Method and system for correcting reliability of power supply system based on independent power grid

Abstract

The invention discloses a correction method and system based on reliability of an independent power grid power supply system, and relates to the technical field of electric power automation. Judging whether the power supply load state of the future independent power grid is in the actual power supply load demand range interval of the independent power grid or not, and if the independent power grid is judged to be abnormal, generating a fault probability distribution diagram of the component and a risk type database of the component; correlation matching is carried out, and risk coefficients of components of the independent power grid are determined; correcting the predicted future power supply load state of the independent power grid according to the risk coefficients of the components of the independent power grid, performing reliability verification according to the simulation function SIM, and generating a reliability score of the independent power grid; generating a maintenance task of the independent power grid power supply system according to the stability level corresponding to the reliability score of the independent power grid; the method has the advantages that the power supply reliability of an independent power grid is improved, and the stability of a power system is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power automation, and in particular to a correction method and system based on the reliability of an independent power grid power supply system. Background Art

[0002] As the components of an independent power grid progress in service, their overall stability gradually decreases. However, when there is no substantial power grid disaster, components facing abnormal fluctuations can still request restarts until all independent power grid components are online to ensure the normal operation of the independent power grid. However, when the overall stability of the components of the independent power grid is at a relatively low level, repeated restart requests will further consume the resources of the overall system and cause the independent power grid power supply system to collapse. Summary of the invention

[0003] In order to solve the above technical problems, a method and system for correcting the reliability of an independent power grid power supply system are provided. This technical solution solves the above problems.

[0004] In order to achieve the above purpose, the technical solution adopted by the present invention is: A correction method and system based on reliability of an independent power grid power supply system, comprising: Based on the historical power supply data of the independent power grid, analyze the power supply trend of the historical power supply data and predict the power supply load status of the independent power grid in the future; Determine whether the predicted future power supply load state of the independent power grid is within the actual power supply load demand range of the independent power grid. If so, determine that the independent power grid is normal; if not, determine that the independent power grid is abnormal; If it is determined that the independent power grid is abnormal, the operation log of the independent power grid is obtained, the historical operation parameters of the components of the independent power grid are marked, the deviation distribution of the components is analyzed, and a failure probability distribution diagram of the components is generated; Based on the known risk types of power grid components, a risk category database of component components is established; Determine the risk coefficient of the components of the independent power grid by associating and matching the failure probability distribution map of the components of the independent power grid with the risk category database of the components; According to the risk coefficients of the components of the independent power grid, the power supply load state of the predicted future independent power grid is corrected, and the reliability verification is performed according to the simulation function SIM to generate the reliability score of the independent power grid; Generate maintenance tasks for the independent power grid power supply system according to the stability level corresponding to the reliability score of the independent power grid; The correction according to the risk coefficient of the components of the independent power grid for predicting the power supply load state of the future independent power grid is specifically: ; In the formula, To correct and predict the power supply load state index of the future independent power grid for t unit time, To predict the power supply load state index of the future independent power grid for t unit time, is the risk factor of the vth component of the independent power grid; The reliability score of the independent power grid is specifically: ; In the formula, is the reliability score of the node in the independent power grid at t unit time.

[0005] Preferably, based on the historical power supply data of the independent power grid, analyzing the power supply trend of the historical power supply data, and predicting the power supply load state of the future independent power grid specifically includes: Based on the historical power supply data of the independent power grid, pre-process the missing values ​​and abnormal values ​​in the historical power supply data; According to the historical power supply data of the independent power grid, the data is divided according to the time attribute to obtain the historical power supply time series data of the independent power grid; Based on the historical power supply time series data of the independent power grid, the STL time series analysis method is used to extract the features of the historical power supply time series data to obtain the historical power supply time series feature data of the independent power grid; Based on the ARIMA autoregressive moving average model, an independent power grid power supply load analysis model is constructed; The historical power supply time series characteristic data of the independent power grid is substituted into the power supply load analysis model of the independent power grid, with the historical power supply time series characteristics as input, the minimum model error function under the optimal autoregressive coefficient and moving average coefficient of the model as the training end target, and the predicted power supply load state index of the future independent power grid as the output; The power supply load analysis model of the independent power grid is specifically as follows: ; In the formula, To predict the power supply load state index of the future independent power grid for t unit time, is a constant term, is the i-th lag autoregressive coefficient, For independent power grid The historical power supply timing characteristic value of unit time lag order, is the jth moving average coefficient, For independent power grid The random error of the moving average order per unit time, n is the total number of lag orders, and m is the total number of moving average orders.

[0006] Preferably, determining whether the predicted future power supply load state of the independent power grid is within the actual power supply load demand range of the independent power grid specifically includes: According to the time series, the power supply load state index of the predicted future independent power grid and the actual power supply load demand range of the independent power grid are marked, and the power supply load state index of each unit time node of the predicted future independent power grid and the actual power supply load demand range of each unit time node of the independent power grid are determined; The linear mapping is used to perform vector conversion on the power supply load state index of each unit time node of the predicted future independent power grid and the actual power supply load demand range of each unit time node of the independent power grid, so as to obtain the power supply load state vector index of each unit time node of the predicted future independent power grid and the actual power supply load vector demand range of each unit time node of the independent power grid; The Euclidean distance formula is used to calculate and predict the spatial range between the power supply load state vector index of each unit time node of the future independent power grid and the actual power supply load vector demand range of each unit time node of the independent power grid, and to determine whether the power supply load state vector index of each unit time node is within the actual power supply load vector demand range.

[0007] Preferably, the step of determining that the power supply load state vector indicator of each unit time node is within the actual power supply load vector requirement range is specifically: ; In the formula, To predict the actual power supply load vector demand range of the future independent power grid at the t-th unit time node, To predict the power supply load state vector index of the t-th unit time node of the future independent power grid, is the lower limit of the actual power supply load vector demand range of the t-th unit time node of the independent power grid, It is the upper limit of the actual power supply load vector demand range of the t-th unit time node of the independent power grid.

[0008] Preferably, if it is determined that the independent power grid is abnormal, obtaining the operation log of the independent power grid, marking the historical operation parameters of the components of the independent power grid, analyzing the deviation distribution of the components, and generating the fault probability distribution diagram of the components specifically includes: Based on the service life of the independent power grid, determine the basic performance loss status of the components of the independent power grid; Initialize the standardized operation parameter array of the components of the independent power grid according to the basic performance loss status of the components of the independent power grid; Based on the operation log of the independent power grid, the historical operation parameters of the components of the independent power grid are obtained and the standardized operation parameter array of the components of the initialized independent power grid is used to perform deviation operation, and the deviation parameters of the components of the independent power grid are marked and recorded as abnormal characteristic data of the components of the independent power grid; The abnormal characteristic data of the components of the independent power grid are taken as positive samples, and the standardized operating parameter arrays of the components of the initialized independent power grid are taken as negative samples, so as to establish the original data set of the components of the independent power grid; Based on Logistic regression, a failure probability prediction model for components of an independent power grid is constructed; The original data set of the components of the independent power grid is substituted into the failure probability prediction model of the components of the independent power grid, the abnormal characteristic data of the components of the independent power grid is taken as input, and the failure probability value of the components of the independent power grid is taken as output to generate a failure probability distribution map of the components of the independent power grid.

[0009] Preferably, the failure probability prediction model of the components of the independent power grid is specifically: ; In the formula, is the failure probability value of the components of the independent power grid under the given abnormal characteristic data U of the components of the independent power grid, is the kth abnormal characteristic data of the vth component of the independent power grid, is the intercept term, , , is the regression coefficient, is a logarithmic function.

[0010] Preferably, the risk coefficient of the components of the independent power grid is determined by associating and matching the failure probability distribution map of the components of the independent power grid with the risk category database of the components; Based on the failure probability distribution diagram of the components of the independent power grid, the failure probability distribution value of each component of the independent power grid is standardized; Using one-hot encoding to perform vector conversion on the fault probability distribution graph of the components of the independent power grid, the fault probability distribution vector of each component of the independent power grid is obtained; Analyze the matching degree between the failure probability distribution vector of the components of each independent power grid and the risk types in the risk type database of the components, and determine the failure risk type matching index of the components of each independent power grid; According to the fault risk type matching index of each component of the independent power grid, weights are assigned to the fault risk types of the components of the independent power grid; The risk coefficients of the components of the independent power grid are calculated according to the failure probability values ​​of the components of the independent power grid and the failure risk type weights of the components of the independent power grid.

[0011] Preferably, the determination of the fault risk type matching index of the components of each independent power grid specifically includes: ; In the formula, is the z-th risk type matching index of the v-th component failure of the independent power grid, is the failure probability distribution vector of the vth component of the independent power grid, is the zth risk category of the vth component in the risk category database; The risk factors of the components of the independent power grid specifically include: ; In the formula, is the risk factor of the vth component of the independent power grid, is the failure probability value of the vth component of the independent power grid, is the zth risk type weight of the vth component failure of the independent power grid.

[0012] The present invention also discloses a correction system based on the reliability of an independent power grid power supply system, comprising: The load forecasting module analyzes the power supply trend of the historical power supply data based on the independent power grid and predicts the power supply load status of the independent power grid in the future; The abnormality judgment module is electrically connected to the load prediction module. The abnormality judgment module is used to judge whether the power supply load state of the predicted future independent power grid is within the actual power supply load demand range of the independent power grid. If so, it is determined that the independent power grid has no abnormality. If not, it is determined that the independent power grid has an abnormality. A fault probability map module, the fault probability map module is electrically connected to the abnormality judgment module, and the fault probability map module is used to obtain the operation log of the independent power grid, mark the historical operation parameters of the components of the independent power grid, analyze the deviation distribution of the components, and generate a fault probability distribution map of the components if it is determined that the independent power grid has an abnormality; Risk database, based on the known risk types of power grid components, establish a risk category database of component components; A risk coefficient module, the risk coefficient module is electrically connected to the risk database and the fault probability map module, and the risk coefficient module is used to determine the risk coefficient of the component of the independent power grid by associating and matching the fault probability distribution map of the component of the independent power grid with the risk category database of the component; A reliability verification module, the reliability verification module is electrically connected to the risk coefficient module, and the reliability verification module is used to correct the power supply load state of the predicted future independent power grid according to the risk coefficient of the components of the independent power grid, perform reliability verification according to the simulation function SIM, and generate a reliability score of the independent power grid; A maintenance task module is electrically connected to the reliability verification module, and is used to generate maintenance tasks for the independent power grid power supply system according to the stability level corresponding to the reliability score of the independent power grid.

[0013] Compared with the prior art, the present invention has the following beneficial effects: The present invention proposes a correction scheme based on the reliability of the independent power grid power supply system. By analyzing the historical power supply data and real-time power supply data of the independent power grid, the future power supply load state is predicted, and compared with the actual demand to determine whether there is an abnormality in the power grid. Once an abnormality is found, the system evaluates the risk factor of the power grid by analyzing the operation log and fault probability distribution of the power grid components, combined with a known risk type database. The risk factor is used to correct the future load state, and the reliability is verified through a simulation function, and finally a reliability score of the power grid is generated. The beneficial effect of this scheme is to improve the power supply reliability of the independent power grid and enhance the stability of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is a flow chart of a correction method based on the reliability of an independent power grid power supply system; Figure 2 A flow chart of a method for predicting the power supply load state of a future independent power grid; Figure 3 A flow chart of a method for judging whether the power supply load state of the future independent power grid is within the actual power supply load demand range of the independent power grid; Figure 4 A flow chart of a method for generating a failure probability distribution diagram of a component; Figure 5 Flowchart of a method for determining risk factors of components of an independent power grid; Figure 6 It is a framework diagram of a correction system based on the reliability of an independent power grid power supply system. DETAILED DESCRIPTION

[0015] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0016] Reference Figure 1 As shown, a correction method based on the reliability of an independent power grid power supply system includes: Based on the historical power supply data of the independent power grid, analyze the power supply trend of the historical power supply data and predict the power supply load status of the independent power grid in the future; Determine whether the predicted future power supply load state of the independent power grid is within the actual power supply load demand range of the independent power grid. If so, determine that the independent power grid is normal; if not, determine that the independent power grid is abnormal; If it is determined that the independent power grid is abnormal, the operation log of the independent power grid is obtained, the historical operation parameters of the components of the independent power grid are marked, the deviation distribution of the components is analyzed, and a failure probability distribution diagram of the components is generated; Based on the known risk types of power grid components, a risk category database of component components is established; Determine the risk coefficient of the components of the independent power grid by associating and matching the failure probability distribution map of the components of the independent power grid with the risk category database of the components; According to the risk coefficients of the components of the independent power grid, the power supply load state of the predicted future independent power grid is corrected, and the reliability verification is performed according to the simulation function SIM to generate the reliability score of the independent power grid; Generate maintenance tasks for the independent power grid power supply system according to the stability level corresponding to the reliability score of the independent power grid; The correction according to the risk coefficient of the components of the independent power grid for predicting the power supply load state of the future independent power grid is specifically: ; In the formula, To correct and predict the power supply load state index of the future independent power grid for t unit time, To predict the power supply load state index of the future independent power grid for t unit time, is the risk factor of the vth component of the independent power grid; The reliability score of the independent power grid is specifically: ; In the formula, is the reliability score of the node in the independent power grid at t unit time.

[0017] It should be noted that the simulation function SIM here is the interface of the simulink model.

[0018] This solution predicts the future power supply load status by analyzing the historical power supply data and real-time power supply data of the independent power grid, and compares it with the actual demand to determine whether there is an abnormality in the power grid. Once an abnormality is found, the system analyzes the operation logs and fault probability distribution of the power grid components, combined with the known risk type database, to evaluate the risk factor of the power grid. The risk factor is used to correct the future load status, and the reliability is verified through simulation functions, and finally the reliability score of the power grid is generated. The beneficial effect of this solution is to improve the power supply reliability of the independent power grid and enhance the stability of the power system.

[0019] Reference Figure 2 As shown, based on the historical power supply data of the independent power grid, the power supply trend of the historical power supply data is analyzed, and the power supply load state of the future independent power grid is predicted, which specifically includes: Based on the historical power supply data of the independent power grid, pre-process the missing values ​​and abnormal values ​​in the historical power supply data; According to the historical power supply data of the independent power grid, the data is divided according to the time attribute to obtain the historical power supply time series data of the independent power grid; Based on the historical power supply time series data of the independent power grid, the STL time series analysis method is used to extract the features of the historical power supply time series data to obtain the historical power supply time series feature data of the independent power grid; Based on the ARIMA autoregressive moving average model, an independent power grid power supply load analysis model is constructed; The historical power supply time series characteristic data of the independent power grid is substituted into the power supply load analysis model of the independent power grid, with the historical power supply time series characteristics as input, the minimum model error function under the optimal autoregressive coefficient and moving average coefficient of the model as the training end target, and the predicted power supply load state index of the future independent power grid as the output; The power supply load analysis model of the independent power grid is specifically as follows: ; In the formula, To predict the power supply load state index of the future independent power grid for t unit time, is a constant term, is the i-th lag autoregressive coefficient, For independent power grid The historical power supply timing characteristic value of unit time lag order, is the jth moving average coefficient, For independent power grid The random error of the moving average order per unit time, n is the total number of lag orders, and m is the total number of moving average orders.

[0020] Reference Figure 3As shown, judging whether the power supply load state of the predicted future independent power grid is within the actual power supply load demand range of the independent power grid specifically includes: According to the time series, the power supply load state index of the predicted future independent power grid and the actual power supply load demand range of the independent power grid are marked, and the power supply load state index of each unit time node of the predicted future independent power grid and the actual power supply load demand range of each unit time node of the independent power grid are determined; The linear mapping is used to perform vector conversion on the power supply load state index of each unit time node of the predicted future independent power grid and the actual power supply load demand range of each unit time node of the independent power grid, so as to obtain the power supply load state vector index of each unit time node of the predicted future independent power grid and the actual power supply load vector demand range of each unit time node of the independent power grid; Using the Euclidean distance formula, the spatial range between the power supply load state vector index of each unit time node of the future independent power grid and the actual power supply load vector demand range of each unit time node of the independent power grid is calculated and predicted, and the power supply load state vector index of each unit time node is determined to be within the actual power supply load vector demand range interval; The method of determining that the power supply load state vector indicator of each unit time node is within the actual power supply load vector demand range is specifically: ; In the formula, To predict the actual power supply load vector demand range of the future independent power grid at the t-th unit time node, To predict the power supply load state vector index of the t-th unit time node of the future independent power grid, is the lower limit of the actual power supply load vector demand range of the t-th unit time node of the independent power grid, It is the upper limit of the actual power supply load vector demand range of the t-th unit time node of the independent power grid.

[0021] This solution uses time series to mark the predicted power supply load and actual power supply load demand, uses linear mapping to convert these data into vector form, and then calculates the Euclidean distance to determine whether the predicted power supply load state is within the actual power supply load demand range, thereby determining whether the independent power grid is abnormal. The beneficial effect is that it can accurately and efficiently evaluate the power supply load state of the independent power grid, and timely discover and solve potential power grid abnormalities.

[0022] Reference Figure 4 As shown, if it is determined that the independent power grid is abnormal, the operation log of the independent power grid is obtained, the historical operation parameters of the components of the independent power grid are marked, the deviation distribution of the components is analyzed, and the fault probability distribution diagram of the components is generated, which specifically includes: Based on the service life of the independent power grid, determine the basic performance loss status of the components of the independent power grid; Initialize the standardized operation parameter array of the components of the independent power grid according to the basic performance loss status of the components of the independent power grid; Based on the operation log of the independent power grid, the historical operation parameters of the components of the independent power grid are obtained and the standardized operation parameter array of the components of the initialized independent power grid is used to perform deviation operation, and the deviation parameters of the components of the independent power grid are marked and recorded as abnormal characteristic data of the components of the independent power grid; The abnormal characteristic data of the components of the independent power grid are taken as positive samples, and the standardized operating parameter arrays of the components of the initialized independent power grid are taken as negative samples, so as to establish the original data set of the components of the independent power grid; Based on Logistic regression, a failure probability prediction model for components of an independent power grid is constructed; Substituting the original data set of the components of the independent power grid into the failure probability prediction model of the components of the independent power grid, taking the abnormal characteristic data of the components of the independent power grid as input and the failure probability value of the components of the independent power grid as output, a failure probability distribution map of the components of the independent power grid is generated; The failure probability prediction model of the components of the independent power grid is specifically: ; In the formula, is the failure probability value of the components of the independent power grid under the given abnormal characteristic data U of the components of the independent power grid, is the kth abnormal characteristic data of the vth component of the independent power grid, is the intercept term, , , is the regression coefficient, is a logarithmic function.

[0023] Based on the service life of the independent power grid, the basic performance loss status of its components is evaluated, and the standardized operating parameter array is initialized. Next, the historical operating parameters of the components are obtained using the power grid operation log, and the deviation operation is performed with the standardized parameters to mark the abnormal feature data. Then, an original data set containing positive samples of abnormal feature data and negative samples of standardized parameters is constructed. Based on these data, a fault probability prediction model is constructed using Logistic regression. Finally, the original data set is substituted into the model to generate a fault probability distribution map of the components, providing a reference vector for the subsequent identification of specific faults of the components.

[0024] Reference Figure 5As shown, the risk coefficients of the components of the independent power grid are determined by associating and matching the failure probability distribution diagram of the components of the independent power grid with the risk category database of the components; Based on the failure probability distribution diagram of the components of the independent power grid, the failure probability distribution value of each component of the independent power grid is standardized; Using one-hot encoding to perform vector conversion on the fault probability distribution graph of the components of the independent power grid, the fault probability distribution vector of each component of the independent power grid is obtained; Analyze the matching degree between the failure probability distribution vector of the components of each independent power grid and the risk types in the risk type database of the components, and determine the failure risk type matching index of the components of each independent power grid; According to the fault risk type matching index of each component of the independent power grid, weights are assigned to the fault risk types of the components of the independent power grid; Calculate the risk coefficient of the components of the independent power grid according to the failure probability value of the components of the independent power grid and the failure risk type weight of the components of the independent power grid; The determination of the fault risk type matching index of each component of the independent power grid specifically includes: ; In the formula, is the z-th risk type matching index of the v-th component failure of the independent power grid, is the failure probability distribution vector of the vth component of the independent power grid, is the zth risk category of the vth component in the risk category database; The risk factors of the components of the independent power grid specifically include: ; In the formula, is the risk factor of the vth component of the independent power grid, is the failure probability value of the vth component of the independent power grid, is the zth risk type weight of the vth component failure of the independent power grid.

[0025] This scheme standardizes the fault probability distribution diagrams of the components of the independent power grid to ensure that the fault probability values ​​of each component are on the same scale, and uses the unique hot encoding technology to convert these fault probability distribution diagrams into vectors to match them with the risk type database of the components, and determine the fault risk type matching index of each component. Based on the fault risk type matching index, each fault risk type is assigned a corresponding weight. Finally, combined with the fault probability value of the component and the fault risk type weight, the risk coefficient of the components of the independent power grid is comprehensively calculated, which can more truly reflect the actual risk situation of the power grid.

[0026] Reference Figure 6 As shown, based on the same inventive concept of a correction method based on the reliability of an independent power grid power supply system, a correction system based on the reliability of an independent power grid power supply system is proposed, comprising: The load forecasting module analyzes the power supply trend of the historical power supply data based on the independent power grid and predicts the power supply load status of the independent power grid in the future; The abnormality judgment module is electrically connected to the load prediction module. The abnormality judgment module is used to judge whether the power supply load state of the predicted future independent power grid is within the actual power supply load demand range of the independent power grid. If so, it is determined that the independent power grid has no abnormality. If not, it is determined that the independent power grid has an abnormality. A fault probability map module, the fault probability map module is electrically connected to the abnormality judgment module, and the fault probability map module is used to obtain the operation log of the independent power grid, mark the historical operation parameters of the components of the independent power grid, analyze the deviation distribution of the components, and generate a fault probability distribution map of the components if it is determined that the independent power grid has an abnormality; Risk database, based on the known risk types of power grid components, establish a risk category database of component components; A risk coefficient module, the risk coefficient module is electrically connected to the risk database and the fault probability map module, and the risk coefficient module is used to determine the risk coefficient of the component of the independent power grid by associating and matching the fault probability distribution map of the component of the independent power grid with the risk category database of the component; A reliability verification module, the reliability verification module is electrically connected to the risk coefficient module, and the reliability verification module is used to correct the power supply load state of the predicted future independent power grid according to the risk coefficient of the components of the independent power grid, perform reliability verification according to the simulation function SIM, and generate a reliability score of the independent power grid; A maintenance task module is electrically connected to the reliability verification module, and is used to generate maintenance tasks for the independent power grid power supply system according to the stability level corresponding to the reliability score of the independent power grid.

[0027] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A correction method based on the reliability of an independent power grid power supply system, characterized in that: include: Based on the historical power supply data of the independent power grid, analyze the power supply trend of the historical power supply data and predict the power supply load status of the independent power grid in the future; Determine whether the predicted future power supply load state of the independent power grid is within the actual power supply load demand range of the independent power grid. If so, determine that the independent power grid is normal; if not, determine that the independent power grid is abnormal; If it is determined that the independent power grid is abnormal, the operation log of the independent power grid is obtained, the historical operation parameters of the components of the independent power grid are marked, the deviation distribution of the components is analyzed, and a failure probability distribution diagram of the components is generated; Based on the known risk types of power grid components, a risk category database of component components is established; Determine the risk coefficient of the components of the independent power grid by associating and matching the failure probability distribution map of the components of the independent power grid with the risk category database of the components; According to the risk coefficients of the components of the independent power grid, the power supply load state of the predicted future independent power grid is corrected, and the reliability verification is performed according to the simulation function SIM to generate the reliability score of the independent power grid; Generate maintenance tasks for the independent power grid power supply system according to the stability level corresponding to the reliability score of the independent power grid; The correction according to the risk coefficient of the components of the independent power grid for predicting the power supply load state of the future independent power grid is specifically: ； In the formula, To correct and predict the power supply load state index of the future independent power grid for t unit time, To predict the power supply load state index of the future independent power grid for t unit time, is the risk factor of the vth component of the independent power grid; The reliability score of the independent power grid is specifically: ； In the formula, is the reliability score of the node in the independent power grid at t unit time.

2. A method for correcting the reliability of an independent power grid power supply system according to claim 1, characterized in that: Based on the historical power supply data of the independent power grid, the power supply trend of the historical power supply data is analyzed, and the power supply load status of the independent power grid in the future is predicted, including: Based on the historical power supply data of the independent power grid, pre-process the missing values ​​and abnormal values ​​in the historical power supply data; According to the historical power supply data of the independent power grid, the data is divided according to the time attribute to obtain the historical power supply time series data of the independent power grid; Based on the historical power supply time series data of the independent power grid, the STL time series analysis method is used to extract the features of the historical power supply time series data to obtain the historical power supply time series feature data of the independent power grid; Based on the ARIMA autoregressive moving average model, an independent power grid power supply load analysis model is constructed; The historical power supply time series characteristic data of the independent power grid is substituted into the power supply load analysis model of the independent power grid, with the historical power supply time series characteristics as input, the minimum model error function under the optimal autoregressive coefficient and moving average coefficient of the model as the training end target, and the predicted power supply load state index of the future independent power grid as the output; The power supply load analysis model of the independent power grid is specifically as follows: ； In the formula, To predict the power supply load state index of the future independent power grid for t unit time, is a constant term, is the i-th lag autoregressive coefficient, For independent power grid The historical power supply timing characteristic value of unit time lag order, is the jth moving average coefficient, For independent power grid The random error of the moving average order per unit time, n is the total number of lag orders, and m is the total number of moving average orders.

3. A correction method based on the reliability of an independent power grid power supply system according to claim 2, characterized in that: The judgment of whether the power supply load state of the predicted future independent power grid is within the actual power supply load demand range of the independent power grid specifically includes: According to the time series, the power supply load state index of the predicted future independent power grid and the actual power supply load demand range of the independent power grid are marked, and the power supply load state index of each unit time node of the predicted future independent power grid and the actual power supply load demand range of each unit time node of the independent power grid are determined; The linear mapping is used to perform vector conversion on the power supply load state index of each unit time node of the predicted future independent power grid and the actual power supply load demand range of each unit time node of the independent power grid, so as to obtain the power supply load state vector index of each unit time node of the predicted future independent power grid and the actual power supply load vector demand range of each unit time node of the independent power grid; The Euclidean distance formula is used to calculate and predict the spatial range between the power supply load state vector index of each unit time node of the future independent power grid and the actual power supply load vector demand range of each unit time node of the independent power grid, and to determine whether the power supply load state vector index of each unit time node is within the actual power supply load vector demand range.

4. A method for correcting the reliability of an independent power grid power supply system according to claim 3, characterized in that: The specific determination of the power supply load state vector index of each unit time node being within the actual power supply load vector demand range is: ； In the formula, To predict the actual power supply load vector demand range of the future independent power grid at the t-th unit time node, To predict the power supply load state vector index of the t-th unit time node of the future independent power grid, is the lower limit of the actual power supply load vector demand range of the t-th unit time node of the independent power grid, It is the upper limit of the actual power supply load vector demand range of the t-th unit time node of the independent power grid.

5. A method for correcting the reliability of an independent power grid power supply system according to claim 4, characterized in that: If it is determined that the independent power grid is abnormal, the operation log of the independent power grid is obtained, the historical operation parameters of the components of the independent power grid are marked, the deviation distribution of the components is analyzed, and the fault probability distribution diagram of the components is generated. Specifically, the following steps are performed: Based on the service life of the independent power grid, determine the basic performance loss status of the components of the independent power grid; Initialize the standardized operation parameter array of the components of the independent power grid according to the basic performance loss status of the components of the independent power grid; Based on the operation log of the independent power grid, the historical operation parameters of the components of the independent power grid are obtained and the standardized operation parameter array of the components of the initialized independent power grid is used to perform deviation operation, and the deviation parameters of the components of the independent power grid are marked and recorded as abnormal characteristic data of the components of the independent power grid; The abnormal characteristic data of the components of the independent power grid are taken as positive samples, and the standardized operating parameter arrays of the components of the initialized independent power grid are taken as negative samples, so as to establish the original data set of the components of the independent power grid; Based on Logistic regression, a failure probability prediction model for components of an independent power grid is constructed; The original data set of the components of the independent power grid is substituted into the failure probability prediction model of the components of the independent power grid, the abnormal characteristic data of the components of the independent power grid is taken as input, and the failure probability value of the components of the independent power grid is taken as output to generate a failure probability distribution map of the components of the independent power grid.

6. A method for correcting the reliability of an independent power grid power supply system according to claim 5, characterized in that: The failure probability prediction model of the components of the independent power grid is specifically: ； In the formula, is the failure probability value of the components of the independent power grid under the given abnormal characteristic data U of the components of the independent power grid, is the kth abnormal characteristic data of the vth component of the independent power grid, is the intercept term, , , is the regression coefficient, is a logarithmic function.

7. A method for correcting the reliability of an independent power grid power supply system according to claim 6, characterized in that: Determine the risk coefficient of the components of the independent power grid by associating and matching the failure probability distribution map of the components of the independent power grid with the risk category database of the components; Based on the failure probability distribution diagram of the components of the independent power grid, the failure probability distribution value of each component of the independent power grid is standardized; Using one-hot encoding to perform vector conversion on the fault probability distribution graph of the components of the independent power grid, the fault probability distribution vector of each component of the independent power grid is obtained; Analyze the matching degree between the failure probability distribution vector of the components of each independent power grid and the risk types in the risk type database of the components, and determine the failure risk type matching index of the components of each independent power grid; According to the fault risk type matching index of each component of the independent power grid, weights are assigned to the fault risk types of the components of the independent power grid; The risk coefficients of the components of the independent power grid are calculated according to the failure probability values ​​of the components of the independent power grid and the failure risk type weights of the components of the independent power grid.

8. The method for correcting the reliability of an independent power grid power supply system according to claim 7, characterized in that: Determine the fault risk type matching index of each independent power grid component, including: ； In the formula, is the z-th risk type matching index of the v-th component failure of the independent power grid, is the failure probability distribution vector of the vth component of the independent power grid, is the zth risk category of the vth component in the risk category database; The risk factors of the components of the independent power grid specifically include: ； In the formula, is the risk factor of the vth component of the independent power grid, is the failure probability value of the vth component of the independent power grid, is the zth risk type weight of the vth component failure of the independent power grid.

9. A correction system based on the reliability of an independent power grid power supply system, characterized in that: A method for correcting the reliability of an independent power grid power supply system according to any one of claims 1 to 8, comprising: The load forecasting module analyzes the power supply trend of the historical power supply data based on the independent power grid and predicts the power supply load status of the independent power grid in the future; The abnormality judgment module is electrically connected to the load prediction module. The abnormality judgment module is used to judge whether the power supply load state of the predicted future independent power grid is within the actual power supply load demand range of the independent power grid. If so, it is determined that the independent power grid has no abnormality. If not, it is determined that the independent power grid has an abnormality. A fault probability map module, the fault probability map module is electrically connected to the abnormality judgment module, and the fault probability map module is used to obtain the operation log of the independent power grid, mark the historical operation parameters of the components of the independent power grid, analyze the deviation distribution of the components, and generate a fault probability distribution map of the components if it is determined that the independent power grid has an abnormality; Risk database, based on the known risk types of power grid components, establish a risk category database of component components; A risk coefficient module, the risk coefficient module is electrically connected to the risk database and the fault probability map module, and the risk coefficient module is used to determine the risk coefficient of the component of the independent power grid by associating and matching the fault probability distribution map of the component of the independent power grid with the risk category database of the component; A reliability verification module, the reliability verification module is electrically connected to the risk coefficient module, and the reliability verification module is used to correct the power supply load state of the predicted future independent power grid according to the risk coefficient of the components of the independent power grid, perform reliability verification according to the simulation function SIM, and generate a reliability score of the independent power grid; A maintenance task module is electrically connected to the reliability verification module, and is used to generate maintenance tasks for the independent power grid power supply system according to the stability level corresponding to the reliability score of the independent power grid.

Images