Singular value decomposition expansion for electrical demand analysis

To analyse electricity-demand data, appropriate mathematical models and algorithms have been developed. Some theoretical properties of the singular value decomposition (SVD) and SVD expansion have been found very useful during these developments. These properties, which have generic implications in data mining and numerical analysis, are presented in this paper. Following a discussion of the theoretical development, this paper reports applications of the SVD expansion in demand analysis and forecasting with three illustrative practical algorithms, that have been developed by the National Grid Company in recent years.