A power grid risk identification system based on probabilistic power flow calculation

Abstract

The application discloses a power grid risk identification system based on probability power flow calculation and belongs to the technical field of power grid risk identification. A probability power flow calculation model is established based on probability power flow calculation and an identification scheme. A probability power flow estimation model is established according to the identification scheme and arranged in a corresponding user terminal. The probability input data of a corresponding unit area are acquired in real time. The corresponding probability input data are analyzed according to the preset probability power flow estimation model, and identification result data of the corresponding unit area are obtained. The probability input data are sent to a platform terminal, the received probability input data are analyzed by the corresponding probability power flow calculation model, and corresponding standard result data are obtained. The standard result data are sent to the corresponding user terminal. The corresponding probability input data, the identification result data and the standard result data are integrated into optimization learning data, and the corresponding probability power flow estimation model is optimized and adjusted according to the optimization learning data.

Description

Technical Field

[0001] This invention belongs to the field of power grid risk identification technology, specifically a power grid risk identification system based on probabilistic power flow calculation. Background Technology

[0002] With the large-scale integration of new energy sources into the grid (such as wind power and photovoltaics), their power output exhibits significant intermittency and randomness. Simultaneously, the volatility of power load increases, and the network structure dynamically changes due to faults or maintenance. These uncertainties complicate the grid's operational status, making it difficult for traditional deterministic power flow calculation methods to comprehensively assess system risks and provide a reliable basis for grid planning, operation, and dispatch. For instance, traditional risk assessments often rely on deterministic assumptions, neglecting the dynamic impact of uncertainties, thus failing to accurately reflect the actual risk level of the grid during operation.

[0003] Probabilistic power flow calculation, by introducing a probabilistic model, can quantify the impact of uncertainties on the power grid state and provide probability distribution information for state variables such as node voltage and branch power flow. This method has significant application value in power grid planning, static security analysis, and real-time monitoring of operational status; however, existing technologies still have shortcomings in terms of computational efficiency, computational resources, and real-time requirements.

[0004] Based on this, the present invention provides a power grid risk identification system based on probabilistic power flow calculation. Summary of the Invention

[0005] To address the problems of the above solutions, this invention provides a power grid risk identification system based on probabilistic power flow calculation.

[0006] The objective of this invention can be achieved through the following technical solutions:

[0007] A power grid risk identification system based on probabilistic power flow calculation, comprising a platform and a user end;

[0008] The platform includes a regional information module, a model module, and a standard analysis module;

[0009] The regional information module is used to analyze the power grid area and generate a regional information map, which includes user areas and corresponding unit areas.

[0010] Furthermore, the generation of regional information maps:

[0011] Identify each user area within the power grid area, obtain the power grid information of the user area, and segment the user area based on the power grid information to obtain several unit areas corresponding to the user area;

[0012] Obtain a power grid area map, mark the corresponding user areas and unit areas in the power grid area map, and mark the power grid area map as an area information map.

[0013] Furthermore, the user area is segmented based on power grid information, including:

[0014] Determine the unit area standard, wherein the unit area standard is that the segmented unit area meets the requirements of probabilistic power flow calculation; perform segmentation simulation on the user area according to the unit area standard to obtain several simulated segmentation methods;

[0015] The simulated segmentation methods are filtered to obtain the target segmentation method. The user area is then segmented according to the target segmentation method to obtain the corresponding unit areas.

[0016] Furthermore, the simulated segmentation methods are screened, including:

[0017] Obtain an identification scheme based on probabilistic power flow calculation; perform simulation analysis on each simulated segmentation method according to the identification scheme, and obtain the computational representativeness and implementation cost of the simulated segmentation method;

[0018] The priority value of the simulated segmentation method is calculated by inputting the calculation of representative efficiency and implementation cost into the priority formula, which is as follows:

[0019] ;

[0020] In the formula: QS is the priority value; CB is the implementation cost; e is the natural constant; η´ is the calculation representative efficiency;

[0021] The target segmentation method is determined based on the priority value.

[0022] Furthermore, the calculation of the representative efficiency includes:

[0023] The simulation is divided according to the simulation segmentation method to obtain the corresponding simulation unit region, and the computational efficiency of the probabilistic power flow calculation in the corresponding simulation unit region is estimated.

[0024] Let the simulation cell region be labeled i, i = 1, 2, ..., n, where n is the number of simulation cell regions; let the computational efficiency be labeled η. i ;

[0025] The computational efficiency of the simulation segmentation method is calculated by substituting the computational efficiency of each simulation unit region into a preset representative formula, which is:

[0026] ;

[0027] In the formula: η´ represents the calculated efficiency; Ai represents the area or number of nodes of the corresponding simulation unit region.

[0028] The model module is used to build a model, obtain the identification scheme of the power grid area, build a probabilistic power flow calculation model based on the probabilistic power flow calculation and the identification scheme, build a probabilistic power flow prediction model according to the identification scheme, and deploy the probabilistic power flow prediction model in the corresponding user terminal.

[0029] Furthermore, the identification scheme is as follows:

[0030] Determine the cell regions based on probabilistic power flow calculation and the cell regions corresponding to the probabilistic power flow prediction model;

[0031] Risk analysis is performed on the corresponding unit area based on probabilistic power flow calculation to obtain the identification result data of the corresponding unit area; the corresponding probabilistic power flow prediction model is learned and adjusted based on the identification result data.

[0032] The probabilistic power flow prediction model is used to predict real-time identification results within the corresponding unit area.

[0033] Furthermore, the cell regions based on probabilistic power flow calculations and the cell regions corresponding to the probabilistic power flow prediction models are determined, including:

[0034] Estimate whether the correlation assessment requirements are met between each unit area, and obtain the correlation assessment results between the corresponding unit areas; the correlation assessment requirements are that the corresponding unit areas can be predicted by the same probabilistic power flow prediction model.

[0035] The unit regions are classified according to the correlation assessment results to obtain the corresponding correlation classifications; each correlation classification corresponds to a probabilistic power flow prediction model.

[0036] The corresponding unit region is determined based on the association classification and probabilistic power flow calculation.

[0037] Furthermore, based on the association classification, the corresponding unit regions calculated based on probabilistic power flow are determined, including:

[0038] Select one or more cell regions from the association classification as cell regions for probabilistic power flow calculation.

[0039] The standard analysis module is used to analyze the received probability input data according to the corresponding probability power flow calculation model, obtain the corresponding standard result data, and send the standard result data to the corresponding user terminal.

[0040] The user terminal includes a monitoring module, a monitoring analysis module, and an optimization module;

[0041] The monitoring module is used to perform real-time power monitoring of the user area and obtain corresponding power grid monitoring data; it also performs feature acquisition on the power monitoring data according to preset input data items to obtain probability input data for the corresponding unit area.

[0042] The monitoring and analysis module is used to perform power grid risk analysis in user areas, obtain probability input data for each unit area, analyze the corresponding probability input data according to the preset probability power flow prediction model, and obtain the identification result data of the corresponding unit area.

[0043] The optimization module is used to optimize and adjust the preset probabilistic power flow prediction model, send the probability input data to the platform in real time, and receive the standard result data sent by the platform; integrate the probability input data, identification result data and received standard result data corresponding to the probabilistic power flow prediction model into optimization learning data, and optimize and adjust the corresponding probabilistic power flow prediction model according to the optimization learning data.

[0044] Compared with the prior art, the beneficial effects of the present invention are:

[0045] The power grid risk identification system provided by this invention addresses the low computational efficiency of existing probabilistic power flow calculation methods by designing a highly efficient algorithm. This algorithm optimizes the calculation process, reducing unnecessary repetitive calculation steps and significantly shortening the computation time. Tasks that originally required hours or even days can be completed in a shorter time, improving the timeliness of power grid risk assessment and meeting the needs of real-time power grid risk assessment. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0047] Figure 1 This is a block diagram illustrating the principle of the present invention. Detailed Implementation

[0048] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0049] like Figure 1 As shown, a power grid risk identification system based on probabilistic power flow calculation includes a platform and a user terminal.

[0050] The platform is used by the platform provider or the higher-level power department to assist the entities responsible for identifying power grid risks in various localities or regions, i.e., users; it includes a regional information module, a model module, and a standard analysis module.

[0051] The regional information module is used to analyze the power grid region and generate a regional information map. The regional information map is used to represent each user region and the corresponding unit region. The unit region is derived from the user region, that is, each unit region constitutes a user region. The user region is the power grid region where the user needs to perform power grid risk identification and management, and is included in the power grid region corresponding to the platform.

[0052] In one embodiment, the generation of a regional information map:

[0053] Identify each user area within the power grid area, obtain the power grid information of the user area, and segment the user area based on the power grid information to obtain several unit areas corresponding to the user area;

[0054] Obtain the power grid area map, which is the existing power grid information map of the power grid area. Mark the corresponding user areas and unit areas in the power grid area map, and mark the current power grid area map as the area information map.

[0055] In one embodiment, the user area is segmented based on power grid information, that is, the user area is segmented according to the unit area standard, which is that the segmented unit area can meet the requirements of probabilistic power flow calculation. Therefore, there are multiple segmentation methods, which can be performed according to existing methods or platform requirements; or other methods can be used for segmentation, such as clustering algorithms.

[0056] In one embodiment, segmenting the user area based on power grid information includes:

[0057] Determine the unit area standard, and perform segmentation simulation of the user area according to the unit area standard to obtain several simulated segmentation methods; filter the simulated segmentation methods to obtain the target segmentation method, and segment the user area according to the target segmentation method to obtain the corresponding unit areas of the user area.

[0058] In one embodiment, the screening of simulated segmentation methods can be based on existing screening methods, priority algorithms, etc.

[0059] In one embodiment, filtering the simulated segmentation method includes:

[0060] The identification scheme based on probabilistic power flow calculation is preset by the platform and is reflected in the identification process as follows: For example, a corresponding probabilistic power flow prediction model is configured for each unit area, and risk analysis is performed on each unit area based on probabilistic power flow calculation to obtain corresponding identification result data. The identification result data is used to learn and adjust the corresponding probabilistic power flow prediction model in real time. The probabilistic power flow prediction model is responsible for predicting the real-time identification result data in the unit area, which solves the problem of long probabilistic power flow calculation time. In other schemes, it is also possible not to analyze all unit areas in the user area simultaneously based on probabilistic power flow calculation. The analysis can be carried out by replacement, equivalence, rotation and other methods.

[0061] Based on the identification scheme, simulation analysis is performed on each simulated segmentation method to determine the computational representative efficiency and implementation cost corresponding to the simulated segmentation method. The computational representative efficiency is the integration of the computational probabilities of the probability flow calculation of each simulated unit region to form a representative computational efficiency, such as the mean or mode. The implementation cost is the estimated cost of implementing the identification scheme by applying the simulated segmentation method, specifically estimated using existing simulation prediction techniques.

[0062] The calculation of representative efficiency and implementation cost is input into the priority formula to calculate the priority value of the corresponding simulated segmentation method. The target segmentation method is determined based on the priority value, i.e., the one with the highest priority value. The priority formula is:

[0063] ;

[0064] In the formula: QS is the priority value; CB is the implementation cost; e is the natural constant; η´ is the calculation representative efficiency.

[0065] In one embodiment, the priority formula is:

[0066] ;

[0067] In the formula: QS is the priority value; CB is the implementation cost; e is the natural constant; η´ is the calculation representative efficiency.

[0068] In one embodiment, the priority formula may also be in other forms.

[0069] In one embodiment, calculating the representative efficiency includes:

[0070] The simulation is divided according to the simulation segmentation method to obtain the corresponding simulation unit regions. The computational efficiency of probabilistic power flow calculation in the corresponding simulation unit regions is estimated, such as based on the amount of data and the number of nodes. The simulation unit regions are labeled as i, i = 1, 2, ..., n, where n is the number of simulation unit regions. The computational efficiency is labeled as η. i ;

[0071] The computational efficiency of each simulation unit region is substituted into a preset representative formula to calculate the corresponding computational representative efficiency. The representative formula is as follows:

[0072] ;

[0073] In the formula: η´ represents the calculated representative efficiency; Ai represents the area or number of nodes of the corresponding simulation unit region, which is selected as needed and serves as the reference data for the proportion of the corresponding simulation unit region.

[0074] The model module is used to build a model, obtain an identification scheme for the power grid area, establish a probabilistic power flow calculation model based on the probabilistic power flow calculation and the identification scheme, analyze the probabilistic input data of the corresponding unit area, and obtain the corresponding identification result data; establish a probabilistic power flow prediction model according to the identification scheme, and predict the identification result data of the probabilistic power flow calculation model based on the probabilistic input data of the corresponding unit area; and deploy the probabilistic power flow prediction model in the corresponding user terminal.

[0075] In one embodiment, both the probabilistic power flow calculation model and the probabilistic power flow prediction model are built based on existing technologies. For example, the probabilistic power flow prediction model can be built based on machine learning, deep learning algorithms, etc. This involves collecting a large amount of historical data (load, renewable energy output, grid status), using intelligent models (such as LSTM, Transformer) to learn the nonlinear relationship between input and output, and directly predicting the probabilistic power flow results. Probabilistic power flow calculation generates partial training data (such as key scenarios), and intelligent models are used to supplement the predictions for other scenarios. Combining the results of both improves accuracy and efficiency.

[0076] In one embodiment, the identification scheme can be set by the platform in various ways as described in the above embodiments, as long as the unit region that needs to be analyzed based on probabilistic power flow calculation and the unit region that the probabilistic power flow prediction model can predict are determined.

[0077] In one embodiment, the identification scheme includes the determination of cell regions based on probabilistic power flow calculation and the setting of a probabilistic power flow prediction model, including:

[0078] Whether the same probabilistic power flow prediction model can be used to predict the power flow between different unit areas is mainly assessed based on the historical probability input data and corresponding historical identification results of the respective unit areas. Simulation analysis can also be performed, and the differences between simulation results are used for judgment. Various methods can be applied to make this judgment. If the corresponding correlation assessment results are obtained, the same probabilistic power flow prediction model can be used for prediction, which is considered to meet the correlation assessment requirements. Based on the correlation assessment results, the unit areas are classified to obtain the corresponding correlation classifications. Each correlation classification corresponds to a probabilistic power flow prediction model.

[0079] Based on the association classification, the corresponding unit region is determined and identified and analyzed using probabilistic power flow calculation. One or more unit regions within the association classification can be selected for analysis, or the selection can be made from the perspective of computational efficiency. The specific adjustments can be made according to the actual situation.

[0080] In one embodiment, as the prediction accuracy of the probabilistic power flow prediction model increases, the time interval for probabilistic power flow calculations can be increased to save computing resources.

[0081] The standard analysis module is used to analyze the received probability input data according to the probability flow calculation model, obtain the corresponding standard result data, and send the standard result data to the corresponding user terminal.

[0082] The user terminal includes a monitoring module, a monitoring analysis module, and an optimization module;

[0083] The monitoring module is used to perform real-time power monitoring of the user area and obtain corresponding power grid monitoring data; determine the probability input data required for probability power flow calculation, set input acquisition items according to the probability input data, and collect corresponding probability input data from the power grid monitoring data; and perform feature acquisition on the power monitoring data according to the preset input data items to obtain the probability input data of the corresponding unit area.

[0084] The monitoring and analysis module is used to perform power grid risk analysis in user areas, obtain probability input data for each unit area, analyze the corresponding probability input data according to the preset probability power flow prediction model, and obtain the identification result data of the corresponding unit area.

[0085] The optimization module is used to optimize and adjust the preset probabilistic power flow prediction model, and sends the probabilistic input data to the platform in real time. The corresponding probabilistic power flow calculation model on the platform analyzes the probabilistic input data, obtains the corresponding identification result data, marks it as standard result data, and the platform feeds back the standard result data to the user. The probabilistic input data, identification result data and received standard result data corresponding to the probabilistic power flow prediction model are integrated into optimization learning data, and the corresponding probabilistic power flow prediction model is optimized and adjusted based on the optimization learning data.

[0086] In one embodiment, the probabilistic input data transmitted to the platform is subjected to data security processing such as encryption and de-hiding.

[0087] The above formulas are all numerical calculations after removing dimensions. The formulas are obtained by software simulation based on a large amount of data and are closest to the real situation. The preset parameters and preset thresholds in the formulas are set by those skilled in the art according to the actual situation or obtained by simulation based on a large amount of data.

[0088] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A power grid risk identification system based on probabilistic power flow calculation, characterized in that, Including both the platform side and the user side; The platform includes a regional information module, a model module, and a standard analysis module; The regional information module is used to analyze the power grid region and generate a regional information map, which includes user areas and corresponding unit areas; determine the unit area standard, which is that the segmented unit areas meet the requirements of probabilistic power flow calculation; and perform segmentation simulation of the user area according to the unit area standard to obtain several simulated segmentation methods. The simulated segmentation methods are selected based on their computational efficiency and implementation cost to obtain the target segmentation method. The user area is then segmented according to the target segmentation method to obtain the corresponding unit areas. The model module is used to build a model and obtain an identification scheme for the power grid area. The identification scheme includes determining the unit area based on probabilistic power flow calculation and the unit area corresponding to the probabilistic power flow prediction model. A probabilistic power flow calculation model is established based on the probabilistic power flow calculation and identification scheme. A probabilistic power flow prediction model is established based on the identification scheme. The probabilistic power flow prediction model is then deployed in the corresponding user terminals. The standard analysis module is used to analyze the received probability input data according to the corresponding probability flow calculation model, obtain the corresponding standard result data, and send the standard result data to the corresponding user terminal. The user terminal includes a monitoring module, a monitoring analysis module, and an optimization module; The monitoring module is used to perform real-time power monitoring of the user area and obtain corresponding power grid monitoring data; it also performs feature acquisition on the power monitoring data according to preset input data items to obtain probability input data for the corresponding unit area. The monitoring and analysis module is used to perform power grid risk analysis in user areas, obtain probability input data for each unit area, analyze the corresponding probability input data according to the preset probability power flow prediction model, and obtain the identification result data of the corresponding unit area. The optimization module is used to optimize and adjust the preset probabilistic power flow prediction model, send the probabilistic input data to the platform, and receive the standard result data sent by the platform; integrate the corresponding probabilistic input data, identification result data and standard result data into optimization learning data, and optimize and adjust the corresponding probabilistic power flow prediction model according to the optimization learning data.

2. The power grid risk identification system based on probabilistic power flow calculation according to claim 1, characterized in that, Generation of regional information maps: Identify each user area within the power grid area, obtain the power grid information of the user area, and segment the user area based on the power grid information to obtain several unit areas corresponding to the user area; Obtain a power grid area map, mark the corresponding user areas and unit areas in the power grid area map, and mark the power grid area map as an area information map.

3. The power grid risk identification system based on probabilistic power flow calculation according to claim 1, characterized in that, The simulated segmentation methods were filtered, including: Obtain an identification scheme based on probabilistic power flow calculation; perform simulation analysis on each simulated segmentation method according to the identification scheme, and obtain the computational representativeness and implementation cost of the simulated segmentation method; The priority value of the simulated segmentation method is calculated by inputting the calculation of representative efficiency and implementation cost into the priority formula, which is as follows: ； In the formula: QS is the priority value; CB is the implementation cost; e is the natural constant; η´ is the calculation representative efficiency; The target segmentation method is determined based on the priority value.

4. The power grid risk identification system based on probabilistic power flow calculation according to claim 3, characterized in that, Calculations representing efficiency include: The simulation is divided according to the simulation segmentation method to obtain the corresponding simulation unit region, and the computational efficiency of the probabilistic power flow calculation in the corresponding simulation unit region is estimated. Let the simulation cell region be labeled i, i = 1, 2, ..., n, where n is the number of simulation cell regions; let the computational efficiency be labeled η. i ; The computational efficiency of the simulation segmentation method is calculated by substituting the computational efficiency of each simulation unit region into a preset representative formula, which is: ； In the formula: η´ represents the calculated efficiency; Ai represents the area or number of nodes of the corresponding simulation unit region.

5. A power grid risk identification system based on probabilistic power flow calculation according to claim 1, characterized in that, The identification scheme is as follows: Determine the cell regions based on probabilistic power flow calculation and the cell regions corresponding to the probabilistic power flow prediction model; Risk analysis is performed on the corresponding unit area based on probabilistic power flow calculation to obtain the identification result data of the corresponding unit area; the corresponding probabilistic power flow prediction model is learned and adjusted based on the identification result data. The probabilistic power flow prediction model is used to predict real-time identification results within the corresponding unit area.

6. A power grid risk identification system based on probabilistic power flow calculation according to claim 5, characterized in that, Determine the cell regions based on probabilistic power flow calculations and the cell regions corresponding to the probabilistic power flow prediction models, including: Estimate whether the correlation assessment requirements are met between each unit area, and obtain the correlation assessment results between the corresponding unit areas; the correlation assessment requirements are that the corresponding unit areas can be predicted by the same probabilistic power flow prediction model. The unit regions are classified according to the correlation assessment results to obtain the corresponding correlation classifications; each correlation classification corresponds to a probabilistic power flow prediction model. The corresponding unit region is determined based on the association classification and probabilistic power flow calculation.

7. A power grid risk identification system based on probabilistic power flow calculation according to claim 6, characterized in that, Based on the association classification, the corresponding unit regions calculated based on probabilistic power flow are determined, including: Select one or more cell regions from the association classification as cell regions for probabilistic power flow calculation.

Images