A chaotic genetic-BP neural network forecasting method for wind power in microgrid

Abstract

The invention relates to a chaotic inheritance of wind power in a microgrid. The BP neural network prediction method comprises the following steps: S1, collecting the historical data of the output power of the wind turbine generator system in the microgrid, and dividing the data set into training data and test data; S2, preprocessing the mixed normalization according to the distribution characteristics of the data set, so that the data distribution becomes uniform; S3, constructing a BP neural network and initializing weights, thresholds and other parameter values of the neural network; S4: optimizing the weights and thresholds of the neural network by using the chaotic genetic algorithm, and searching for the optimal neural network parameters; S5:using the processed training data to trainchaotic inheritance-BP neural network, and then the prediction data is output, and the prediction error is calculated. The invention can reduce the influence of the data distribution characteristic on the model, improve the prediction accuracy of the fan output in the micro grid, and provide certain reference for accurate wind power prediction of the micro grid.

Description

technical field [0001] The invention relates to the technical field of wind power prediction of microgrid, in particular to a chaotic genetic-BP neural network prediction method of wind power of microgrid. Background technique [0002] With the development of micro-grid technology, the most mature and pollution-free wind turbines in the renewable field are applied to the micro-grid. At the same time, microgrids connected to wind turbines are more susceptible to random, intermittent, and uneven power distribution characteristics of wind power generation, which creates difficulties in operation and configuration. Therefore, accurately predicting the wind power output of the microgrid is an important means to effectively reduce the impact of wind energy on the microgrid, and to ensure the safe, reliable and economical operation of the microgrid system. [0003] Wind power forecasting refers to the forecasting of the output power of wind power in a period of time in the future....

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to overcome the deficiencies of the prior art, provide a method that can effectively avoid the influence of the distribution characteristics of historical data on the prediction model, solve the precocious problem of easily falling into local optimum when using the genetic algorithm to optimize the parameter value, and improve the efficiency of the microgrid. Chaotic Genetic-BP Neural Network Prediction Method of Wind Power Power in Microgrid Based on Prediction Accuracy of Internal Fan Output

Images