Application Study On Short-term Load Forecasting Based On Support Vector Regression

In the economic conditions of electricity market, load forecasting face new challenges, its impact and role of the power system more and more obvious, good load forecasting is the basis for the safe operation of power system. Short-term load forecasting has become increasingly important since the rise of the competitive energy markets and has become one of the major areas of research in recent years. Power system load forecasting refers to itself form the changes in load and meteorological factors such as laws of the economy, through analysis of historical data and research to explore the link load and development of the internal changes of the future electricity demand pre-estimates and projections. Load forecasting is usually made by constructing models on relative information, such as climate and previous load demand data. Today more and more papers applied support vector machines in short-term load forecasting and get good results. This paper provides a new regression algorithm to predict the electricity load to load, which we mentioned algorithm based on support vector machine regression, we applied this method to load forecasting work, the better the supply areas to determine for consumption on an annual electricity load and the maximum power load to provide a new approach and ideas. This paper did study on Short-term load forecasting, Support Vector Machines (SVM), Sequential Minimal Optimization (SMO) theory and did experiments on the load data by using linear regression, SVM and SMO algorithms. The results show that SMO algorithm has better adaptability and excellent forecasting accuracy.