The specific implementation method can refer to the above method embodiment and will not be repeated here.

[0079] Alternatively, the memory may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, a random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, an optical disc storage (including a compact disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and capable of being accessed by a computer, but is not limited thereto. The memory may be integrated with the processor or exist independently and be coupled to the processing unit through the processor's interface circuit, and this is not specifically limited in the embodiments of the present application.

[0080] It should be noted that the structure of the processor shown in this embodiment does not constitute a limitation on the device. The actual device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0081] In addition, the technical effects of the processor can refer to the technical effects of the method described in the above method embodiment, and will not be repeated here.

[0082] It should be understood that the processor in the embodiments of the present application may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0083] It should also be understood that the memory in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0084] The above embodiments can be implemented in whole or in part via software, hardware (e.g., circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product comprises one or more computer instructions or computer programs. When loaded or executed on a computer, the processes or functions described in the embodiments of this application are fully or partially generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired means (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium accessible by a computer or a data storage device such as a server or data center that contains a collection of one or more available media. The available medium can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media. The semiconductor media can be a solid-state drive.

[0085] In this application, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, "at least one of a, b, or c" can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or plural.

[0086] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0087] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0088] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0089] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0090] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0091] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0092] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program code, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0093] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A method for evaluating the operating status of key equipment in a main power grid, characterized in that: Applied to the condition assessment of a power transformer, the method comprises: Acquire a multi-index data sequence corresponding to the subsystem of the power transformer, and perform data cleaning on the data to remove abnormal data; determine a continuous distribution state of the data, and determine whether the operating state meets the initial preset conditions based on the continuous distribution state; When the running state meets the initial preset conditions, the feature sequence of the data sequence corresponding to each indicator is extracted, and the features corresponding to multiple indicators are fused to obtain a fused feature sequence; The fused feature sequence is input into a prediction model to obtain a state assessment result of the power transformer; the prediction model is constructed by a gated recurrent network.

2. The method for evaluating the operating status of key equipment in the main power grid according to claim 1, characterized in that: The subsystem includes a bushing, a winding, an iron core and a tap, and the multi-index data is used to characterize the system status corresponding to multiple subsystems; the system status includes the bushing insulation performance, the bushing electrical performance, the winding insulation performance, the winding electrical performance, the winding mechanical performance, the iron core insulation performance to ground and the tap electrical performance; the data includes the bushing dielectric loss value, the bushing insulation resistance value, the bushing DC resistance value, the winding insulation resistance value, the winding DC leakage current value, the winding capacitance, the winding short-circuit impedance value, the iron core grounding current value, the iron core insulation resistance value, and the tap voltage ratio.

3. The method for evaluating the operating status of key equipment in the main power grid according to claim 2, characterized in that: The determining whether the operating state meets the initial preset conditions based on the continuous distribution state includes: determining the associated indicator corresponding to the data and the continuous data distribution state corresponding to the associated indicator, and establishing a data distribution state relationship between the indicator and the associated indicator, determining the deviation between the data distribution state relationship and the standard data distribution state relationship, and determining whether the current operating state meets the preset conditions based on whether the deviation meets the standard deviation.

4. The method for evaluating the operating status of key equipment in the main power grid according to claim 3, characterized in that: The method also includes determining the weights of the indicator and the associated indicator, and obtaining a joint weight based on the weights of the two, and updating the deviation based on the joint weight to obtain a deviation update value, and determining whether the current operating state meets the preset conditions based on whether the deviation update value meets the standard deviation.

5. The method for evaluating the operating status of key equipment in the main power grid according to claim 4, characterized in that: The determination of the weight value of each of the indicators includes: converting the multiple data corresponding to each of the indicators into a probability matrix, and determining the contribution of each data to the indicator, determining the information entropy and information entropy redundancy corresponding to each of the indicators based on the contribution of each data, and obtaining the weight of the indicator based on the information entropy redundancy.

6. The method for evaluating the operating status of key equipment in the main power grid according to claim 5, characterized in that: The method also includes: updating the weight to obtain a target weight; specifically including: retrieving a second weight corresponding to the indicator, the second weight being obtained based on expert decision-making; integrating the second weight with the calculated weight value based on a combination coefficient to obtain the target weight.

7. The method for evaluating the operating status of key equipment in the main power grid according to claim 1, characterized in that: The data is cleaned to remove abnormal data, including: obtaining multiple data corresponding to each type of data sequence, and calculating the path length corresponding to each data in the isolation tree and the average path length corresponding to the multiple data, and removing the data with a path length less than the average path length as abnormal data.

8. The method for evaluating the operating status of key equipment in the main power grid according to claim 6, characterized in that: The features of the data sequence corresponding to each indicator are extracted through a one-dimensional convolutional neural network configured with an attention mechanism; the features corresponding to multiple indicators are fused to obtain a fused feature sequence, including: fusing the weight values corresponding to each indicator based on a fully connected layer to obtain a fused feature.

9. A main power grid key equipment operation status assessment system, characterized in that: Applied to the condition assessment of power transformers, including: A data acquisition unit, used to obtain multi-index data corresponding to each subsystem of the power transformer; A data storage unit, configured to store the multi-index data in a corresponding manner to form multiple data sequences; A data processing unit, configured to retrieve data from the data storage unit and execute the method according to any one of claims 1 to 9.

10. The main power grid key equipment operation status assessment system according to claim 9, characterized in that: The data processing unit includes: A first evaluation module is configured to obtain a multi-index data sequence corresponding to the subsystem of the power transformer, clean the data and remove abnormal data; determine a continuous distribution state of the data, and determine whether the operating state meets an initial preset condition based on the continuous distribution state; The data processing module is used to extract the feature sequence of the data sequence corresponding to each indicator when the operating state meets the initial preset conditions, and fuse the features corresponding to multiple indicators to obtain a fused feature sequence; The second evaluation module is used to input the fused feature sequence into a prediction model to obtain a status evaluation result of the power transformer.

Images