Price-load elasticity analysis by hybrid algorithm

Due to the short history of electricity market, price- load elasticity is along one of focused and unsolved problems in the researches of electricity market. This paper describes a novel model for price-load elasticity analysis, where support vector machine (abbr. SVM) is applied for pattern identification of electricity price. First the correlation analysis of loads and prices are performed to choose the proper influence factors. Second, to overcome the limitations of traditional classification, intelligence classification of electricity price is accomplished by SVM that is a strong nonlinear classification by solving a quadratic programming problem. Based on that model, numerical analysis is performed to measure the change ratio of price and its corresponding load in consistency with the classification. Then price-load elasticity models are deduced, whose elasticity coefficients are computed by using the obtained regression model. The method embodies clear concepts, so it can be easily understood theoretically and realized computationally. Case studies are used to test the proposed model.