Intelligent normal pool level optimal selection method based on genetic neural network models

Abstract

The invention discloses an intelligent normal pool level optimal selection method based on genetic neural network models. The method includes the steps of firstly, determining high-low limit of normal pool level; secondly, providing a normal pool level elevation threshold, acquiring area and volume index to form a draft selection scheme, and repeating the second step to form a comparison and selection scheme set; thirdly, selecting index evaluation factors, and selecting evident difference indexes to participate in model calculation; fourthly, forming network learning samples according to required sample number, and normalizing training samples and test samples; fifthly determining a BP network learning structure and initial genetic algorithm population; sixthly, using genetic algorithm to optimize neural network weights and thresholds; seventhly, fine adjusting BP neural network weights, and using built models to evaluate to-be-selected schemes. The method is significantly practical in the field of hydraulic and hydroelectric engineering, the influence of subjective factors is reduced effectively, and objectivity of index weight determination is increased.

Description

technical field [0001] The invention relates to the field of information technology of genetic neural network model and GIS technology in hydropower engineering construction, and in particular to a method for optimizing the normal water storage level of a hydropower station. Background technique [0002] With the rapid growth of the national economy, the demand for electricity continues to increase, and the energy structure dominated by coal will bring constraints on resources and the environment. Hydropower is a pollution-free renewable resource, and hydropower has entered a period of accelerated development. Therefore, to give priority to the development of hydropower projects, more advanced and scientific methods are needed to ensure the improvement of hydropower project construction efficiency and reduce construction costs. It is open and transparent, and reduces social and political influence. , Improve the scientific and technological content of water conservancy work....

AI Technical Summary

Problems solved by technology

However, there are still some deficiencies: (1) Emphasis on the selection of programs, ignoring the selection of indicator effectiveness

(2) The optimization of the scheme relies too much on engineering experience and expert scoring, and the distribution of the weighted values ​​of each target is highly subjective. The determination of the weighted values ​​of each factor in the model becomes a bottleneck

(3) It is based on the fact that several alternative water storage level schemes already exist, how to scientifically draw up the alternative set of water level control elevations to form the water level scheme, and select the optimal scheme from the alternative set is rarely studied

Images