Design of a Model for Heat Demand Prediction Using the Neural Network Synthesis

This paper deals with design of a model for short-term heat demand forecasting. Forecast of this heat demand course is significant for short-term planning of heat production and it is most important for technical and economic consideration. In this paper we propose the forecast model of heat demand based on the assumption that the course of heat demand can be described sufficiently well as a function of the outdoor temperature and the weather independent component (social components). Forecast of social component is realized by means of Box-Jenkins methodology. For inclusion of outdoor temperature influence in calculation of prediction of heat demand is used the heating characteristic (function that describes the temperaturedependent part of heat consumption). The principal aim is to derive an explicit expression for the heating characteristics. The Neural Network Synthesis is used for optimal finding of the expression. Key-Words: Modelling, Prediction, Box-Jenkins, Control algorithms, Neural Network Synthesis