Power grid operation risk prediction method, system, computer equipment and storage medium

Abstract

The present invention belongs to the field of power system automation and discloses a grid operation risk prediction method, system, computer device, and storage medium. The method comprises: selecting, from a plurality of preset grid operation risk prediction models, a grid operation risk prediction model having the same grid operation gradient level as the current grid operation gradient level and the same grid operation safety evaluation index as the grid operation safety evaluation index to be predicted, based on the current grid new energy access gradient level and the grid operation safety evaluation index to be predicted, to obtain a target grid operation risk prediction model; determining, based on the grid operation safety evaluation index to be predicted, the current prediction characteristic variables of the grid and inputting them into the target grid operation risk prediction model, obtaining a predicted value, and determining the grid operation risk based on the predicted value. By classifying the new energy access gradient level and corresponding the grid operation safety evaluation index with the prediction characteristic variables, the accuracy of grid operation risk prediction is effectively improved.

Description

Technical Field

[0001] The present invention belongs to the field of power system automation and relates to a power grid operation risk prediction method, system, computer equipment and storage medium. Background Art

[0002] A core characteristic of the new power system is the dominance of renewable energy as the primary energy source. With the goals of achieving carbon peak and carbon neutrality, the proportion of renewable energy in primary energy consumption continues to increase, accelerating the replacement of fossil fuels. In the future, the installed capacity of power generation systems will maintain steady and rapid growth, with wind and solar power leading the way, and multi-source coordinated power generation expected to be the fastest-growing power source.

[0003] However, renewable energy output is highly uncertain, characterized by randomness, volatility, and peak-shaving. "No wind during extreme heat," "no solar power during late peak hours," and "large installed capacity, low power consumption" have become industry drawbacks. Furthermore, with the continuous development of renewable energy and the electricity market, the random fluctuations and frequent market transactions of renewable energy power sources have made the grid's operating environment increasingly unpredictable. This is especially true after large-scale renewable energy is connected to the grid. The intermittent, fluctuating, and random characteristics of renewable energy significantly impact the power system's active-frequency and reactive-voltage characteristics, as well as power quality. This reduces the power system's adjustability, easily leading to large-scale random power flow shifts and causing grid operational risks such as power limit violations and voltage instability. Summary of the Invention

[0004] The purpose of the present invention is to overcome the shortcomings of the above-mentioned prior art and provide a power grid operation risk prediction method, system, computer equipment and storage medium.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] A first aspect of the present invention provides a method for predicting power grid operation risks, comprising:

[0007] Obtain the current renewable energy access gradient level of the power grid and the power grid operation safety evaluation indicators to be predicted;

[0008] According to the current new energy access gradient level and the grid operation safety evaluation index to be predicted, a grid operation risk prediction model with the same new energy access gradient level as the current new energy access gradient level and the same grid operation safety evaluation index as the grid operation safety evaluation index to be predicted is selected from several preset grid operation risk prediction models to obtain a target grid operation risk prediction model;

[0009] According to the power grid operation safety evaluation index to be predicted, the current prediction characteristic variables of the power grid are determined, and the data of the current prediction characteristic variables are obtained and input into the target power grid operation risk prediction model to obtain the predicted value of the power grid operation safety evaluation index to be predicted, and the power grid operation risk is determined based on the predicted value.

[0010] Optionally, the preset plurality of power grid operation risk prediction models are constructed by the following construction method, which includes:

[0011] Obtaining historical operation data of the power grid within a preset time period; wherein the historical operation data includes historical data of each preset characteristic variable of the power grid;

[0012] Obtain the new energy scenario coefficient at each historical operation data sampling time, and obtain the new energy scenario coefficient range for each new energy access gradient level of the power grid based on the new energy scenario coefficient at each historical operation data sampling time and the number of new energy access gradient levels preset by the power grid;

[0013] Determine the historical operating data for each new energy access gradient level based on the new energy scenario coefficient range for each new energy access gradient level and the new energy scenario coefficient at the time of sampling each historical operating data;

[0014] Obtaining correlation coefficients between each grid operation safety evaluation index and each preset characteristic variable of the power grid, and obtaining characteristic variables for predicting each grid operation safety evaluation index based on the correlation coefficients between each grid operation safety evaluation index and each preset characteristic variable;

[0015] From the historical operation data of each renewable energy access gradient level, the historical data of the characteristic variables used for prediction of each power grid operation safety evaluation index are selected to obtain the predicted sample data of each power grid operation safety evaluation index under each renewable energy access gradient level;

[0016] The predicted sample data of each power grid operation safety evaluation index under each new energy access gradient level are used to train the preset machine learning model to obtain the power grid operation risk prediction model of each power grid operation safety evaluation index under each new energy access gradient level.

[0017] Optionally, obtaining historical operation data of the power grid within a preset time period includes:

[0018] Obtain historical operating data of the power grid for several days before and after the day with the maximum renewable energy power generation ratio within a preset time period, and eliminate historical operating data where the renewable energy power generation ratio at the time of sampling is less than the preset renewable energy power generation ratio threshold.

[0019] Optionally, the new energy scenario coefficients obtained when sampling each historical operation data include:

[0020] The new energy scenario coefficient τ at the time of sampling each historical operating data is obtained by the following formula:

[0021] τ=F×Q×W

[0022] Among them, F is the proportion of renewable energy power generation in the power grid when the historical operation data is sampled, Q is the average light intensity of the power grid, and W is the average wind speed of the power grid.

[0023] Optionally, the range of the new energy scenario coefficient for each new energy access gradient level of the power grid obtained based on the new energy scenario coefficient at each sampling time of the historical operation data and the number of new energy access gradient levels preset by the power grid includes:

[0024] Obtain the maximum and minimum values ​​of the new energy scenario coefficients when sampling each historical operating data;

[0025] The interval between the maximum value and the minimum value is equally divided into the number of new energy access gradient levels preset by the power grid, as the new energy scenario coefficient range of each new energy access gradient level of the power grid.

[0026] Optionally, obtaining the correlation coefficients between each grid operation safety evaluation index and each preset characteristic variable of the grid includes:

[0027] The first-order correlation coefficient R1 of each grid operation safety evaluation index and each preset characteristic variable is obtained by the following formula: i…n and quadratic correlation coefficient R2 i…n :

[0028]

[0029] Among them, Z i…n is each preset characteristic variable, n is the number of preset characteristic variables, σ 2 Z i…n The variance of continuous sampling data, S is the power grid operation safety evaluation index value;

[0030] The characteristic variables for predicting each power grid operation safety evaluation index obtained based on the correlation coefficient between each power grid operation safety evaluation index and each preset characteristic variable include:

[0031] Arrange the primary correlation coefficient and the secondary correlation coefficient of each power grid operation safety evaluation index and each preset characteristic variable in descending order to obtain a first sorting sequence and a second sorting sequence of each power grid operation safety evaluation index;

[0032] Preset characteristic variables are selected alternately from the first sorting sequence and the second sorting sequence of the power grid operation safety evaluation index according to the sorting order, and repeated preset characteristic variables are eliminated until the number of selected preset characteristic variables reaches the preset number of characteristic variables, so as to obtain the characteristic variables for prediction of each power grid operation safety evaluation index.

[0033] Optionally, obtaining the current new energy access gradient level of the power grid includes: obtaining a current new energy scenario coefficient of the power grid, and obtaining the current new energy access gradient level of the power grid according to the current new energy scenario coefficient and a preset new energy scenario coefficient range of each new energy access gradient level;

[0034] The determining of the current prediction characteristic variables of the power grid according to the power grid operation safety evaluation index to be predicted includes: determining the current prediction characteristic variables of the power grid according to the power grid operation safety evaluation index to be predicted by using the preset prediction characteristic variables of each power grid operation safety evaluation index.

[0035] A second aspect of the present invention provides a power grid operation risk prediction system, comprising:

[0036] The data acquisition module is used to obtain the current renewable energy access gradient level of the power grid and the power grid operation safety evaluation indicators to be predicted;

[0037] a model determination module for selecting, from among several preset grid operation risk prediction models, a grid operation risk prediction model having the same grid operation gradient level as the current grid operation gradient level and the same grid operation safety evaluation index as the grid operation safety evaluation index to be predicted, based on the current grid operation gradient level and the grid operation safety evaluation index to be predicted, to obtain a target grid operation risk prediction model;

[0038] The prediction module is used to determine the current prediction characteristic variables of the power grid based on the power grid operation safety evaluation index to be predicted, and obtain the data of the current prediction characteristic variables to input into the target power grid operation risk prediction model to obtain the predicted value of the power grid operation safety evaluation index to be predicted, and determine the power grid operation risk based on the predicted value.

[0039] Optionally, the model determination module is further configured to:

[0040] Obtaining historical operation data of the power grid within a preset time period; wherein the historical operation data includes historical data of each preset characteristic variable of the power grid;

[0041] Obtain the new energy scenario coefficient at each historical operation data sampling time, and obtain the new energy scenario coefficient range for each new energy access gradient level of the power grid based on the new energy scenario coefficient at each historical operation data sampling time and the number of new energy access gradient levels preset by the power grid;

[0042] Determine the historical operating data for each new energy access gradient level based on the new energy scenario coefficient range for each new energy access gradient level and the new energy scenario coefficient at the time of sampling each historical operating data;

[0043] Obtaining correlation coefficients between each grid operation safety evaluation index and each preset characteristic variable of the power grid, and obtaining characteristic variables for predicting each grid operation safety evaluation index based on the correlation coefficients between each grid operation safety evaluation index and each preset characteristic variable;

[0044] From the historical operation data of each renewable energy access gradient level, the historical data of the characteristic variables used for prediction of each power grid operation safety evaluation index are selected to obtain the predicted sample data of each power grid operation safety evaluation index under each renewable energy access gradient level;

[0045] The predicted sample data of each power grid operation safety evaluation index under each new energy access gradient level are used to train the preset machine learning model to obtain the power grid operation risk prediction model of each power grid operation safety evaluation index under each new energy access gradient level.

[0046] Optionally, the model determination module is specifically used to:

[0047] The first-order correlation coefficient R1 of each grid operation safety evaluation index and each preset characteristic variable is obtained by the following formula: i…n and quadratic correlation coefficient R2 i…n :

[0048]

[0049] Among them, Z i…n is each preset characteristic variable, n is the number of preset characteristic variables, σ 2 Z i…n The variance of continuous sampling data, S is the power grid operation safety evaluation index value;

[0050] The characteristic variables for predicting each power grid operation safety evaluation index obtained based on the correlation coefficient between each power grid operation safety evaluation index and each preset characteristic variable include:

[0051] Arrange the primary correlation coefficient and the secondary correlation coefficient of each power grid operation safety evaluation index and each preset characteristic variable in descending order to obtain a first sorting sequence and a second sorting sequence of each power grid operation safety evaluation index;

[0052] Preset characteristic variables are selected alternately from the first sorting sequence and the second sorting sequence of the power grid operation safety evaluation index according to the sorting order, and repeated preset characteristic variables are eliminated until the number of selected preset characteristic variables reaches the preset number of characteristic variables, so as to obtain the characteristic variables for prediction of each power grid operation safety evaluation index.

[0053] In a third aspect of the present invention, a computer device comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned method for predicting power grid operation risks when executing the computer program.

[0054] A fourth aspect of the present invention is a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the steps of the above-mentioned method for predicting power grid operation risks.

[0055] Compared with the prior art, the present invention has the following beneficial effects:

[0056] The grid operation risk prediction method of the present invention selects a grid operation risk prediction model whose new energy access gradient level is the same as the current new energy access gradient level and whose grid operation safety evaluation index is the same as the grid operation safety evaluation index to be predicted as a target grid operation risk prediction model according to the current new energy access gradient level of the grid and the grid operation safety evaluation index to be predicted. The grid operation state is refined by the new energy access gradient to ensure that the target grid operation risk prediction model is adapted to the current grid operation state. The current prediction feature variables of the grid are determined according to the grid operation safety evaluation index to be predicted to ensure the correspondence between the prediction feature variables and the grid operation safety evaluation index to be predicted, thereby effectively improving the pertinence and prediction accuracy of the target grid operation risk prediction model for the grid operation safety evaluation index to be predicted, and then ensuring the accuracy of determining the grid operation risk according to the predicted value. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 This is the process of the power grid operation risk prediction method of the present invention;

[0058] Figure 2 This is a structural block diagram of the power grid operation risk prediction system of the present invention. DETAILED DESCRIPTION

[0059] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0060] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0061] First, the relevant terms involved in the embodiments of the present invention are introduced:

[0062] Feature selection: As a data preprocessing process, it plays an important role in data mining, pattern recognition, and machine learning. Feature selection can reduce the complexity of the problem and improve the prediction accuracy, robustness, and interpretability of the learning algorithm.

[0063] Feature engineering: The process of using professional background knowledge and techniques to process data so that features can play a better role in machine learning algorithms.

[0064] Feature extraction: Feature extraction generally maps data from a high-dimensional feature space to a low-dimensional feature space through mathematical methods (such as projection).

[0065] Correlation analysis: Correlation analysis is a simple and practical analytical technique used to discover correlations or relationships in large data sets.

[0066] The present invention is described in further detail below with reference to the accompanying drawings:

[0067] See also Figure 1 In one embodiment of the present invention, a method for predicting the operation risk of a power grid is provided, specifically an operation risk prediction method for a typical scenario of a high-proportion new energy power grid based on artificial intelligence, which conducts detailed research on big data power grid operation data analysis, feature extraction and correlation analysis.

[0068] Specifically, the power grid operation risk prediction method includes the following steps:

[0069] S1: Obtain the current renewable energy access gradient level of the power grid and the power grid operation safety evaluation index to be predicted.

[0070] Specifically, obtaining the current new energy access gradient level of the power grid includes: obtaining the current new energy scenario coefficient of the power grid, and obtaining the current new energy access gradient level of the power grid according to the current new energy scenario coefficient and the preset new energy scenario coefficient range of each new energy access gradient level.

[0071] When obtaining the current renewable energy scenario coefficient of the power grid, the current renewable energy scenario coefficient τ′ of the power grid is obtained by the following formula:

[0072] τ′=F′×Q×W

[0073] Among them, F′ is the current proportion of renewable energy power generation in the power grid, Q is the average sunlight intensity of the power grid, and W is the average wind speed of the power grid.

[0074] After determining the current new energy scenario coefficient of the power grid, based on the preset new energy scenario coefficient range of each new energy access gradient level, the current new energy scenario coefficient of the power grid can be judged to which new energy access gradient level the current new energy scenario coefficient of the power grid belongs, so as to determine the current new energy access gradient level of the power grid.

[0075] Predicted grid operation safety evaluation indicators are pre-selected before grid operation risk forecasting to assess grid operation safety. Generally, these indicators include structural safety indicators, adequacy indicators, operational safety indicators, and reliability indicators. Structural safety indicators include the N-1 verification pass rate, transformer load factor, transformer capacity ratio, and unit line power supply. Adequacy indicators include the expected power shortage value. Operational safety indicators include the number of substation accidents and transmission accidents. Reliability indicators include frequency stability, voltage stability, and spinning reserve ratio.

[0076] S2: According to the current new energy access gradient level and the grid operation safety evaluation index to be predicted, from several preset grid operation risk prediction models, select a grid operation risk prediction model whose new energy access gradient level is the same as the current new energy access gradient level and whose grid operation safety evaluation index is the same as the grid operation safety evaluation index to be predicted, and obtain the target grid operation risk prediction model.

[0077] Specifically, the preset several power grid operation risk prediction models are constructed under the condition of new energy access gradient level, using different power grid operation safety evaluation indicators as the power grid operation safety evaluation indicators to be predicted. Therefore, for the current new energy access gradient level and the power grid operation safety evaluation indicators to be predicted, the corresponding power grid operation risk prediction model, that is, the target power grid operation risk prediction model, can be determined according to the new energy access gradient level and the specific type of power grid operation safety evaluation indicators.

[0078] In a possible implementation, the preset plurality of power grid operation risk prediction models are constructed by the following construction method, which includes:

[0079] S21: Acquire historical operation data of the power grid within a preset time period; wherein the historical operation data includes historical data of various preset characteristic variables of the power grid.

[0080] Specifically, when acquiring historical operating data for the power grid within a preset time period, 500kV substations are generally selected as sampling nodes based on the power grid's current trends. Preset characteristic variables generally include current data for equipment at different voltage levels, general power generation data, and meteorological data. Equipment at different voltage levels primarily includes 500kV busbars, 220kV busbars, 500kV transformers, wind farm units, thermal power plant units, photovoltaic power station units, and 220kV transformers. General power generation data primarily includes total power generation, total direct-controlled power generation, total real-time active power from local power plants, and total wind power connected to the grid. Meteorological data primarily includes wind power synchronicity and average irradiation intensity.

[0081] In one possible embodiment, obtaining the historical operating data of the power grid within a preset time period includes: obtaining the historical operating data of the power grid for several days before and after the day with the maximum proportion of renewable energy power generation within the preset time period, and eliminating the historical operating data in which the proportion of renewable energy power generation at the time of sampling is less than a preset threshold value of the proportion of renewable energy power generation.

[0082] Specifically, the historical operating data is extracted for the day with the highest renewable energy generation share, totaling five days of historical operating data before and after. Data from 7,200 sampling points are collected for each preset characteristic variable. Furthermore, to further ensure the characteristics of high-proportion renewable energy access scenarios and reduce the complexity of data processing, a renewable energy generation share threshold is set. For example, the renewable energy generation share threshold is set at 15%. Historical operating data with a renewable energy generation share of less than 15% at the time of sampling are eliminated to obtain the final historical operating data. Optionally, the final historical operating data is constructed as a time-based data series to prepare for subsequent analysis and calculations. The calculation accuracy is standardized, and two digits of significant data are retained.

[0083] S22: Obtain the new energy scenario coefficient at each historical operation data sampling time, and obtain the new energy scenario coefficient range of each new energy access gradient level of the power grid based on the new energy scenario coefficient at each historical operation data sampling time and the number of new energy access gradient levels preset by the power grid.

[0084] In a possible implementation, obtaining the new energy scenario coefficient at each sampling time of the historical operation data includes obtaining the new energy scenario coefficient τ at each sampling time of the historical operation data by the following formula:

[0085] τ=F×Q×W

[0086] Among them, F is the proportion of renewable energy power generation in the power grid when the historical operation data is sampled, Q is the average light intensity of the power grid, and W is the average wind speed of the power grid.

[0087] The method of obtaining the new energy scenario coefficient range for each new energy access gradient level of the power grid based on the new energy scenario coefficient when each historical operation data is sampled and the number of new energy access gradient levels preset by the power grid includes: obtaining the maximum value and the minimum value of the new energy scenario coefficient when each historical operation data is sampled; dividing the interval between the maximum value and the minimum value into equal parts according to the number of new energy access gradient levels preset by the power grid, as the new energy scenario coefficient range for each new energy access gradient level of the power grid.

[0088] Referring to Table 1, in this embodiment, the number of new energy access gradient levels preset by the power grid is set to 5, and the interval between the maximum value and the minimum value is divided into 5 equal parts to obtain the new energy scenario coefficient range corresponding to each of the 5 new energy access gradient levels, and each new energy access gradient level is identified as a scenario.

[0089] Table 1 New energy access gradient classification table

[0090]

[0091]

[0092] The new energy scenario coefficient ranges for each new energy access gradient level of the power grid obtained in this way can be used as preset data for use when obtaining the current new energy access gradient level of the power grid in S1.

[0093] S23: Determine the historical operation data of each new energy access gradient level according to the new energy scenario coefficient range of each new energy access gradient level and the new energy scenario coefficient when each historical operation data is sampled.

[0094] Specifically, the historical operating data are arranged in descending order according to the new energy scenario coefficient at the time of sampling, and then the new energy scenario coefficient range of each new energy access gradient level is corresponded respectively, and the historical operating data are allocated to each new energy access gradient level.

[0095] S24: Obtaining correlation coefficients between each grid operation safety evaluation index and each preset characteristic variable, and obtaining characteristic variables for prediction of each grid operation safety evaluation index based on the correlation coefficients between each grid operation safety evaluation index and each preset characteristic variable.

[0096] In a possible implementation manner, obtaining the correlation coefficient between each grid operation safety evaluation index and each preset characteristic variable of the grid includes:

[0097] The first-order correlation coefficient R1 of each grid operation safety evaluation index and each preset characteristic variable is obtained by the following formula: i…n and quadratic correlation coefficient R2 i…n :

[0098]

[0099] Among them, Z i…n is each preset characteristic variable, n is the number of preset characteristic variables, σ 2 Z i…n is the variance of the continuous sampling data, and S is the grid operation safety evaluation index value.

[0100] Among them, σ 2 As a measurement quality factor, the result of the above calculation can be double-verified by introducing the measurement quality factor, that is, two rounds of successive calculations are performed. The first round is based on σ 2 Arrange the data in positive order, and then calculate R1 for each preset characteristic variable and the power grid operation safety evaluation index i…n , n first-order correlation coefficients, the second one is based on σ 2 Arrange the data in reverse order, and then calculate the R2 of each preset characteristic variable and the power grid operation safety evaluation index respectively. i…n , n quadratic correlation coefficients, integrating the two settlement results to achieve double verification, effectively reducing the impact of data errors caused by the sampling and measurement characteristics in scenarios with a high proportion of new energy access, achieving the purpose of improving accuracy, and thus greatly improving the accuracy of power grid operation risk prediction.

[0101] The method of obtaining the predictive characteristic variables of each power grid operation safety evaluation index based on the correlation coefficient between each power grid operation safety evaluation index and each preset characteristic variable includes: arranging the primary correlation coefficient and the secondary correlation coefficient between each power grid operation safety evaluation index and each preset characteristic variable in descending order to obtain a first sorting sequence and a second sorting sequence of each power grid operation safety evaluation index; alternately selecting the preset characteristic variables from the first sorting sequence and the second sorting sequence of the power grid operation safety evaluation index according to the sorting order, and eliminating duplicate preset characteristic variables until the number of selected preset characteristic variables reaches the preset number of characteristic variables, thereby obtaining the predictive characteristic variables of each power grid operation safety evaluation index.

[0102] Specifically, the primary and secondary correlation coefficients between each grid operation safety evaluation indicator and each preset characteristic variable are arranged in descending order. These preset characteristic variables are then combined and analyzed, i.e., the higher-ranking preset characteristic variables are alternately selected until a certain number of preset characteristic variables are selected. Alternatively, a number of preset characteristic variables with relatively high primary and secondary correlation coefficients can be retained, for example, 15, and then duplicate preset characteristic variables are deleted after the combination, until the number of retained preset characteristic variables is within 20.

[0103] S25: From the historical operation data of each new energy access gradient level, select the historical data of the characteristic variables for prediction of each power grid operation safety evaluation index, and obtain the predicted sample data of each power grid operation safety evaluation index under each new energy access gradient level.

[0104] Specifically, according to the characteristic variables for prediction of each power grid operation safety evaluation index, historical data of the characteristic variables for prediction of each power grid operation safety evaluation index are selected from the historical operation data of each new energy access gradient level and combined to obtain the historical data of the characteristic variables for prediction of each power grid operation safety evaluation index under each new energy access gradient level, which are used as prediction sample data and as the basis for predicting the operation safety evaluation index of each power grid under each new energy access gradient level.

[0105] S26: Using the predicted sample data of each power grid operation safety evaluation index under each new energy access gradient level, a preset machine learning model is trained to obtain a power grid operation risk prediction model of each power grid operation safety evaluation index under each new energy access gradient level.

[0106] Specifically, the method first obtains the grid operation safety evaluation index values ​​corresponding to the predicted sample data for each grid operation safety evaluation index at each renewable energy access gradient level. The predicted sample data and grid operation safety evaluation index values ​​are then preprocessed, typically by normalization and standardization. Next, a pre-set machine learning model, typically a logistic regression model, is trained using the predicted sample data as input and the predicted values ​​of the grid operation safety evaluation index as output to obtain a grid operation risk prediction model for each grid operation safety evaluation index at each renewable energy access gradient level.

[0107] When training a preset machine learning model, 80% of the test sample data is randomly selected to be defined as the training feature dataset, and the corresponding power grid operation safety evaluation index values ​​are defined as the training target dataset. The remaining 20% ​​of the test sample data is defined as the testing feature dataset, and its corresponding power grid operation safety evaluation index values ​​are defined as the testing target dataset. The machine learning model is optimized by directly comparing the error between the predicted values ​​of the power grid operation safety evaluation index of the training feature dataset and the power grid operation safety evaluation index values ​​of the training target dataset. The prediction accuracy of the trained machine learning model is calculated by directly comparing the predicted values ​​of the power grid operation safety evaluation index of the testing feature dataset with the power grid operation safety evaluation index values ​​of the testing target dataset to verify the accuracy of the results. After meeting the preset requirements, it is used as a power grid operation risk prediction model.

[0108] S3: Based on the power grid operation safety evaluation index to be predicted, determine the current prediction characteristic variables of the power grid, obtain the data of the current prediction characteristic variables and input them into the target power grid operation risk prediction model to obtain the predicted value of the power grid operation safety evaluation index to be predicted, and determine the power grid operation risk based on the predicted value.

[0109] Specifically, determining the current prediction characteristic variables of the power grid based on the power grid operation safety evaluation index to be predicted includes: determining the current prediction characteristic variables of the power grid based on the power grid operation safety evaluation index to be predicted using preset prediction characteristic variables for each power grid operation safety evaluation index. The prediction characteristic variables for each power grid operation safety evaluation index may be the prediction characteristic variables for each power grid operation safety evaluation index obtained in the method for constructing the power grid operation risk prediction model.

[0110] By inputting the data of the current prediction characteristic variables into the target power grid operation risk prediction model, the predicted value of the power grid operation safety evaluation index to be predicted under the current new energy access gradient level is obtained, and then the power grid operation risk is judged according to the size of the predicted value.

[0111] For example, based on the actual grid operation data of a certain power grid, the "frequency deviation rate Δf" is selected as the safety evaluation index of the grid operation to be predicted:

[0112]

[0113] Where: f N is the rated power, usually 50 Hz, f is the grid frequency at the current moment, and 0.2 is the maximum allowable range of frequency deviation in the frequency standard.

[0114] By adopting the grid operation risk prediction method of the present invention, i.e., dividing the new energy access gradient level for prediction, and the existing grid operation risk prediction method, i.e., not dividing the new energy access gradient level for prediction, grid operation risk prediction is performed respectively. Referring to Table 2, it can be seen that the prediction accuracy of the grid operation risk prediction method of the present invention is significantly higher than that of the existing grid operation risk prediction method.

[0115] Table 2 Comparison of the accuracy of power grid operation risk prediction between the present invention and the existing method

[0116]

[0117] To sum up, the grid operation risk prediction method of the present invention selects a grid operation risk prediction model whose new energy access gradient level is the same as the current new energy access gradient level and whose grid operation safety evaluation index is the same as the grid operation safety evaluation index to be predicted as the target grid operation risk prediction model according to the current new energy access gradient level of the grid and the grid operation safety evaluation index to be predicted. The grid operation state is refined by the new energy access gradient to ensure that the target grid operation risk prediction model is adapted to the current operation state of the grid. The current prediction feature variables of the grid are determined according to the grid operation safety evaluation index to be predicted to ensure the correspondence between the prediction feature variables and the grid operation safety evaluation index to be predicted, thereby effectively improving the pertinence and prediction accuracy of the target grid operation risk prediction model for the grid operation safety evaluation index to be predicted, and then ensuring the accuracy of determining the grid operation risk according to the predicted value.

[0118] The following are device embodiments of the present invention, which can be used to implement the method embodiments of the present invention. For details not disclosed in the device embodiments, please refer to the method embodiments of the present invention.

[0119] See also Figure 2In another embodiment of the present invention, a power grid operation risk prediction system is provided, which can be used to implement the above-mentioned power grid operation risk prediction method. Specifically, the power grid operation risk prediction system includes a data acquisition module, a model determination module, and a prediction module. Among them, the data acquisition module is used to obtain the current new energy access gradient level of the power grid and the power grid operation safety evaluation index to be predicted; the model determination module is used to select a power grid operation risk prediction model with the same new energy access gradient level as the current new energy access gradient level and the power grid operation safety evaluation index to be predicted from a number of preset power grid operation risk prediction models based on the current new energy access gradient level and the power grid operation safety evaluation index to be predicted, and obtain a target power grid operation risk prediction model; the prediction module is used to determine the current prediction feature variables of the power grid based on the power grid operation safety evaluation index to be predicted, and obtain the data of the current prediction feature variables and input them into the target power grid operation risk prediction model to obtain the predicted value of the power grid operation safety evaluation index to be predicted, and determine the power grid operation risk based on the predicted value.

[0120] In a possible implementation, the model determination module is further configured to:

[0121] Obtain historical operation data of the power grid within a preset time period; wherein the historical operation data includes historical data of each preset characteristic variable of the power grid; obtain the new energy scenario coefficient when each historical operation data is sampled, and obtain the new energy scenario coefficient range of each new energy access gradient level of the power grid based on the new energy scenario coefficient when each historical operation data is sampled and the number of new energy access gradient levels preset by the power grid; determine the historical operation data of each new energy access gradient level based on the new energy scenario coefficient range of each new energy access gradient level and the new energy scenario coefficient when each historical operation data is sampled; obtain the correlation coefficient between each power grid operation safety evaluation index and each preset characteristic variable, and obtain the characteristic variable for predicting each power grid operation safety evaluation index based on the correlation coefficient between each power grid operation safety evaluation index and each preset characteristic variable; select the historical data of the characteristic variable for predicting each power grid operation safety evaluation index from the historical operation data of each new energy access gradient level, and obtain predicted sample data of each power grid operation safety evaluation index under each new energy access gradient level; use the predicted sample data of each power grid operation safety evaluation index under each new energy access gradient level to train a preset machine learning model, and obtain a power grid operation risk prediction model for each power grid operation safety evaluation index under each new energy access gradient level.

[0122] In a possible implementation, the model determination module is specifically configured to:

[0123] The first-order correlation coefficient R1 of each grid operation safety evaluation index and each preset characteristic variable is obtained by the following formula: i…nand quadratic correlation coefficient R2 i…n :

[0124]

[0125] Among them, Z i…n is each preset characteristic variable, n is the number of preset characteristic variables, σ 2 Z i…n is the variance of the continuous sampling data, and S is the grid operation safety evaluation index value.

[0126] The model determination module obtains the characteristic variables for prediction of each power grid operation safety evaluation index based on the correlation coefficient between each power grid operation safety evaluation index and each preset characteristic variable, including: arranging the primary correlation coefficient and the secondary correlation coefficient between each power grid operation safety evaluation index and each preset characteristic variable in order from large to small to obtain a first sorting sequence and a second sorting sequence of each power grid operation safety evaluation index; alternately selecting the preset characteristic variables from the first sorting sequence and the second sorting sequence of the power grid operation safety evaluation index according to the sorting order, and eliminating repeated preset characteristic variables until the number of selected preset characteristic variables reaches the preset number of characteristic variables, thereby obtaining the characteristic variables for prediction of each power grid operation safety evaluation index.

[0127] All relevant contents of each step involved in the embodiment of the aforementioned power grid operation risk prediction method can be referred to the functional description of the functional modules corresponding to the power grid operation risk prediction system in the embodiment of the present invention, and will not be repeated here.

[0128] The module division in the embodiments of the present invention is illustrative and represents only one logical functional division. In actual implementation, other division methods may be used. Furthermore, the functional modules in various embodiments of the present invention may be integrated into a single processor, exist physically as separate modules, or two or more modules may be integrated into a single module. The integrated modules may be implemented in either hardware or software functional modules.

[0129] In another embodiment of the present invention, a computer device is provided, which includes a processor and a memory, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or may be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions in the computer storage medium to implement the corresponding method flow or corresponding function; the processor described in the embodiment of the present invention can be used for the operation of the power grid operation risk prediction method.

[0130] In another embodiment of the present invention, the present invention further provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a computer device for storing programs and data. It is understandable that the computer-readable storage medium here can include both built-in storage media in the computer device and, of course, extended storage media supported by the computer device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space. These instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the power grid operation risk prediction method in the above embodiment.

[0131] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0132] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0133] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0134] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0135] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A method for predicting power grid operation risk, characterized in that: include: Obtain the current renewable energy access gradient level of the power grid and the power grid operation safety evaluation indicators to be predicted; According to the current new energy access gradient level and the grid operation safety evaluation index to be predicted, a grid operation risk prediction model with the same new energy access gradient level as the current new energy access gradient level and the same grid operation safety evaluation index as the grid operation safety evaluation index to be predicted is selected from several preset grid operation risk prediction models to obtain a target grid operation risk prediction model; Determine the current prediction characteristic variables of the power grid based on the power grid operation safety evaluation index to be predicted, obtain data of the current prediction characteristic variables and input them into the target power grid operation risk prediction model to obtain a predicted value of the power grid operation safety evaluation index to be predicted, and determine the power grid operation risk based on the predicted value; The preset plurality of power grid operation risk prediction models are constructed by the following construction method, which includes: Acquiring historical operation data of the power grid within a preset time period; wherein the historical operation data includes historical data of each preset characteristic variable of the power grid; Obtain the new energy scenario coefficient at each historical operation data sampling time, and obtain the new energy scenario coefficient range for each new energy access gradient level of the power grid based on the new energy scenario coefficient at each historical operation data sampling time and the number of new energy access gradient levels preset by the power grid; Determine the historical operating data for each new energy access gradient level based on the new energy scenario coefficient range for each new energy access gradient level and the new energy scenario coefficient at the time of sampling each historical operating data; Obtaining correlation coefficients between each grid operation safety evaluation index and each preset characteristic variable of the power grid, and obtaining characteristic variables for predicting each grid operation safety evaluation index based on the correlation coefficients between each grid operation safety evaluation index and each preset characteristic variable; From the historical operation data of each renewable energy access gradient level, the historical data of the characteristic variables used for prediction of each power grid operation safety evaluation index are selected to obtain the predicted sample data of each power grid operation safety evaluation index under each renewable energy access gradient level; The predicted sample data of each power grid operation safety evaluation index under each new energy access gradient level are used to train the preset machine learning model to obtain the power grid operation risk prediction model of each power grid operation safety evaluation index under each new energy access gradient level.

2. The method for predicting power grid operation risk according to claim 1, characterized in that: The obtaining of historical operation data of the power grid within a preset time period includes: Obtain historical operating data of the power grid for several days before and after the day with the maximum renewable energy power generation ratio within a preset time period, and eliminate historical operating data where the renewable energy power generation ratio at the time of sampling is less than the preset renewable energy power generation ratio threshold.

3. The method for predicting power grid operation risk according to claim 1, characterized in that: The new energy scenario coefficients for obtaining each historical operation data sampling include: The new energy scenario coefficient τ at the time of sampling each historical operating data is obtained by the following formula: τ=F×Q×W Among them, F is the proportion of renewable energy power generation in the power grid when the historical operation data is sampled, Q is the average light intensity of the power grid, and W is the average wind speed of the power grid.

4. The method for predicting power grid operation risk according to claim 1, characterized in that: The range of the new energy scenario coefficient for each new energy access gradient level of the power grid obtained based on the new energy scenario coefficient at the time of sampling each historical operation data and the number of new energy access gradient levels preset by the power grid includes: Obtain the maximum and minimum values ​​of the new energy scenario coefficients when sampling each historical operating data; The interval between the maximum value and the minimum value is equally divided into the number of new energy access gradient levels preset by the power grid, as the new energy scenario coefficient range of each new energy access gradient level of the power grid.

5. The method for predicting power grid operation risk according to claim 1, characterized in that: The correlation coefficients between the grid operation safety evaluation indicators and the preset characteristic variables are obtained as follows: The first-order correlation coefficient R1 of each grid operation safety evaluation index and each preset characteristic variable is obtained by the following formula: i…n and quadratic correlation coefficient R2 i…n : Among them, Z i…n is each preset characteristic variable, n is the number of preset characteristic variables, σ 2 Z i…n The variance of continuous sampling data, S is the power grid operation safety evaluation index value; The characteristic variables for predicting each power grid operation safety evaluation index obtained based on the correlation coefficient between each power grid operation safety evaluation index and each preset characteristic variable include: Arrange the primary correlation coefficient and the secondary correlation coefficient of each power grid operation safety evaluation index and each preset characteristic variable in descending order to obtain a first sorting sequence and a second sorting sequence of each power grid operation safety evaluation index; Preset characteristic variables are selected alternately from the first sorting sequence and the second sorting sequence of the power grid operation safety evaluation index according to the sorting order, and repeated preset characteristic variables are eliminated until the number of selected preset characteristic variables reaches the preset number of characteristic variables, so as to obtain the characteristic variables for prediction of each power grid operation safety evaluation index.

6. The method for predicting power grid operation risk according to claim 1, characterized in that: The obtaining of the current renewable energy access gradient level of the power grid includes: obtaining the current renewable energy scenario coefficient of the power grid, and obtaining the current renewable energy access gradient level of the power grid according to the current renewable energy scenario coefficient and a preset renewable energy scenario coefficient range of each renewable energy access gradient level; The determining of the current prediction characteristic variables of the power grid according to the power grid operation safety evaluation index to be predicted includes: determining the current prediction characteristic variables of the power grid according to the power grid operation safety evaluation index to be predicted by using the preset prediction characteristic variables of each power grid operation safety evaluation index.

7. A power grid operation risk prediction system, characterized in that: include: The data acquisition module is used to obtain the current renewable energy access gradient level of the power grid and the power grid operation safety evaluation indicators to be predicted; a model determination module for selecting, from among several preset grid operation risk prediction models, a grid operation risk prediction model having the same grid operation gradient level as the current grid operation gradient level and the same grid operation safety evaluation index as the grid operation safety evaluation index to be predicted, based on the current grid operation gradient level and the grid operation safety evaluation index to be predicted, to obtain a target grid operation risk prediction model; A prediction module is used to determine the current prediction characteristic variables of the power grid based on the power grid operation safety evaluation index to be predicted, obtain data of the current prediction characteristic variables and input them into the target power grid operation risk prediction model to obtain the predicted value of the power grid operation safety evaluation index to be predicted, and determine the power grid operation risk based on the predicted value; The model determination module is further configured to: Obtaining historical operation data of the power grid within a preset time period; wherein the historical operation data includes historical data of each preset characteristic variable of the power grid; Obtain the new energy scenario coefficient at each historical operation data sampling time, and obtain the new energy scenario coefficient range for each new energy access gradient level of the power grid based on the new energy scenario coefficient at each historical operation data sampling time and the number of new energy access gradient levels preset by the power grid; Determine the historical operating data for each new energy access gradient level based on the new energy scenario coefficient range for each new energy access gradient level and the new energy scenario coefficient at the time of sampling each historical operating data; Obtaining correlation coefficients between each grid operation safety evaluation index and each preset characteristic variable of the power grid, and obtaining characteristic variables for predicting each grid operation safety evaluation index based on the correlation coefficients between each grid operation safety evaluation index and each preset characteristic variable; From the historical operation data of each renewable energy access gradient level, the historical data of the characteristic variables used for prediction of each power grid operation safety evaluation index are selected to obtain the predicted sample data of each power grid operation safety evaluation index under each renewable energy access gradient level; The predicted sample data of each power grid operation safety evaluation index under each new energy access gradient level are used to train the preset machine learning model to obtain the power grid operation risk prediction model of each power grid operation safety evaluation index under each new energy access gradient level.

8. The power grid operation risk prediction system according to claim 7, characterized in that: The model determination module is specifically used for: The first-order correlation coefficient R1 of each grid operation safety evaluation index and each preset characteristic variable is obtained by the following formula: i…n and quadratic correlation coefficient R2 i…n : Among them, Z i…n is each preset characteristic variable, n is the number of preset characteristic variables, σ 2 Z i…n The variance of continuous sampling data, S is the power grid operation safety evaluation index value; The characteristic variables for predicting each power grid operation safety evaluation index obtained based on the correlation coefficient between each power grid operation safety evaluation index and each preset characteristic variable include: Arrange the primary correlation coefficient and the secondary correlation coefficient of each power grid operation safety evaluation index and each preset characteristic variable in descending order to obtain a first sorting sequence and a second sorting sequence of each power grid operation safety evaluation index; Preset characteristic variables are selected alternately from the first sorting sequence and the second sorting sequence of the power grid operation safety evaluation index according to the sorting order, and repeated preset characteristic variables are eliminated until the number of selected preset characteristic variables reaches the preset number of characteristic variables, so as to obtain the characteristic variables for prediction of each power grid operation safety evaluation index.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the power grid operation risk prediction method according to any one of claims 1 to 6 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the power grid operation risk prediction method according to any one of claims 1 to 6 are implemented.

Images