optimized by genetic algorithm

 

 

HUANG Min min, YAN Wen jun

 

(College of Electrical Engineering, Zhejiang University, Hangzhou 310027, China)

 

 

Abstract: Aiming at the influence of output fluctuation of photovoltaic power generation system on the grid, the methods of system output power forecast was researched.A neural network power prediction model optimized by the genetic algorithm was proposed.To find the influencing factors, the relevance between output power and meteorological data was analyzed.The BP neural network model improved by genetic algorithm was established to forecast power output one day in advance.The dynamic correction was used during the prediction to improve model accuracy.Based on the Matlab toolboxes,the trained model was put into use under different weather types, and the data was compared to the results of traditional method.The results indicate that the apply of genetic algorithm optimization is helpful to improve the precision of the neural network model and that the percentage of average error between the prediction results and measured results decreases.The proposed forecasting method has significance in engineering applications.

 

Key words: photovoltaic(PV) power generation; power prediction; genetic algorithm; neural network