Reactive power optimization method based on integrated learning

Abstract

The present invention discloses a reactive power optimization method based on integrated learning. The method comprises the steps of: (1) determining system parameters; (2) establishing a reactive power optimization model of a power grid; (3) constructing an integrated learning optimizer, comprising: (3-1) selecting a sub optimizer, and (3-2) constructing an integrated optimizer; and (4) adoptingan integrated learning optimizer to solve a reactive power optimization model to make a reactive power optimization strategy. The reactive power optimization method based on integrated learning establishes the integrated learning optimizer to provide a novel optimization solution idea.

Description

technical field [0001] The invention relates to the field of optimal operation and control of electronic systems, in particular to a reactive power optimization method based on integrated learning. Background technique [0002] In modern society, people pay more and more attention to energy conservation and emission reduction. The Chinese government attaches great importance to energy conservation, and power supply enterprises should also respond to national policies. Optimizing the reactive power of the power grid is an urgent measure to improve the power quality, reduce the loss of the power grid, and enhance the power transmission capacity of the power grid. [0003] In the current study, the objective function of reactive power optimization. Existing studies mainly focus on reducing network loss and improving voltage quality. Traditional mathematical solutions are adopted for this, mainly including: linear programming method, nonlinear programming method, etc. These met...

AI Technical Summary

Problems solved by technology

Traditional mathematical solutions are used for this, mainly including: linear programming method, nonlinear programming method, etc. These methods have their own characteristics and advantages, but they have certain restrictions on the processing of discrete variables.

However, for the nonlinearity and multi-objectives of reactive power optimization and the complexity of large power grids, it is easy to have poor convergence when applying nonlinear programming methods.

Another solution is to use classical intelligent algorithms. People get inspiration from observing some animal behaviors in nature (generally group species), and then summarize intelligent algorithms. However, intelligent algorithms also have slow search speed and are easy to The disadvantage of being stuck in a local optimum

Images