基于减法聚类及自适应模糊神经网络的短期电价预测.pdfVIP

31 卷 19 期 电 网 技 术 Vol. 31No. 19 2007 年10 月 Power System Technology Oct. 2007 章编号：1000-3673 （2007 ）19-0069-05 中图分类号：F123.9 献标识码：A 学科代码：520 2060 基于减法聚类及自适应模糊神经网络的 短期电价预测 兴华，周 晖 （北京交通大学 电气工程学院，北京市 海淀区 100044 ） Short-Term Electricity Price Forecasting Based on Subtractive Clustering and Adaptive Neuro-Fuzzy Inference System WU Xing-hua ，ZHOU Hui （School of Electrical Engineering ，Beijing Jiaotong University，Haidian District，Beijing 100044，China ） ABSTRACT: A short-term electricity price forecasting method 区域电力市场为重点的方向。电价是电力市场的 based on Takagi-Sugeno model and adaptive neuro-fuzzy 核心和杠杆，是电力市场中传递供求变化的最敏 inference system (ANFIS) is proposed. Firstly, the structure of 感的信号，电价预测越来越受到电力市场中各利 ANFIS is decided by subtractive clustering; then the premise 益主体的关注。在我国电力市场中，特别是发电 parameters and consequent parameters of ANFIS are trained by 商可以根据预测电价制定最优投标策略，以获取 hybrid learning algorithm; finally, related factors that impact 最大利润。 futural daily electricity price are input into the trained ANFIS 电价预测可以分为短期电价预测和中长期电 to forecast electricity price. By use of the published load and electricity price data of California Electricity Market in 1999, 价预测。短期电价预测指周电价预测、日前电价预 the model training and price forecasting are carried out. 测和小时前电价预测，其中日前和小时前电价预测 Forecasting results of day-ahead Market Clearing Prices 是研究的重点。中长期电价预测指月度和年度电价 (MCPs) show that the forecasting model established by the 预测，由于中长期电价预测包含太多不确定的影响 proposed method is available. [1]