Time-Variant Slide Fuzzy Time-series Method for Short-Term Load Forecasting
Power load forecasting is important in energy man agement with great influence on generation scheduling, operation and controlling electric power systems. Economic and precise short-term load forecasting is needed in power system planning and distribution and it can result in costing saving and better operation conditions. In order to reduce the load forecasting error, the concept of fuzzy time series is introduced in the short term load forecasting. The proposed forecasting method adapts an analysis slide window of fuzzy time series to train the trend predictor in the training phase, and uses these trend predictor to generate forecasting values in the forecasting phase. By using the data from the National Electric Power Company in Jordan (used in 10), the numerical examples are employed to illustrate the proposed method, as well as to compare the training accuracy of the proposed method with the fuzzy inference model. The results show that the maximum the mean absolute percentage(MAPE) in proposed method is less than 1％, which produces more accurate training results as compared to the fuzzy inference model. The MAPE in forecasting phase in our proposed model is less than 10％.
load forecasting fuzzy time-series fuzzy sets
Xiaojuan Liu Enjian Bai Jianan Fang Lunhan Luo
College of Information Science & Technology Donghua University Shanghai, China Department of Mathema College of Information Science & Technology Donghua University Shanghai, China
国际会议
厦门
英文
65-68
2010-10-29（万方平台首次上网日期，不代表论文的发表时间）