Daily Power Demand Forecast Models of the Differential Polynomial Neural Network

Abstract

Short-term electric energy estimations of a future demand are needful for the planning of generating electricity of regional grid systems and operating power systems. In order to guarantee a regular supply, it is necessary to keep a reserve. However an over-estimating of a future load results in an unused spinning reserve. Under-estimating a future load is equally detrimental because buying at the last minute from other suppliers is obviously too expensive. Cooperation on the electricity grid requires from all providers to foresee the demands within a sufficient accuracy. Differential polynomial neural network is a new neural network type, which forms and solves an unknown general partial differential equation of an approximation of a searched function, described by data observations. It generates convergent sum series of fractional polynomial derivative terms. This operating principle differs by far from other common neural network techniques. In the case of a prediction of only 1-parametric function, described by real data time-series, an ordinary differential equation is constructed and substituted with partial derivatives.