Progressive planning-based power grid evaluation method and system based on combined assessment theory
By combining assessment theory and deep learning models, the problem of subjective bias in traditional power grid assessment methods is solved, achieving objectivity and accuracy in power grid assessment, and making it applicable to assessment and analysis in power grids and other fields.
Patent Information
- Application Number
- PCT/CN2024/135897
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2024-10-08
- Filing Date
- 2024-11-29
- Publication Date
- 2026-04-16
AI Technical Summary
Traditional power grid assessment methods are influenced by subjective biases, leading to inaccurate assessment results and making it difficult to comprehensively and accurately measure the effectiveness of distribution network planning.
A progressive planning power grid evaluation method based on combined evaluation theory is adopted, which combines subjective and objective weighting methods, similarity theory evaluation, nonparametric regression and deep learning models, and performs power grid data analysis through data processing, feature engineering, LSTM network and attention mechanism.
It improves the objectivity and accuracy of power grid assessment, can handle large amounts of complex data, enhances assessment efficiency and practicality, and is applicable to power grid assessment and can be extended to other fields.
Smart Images
Figure CN2024135897_16042026_PF_FP_ABST
Abstract
Description
An Incremental Planning Power Grid Evaluation Method and System Based on Combinatorial Evaluation Theory Technical Field
[0001] This invention relates to the technical field of power grid assessment and optimization, specifically to an incremental planning power grid evaluation method and system based on combined assessment theory. Background Technology
[0002] Guided by electrification, clean energy, digitalization, and standardization, building a clean, efficient, safe, and low-carbon energy system is a crucial pathway to achieving the goals of "carbon peaking and carbon neutrality." With the rapid development of the power industry, the demand for distribution network construction is increasing daily. Simultaneously, distribution network construction is characterized by its wide coverage, large scale, and numerous related factors, making its planning, renovation, and construction a massive and complex systems engineering project. Located at the end of the power system and directly connected to users, the distribution network reflects the operational capacity of the entire power system. As the scale of distribution networks gradually increases, distribution network planning and design issues have become an increasingly important concern in the power industry. To achieve optimal economic or safety in distribution network construction, a comprehensive and accurate assessment of the network's operational level is necessary. Therefore, a comprehensive and precise evaluation index system is needed to objectively measure the effectiveness and achievements of the planning. The weighting of evaluation indicators is a key issue in the evaluation process. Different indicators have different levels of importance in distribution network planning, and traditional weighting methods are often based on subjective judgment or expert opinions, easily influenced by subjective biases and uncertainties, leading to biased evaluation results. Therefore, studying objective weighting methods and mixed subjective-objective interval affine weighting methods can improve the accuracy and reliability of evaluation results by considering the interrelationships and weight allocation between indicators under conditions of uncertainty. Summary of the Invention
[0003] In view of the above-mentioned problems, the present invention is proposed.
[0004] Therefore, the technical problem solved by this invention is: how to more accurately and efficiently evaluate the performance and status of the power grid.
[0005] To address the aforementioned technical problems, this invention provides the following technical solution: a progressive planning power grid evaluation method based on combined evaluation theory, comprising the following steps:
[0006] Data from the power grid system is collected, and the collected data is weighted using a combined weighting method. Evaluation values are derived based on similarity theory. Dimensionless processing is used to standardize the indicators. Evaluation is then performed based on nonparametric regression. As a preferred embodiment of the progressive planning power grid evaluation method based on combined evaluation theory described in this invention, the power grid system data includes power grid operating status data, power grid equipment operating data, power grid load data, power grid fault data, and power grid environmental data.
[0007] The combined weighting method includes subjective weighting and objective weighting. The weighting method is selected based on the performance analysis of the power grid system.
[0008] The data indicators of the power grid system are evaluated using the subjective weighting method, a judgment matrix is generated, the judgment matrix is normalized and verified, and the system weight of the matrix is obtained through iterative calculation.
[0009] The data indicators of the power grid system are weighted by objective weighting method to generate the original data matrix. The weight values are then dimensionless and normalized to obtain the comprehensive weight of the power grid system.
[0010] The process of generating the judgment matrix includes comparing the evaluation factors and filling the comparison results into the judgment matrix.
[0011] The normalized check is represented as follows:
[0012] If CR is less than 0.1, it means that the normalization check has been passed.
[0013] Where CI represents the consistency index, and λ mad represents the largest eigenvalue of the judgment matrix, CR represents the consistency ratio, RI represents the random consistency index, and n represents the number of evaluation factors.
[0014] As a preferred embodiment of the incremental planning power grid evaluation method based on the combined evaluation theory described in this invention, the method for obtaining the comprehensive weight of the power grid system includes: calculating the weight of each data indicator according to the objective weighting method and verifying it according to the Kendall concordance coefficient test method.
[0015] When d(W) (1) W (2) If the value is greater than the test threshold, the test is passed. If the test is passed, then use... Calculate the overall weight of the power grid system.
[0016] The Kendall concordance coefficient verification method is expressed as follows:
[0017] Wherein d(W (1) W(2) ) represents the weight vector W (1) and W (2) The Euclidean distance between them, (W) j (1) W j (2) ) 2 This represents the square of the difference between the weight values of two weight vectors at position j. This represents the total weight difference calculated, where W represents the overall weight of the power grid system, and λ represents the total weight difference calculated. k q represents the weight of the k-th evaluation criterion, and q represents the total number of evaluation criteria.
[0018] As a preferred embodiment of the incremental planning power grid evaluation method based on combinatorial evaluation theory described in this invention, the evaluation method based on similarity theory includes: constructing a new model for system level evaluation using the Shepard similarity interpolation method based on accelerated genetic algorithm and the ideal interval method based on genetic algorithm.
[0019] The new model for constructing system level evaluation includes randomly generating evaluation level sample series x(i,j) and y(i) according to the system evaluation standard table, where i = 1 to n and j = 1 to m.
[0020] Based on the sample parameter b, an optimization estimate is performed. Sample i is taken from the sample series, and Shepard interpolation is performed on the other n-1 samples to obtain the interpolation value corresponding to the evaluation level y(i), denoted as y. c (i).
[0021] The accelerated genetic algorithm is expressed as follows: st1≤b≤5
[0022] in, Indicates that the summation is true, y c (i) represents the i-th observation, y(i) represents the i-th actual value, and st represents the constraint condition.
[0023] The ideal interval method based on genetic algorithm includes generating a series of evaluation standard samples and performing dimensionless processing, calculating the distance between each standard sample and the ideal interval of the standard level, calculating the relative membership value of each standard sample to the ideal interval of the standard level, and the equivalent comprehensive evaluation.
[0024] As a preferred embodiment of the incremental planning power grid evaluation method based on combinatorial evaluation theory described in this invention, the dimensionless processing is expressed as: x(k,j)=x*(k,j) / x max (j)(k=1~n k j = 1 to n j a(i,j)=a*(i,j) / xmax (j)(i=1~n i j = 1 to n j b(i,j)=b*(i,j) / x max (j)(i=1~n i j = 1 to n j )
[0025] Where x(k,j), a(i,j), and b(i,j) represent two-dimensional variables, x * (k,j), a * (i, j), b * (i, j) represent the original values of x(k, j), a(i, j), and b(i, j), respectively. max (j) represents the maximum value of the j-th evaluation index in the sample series of acceptable evaluation criteria, denoted as,
[0026] The distance between each standard sample and the ideal interval of the standard level is expressed as follows:
[0027] Where D(k, i) represents the total distance between the k-th observation and the i-th reference value, w(j) represents the weight of the j-th indicator, d(k, i, j) represents the distance between the k-th observation and the i-th reference value on the j-th feature, a(i, j) and b(i, j) represent the lower and upper bounds of the i-th reference value on the j-th feature, and x(k, j) represents the value of the k-th observation on the j-th feature.
[0028] As a preferred embodiment of the progressive planning power grid evaluation method based on combinatorial evaluation theory described in this invention, the evaluation based on nonparametric regression includes obtaining a one-dimensional projection value z(i) of the evaluation value, expressed as follows:
[0029] The projection index function is derived from the one-dimensional projection value, expressed as Q(a)=S z |R xy |
[0030] Where || represents absolute value, S z This represents the standard deviation of the projected value z(i);
[0031] The optimal projection direction, maxQ(a) = S, is estimated by solving the problem of maximizing the projection index function. z |R xy |
[0032] Optimization is performed using an accelerated genetic algorithm.
[0033] As a preferred embodiment of the incremental planning power grid evaluation method based on combinatorial evaluation theory described in this invention, the evaluation based on nonparametric regression further includes establishing a corresponding Nadaraya-Watson nonparametric model, expressed as follows:
[0034] Using standard evaluation object {z * (i) |i=1~n} series and accelerated genetic algorithm are optimized by solving the following optimization problem, expressed as,
[0035] Another objective of this invention is to provide a progressive planning power grid evaluation system based on combinatorial evaluation theory. This system can systematically collect, process, and analyze power grid data, and combine LSTM networks and attention mechanisms to construct and train deep learning models, thereby solving the problems of low efficiency and low accuracy of existing evaluation methods when dealing with large amounts of complex power grid data.
[0036] To address the aforementioned technical problems, this invention provides the following technical solution: a progressive planning power grid evaluation system based on combinatorial evaluation theory, comprising: a data processing module, a feature engineering module, a deep learning model training module, and a regularization and optimization module.
[0037] The data processing module is responsible for collecting historical and real-time data from the power grid system.
[0038] The feature engineering module is responsible for selecting and constructing the features required for evaluation.
[0039] The deep learning model training module is responsible for building and training an LSTM network combined with an attention mechanism.
[0040] The regularization and optimization module is responsible for preventing the model from overfitting.
[0041] A computer device includes a memory and a processor, the memory storing a computer program, characterized in that the processor executes the computer program to implement the steps of the incremental planning power grid evaluation method based on the combined evaluation theory as described above.
[0042] A computer-readable storage medium having a computer program stored thereon, characterized in that, when the computer program is executed by a processor, it implements the steps of the incremental planning power grid evaluation method based on the combined evaluation theory as described above.
[0043] The beneficial effects of this invention are as follows: By employing a combined subjective and objective weighting method, this invention ensures the objectivity and fairness of the evaluation results, making the evaluation more comprehensive. Through deep learning models, particularly LSTM networks and attention mechanisms, this invention can process large amounts of complex power grid data, improving the efficiency and accuracy of the evaluation. The systematic method of this invention is not only applicable to power grid evaluation but can also be extended to evaluation and analysis in other fields, showing broad application prospects. Through the design of automatically optimized evaluation parameters, this invention can adaptively adjust based on historical and real-time data, making the evaluation results more realistic and improving the practicality of the evaluation. Attached Figure Description
[0044] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:
[0045] Figure 1 is an overall flowchart of the incremental planning power grid evaluation method based on the combined evaluation theory provided in the first embodiment of the present invention.
[0046] Figure 2 is an overall framework diagram of the incremental planning power grid evaluation system based on the combined evaluation theory provided in the second embodiment of the present invention. Detailed Implementation
[0047] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.
[0048] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.
[0049] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.
[0050] This invention is described in detail with reference to the schematic diagrams. When describing the embodiments of this invention, for ease of explanation, the cross-sectional views illustrating the device structure may be partially enlarged, not according to the usual scale. Furthermore, the schematic diagrams are merely examples and should not be construed as limiting the scope of protection of this invention. In actual fabrication, the three-dimensional spatial dimensions of length, width, and depth should be included.
[0051] Furthermore, in the description of this invention, it should be noted that the terms "upper," "lower," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. These terms are used solely for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. In addition, the terms "first," "second," or "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.
[0052] Unless otherwise explicitly specified and limited, the terms "installation," "connection," and "joining" in this invention should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; similarly, they can refer to mechanical connections, electrical connections, or direct connections, or indirect connections through an intermediate medium, or internal connections between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.
[0053] Example 1
[0054] Referring to Figure 1, an embodiment of the present invention provides a progressive planning power grid evaluation method based on combined evaluation theory, characterized in that:
[0055] S1: Collect power grid system data and adjust the weights of the collected power grid system data based on the combined weighting method.
[0056] Power grid system data includes power grid operating status data, power grid equipment operating data, power grid load data, power grid fault data, and power grid environmental data.
[0057] The combined weighting method includes subjective weighting and objective weighting. The weighting method is selected based on the performance analysis of the power grid system.
[0058] The data indicators of the power grid system are evaluated using the subjective weighting method, a judgment matrix is generated, the judgment matrix is normalized and verified, and the system weight of the matrix is obtained through iterative calculation.
[0059] The data indicators of the power grid system are weighted by objective weighting method to generate the original data matrix. The weight values are then dimensionless and normalized to obtain the comprehensive weight of the power grid system.
[0060] The process of generating the judgment matrix includes comparing the evaluation factors and filling the comparison results into the judgment matrix.
[0061] The normalized check is represented as follows:
[0062] If CR is less than 0.1, it means that the normalization check has been passed.
[0063] Where CI represents the consistency index, and λ mad represents the largest eigenvalue of the judgment matrix, CR represents the consistency ratio, RI represents the random consistency index, and n represents the number of evaluation factors.
[0064] Furthermore, the comprehensive weight of the power grid system is obtained by calculating the weight of each data indicator according to the objective weighting method and verifying it according to the Kendall concordance coefficient test method.
[0065] When d(W) (1) W (2) If the value is greater than the test threshold, the test is passed. If the test is passed, then use... Calculate the overall weight of the power grid system.
[0066] The Kendall concordance coefficient verification method is expressed as follows:
[0067] Wherein d(W (1) W (2) ) represents the weight vector W (1) and W (2) The Euclidean distance between them, (W) j (1) W j (2) ) 2 This represents the square of the difference between the weight values of two weight vectors at position j. This represents the total weight difference calculated, where W represents the overall weight of the power grid system, and λ represents the total weight difference calculated. k q represents the weight of the k-th evaluation criterion, and q represents the total number of evaluation criteria.
[0068] S2: The evaluation value is obtained based on the similarity theory evaluation method.
[0069] Furthermore, similarity theory evaluation methods include constructing a new model for system ranking evaluation using the Shepard similarity interpolation method based on accelerated genetic algorithms, and the ideal interval method based on genetic algorithms.
[0070] The new model for system level evaluation includes randomly generating evaluation level sample series x(i,j) and y(i) based on the system evaluation standard table, where i = 1 to n and j = 1 to m.
[0071] Based on the sample parameter b, an optimization estimate is performed. Sample i is taken from the sample series, and Shepard interpolation is performed on the other n-1 samples to obtain the interpolation value corresponding to the evaluation level y(i), denoted as y. c (i).
[0072] The accelerated genetic algorithm is represented as follows: st1≤b≤5
[0073] in, Indicates that the summation is true, y c (i) represents the i-th observation, y(i) represents the i-th actual value, and st represents the constraint condition.
[0074] The ideal interval method based on genetic algorithms includes generating a series of evaluation standard samples and performing dimensionless processing, calculating the distance between each standard sample and the ideal interval of the standard level, calculating the relative membership value of each standard sample to the ideal interval of the standard level, and performing an equivalent comprehensive evaluation.
[0075] S3: Standardize the indicators by using dimensionless processing.
[0076] Further dimensionless processing can be expressed as: x(k,j)=x*(k,j) / x max (j)(k=1~n k j = 1 to n j a(i,j)=a*(i,j) / x max (j)(i=1~n i j = 1 to n j b(i,j)=b*(i,j) / x max (j)(i=1~n i j = 1 to n j )
[0077] Where x(k,j), a(i,j), and b(i,j) represent two-dimensional variables, x * (k,j), a * (i, j), b * (i, j) represent the original values of x(k, j), a(i, j), and b(i, j), respectively. max (j) represents the maximum value of the j-th evaluation index in the sample series of acceptable evaluation criteria, denoted as,
[0078] The distance between each standard sample and the ideal interval of the standard level is calculated and expressed as follows:
[0079] Where D(k, i) represents the total distance between the k-th observation and the i-th reference value, w(j) represents the weight of the j-th indicator, d(k, i, j) represents the distance between the k-th observation and the i-th reference value on the j-th feature, a(i, j) and b(i, j) represent the lower and upper bounds of the i-th reference value on the j-th feature, and x(k, j) represents the value of the k-th observation on the j-th feature.
[0080] S4: Evaluation based on nonparametric regression.
[0081] Evaluation based on nonparametric regression includes obtaining the one-dimensional projection value z(i) of the evaluation value, denoted as,
[0082] The projection index function is derived from the one-dimensional projection value, expressed as Q(a)=S z |R xy |
[0083] Where || represents absolute value, S z This represents the standard deviation of the projected value z(i);
[0084] The optimal projection direction, maxQ(a) = S, is estimated by solving the problem of maximizing the projection index function. z |R xy |
[0085] Optimization is performed using an accelerated genetic algorithm.
[0086] Evaluation based on nonparametric regression also includes establishing a corresponding Nadaraya-Watson nonparametric model, denoted as...
[0087] Using standard evaluation object {z * (i) |i=1~n} series and accelerated genetic algorithm are optimized by solving the following optimization problem, expressed as,
[0088] Example 2
[0089] Referring to Figure 2, an embodiment of the present invention provides a system for an incremental planning power grid evaluation method based on combinatorial evaluation theory. The incremental planning power grid evaluation system based on combinatorial evaluation theory includes a data processing module, a feature engineering module, a deep learning model training module, and a regularization and optimization module.
[0090] The data processing module is responsible for collecting historical and real-time data from the power grid system; the feature engineering module is responsible for selecting and constructing the features required for evaluation; the deep learning model training module is responsible for building and training an LSTM network combined with an attention mechanism; and the regularization and optimization module is responsible for preventing model overfitting.
[0091] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0092] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.
[0093] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.
[0094] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.
[0095] Example 3
[0096] In this embodiment, in order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.
[0097] Data source: Historical and real-time data of the power grid in a certain region were selected, including key indicators such as current, voltage, power, frequency, and harmonic content.
[0098] Evaluation objectives: stability, efficiency, reliability, response speed, fault tolerance, and economy of the power grid.
[0099] Evaluation methods: On the one hand, the method of this invention is used; on the other hand, traditional power grid evaluation methods are used.
[0100] This embodiment conducted experiments on both the existing traditional method and the method of this embodiment, as shown in Table 1.
[0101] Table 1 Comparison of Experimental Results
[0102] As can be seen from the table above, the evaluation method of this invention outperforms traditional methods in all evaluation indicators. This indicates that the method of this invention is more accurate and comprehensive, and can better reflect the actual performance of the power grid.
[0103] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.
Claims
1. A progressive planning grid evaluation method based on the theory of portfolio evaluation, characterized in that, include: Data from the power grid system is collected, and the collected data is weighted and corrected based on a combined weighting method. The evaluation value is derived based on the similarity theory evaluation method; The indicators are standardized by using dimensionless processing; Evaluation is based on nonparametric regression.
2. The method of claim 1, wherein the method is based on a combination of assessment theory and progressive planning of the grid evaluation. The power grid system data includes power grid operating status data, power grid equipment operating data, power grid load data, power grid fault data, and power grid environment data. The combined weighting method includes subjective weighting and objective weighting, and the weighting method is selected based on the performance analysis of the power grid system. The data indicators of the power grid system are evaluated based on the subjective weighting method, a judgment matrix is generated, the judgment matrix is normalized and verified, and the system weight of the matrix is obtained through iterative calculation. The data indicators of the power grid system are weighted by objective weighting method to generate the original data matrix. The weight values are then dimensionless and normalized to obtain the comprehensive weight of the power grid system. The process of generating the judgment matrix includes comparing the evaluation factors and filling the comparison results into the judgment matrix. The normalized check is represented as follows: If CR is less than 0.1, it means that the normalization check has been passed; Where CI represents the consistency index, and λ mad represents the largest eigenvalue of the judgment matrix, CR represents the consistency ratio, RI represents the random consistency index, and n represents the number of evaluation factors.
3. The incremental planning power grid evaluation method based on combined evaluation theory as described in claim 2, characterized in that: The process of obtaining the comprehensive weight of the power grid system includes calculating the weight of each data indicator according to the objective weighting method and verifying it according to the Kendall concordance coefficient test method. When d(W) (1) W (2) If the value is greater than the test threshold, the test is passed. If the test is passed, then use... Calculate the overall weight of the power grid system; The Kendall concordance coefficient verification method is expressed as follows: Wherein d(W (1) W (2) ) represents the weight vector W (1) and W (2) The Euclidean distance between them, (W) j (1) W j (2) ) 2 This represents the square of the difference between the weight values of two weight vectors at position j. This represents the total weight difference calculated, where W represents the overall weight of the power grid system, and λ represents the total weight difference calculated. k q represents the weight of the k-th evaluation criterion, and q represents the total number of evaluation criteria.
4. The incremental planning power grid evaluation method based on combined evaluation theory as described in claim 3, characterized in that: The similarity theory evaluation methods include a new model for power grid system level evaluation constructed by the Shepard similarity interpolation method based on accelerated genetic algorithm and the ideal interval method based on genetic algorithm; The new model for constructing system level evaluation includes randomly generating a series of power grid data evaluation level samples x(i,j) and y(i) based on a system evaluation standard table, where i = 1 to n. i j = 1 to n j ; Based on the sample parameter b, an optimization estimate is performed. Sample i is taken from the sample series, and Shepard interpolation is performed on the other n-1 samples to obtain the interpolation value corresponding to the evaluation level y(i), denoted as y. c (i); The accelerated genetic algorithm is expressed as follows: st1≤b≤5 in, Indicates that the summation is true, y c (i) represents the i-th observation, y(i) represents the i-th actual value, and st represents the constraint condition; The ideal interval method based on genetic algorithm includes generating a series of evaluation standard samples and performing dimensionless processing, calculating the distance between each standard sample and the ideal interval of the standard level, and calculating the relative membership value and evaluation value of each standard sample to the ideal interval of the standard level.
5. The incremental planning power grid evaluation method based on combined evaluation theory as described in claim 4, characterized in that: The dimensionless processing is expressed as x(k,j)=x*(k,j) / x max (j)(k=1~n k j = 1 to n j a(i,j)=a*(i,j) / x max (j)(i=1~n i j = 1 to n j b(i,j)=b*(i,j) / x max (j)(i=1~n i j = 1 to n j ) Where x(k,j), a(i,j), and b(i,j) represent two-dimensional variables, x * (k,j), a * (i, j), b * (i, j) represent the original values of x(k, j), a(i, j), and b(i, j), respectively. max (j) represents the maximum value of the j-th evaluation index in the sample series of acceptable evaluation criteria, denoted as, The distance between each standard sample and the ideal interval of the standard level is expressed as follows: Where D(k, i) represents the total distance between the k-th observation and the i-th reference value, w(j) represents the weight of the j-th indicator, d(k, i, j) represents the distance between the k-th observation and the i-th reference value on the j-th feature, a(i, j) and b(i, j) represent the lower and upper bounds of the i-th reference value on the j-th feature, and x(k, j) represents the value of the k-th observation on the j-th feature, k = 1 to n k ; The relative membership value is expressed as follows: Where h(k) represents the quality grade value of the observation k, and y(k) represents the true value of the observation k; The evaluation value is represented as {z} * (k,j)|k=1~n z j = 1 to n j } 6. The incremental planning power grid evaluation method based on combined evaluation theory as described in claim 5, characterized in that: The evaluation based on nonparametric regression includes obtaining the one-dimensional projection value z(i) of the evaluation value, denoted as, The projection index function is derived from the one-dimensional projection value, and is expressed as follows: Q(a)=S z |R xy | where || denotes the absolute value, S z denotes the standard deviation of the projection values z(i); The optimal projection direction is estimated by solving the problem of maximizing the projection index function. maxQ(a)=S z |R xy | Optimization is performed using an accelerated genetic algorithm.
7. The incremental planning power grid evaluation method based on combined evaluation theory as described in claim 6, characterized in that: The evaluation based on nonparametric regression also includes establishing a corresponding Nadaraya-Watson nonparametric model, denoted as follows: Using standard evaluation object {z * (i) |i=1~n} series and accelerated genetic algorithm are optimized by solving the following optimization problem, expressed as, 8. A system employing the incremental planning power grid evaluation method based on the combined evaluation theory as described in any one of claims 1 to 7, characterized in that: It includes a data processing module, a feature engineering module, a deep learning model training module, and a regularization and optimization module; The data processing module is responsible for collecting historical and real-time data from the power grid system. The feature engineering module is responsible for selecting and constructing the features required for evaluation; The deep learning model training module is responsible for building and training an LSTM network combined with an attention mechanism. The regularization and optimization module is responsible for preventing the model from overfitting.
9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the incremental planning power grid evaluation method based on the combined evaluation theory as described in any one of claims 1 to 7.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the incremental planning power grid evaluation method based on the combined evaluation theory as described in any one of claims 1 to 7.