Abstract
In this paper, a novel method is proposed for short-term load forecasting. It is one of important tasks in power system operation and planning. The load behavior is so complicated that it is hard to predict the load. The deregulated power market is faced with a new aspect that the degree of uncertainty increases. Thus, power system operators are concerned with the significant level of load forecasting. Namely, probabilistic load forecasting is required to smooth power system operation and planning. In this paper, an IVM (Informative Vector Machine) based method is proposed for short-term load forecasting. IVM is a one of kernel machine techniques that are derived from SVM (Support Vector Machine). The Gaussian process (GP) satisfies the requirements that the prediction results are expressed in distribution rather than point. However, it is inclined to be overfitting for noise due to the basis function with N2 elements for N data. To overcome the problem, this paper makes use of IVM that selects necessary data for the model approximation with posteriori distribution of entropy. That has a useful function to suppress the overfitting. The proposed method is tested for real data of short-term load forecasting.
