CN-122000988-A - Multi-scene-oriented source load storage aggregation scheme generation method

CN122000988ACN 122000988 ACN122000988 ACN 122000988ACN-122000988-A

Abstract

The invention discloses a multi-scene-oriented source charge storage aggregation scheme generation method. The method comprises the steps of obtaining a plurality of characteristic index data, obtaining a plurality of preset scene adaptability evaluation indexes, obtaining scores of a plurality of resources and/or resource combinations on each preset scene adaptability evaluation index respectively to serve as first evaluation groups of each resource and/or resource combination, calculating to obtain a plurality of scenes and scene characteristics corresponding to each scene according to the plurality of index data, calculating weights for the plurality of preset scene adaptability evaluation indexes by using a random forest algorithm according to the plurality of scenes and the corresponding scene characteristics to obtain weights of the plurality of preset scene adaptability evaluation indexes under each scene, calculating to obtain an adaptation matrix of the plurality of scenes and the plurality of resources and/or resource combinations according to first evaluation groups of the resources and/or resource combinations and weights of the plurality of preset scene adaptability evaluation indexes under each scene, and obtaining an optimal scheme under each scene according to the adaptation matrix of the plurality of scenes and the plurality of resources and/or resource combinations. The method has the characteristic of high resource utilization efficiency.

Inventors

LI HANYI
ZHANG HONG
CAO JIANDONG
HAN SHUAI
FU ZEBIN
WEN CHUNYAN
DAI YUN
WEN XIANGFENG
BAI YANLING

Assignees

南网综能数字服务(广州)有限公司

Dates

Publication Date: 20260508
Application Date: 20251230

Claims (10)

1. The method for generating the multi-scene-oriented source load storage aggregation scheme is characterized by comprising the following steps of: S1, acquiring a plurality of characteristic index data, a plurality of preset scene adaptability evaluation indexes, and acquiring scores of a plurality of resources and/or resource combinations on each preset scene adaptability evaluation index respectively as a first evaluation group of each resource and/or resource combination; S2, calculating a plurality of scenes and scene characteristics corresponding to each scene according to a plurality of index data; S3, calculating weights for a plurality of preset scene fitness evaluation indexes by using a random forest algorithm according to a plurality of scenes and corresponding scene characteristics to obtain weights of a plurality of preset scene fitness evaluation indexes under each scene; s4, calculating to obtain a fitness matrix of a plurality of scenes and a plurality of resources and/or resource combinations according to the first evaluation group of each resource and/or resource combination and the weight of a plurality of preset scene fitness evaluation indexes in each scene; And S5, obtaining an optimal scheme under each scene according to the adaptation degree matrix of the plurality of scenes and the plurality of resources and/or the resource combination.
2. The method for generating a multi-scene-oriented source payload aggregation scheme according to claim 1, wherein in step S2, a plurality of scenes and scene features corresponding to each scene are calculated according to a plurality of index data, and the method comprises the following steps: S201, carrying out standardization processing on each characteristic index data to obtain a plurality of standardized characteristic index data; s202, performing main component dimension reduction on each piece of standardized characteristic index data to obtain a plurality of dimension reduced characteristic index data; and S203, performing cluster analysis on the plurality of feature index data subjected to dimension reduction to obtain a plurality of scenes and scene features corresponding to each scene.
3. The method for generating a multi-scenario-oriented source payload aggregation scheme according to claim 2, wherein in step S203, the cluster analysis includes: s20301, processing the plurality of feature index data subjected to dimension reduction by using an elbow method to obtain the clustering number; S20302, clustering a plurality of the feature index data subjected to dimension reduction by using a K-Means clustering algorithm according to the clustering number to obtain a plurality of scenes and a plurality of feature index data corresponding to each scene; S20303, carrying out averaging on a plurality of characteristic index data corresponding to each scene to form clustering center characteristic index data corresponding to the plurality of scenes; S20304, comparing the cluster center feature index data corresponding to each scene with a first preset threshold vector to obtain scene features corresponding to each scene.
4. A multi-scene-oriented source load storage aggregation scheme generation method according to any one of claims 1-3 is characterized in that the characteristic index data comprise characteristic data, environment data and policy data of each resource.
5. The method for generating a multi-scene oriented source load storage aggregation scheme as claimed in claim 1, wherein the preset scene fitness evaluation indexes comprise a resource availability index, a response speed index and a cost benefit index.
6. The method for generating a multi-scenario-oriented source payload aggregation scheme according to claim 1, wherein in step S4, an fitness matrix of a plurality of scenarios and a plurality of resources and/or resource combinations is calculated, and the method comprises: S401, calculating to obtain comprehensive fitness scores of each resource and/or resource combination in each scene according to the first evaluation group of each resource and/or resource combination and the weights of a plurality of preset scene fitness evaluation indexes in each scene; and S402, normalizing the first intermediate matrix to obtain a normalized first intermediate matrix which is used as an adaptation degree matrix of a plurality of scenes and a plurality of resources and/or resource combinations.
7. The method for generating a multi-scenario-oriented source payload aggregation scheme according to claim 6, wherein in step S401, a calculation formula of a comprehensive fitness score in each scenario is as follows: a total number of first scoring packets representing resource and/or resource combinations, A sequence number of a first scoring packet representing a resource and/or a combination of resources, A weight indicating an a-th preset scene fitness evaluation index, An a-th value in a first scoring group representing a resource and/or a combination of resources.
8. The method for generating a multi-scenario-oriented source payload aggregation scheme according to claim 1, wherein in step S5, an optimal scheme under each scenario is obtained according to a fitness matrix of the multiple scenarios and multiple resources and/or resource combinations, including: s501, selecting resources and/or resource combinations with the adaptation degree larger than a second preset threshold value in each scene as a first resource group of each scene according to the adaptation degree matrix of the scenes and the resources and/or resource combinations; s502, carrying out scheme optimization generation on the first resource group of each scene by adopting a multi-objective optimization algorithm to obtain an optimal scheme under each scene.
9. The method for generating a multi-scenario-oriented source payload aggregation scheme according to claim 8, wherein in step S502, the multi-objective optimization algorithm includes: s50201, generating a plurality of resource allocation genes, scheduling strategy genes and economic genes corresponding to each scene according to the first resource group of each scene; s50202, generating a plurality of chromosomes corresponding to each scene according to the plurality of resource allocation genes, the scheduling strategy genes and the economic genes corresponding to each scene; s50203, optimizing a plurality of chromosomes corresponding to each scene by utilizing a genetic algorithm to obtain an optimal chromosome in each scene; And S50204, obtaining an optimal scheme under each scene according to the optimal chromosome under each scene.
10. A multi-scenario-oriented source load storage aggregation scheme generation system applied to the generation method of any one of claims 1 to 9, comprising: The data acquisition module is used for acquiring a plurality of characteristic index data, a plurality of preset scene adaptability evaluation indexes, and a plurality of resources and/or resource combinations, which are respectively scored on each preset scene adaptability evaluation index, as a first scoring group of each resource and/or resource combination; The scene calculation module is used for calculating a plurality of scenes and scene characteristics corresponding to each scene according to the index data; The weight calculation module is used for calculating weights of the preset scene fitness evaluation indexes by utilizing a random forest algorithm according to the scenes and the corresponding scene characteristics to obtain weights of the preset scene fitness evaluation indexes in each scene; The adaptation degree matrix calculation module is used for calculating and obtaining an adaptation degree matrix of a plurality of scenes and a plurality of resources and/or resource combinations according to the first evaluation group of each resource and/or resource combination and the weights of a plurality of preset scene adaptation degree evaluation indexes under each scene; And the scheme generating module is used for obtaining an optimal scheme under each scene according to the adaptation degree matrix of the plurality of scenes and the plurality of resources and/or the resource combination.

Description

Multi-scene-oriented source load storage aggregation scheme generation method Technical Field The invention relates to the technical field of power systems, in particular to a multi-scene-oriented source charge storage aggregation scheme generation method. Background With the wide application of resources such as distributed photovoltaic, energy storage and flexible load, specific scenes are required to be matched for the resources in an electric power system, so that the efficient utilization of the resources is realized. The traditional scene division method is often used for carrying out scene division based on only one or a few dimensions, such as meteorological types, load curve forms and resource types, and cannot systematically integrate complex coupling relations among physical characteristics of resources, external environment conditions and market policy rules, and a simplified division mode is difficult to truly reflect typical operation states formed under interaction of source-load-storage-environment-policy multiple elements, so that constructed scenes are disjointed from actual operation requirements, and resource utilization efficiency is low. The prior art discloses a hierarchical cooperative control method for a plurality of aggregation distributed resource clusters of source load storage, which comprises the steps of classifying control grades of the plurality of aggregation distributed resource clusters of the source load storage, aggregating resources of the same control grade into the same resource cluster based on a control grade classification result, constructing the aggregation distributed resource clusters of the source load storage, establishing a distributed resource cluster cooperative control model, and solving the distributed resource cluster cooperative control model by introducing a graph neural network model. The method does not consider the factors of the application scene, so that the method cannot keep high resource utilization efficiency in a plurality of scenes. Disclosure of Invention Aiming at the defect of low resource utilization efficiency in multiple scenes in the prior art, the invention provides a method for generating a source load storage aggregation scheme for multiple scenes. The method has the characteristic of high resource utilization efficiency. The primary purpose of the invention is to solve the technical problems, and the technical scheme of the invention is as follows: A multi-scene-oriented source charge storage aggregation scheme generation method comprises the following steps: S1, acquiring a plurality of characteristic index data, a plurality of preset scene adaptability evaluation indexes, and acquiring scores of a plurality of resources and/or resource combinations on each preset scene adaptability evaluation index respectively as a first evaluation group of each resource and/or resource combination; S2, calculating a plurality of scenes and scene characteristics corresponding to each scene according to a plurality of index data; S3, calculating weights for a plurality of preset scene fitness evaluation indexes by using a random forest algorithm according to a plurality of scenes and corresponding scene characteristics to obtain weights of a plurality of preset scene fitness evaluation indexes under each scene; s4, calculating to obtain a fitness matrix of a plurality of scenes and a plurality of resources and/or resource combinations according to the first evaluation group of each resource and/or resource combination and the weight of a plurality of preset scene fitness evaluation indexes in each scene; And S5, obtaining an optimal scheme under each scene according to the adaptation degree matrix of the plurality of scenes and the plurality of resources and/or the resource combination. Further, in step S2, according to the plurality of index data, a plurality of scenes and scene features corresponding to each scene are calculated, including: S201, carrying out standardization processing on each characteristic index data to obtain a plurality of standardized characteristic index data; s202, performing main component dimension reduction on each piece of standardized characteristic index data to obtain a plurality of dimension reduced characteristic index data; and S203, performing cluster analysis on the plurality of feature index data subjected to dimension reduction to obtain a plurality of scenes and scene features corresponding to each scene. Further, in step S203, the cluster analysis includes: s20301, processing the plurality of feature index data subjected to dimension reduction by using an elbow method to obtain the clustering number; S20302, clustering a plurality of the feature index data subjected to dimension reduction by using a K-Means clustering algorithm according to the clustering number to obtain a plurality of scenes and a plurality of feature index data corresponding to each scene; S20303, carrying out averaging on a plural