Active power distribution network economic operation evaluation method based on hypersphere support vector machine

Abstract

A method for evaluating the economic operation of active distribution networks based on hypersphere support vector machines, including the following steps: define six indicators under the background of large-scale distributed energy access to active distribution networks; compare the defined six indicators with the actual Combining existing indicators to build an evaluation index system for the economic operation of the active distribution network, and improving some existing indicators; using the system gap index and system coordination index to reduce the dimensionality of the index system; establishing an active distribution network based on hyperspherical support vector machines The hypersphere model of each level; the overlapping area is improved, and the evaluation level is judged by the judgment function. The present invention fully considers the impact of large-scale distributed energy access to the active distribution network, corrects some existing indicators, uses dimension reduction processing to solve the problem of redundancy in high-dimensional indicators, and combines the evaluation level with the hypersphere support vector Combined with the computer, the evaluation grade results can be obtained quickly and accurately, which is very reasonable.

Description

technical field [0001] The invention relates to the field of distribution network operation, in particular to a method for evaluating the economic operation of an active distribution network based on a hypersphere support vector machine. Background technique [0002] The economic operation of the distribution network refers to adjusting the unreasonable network structure and organizing the adjustment of the power supply and consumption load in a targeted manner, so that the load delivered by the lines and equipment can reach a reasonable value as much as possible within the operating time, reduce power loss, and reduce the network load. Loss rate, improve economic efficiency. [0003] In recent years, with the increasing penetration of distributed energy resources (DER) such as distributed generation, energy storage systems, and controllable loads, traditional distribution networks are facing challenges in terms of operational flexibility, safety, and economy. This is a big...

AI Technical Summary

Problems solved by technology

[0004] At present, a variety of evaluation methods have been widely used in power grid operation evaluation, mainly including analytic hierarchy process, gray relational degree method, BP neural network method, fuzzy comprehensive evaluation method, TOPSIS method and data envelopment method. The method is mostly a combination or improvement of the above methods, which makes the calculation process too complicated, and the calculation amount is large, which affects the calculation speed and is limited in practical applications.

Images