Maximum daily load forecasting based on support vector regression considering accumulated temperature effect

Maximum daily load forecasting is of great significance in power system dispatching. First, electric load characteristics are analysed in this paper. Second, maximum load and weather factors are selected as the input of the maximum incremental load forecasting regression model, and the mapping relationship between input and output is established by least squares support vector machine (LS-SVM). Then, the modified date type normalization of rest days is proposed according to load change regulation in summer. Moreover, the effect of accumulated temperature is considered to reduce the model prediction error. Finally, numerical tests demonstrated the efficiency of the proposed model.