Model of Forecasting System Marginal Price for Power Market in Southwestern China

System marginal price (SMP) is the uniform price that reflects the short-term relationship between demands and supplies in power markets. To forecast SMP accurately, a computational model based on grey theory was proposed. The model can process irregular original data by accumulated generating operation (AGO) and get regular results. According to the regular results, a computational real-time forecasting system can be established. Test and simulation were completed by MATLAB according to the specific data of SMP from California power market. Considering the possible difference between weekend and workdays, two types of data were singled out randomly to calculate. Results show that the model can provide a great deal of flexibility for applications. The forecasting system has been applied in power market in southwestern China and it has been proved that the reliability, accuracy and real-time performance can be used as needed.