Day-ahead campus load interval forecast based on similar day and kernel function estimation

Presently, most of short-term load forecasting are in form of point forecast. It cannot give the fluctuation range of predicted load. Because of complex characteristics of power system, the larger error of point forecast is resulted by uncertain factors. This paper presents a day-ahead interval forecast based on similar day and kernel density estimation for campus load. Similar day unit decomposes point load forecast into the predictions for daily average load and pre-unit curve. Then, these predictive values are synthesized into load interval by kernel density estimation unit. The reason for selecting kernel density estimation is due to its free-form and independence of the sample distribution, which is more fit for the randomness and volatility of campus load. The simulation results collected from North Carolina State University demonstrate the universality and strong robustness of the proposed method.