An evaluation method for situational awareness in power systems

Abstract

The present invention discloses an evaluation method for situation awareness under an electric power system, which belongs to the technical field of electric power system optimization operation, and includes: S1: obtaining the initial weight of an element based on a hierarchical analysis method, and dividing the first element into a first intervention element and a first mixed element, and obtaining the final weight of the first intervention element based on a Dowhy algorithm; S2: obtaining the weight of a process indicator; S3: if the first mixed element does not meet the requirements, executing S4, if it meets the requirements, dividing the first mixed element based on the initial weight, and obtaining the final weight of the second intervention element based on the Dowhy algorithm, taking the second mixed element as the first mixed element, and executing S3; S4: obtaining the final weight of the second element based on the Dowhy algorithm, and obtaining the final weighted causal graph based on the final weight of the second element; S5: evaluating situation awareness based on stability analysis results and data change trends. The present invention solves the problem that it is difficult to accurately obtain the weights of situation awareness evaluation elements, resulting in low accuracy of situation awareness evaluation.

Description

Technical Field

[0001] The present invention relates to the technical field of power system optimization operation, and in particular to an evaluation method for situation awareness in a power system. Background Art

[0002] With the continuous expansion and increasing complexity of regional distribution networks, in order to effectively respond to the challenges faced by power systems under various operating conditions and ensure that the power grid can maintain safe, stable and efficient operation under any circumstances, it is particularly important to accurately and comprehensively evaluate the power grid situation. The existing technology mainly uses the hierarchical analysis method to decompose the elements related to power grid situation awareness into different hierarchical structures such as goals, criteria, and indicators, and then obtains the initial evaluation value by combining the weight of the indicator layer data with the indicator layer data. The initial evaluation value is combined and analyzed with the weight of the criteria in the criterion layer to obtain the final situation awareness evaluation result. However, this method only relies on the subjective evaluation of experts to determine the weight relationship between elements in each level, which is highly subjective. Therefore, there are subjective errors in the weight of the data and the weight of the criterion. It can be seen that the initial evaluation value also has subjective errors. Combining and analyzing the initial evaluation value with subjective errors and the criterion weight with subjective errors will result in a large difference between the final situation awareness evaluation result and the actual situation. Therefore, the accuracy of situation awareness evaluation through the hierarchical analysis method is low. Summary of the invention

[0003] In view of the problem that it is difficult to accurately obtain the weights of situation awareness assessment elements in the prior art, resulting in low accuracy of situation awareness assessment, the present invention provides an assessment method for situation awareness in an electric power system, constructs an initial weighted causal graph through a hierarchical analysis method, identifies intervention elements and mixed elements in the initial weighted causal graph according to the initial weights of the elements in the initial weighted causal graph, and uses a Dowhy algorithm to accurately obtain the final weights of the intervention elements and mixed elements based on the identified intervention elements and mixed elements, significantly reducing the complexity of identifying the intervention elements and mixed elements through the Dowhy algorithm, solving the problem that it is difficult to accurately obtain the weights of situation awareness assessment elements, resulting in low accuracy of situation awareness assessment, and significantly improving the accuracy of situation awareness assessment while also significantly improving the efficiency of situation awareness assessment.

[0004] In order to solve the above technical problems, the present invention provides an evaluation method for situation awareness in a power system, comprising the following steps:

[0005] S1: construct an initial weighted causal diagram based on the analytic hierarchy process and situation awareness factors, divide the first element of the elements into a first intervention element and a first mixed element based on the initial weights of the elements in the initial weighted causal diagram, and obtain the final weight of the first intervention element based on the Dowhy algorithm;

[0006] S2: updating the initial weight based on the final weight of the first intervention element to obtain a process indicator weight;

[0007] S3: Determine whether the first mixed element meets the preset requirements. If not, execute S4. If it meets the requirements, divide the first mixed element into a second intervention element and a second mixed element based on the initial weight, obtain the final weight of the second intervention element based on the process indicator weight and the Dowhy algorithm, update the process indicator weight based on the final weight of the second intervention element, take the second mixed element as the first mixed element, and execute S3.

[0008] S4: obtaining a final weight of a second element in the elements based on the Dowhy algorithm and the process indicator weight, and obtaining a final weighted causal graph based on the process indicator weight and the final weight of the second element;

[0009] S5: Perform situation awareness assessment based on the stability analysis results of the final weighted causal diagram and the changing trend of the real-time data obtained from the power system relative to the historical data of the power system.

[0010] After adopting the above technical solution, the present invention has the following advantages:

[0011] Considering that when evaluating the weight of elements through expert evaluation in the hierarchical analysis method, there are subjective errors that make it difficult to obtain the accurate weight of each element in the causal diagram, and the Dowhy algorithm has high accuracy in obtaining the weight of elements in the causal relationship, the weight of the elements in the causal relationship can be accurately obtained through the Dowhy algorithm. Considering that the complexity of obtaining the intervention elements and the mixed elements in the process of accurately obtaining the weight of the elements in the causal relationship by the Dowhy algorithm is high, and the weight obtained by the hierarchical analysis method can accurately identify the intervention elements and the mixed elements, the situation awareness is evaluated by combining the hierarchical analysis method with the Dowhy algorithm, which significantly improves the accuracy of the situation awareness evaluation and the efficiency of the situation awareness evaluation.

[0012] The situation awareness assessment is performed through the stability analysis results of the final weighted causal graph, which ensures the robustness of the final weighted causal graph and avoids the assessment error caused by the situation assessment through the unrobust final weighted causal graph. This further improves the accuracy of the situation awareness assessment and the utilization rate of computing resources.

[0013] Situational awareness assessment is performed based on the change trend of the data obtained in real time from the power system relative to the historical data of the power system, so that the final weighted causal diagram changes dynamically according to the change trend, which not only enhances the adaptability of situational awareness assessment but also further improves the accuracy of situational awareness assessment;

[0014] Situation awareness assessment is performed by using the analytic hierarchy process, Dowhy algorithm, stability analysis results of the final weighted cause-and-effect diagram, and the change trend of the real-time data obtained by the power system relative to the historical data of the power system, which significantly improves the accuracy of situation awareness assessment and also significantly improves the efficiency and adaptability of situation awareness assessment;

[0015] The problem of difficulty in accurately obtaining the weights of situation awareness assessment elements, resulting in low accuracy of situation awareness assessment, is solved.

[0016] Preferably, in S1, constructing an initial weighted causal diagram based on the analytic hierarchy process and situation awareness factors includes:

[0017] S11: constructing an initial cause-and-effect diagram based on the situation awareness factors and the analytic hierarchy process, comparing the elements pairwise based on the comparison rules in the analytic hierarchy process to obtain a first comparison result, and constructing a fuzzy complementary judgment matrix based on the first comparison result;

[0018] S12: Obtain the weights of the elements based on the eigenvectors of the fuzzy complementary judgment matrix, and then obtain an initial weighted causal graph.

[0019] In this scheme, the initial causal diagram is constructed through situation awareness factors and hierarchical analysis method, and then the complex interaction relationships between all situation awareness factors are divided into a clear hierarchical structure through hierarchical analysis method, which improves the systematic nature of situation awareness assessment. The weight of each situation awareness factor is obtained through hierarchical analysis method, and the complex interaction relationships between factors are quantified. Then, situation awareness assessment is carried out according to the quantified results, which improves the scientificity and rationality of situation awareness assessment.

[0020] Preferably, in S1, the construction of the initial weighted causal diagram based on the hierarchical analysis method and situation awareness factors also includes: performing a consistency check on the fuzzy complementary judgment matrix, if the consistency check result meets the first preset condition, executing S12, if the consistency result does not meet the first preset condition, comparing the logical relationship between the data in the fuzzy complementary judgment matrix to obtain a second comparison result, correcting the fuzzy complementary judgment matrix based on the second comparison result, and executing S12.

[0021] Preferably, in S1, obtaining the final weight of the first intervention element based on the Dowhy algorithm includes:

[0022] S1a: constructing a first initial linear regression model based on the initial weight of the first intervention element and the initial weight of the first mixed element; S1b: inputting the value of the first intervention element and the value of the first mixed element in the historical data into the first initial linear regression model to obtain a first output result, updating the first initial linear regression model based on the first output result and the actual results corresponding to the first intervention element and the first mixed element to obtain a first optimized linear regression model, and obtaining the final weight of the first intervention element based on the first optimized linear regression model.

[0023] In this scheme, the first initial linear regression model is constructed through the initial weights to obtain the preliminary impact of the first intervention element and the first confounding element on the situation awareness assessment, which provides an analysis benchmark for the Dowhy algorithm analysis. Then, through the repeatability and causal inference ability of the Dowhy algorithm, the first output result and the actual result are analyzed to correct the deviation in the first initial linear regression model, so as to obtain a more accurate and reliable final weight of the first intervention element, improve the accuracy and reliability of the final weight of the first intervention element, and then improve the accuracy of the process indicator weight and the accuracy of the final situation awareness assessment.

[0024] Preferably, in S3, the final weight of the second intervention element obtained based on the process indicator weight and the Dowhy algorithm includes:

[0025] The first optimized linear regression model is updated based on the process weight of the second intervention element and the process weight of the second confounding element in the process indicator weight;

[0026] The value of the second intervening element and the value of the second mixed element in the historical data are input into the first optimized linear regression model to obtain a second output result, and the first optimized linear regression model is updated based on the second output result and the actual results corresponding to the second intervening element and the second mixed element to obtain an optimal linear regression model, and the final weight of the second intervening element is obtained based on the optimal linear regression model.

[0027] In this scheme, the first optimized linear regression model is updated through the process weight to obtain the preliminary impact of the second intervention element and the second confounding element on the situation awareness assessment, which provides an analysis benchmark for the Dowhy algorithm analysis. Then, through the repeatability and causal inference ability of the Dowhy algorithm, the second output result and the actual result are analyzed to correct the deviation in the first optimized linear regression model, so as to obtain a more accurate and reliable final weight of the second intervention element, improve the accuracy and reliability of the final weight of the second intervention element, and update the process indicator weight through continuous iteration, so as to improve the accuracy of the process indicator weight and the accuracy of the final situation awareness assessment.

[0028] Preferably, in S4, obtaining the final weight of the second element in the elements based on the Dowhy algorithm and the process indicator weight includes:

[0029] The second element is divided into a dominant element and a secondary element based on the initial weight of the second element, and the final weights of the dominant element and the secondary element are obtained based on the Dowhy algorithm and the process indicator weight.

[0030] In this scheme, considering that the Dowhy algorithm has a high complexity in obtaining the dominant elements and the secondary elements in the process of accurately obtaining the weights of elements in the causal relationship, and the weights obtained by the hierarchical analysis method can accurately identify the intervention elements and the mixed elements, the second element is divided by the weights obtained by the hierarchical analysis method, which significantly improves the efficiency of situation awareness assessment. Considering that when the element weights are evaluated by experts in the hierarchical analysis method, there are subjective errors and it is difficult to obtain the accurate weights of the elements in the causal diagram, and the Dowhy algorithm has high accuracy in obtaining the weights of the elements in the causal relationship, the weight of the second element in the causal relationship can be accurately obtained by the Dowhy algorithm. The accurate weight of the first element in the process indicator weight is combined with the Dowhy algorithm to further significantly improve the accuracy of situation awareness assessment.

[0031] Preferably, the final weights of the dominant elements and the secondary elements obtained based on the Dowhy algorithm and the process indicator weights include:

[0032] A second initial linear regression model is constructed based on the initial weights of the dominant element and the initial weights of the secondary element, and the value of the dominant element and the value of the secondary element are obtained based on the weights of the process indicators;

[0033] The value of the dominant element and the value of the secondary element are input into the second initial linear regression model to obtain a third output result. The second initial linear regression model is updated based on the third output result and the actual results corresponding to the dominant element and the secondary element to obtain a second optimized linear regression model. The final weights of the dominant element and the secondary element are obtained based on the second optimized linear regression model.

[0034] In this scheme, the values ​​of the dominant element and the secondary element are obtained by accurately obtaining the weight of the first element in the process indicator weight, thereby improving the accuracy of the values ​​of the dominant element and the secondary element. The accurate values ​​of the dominant element and the secondary element are input into the second initial linear regression model to obtain the third output result, thereby improving the accuracy of the third output result. Then, the final weights of the dominant element and the secondary element are obtained by comparing the third output result with the actual result, thereby improving the accuracy of the final weights of the dominant element and the secondary element, thereby improving the accuracy of the situation awareness assessment.

[0035] Preferably, S5 includes:

[0036] S51: According to the historical data, a predicted situation awareness assessment result is obtained using the final weighted causal graph, and the predicted situation awareness assessment result is compared with the actual situation awareness assessment result corresponding to the historical data to obtain a stability analysis result. If the stability analysis result does not meet the second preset condition, the historical data is cleaned, outliers are removed, and missing values ​​are removed, and S1b is executed. If the stability analysis result meets the second preset condition, S52 is executed.

[0037] S52: Perform situation awareness assessment based on the changing trend of the data acquired in real time from the power system relative to the historical data of the power system.

[0038] In this scheme, situation awareness assessment is performed through the stability analysis results of the final weighted causal graph, which ensures the robustness of the final weighted causal graph and avoids assessment errors caused by situation assessment through the insecure final weighted causal graph. This further improves the accuracy of situation awareness assessment while improving the utilization rate of computing resources. Situation awareness assessment is performed based on the changing trend of data acquired in real time from the power system relative to the historical data of the power system, so that the final weighted causal graph changes dynamically according to the changing trend, which enhances the adaptability of situation awareness assessment while further improving the accuracy of situation awareness assessment.

[0039] Preferably, the S52 includes:

[0040] S521: Obtaining a change trend of the data acquired in real time by the power system relative to the historical data;

[0041] S522: If the change trend does not satisfy the third preset condition, the data acquired in real time is used as the historical data and S1b is executed. If the change trend satisfies the third preset condition, a situation awareness assessment is performed based on the final weighted cause-effect diagram.

[0042] Beneficial effects of this program:

[0043] The initial cause-and-effect diagram is constructed by situation awareness factors and the hierarchical analysis method, and then the complex interaction relationship between all situation awareness factors is divided into a clear hierarchical structure through the hierarchical analysis method, which improves the systematicness of situation awareness assessment. The weight of each situation awareness factor is obtained through the hierarchical analysis method, and the complex interaction relationship between the factors is quantified. Then, the situation awareness assessment is carried out according to the quantified results, which improves the scientificity and rationality of the situation awareness assessment.

[0044] Considering that when evaluating the weight of elements through expert evaluation in the hierarchical analysis method, there are subjective errors that make it difficult to obtain the accurate weight of each element in the causal diagram, and the Dowhy algorithm has high accuracy in obtaining the weight of elements in the causal relationship, the weight of the elements in the causal relationship can be accurately obtained through the Dowhy algorithm. Considering that the complexity of obtaining the intervention elements and the mixed elements in the process of accurately obtaining the weight of the elements in the causal relationship by the Dowhy algorithm is high, and the weight obtained by the hierarchical analysis method can accurately identify the intervention elements and the mixed elements, the situation awareness is evaluated by combining the hierarchical analysis method with the Dowhy algorithm, which significantly improves the accuracy of the situation awareness evaluation and the efficiency of the situation awareness evaluation.

[0045] The situation awareness assessment is performed through the stability analysis results of the final weighted causal graph, which ensures the robustness of the final weighted causal graph and avoids the assessment error caused by the situation assessment through the unrobust final weighted causal graph. This further improves the accuracy of the situation awareness assessment and the utilization rate of computing resources.

[0046] Situational awareness assessment is performed based on the change trend of the data obtained in real time from the power system relative to the historical data of the power system, so that the final weighted causal diagram changes dynamically according to the change trend, which not only enhances the adaptability of situational awareness assessment but also further improves the accuracy of situational awareness assessment;

[0047] Situation awareness assessment is performed by using the analytic hierarchy process, Dowhy algorithm, stability analysis results of the final weighted cause-and-effect diagram, and the change trend of the real-time data obtained by the power system relative to the historical data of the power system, which significantly improves the accuracy of situation awareness assessment and also significantly improves the efficiency and adaptability of situation awareness assessment;

[0048] The problem of difficulty in accurately obtaining the weights of situation awareness assessment elements, resulting in low accuracy of situation awareness assessment, is solved.

[0049] The present invention also provides a storage medium, in which computer executable instructions are stored. When the computer executable instructions are loaded and executed by a processor, the steps of the method for evaluating situational awareness in a power system are implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Other features, objects and advantages of the present invention will become more apparent by reading the detailed description of non-limiting embodiments made with reference to the following drawings. The drawings are only for the purpose of illustrating preferred embodiments and are not to be considered as limiting the present invention. Also, the same reference symbols are used throughout the drawings to represent the same parts.

[0051] Figure 1A schematic diagram of a flow chart of an evaluation method for situation awareness in a power system according to the present invention;

[0052] Figure 2 This is an initial cause-effect diagram in an evaluation method for situation awareness in an electric power system according to the present invention. DETAILED DESCRIPTION

[0053] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific implementation method described herein is only an optimal embodiment of the present invention, which is only used to explain the present invention and does not limit the scope of protection of the present invention. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0054] Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow charts describe the operations (or steps) as sequential processes, many of the operations (or steps) therein can be implemented in parallel, concurrently, or simultaneously. In addition, the order of the operations can be rearranged. The process can be terminated when its operation is completed, but can also have additional steps not included in the drawings; the process can correspond to a method, function, procedure, subroutine, subprogram, etc.

[0055] Embodiment 1:

[0056] like Figure 1 As shown, a method for evaluating situational awareness in a power system includes the following steps:

[0057] S1: An initial weighted causal diagram is constructed based on the analytic hierarchy process and situation awareness factors. Based on the initial weights of the elements in the initial weighted causal diagram, the first element in the elements is divided into a first intervention element and a first mixed element. The final weight of the first intervention element is obtained based on the Dowhy algorithm.

[0058] In S1, the initial weighted causal diagram constructed based on the analytic hierarchy process and situation awareness factors includes:

[0059] S11: constructing an initial cause-and-effect diagram based on the situation awareness factors and the analytic hierarchy process, comparing the elements pairwise based on the comparison rules in the analytic hierarchy process to obtain a first comparison result, and constructing a fuzzy complementary judgment matrix based on the first comparison result;

[0060] S12: Obtain the weights of the elements based on the eigenvectors of the fuzzy complementary judgment matrix, and then obtain an initial weighted causal graph.

[0061] In S1, the construction of an initial weighted causal diagram based on the analytic hierarchy process and situation awareness factors also includes:

[0062] A consistency check is performed on the fuzzy complementary judgment matrix. If the consistency check result meets the first preset condition, S12 is executed. If the consistency result does not meet the first preset condition, the logical relationship between the data in the fuzzy complementary judgment matrix is ​​compared to obtain a second comparison result. The fuzzy complementary judgment matrix is ​​corrected based on the second comparison result, and S12 is executed.

[0063] In S1, the final weight of the first intervention element obtained based on the Dowhy algorithm includes:

[0064] S1a: constructing a first initial linear regression model based on the initial weight of the first intervention element and the initial weight of the first mixed element; S1b: inputting the value of the first intervention element and the value of the first mixed element in the historical data into the first initial linear regression model to obtain a first output result, updating the first initial linear regression model based on the first output result and the actual results corresponding to the first intervention element and the first mixed element to obtain a first optimized linear regression model, and obtaining the final weight of the first intervention element based on the first optimized linear regression model.

[0065] In this embodiment, the situation awareness factors are divided into a target layer, a criterion layer and an indicator layer through the hierarchical analysis method. Among them, the target layer aims to construct a comprehensive situation evaluation indicator system for the regional distribution network. At the same time, it is divided according to different research criteria. The evaluation factors include six dimensions: low-carbon situation, safety situation, economic situation, regulation situation, power grid structure and regional basic indicators. Screening is performed under these dimensions to form secondary criteria. Each secondary criterion is refined to obtain the corresponding indicators. The constructed initial cause-effect diagram, i.e., the regional distribution network evaluation system diagram, is shown in FIG. Figure 2 As shown, the calculation formula for the proportion of clean energy power generation is: In the formula, P is the proportion of clean energy power generation in the power grid, CE is the clean energy power generation, and TE is the total power generation; the power generation efficiency calculation formula is: In the formula, η is the clean energy power generation efficiency, A c For the actual clean energy generated, I c is the input clean energy; the calculation formula for the density of electric vehicle charging pile facilities is: Where ρ is the density of electric vehicle charging pile facilities, N i is the number of charging piles, A c is the coverage area; the line loss calculation formula is: Where K is the line loss rate of the transmission system, E L is the amount of electricity lost during transmission, E I is the total amount of electricity input into the power system; the calculation formula for the degree of power distribution automation is: Where M is the degree of power distribution automation, A D The number of automation equipment used in the power system, T D is the number of all equipment in the power distribution system; the average power load calculation formula is: In the formula, L is the average power load, E is the total power consumption, and T is the power consumption period. The initial causal diagram is constructed by situation awareness factors and the hierarchical analysis method, and then the complex interaction relationship between all situation awareness factors is divided into a clear hierarchical structure through the hierarchical analysis method, which improves the systematicness of situation awareness assessment.

[0066] The comparison rule in the hierarchical analysis method is to compare the elements of the same level and the same standard in the initial causal diagram in pairs, such as comparing the proportion of clean energy power generation and power generation efficiency on the power generation side, comparing the number of electric vehicle charging devices, the density of electric vehicle charging and replacement facilities, the degree of line loss and the degree of distribution automation on the power grid side, and comparing the power generation side, the power grid side and the demand side under the low-carbon situation in pairs. The fuzzy complementary judgment matrix R = (r ij ) n×n , where r ii =0.5,i=1,2,...,n,r ij +r ji =1, i, j = 1, 2, ..., n, n is the number of elements in the same level and standard in the initial causal diagram. In order to quantitatively explain the relative importance of the two factors, a fuzzy judgment matrix with a scale of 0.1 to 0.9 is used to give a quantitative scale, as shown in Table 1:

[0067] Table 1 0.1~0.9 scaling method and its meaning

[0068]

[0069] The fuzzy complementary judgment matrix is ​​filled in by the scaling method and meaning of Table 1, and then the weights are calculated. If R = (r ij ) n×n is the fuzzy complementary judgment matrix, W = (W 1 ,W 2 ,...,W n ) is the characteristic vector of R, that is, the weight vector, and its expression is: Where W i Factor r iThe weights of each situation awareness factor are obtained through the hierarchical analysis method, and the complex interaction relationship between the factors is quantified. Then, the situation awareness assessment is performed based on the quantified results, which improves the scientificity and rationality of the situation awareness assessment. At the same time, in order to further ensure the accuracy and rationality of the obtained weights, before obtaining the weights through the fuzzy complementary judgment matrix, the characteristic matrix W needs to be calculated. * =(W ij ) n×n , Then through the formula In the formula, A represents the fuzzy complementary matrix, B represents the characteristic matrix, and the fuzzy complementary judgment matrix is ​​subjected to consistency test. When I(A, B)<0.1, it indicates that the consistency test is passed, and when I(A, B)≥0.1, it indicates that the consistency test is not passed. When the consistency test is not passed, the fuzzy complementary judgment matrix is ​​corrected by comparing the logical relationship between the data in the fuzzy complementary judgment matrix. If A is more important than B, and B is more important than C, and the weight of C is greater than the weight of A, there is an obvious logical error at this time. Therefore, the value of C is appropriately reduced or the values ​​of A and B are increased according to the importance, thereby ensuring the rationality of the logical relationship between the data in the fuzzy complementary judgment matrix, and further improving the accuracy of the weights of each element. In this embodiment, the first initial linear regression model is constructed based on the initial weight of the first intervention element and the initial weight of the first mixed element as follows: Y=αS+γU, where α is the initial weight of the first intervention element, S represents the value of the first intervention element, γ is the initial weight of the first mixed element, and U represents the value of the first mixed element. The values ​​of several first intervention elements and the values ​​of the first mixed element in the historical data are input into the first initial linear regression model to obtain the value of the second element, and then the product of the value of the second element and the weight of the second element is calculated, and then the situation awareness assessment result is obtained according to the product. If the situation awareness assessment result does not match the actual situation awareness result, according to the characteristics of the Dowhy algorithm, different historical data are continuously input into the first initial linear regression model to obtain the output result, and then the weight of the first intervention element and the weight of the first mixed element in the first initial linear regression model are continuously corrected according to the difference between the output result and the actual result, so as to obtain the accurate weight of the first intervention element and the first mixed element, thereby improving the accuracy and reliability of the final weight of the first intervention element, and thereby improving the accuracy of the process indicator weight and the accuracy of the final situation awareness assessment.

[0070] S2: Based on the final weight of the first intervention element, the initial weight is updated to obtain a process indicator weight.

[0071] In this embodiment, the initial weight is updated by replacing the initial weight of the first intervention element in the initial weight with the final weight of the first intervention element.

[0072] S3: Determine whether the first mixed element meets the preset requirements. If not, execute S4. If it does, divide the first mixed element into a second intervening element and a second mixed element based on the initial weight, obtain the final weight of the second intervening element based on the process indicator weight and the Dowhy algorithm, update the process indicator weight based on the final weight of the second intervening element, take the second mixed element as the first mixed element, and execute S3.

[0073] In S3, obtaining the final weight of the second intervention element based on the process indicator weight and the Dowhy algorithm includes: updating the first optimized linear regression model based on the process weight of the second intervention element and the process weight of the second confounding element in the process indicator weight;

[0074] The value of the second intervening element and the value of the second mixed element in the historical data are input into the first optimized linear regression model to obtain a second output result, and the first optimized linear regression model is updated based on the second output result and the actual results corresponding to the second intervening element and the second mixed element to obtain an optimal linear regression model, and the final weight of the second intervening element is obtained based on the optimal linear regression model.

[0075] In this embodiment, when the first mixed element is not empty, it meets the preset requirements, and the first mixed element is divided into the second intervening element and the second mixed element based on the initial weight. Specifically, the element with the largest initial weight in the first mixed element is the second intervening element, and the rest are mixed elements. The intervening elements and mixed elements are obtained by the weights obtained by the hierarchical analysis method, which improves the efficiency of identifying the intervening elements and the mixed elements, thereby improving the efficiency of situation awareness assessment.

[0076] The first optimized linear regression model is updated based on the process weight of the second intervention element and the process weight of the second confounding element in the process indicator weight to obtain the first optimized linear regression model: Y = αS + γ 1 U 1 +γ 2 U 2 , where α is the initial weight of the first intervention element, S represents the value of the first intervention element, and γ 1 is the process weight of the second intervention element, U 1 represents the value of the second intervening element, γ 2 is the process weight of the second confounding element, U 2is the value of the second mixed element. The final weight of the second intervening element can be obtained through the above-mentioned method of obtaining the final weight of the first intervening element. The mixed elements are continuously split and the final weights of the elements in the mixed elements are continuously obtained. The final weight of the next element is obtained through the final weight of the obtained element. The accurate weights of all the first elements, that is, all the elements in the indicator layer, are obtained through step-by-step iteration. This solves the problem of difficulty in accurately obtaining the weights of situational awareness assessment elements, resulting in low accuracy of situational awareness assessment, and significantly improves the accuracy of situational awareness assessment.

[0077] S4: obtaining a final weight of a second element in the elements based on the Dowhy algorithm and the process indicator weight, and obtaining a final weighted causal diagram based on the process indicator weight and the final weight of the second element.

[0078] In S4, the obtaining the final weight of the second element in the elements based on the Dowhy algorithm and the process indicator weight includes:

[0079] The second element is divided into a dominant element and a secondary element based on the initial weight of the second element, and the final weights of the dominant element and the secondary element are obtained based on the Dowhy algorithm and the process indicator weight.

[0080] The method of obtaining the final weights of the dominant element and the secondary element based on the Dowhy algorithm and the process indicator weights includes: constructing a second initial linear regression model based on the initial weights of the dominant element and the initial weights of the secondary element, and obtaining the value of the dominant element and the value of the secondary element based on the process indicator weights;

[0081] The value of the dominant element and the value of the secondary element are input into the second initial linear regression model to obtain a third output result. The second initial linear regression model is updated based on the third output result and the actual results corresponding to the dominant element and the secondary element to obtain a second optimized linear regression model. The final weights of the dominant element and the secondary element are obtained based on the second optimized linear regression model.

[0082] In this embodiment, the second element obtained by the analytic hierarchy process, that is, the initial weight of each element in the criterion layer, is divided into a dominant element and a secondary factor, which further improves the accuracy of the situational awareness assessment. The value of the dominant element and the value of the secondary element are obtained based on the process indicator weight: the product of the weight of the first element in the process indicator weight and the value of the first element in the historical data can be used to obtain the value of the dominant element and the secondary element. For example, the second element A has first elements B, C and D, and the weights of B, C and D are 0.5, 0.3, and 0.2 respectively, and the values ​​of B, C and D are 7, 4, and 8 respectively. At this time, the value of the second element A is 0.5*7+0.3*4+0.2*8. The value of the second element is obtained by the accurate weight of the first element, which improves the accuracy of the value of the second element. At the same time, the final weight of the second element is obtained by combining the more accurate value of the second element with the Dowhy algorithm, which improves the accuracy of the final weight of the second element, thereby improving the accuracy of the situational awareness assessment.

[0083] S5: Perform situation awareness assessment based on the stability analysis results of the final weighted causal diagram and the changing trend of the real-time data obtained from the power system relative to the historical data of the power system.

[0084] The S5 includes:

[0085] S51: According to the historical data, a predicted situation awareness assessment result is obtained using the final weighted causal graph, and the predicted situation awareness assessment result is compared with the actual situation awareness assessment result corresponding to the historical data to obtain a stability analysis result. If the stability analysis result does not meet the second preset condition, the historical data is cleaned, outliers are removed, and missing values ​​are removed, and S1b is executed. If the stability analysis result meets the second preset condition, S52 is executed.

[0086] S52: Perform situation awareness assessment based on the changing trend of the data acquired in real time from the power system relative to the historical data of the power system.

[0087] The S52 includes:

[0088] S521: Obtaining a change trend of the data acquired in real time by the power system relative to the historical data;

[0089] S522: If the change trend does not satisfy the third preset condition, the data acquired in real time is used as the historical data and S1b is executed. If the change trend satisfies the third preset condition, a situation awareness assessment is performed based on the final weighted cause-effect diagram.

[0090] In this embodiment, based on a number of historical data, a final weighted causal graph is used to obtain a number of predicted situational awareness assessment results. If a number of predicted situational awareness assessment results greater than a preset number among the several predicted situational awareness assessment results do not match the actual situational awareness assessment results, it means that the final weighted causal graph is unstable. Considering that the instability of the final weighted causal graph may be due to insufficient processing of the acquired historical data, when the stability analysis result does not meet the second preset condition, the historical data is cleaned and other operations are performed, and then the processed historical data is used to obtain the final weighted causal graph, which improves the robustness of the acquired final weighted causal graph, avoids assessment errors caused by situation assessment through the unstable final weighted causal graph, and improves the accuracy of situational awareness assessment while improving the utilization of computing resources.

[0091] In this embodiment, when there is data greater than a preset number in the real-time acquired data and the change trend relative to the historical data is greater than the preset change trend, it means that the third preset condition is not met. The preset number and the preset change trend are set according to actual needs. If it is necessary to improve the efficiency of the situation awareness assessment, the preset number is set to be larger, and the preset change trend is set to be larger. If it is necessary to improve the accuracy of the situation awareness assessment, the preset number is set to be smaller, and the preset change trend is set to be smaller. The situation awareness assessment is performed on the change trend of the data acquired in real time by the power system relative to the historical data of the power system, so that the final weighted causal diagram changes dynamically according to the change trend, which enhances the adaptability of the situation awareness assessment and further improves the accuracy of the situation awareness assessment.

[0092] As an implementation method, in order to improve the efficiency of situation awareness assessment, S51 may be skipped and only S52 may be executed.

[0093] This embodiment also provides a storage medium, in which computer executable instructions are stored. When the computer executable instructions are loaded and executed by a processor, the steps of the method for evaluating situational awareness in a power system are implemented.

[0094] The specific implementation described above is a preferred implementation of a method for evaluating situational awareness in an electric power system of the present invention, and is not intended to limit the specific implementation scope of the present invention. The scope of the present invention includes but is not limited to this specific implementation. All equivalent changes made in accordance with the shape and structure of the present invention are within the protection scope of the present invention.

Claims

1. A method for evaluating situational awareness in a power system, characterized in that: The following steps are involved: S1: construct an initial weighted causal diagram based on the analytic hierarchy process and situation awareness factors, divide the first element of the elements into a first intervention element and a first mixed element based on the initial weights of the elements in the initial weighted causal diagram, and obtain the final weight of the first intervention element based on the Dowhy algorithm; S2: updating the initial weight based on the final weight of the first intervention element to obtain a process indicator weight; S3: Determine whether the first mixed element meets the preset requirements. If not, execute S4. If it meets the requirements, divide the first mixed element into a second intervention element and a second mixed element based on the initial weight, obtain the final weight of the second intervention element based on the process indicator weight and the Dowhy algorithm, update the process indicator weight based on the final weight of the second intervention element, take the second mixed element as the first mixed element, and execute S3. S4: obtaining a final weight of a second element in the elements based on the Dowhy algorithm and the process indicator weight, and obtaining a final weighted causal graph based on the process indicator weight and the final weight of the second element; S5: Perform situation awareness assessment based on the stability analysis results of the final weighted cause-effect diagram and the changing trend of the real-time data acquired from the power system relative to the historical data of the power system.

2. The method for evaluating situational awareness in a power system according to claim 1, characterized in that: In S1, the initial weighted causal diagram constructed based on the analytic hierarchy process and situation awareness factors includes: S11: constructing an initial cause-and-effect diagram based on the situation awareness factors and the analytic hierarchy process, comparing the elements pairwise based on the comparison rules in the analytic hierarchy process to obtain a first comparison result, and constructing a fuzzy complementary judgment matrix based on the first comparison result; S12: Obtain the weights of the elements based on the eigenvectors of the fuzzy complementary judgment matrix, and then obtain an initial weighted causal graph.

3. The method for evaluating situational awareness in a power system according to claim 2, characterized in that: In S1, the construction of an initial weighted causal diagram based on the analytic hierarchy process and situation awareness factors also includes: A consistency check is performed on the fuzzy complementary judgment matrix. If the consistency check result meets the first preset condition, S12 is executed. If the consistency result does not meet the first preset condition, the logical relationship between the data in the fuzzy complementary judgment matrix is ​​compared to obtain a second comparison result. The fuzzy complementary judgment matrix is ​​corrected based on the second comparison result, and S12 is executed.

4. The method for evaluating situational awareness in a power system according to claim 1, characterized in that: In S1, the final weight of the first intervention element obtained based on the Dowhy algorithm includes: S1a: constructing a first initial linear regression model based on the initial weight of the first intervention element and the initial weight of the first mixed element; S1b: inputting the value of the first intervention element and the value of the first mixed element in the historical data into the first initial linear regression model to obtain a first output result, updating the first initial linear regression model based on the first output result and the actual results corresponding to the first intervention element and the first mixed element to obtain a first optimized linear regression model, and obtaining the final weight of the first intervention element based on the first optimized linear regression model.

5. The method for evaluating situational awareness in a power system according to claim 4, characterized in that: In S3, the final weight of the second intervention element obtained based on the process indicator weight and the Dowhy algorithm includes: The first optimized linear regression model is updated based on the process weight of the second intervention element and the process weight of the second confounding element in the process indicator weight; The value of the second intervening element and the value of the second mixed element in the historical data are input into the first optimized linear regression model to obtain a second output result, and the first optimized linear regression model is updated based on the second output result and the actual results corresponding to the second intervening element and the second mixed element to obtain an optimal linear regression model, and the final weight of the second intervening element is obtained based on the optimal linear regression model.

6. The method for evaluating situational awareness in a power system according to claim 1, characterized in that: In S4, the obtaining the final weight of the second element in the elements based on the Dowhy algorithm and the process indicator weight includes: The second element is divided into a dominant element and a secondary element based on the initial weight of the second element, and the final weights of the dominant element and the secondary element are obtained based on the Dowhy algorithm and the process indicator weight.

7. The method for evaluating situational awareness in a power system according to claim 6, characterized in that: The final weights of the dominant elements and the secondary elements obtained based on the Dowhy algorithm and the process indicator weights include: A second initial linear regression model is constructed based on the initial weights of the dominant element and the initial weights of the secondary element, and the value of the dominant element and the value of the secondary element are obtained based on the weights of the process indicators; The value of the dominant element and the value of the secondary element are input into the second initial linear regression model to obtain a third output result. The second initial linear regression model is updated based on the third output result and the actual results corresponding to the dominant element and the secondary element to obtain a second optimized linear regression model. The final weights of the dominant element and the secondary element are obtained based on the second optimized linear regression model.

8. The method for evaluating situational awareness in a power system according to claim 4, characterized in that: The S5 includes: S51: According to the historical data, a predicted situation awareness assessment result is obtained using the final weighted causal graph, and the predicted situation awareness assessment result is compared with the actual situation awareness assessment result corresponding to the historical data to obtain a stability analysis result. If the stability analysis result does not meet the second preset condition, the historical data is cleaned, outliers are removed, and missing values ​​are removed, and S1b is executed. If the stability analysis result meets the second preset condition, S52 is executed. S52: Perform situation awareness assessment based on the changing trend of the data acquired in real time from the power system relative to the historical data of the power system.

9. The method for evaluating situational awareness in a power system according to claim 8, characterized in that: The S52 includes: S521: Obtaining a change trend of the data acquired in real time by the power system relative to the historical data; S522: If the change trend does not satisfy the third preset condition, the data acquired in real time is used as the historical data and S1b is executed. If the change trend satisfies the third preset condition, a situation awareness assessment is performed based on the final weighted cause-effect diagram.

10. A storage medium, characterized in that: The storage medium stores computer executable instructions, and when the computer executable instructions are loaded and executed by the processor, the steps of the method for evaluating situational awareness in a power system as described in any one of claims 1 to 9 are implemented.

Images