Gray Trend Relevant Clustering Based Neural Network Model for Short-Term Load Forecasting
Short-term load forecasting (STLF) is important for real operation of power system. In this paper we demonstrate how a neural network based on gray trend relevant clustering method is used in power load forecasting. This method can effectively reduce the training time and improve convergent speed. According to the features of power load and considering the combined influence of temperature, weather type, humidity, and day type, the historical data can be clustered into several groups. The representative data samples in clustered load data were selected as the training set of the Elman neural network which is a kind of globally feed forward locally recurrent network model with distinguished dynamical characteristics. Using the presented model, the better forecasting accuracy and learning potency can be achieved. Based on daily load data in Hebei province, the model was testified and improved accurate forecasting results were obtained.
Yujun He Dongxing Duan Youchan Zhu
Department of Electronic and Communication Engineering North China Electric Power University Baoding Center of Information and Network Management North China Electric Power University Baoding, Hebei, 0
国际会议
南宁
英文
2007-07-20（万方平台首次上网日期，不代表论文的发表时间）