A BP Artificial Network with Improved Quasi-Newton Self-adaptation Algorithm in Short-term Load Forecasting Application

摘要：

The short term load forecasting is an important part of an energy management system. The short-term load forecast greatly influences the economics and reliability of power system operations. In this paper an improved BP Artificial Neural Network based on quasi-Newton Algorithm application for short-term load forecasting is presented. This algorithm can solve image in version phenomenon, and improve the extending ability of network. It greatly improves BP ANN performance, and the model has a higher convergence and learning speed, experiments and analysis shows that this algorithm has a high application value.

关键词： Artificial Neural Network (ANN) Load Forecasting BP Algorithm Quasi-Newton Algorithm

作者: Li Hai-dong Liu Jia-lu Wu-xin Zhao-bin Xu hong-hua

作者单位: Electric Engineering Institute of CAS, Beijing 100080,China Department of Space Equipment,the Academy of Equipment Command &Technology,Beijing 101416,China

会议类型: 国际会议

会议名称: 第七届国际测试技术研讨会

会议地点: 北京

会议语种:英文

在线出版日期: 2007-08-05（万方平台首次上网日期，不代表论文的发表时间）