A tool for predicting hourly air temperatures and sensibles energy loads in buildings at sketch design stage

Abstract

This paper presents a simplified method, developed for building designers, by which dynamic air temperatures in naturally ventilated buildings and dynamic sensible heating and cooling energy loads in conditioned buildings can be predicted. The philosophy and theoretical reasoning necessary for ensuring that the method is not only accessible to designers during the early design stage but also applicable to a wide range of buildings are discussed. Features of the proposed method, some of which are often omitted in simplified methods, include the following: the modelling of high-mass multi-layered floors which are not insulated from the indoor air; high interior mass such as multi-layered partition walls; an unlimited amount of different multi-layered exterior surfaces with any shading coefficient, orientation and colour; large variations in and distribution of thermal mass and resistance of the different building elements; window shading devices and direct solar gains; on/off interior radiative and convective heat generation or extraction as a result of occupancy, machines, lighting, etc.; as well as natural and mechanical ventilation. To ensure an accessible design tool, the method was implemented in a computer program. It runs on IBM-compatible microcomputers and is fast, highly interactive and extremely easy to use. The success of the computer implementation is evident when considering that, for a typical building zone, the input is in the order of 15 minutes while the computer running times are approximately 15 seconds. Ultimately, the usefulness of a thermal design tool can only be ascertained by comparing its predictions with experimental results from actual buildings. This paper presents the validation studies for temperature predictions: verification of the energy predictions will be the subject of a future paper. A good comparison is observed between temperature measurements and computer predictions for a wide range of real buildings with open and closed windows. The 42 validation studies included the following building types: office blocks, shops, schools, residential buildings, townhouses, medium- and high-mass experimental buildings, low-mass well-insulated structures with ground contact such as factories, and low-mass poorly insulated structures, e.g., agricultural buildings.