Use of meteorological/constructional statistics to estimate building heat needs

Abstract Building heat needs over a month or a year are often estimated on the basis of the relevant degree day value, in which some crude account is taken of solar and other casual gains in the choice of the base temperature. A more realistic assessment of building heat needs including the effects of passive gain (through a window or glazed wall) can be made using the ‘empirical’ approach of Davies. In this article a statistical approach is developed in which use is made of degree day type data, and measures of days which are ‘disadvantageous’ and ‘advantageous’ from the point of view of saving back-up heating, together with the first four moments of the distribution of advantageous days (7 statistics in all). From five of these statistics a curve can be fitted to the distribution. After suitable integration of the fitted curves another estimate of building heat needs is obtained (the ‘curve fitting’ value). The heat needs as obtained empirically and by curve fitting agree closely. The 7 statistics are easily handled by a mini computer. They thus provide a convenient and accurate extension to degree day methods when local meteorological information has been processed in conjunction with passive solar collector characteristics. Window area, or glazed wall resistance, and assumptions about ventilation rate are only needed in the final stage of computing heat needs.