基于减法聚类及自适应模糊神经网络的短期电价预测

提出了基于Takagi-Sugeno模型的自适应模糊神经网络的短期电价预测方法。首先采用减法聚类方法确定自适应模糊神经网络的结构，然后利用混合学习算法训练该网络的前件参数和结论参数，最后将影响未来日电价的相关因素输入到训练好的自适应模糊神经网络中进行电价预测。以美国加州电力市场公布的1999年负荷与电价数据进行模型训练和预测，结果表明采用该方法所建立的预测模型具有较高的预测精度。

Short-Term Electricity Price Forecasting Based on Subtractive Clustering and Adaptive Neuro-Fuzzy Inference System

A short-term electricity price forecasting method based on Takagi-Sugeno model and adaptive neuro-fuzzy inference system (ANFIS) is proposed. Firstly, the structure of ANFIS is decided by subtractive clustering; then the premise parameters and consequent parameters of ANFIS are trained by hybrid learning algorithm; finally, related factors that impact futural daily electricity price are input into the trained ANFIS to forecast electricity price. By use of the published load and electricity price data of California Electricity Market in 1999, the model training and price forecasting are carried out. Forecasting results of day-ahead Market Clearing Prices (MCPs) show that the forecasting model established by the proposed method is available.