混沌时间序列预测在短期电力负荷预测中的应用  

Application of Chaotic Time Series Forecast in Short-term Power Load Forecast

作　　者：张作鹏[1] 刘崇新[1] 逯俊杰[1] 韩俊玉[2] 

机构地区：[1]西安交通大学,陕西西安710049 [2]山西省电力公司,山西太原030001

出　　处：《陕西电力》2007年第1期24-26,共3页Shanxi Electric Power

摘　　要：混沌吸引子具有空间的遍历性、总体的稳定性、吸引性和内部分形性,使得混沌预测成为可能。采用非线性系统理论对电力系统的历史数据进行分析,并基于改进的混沌吸引子模型对山西电网实际负荷数据进行短期预测,预测结果相对误差小、精度高,为电力系统短期负荷预测提供了一种新的预测方法。Chaotic attractor has the characters of spreading all over the space, total stability, attractability and fraction, and it makes the chaotic forecast possible. In this paper, the recording data was analyzed by the method of nonlinear system, and factual power data of Shanxi power network was forecasted based on the improved chaotic attractor model. The results are proved with a small relative error and high precision. It supplies a new method to forecast the short-term load of the electric system.

关 键 词：混沌 负荷预测 混沌时间序列 最大LYAPUNOV指数 

分 类 号：TM714[电气工程—电力系统及自动化]