Prediction of heat consumption in a dwelling on the basis of archival data from residential heat meters

Abstract

The purchase costs of heat, gas fuel or electricity for the dwelling heating are divided into individual residential and commercial premises. For example, heat meters are used to measure the heat consumption for heating individual apartments. However, this device may be damaged and the data cannot be read. The article describes an analysis of the possibility of using historical data to estimate the share in the heating costs of this premise in which a heat meter failure has been found. For this purpose, it was proposed to distinguish a group of homogeneous flats from a flat with a damaged heat meter on the basis of data from previous billing periods. Instead of estimating on the basis of average unit indicators calculated on the basis of records from all apartments, it is possible to determine the missing consumption reading based on the values calculated for this group of homogeneous apartments. The article describes the practical use of this method, on the example of a heat supply system for a group of residential buildings in one of Wrocław’s housing estates.