Study of Power System Short-term Load Forecast Based on Artificial Neural Network and Genetic Algorithm

• Published in: 2010 International Conference on Computational Aspects of Social Networks pp. 725 - 728
• Main Authors: Du Xin-hui, Tian Feng, Tan Shao-qiong
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.09.2010
• Subjects: Artificial neural network; Artificial neural networks; Biological cells; BP algorithm; Genetic algorithm; Load forecasting; Neurons; Short-term load forecasting; Weather forecasting
• ISBN: 9781424487851; 1424487854
• DOI: 10.1109/CASoN.2010.166

The correct schedule, planning and operation of power system has a tight correlation between accurate load forecast. Aims at the variant feature of power system short-term load, the author took a widely study and discussion on the method of artificial neural network applied on power system short-term load forecasting. At the base of three layered BP neural network, the author studied the meteorological factor effect on short-term load forecasting precision, present the short-term load forecasting model made of BP neural network combine with genetic algorithm. According to the load data of area grid and relevant meteorological data, the author forecasted the short-term load with method of three layered BP neural network, four layered BP neural network and four layered BP neural network combine with genetic algorithm. The result shows that four layered BP neural network combine with genetic algorithm have the advantage of fast calculating time and high precision, have value for engineering application and significance for popularization.