A SIMPLIFIED METHOD FOR PREDICTING LONG-TERM AVERAGE PERFORMANCE OF SOLAR-POWERED AIR-CONDITIONING PLANTS

Abstract

The present thesis develops a simplified method for predicting the long-term average daily performance of a solar-powered air-conditioning plant, which consists of an array of solar collectors, a storage tank and an absorption chiller. Using this method, the long­ term average daily performance of the plant can be predicted by determining its performance on an average solar cooling day. In the long-term, the solar-powered air-conditioning plant can only produce solar cooling on days when the daily total solar radiation is higher than the long-term average daily radiation threshold. The amount of cooling that the solar-powered air-conditioning plant can provide is found to be linearly proportional to the amount of the daily solar radiation above the long-term average daily radiation threshold. The coefficient of this linear function is governed by a non-dimensional parameter r and the daily coefficient of performance (COP) of the absorption chiller. The new non-dimensional parameter r introduced in this study is a means to combine the daily utilizability, the daily total radiation, and the long-term average daily radiation threshold into one variable. The value of r is found to be independent of the magnitude of the daily total solar radiation incident on the collector surface. The independence of the r value to the daily total solar radiation allows the prediction of the long-term average daily solar cooling to be based on an average solar cooling day. The method for determining this average solar cooling day is illustrated by an example using the , cumulative frequency distribution of yearly solar radiation in Singapore. Simulation programs are developed for determining the long-term average daily radiation threshold, the r value, and the daily COP of chiller. The simulation model is developed based on the experimental solar-powered air-conditioning plant at the National University of Singapore. Experiments are conducted on a flat-plate heat-pipe collector array and a LiBr-HiO absorption chiller to obtain empirical correlations, which are used in the simulation program to predict the plant performance. Compared to the detailed simulations, the simplified method significantly reduces the calculation efforts. It can predict the daily COP of chiller and the daily radiation on the average solar cooling day with an accuracy of 3%. But the simplified method tends to under-predict the long-term average daily solar cooling. For a plant with an array of horizontal flat-plate collector of 40 m2 a water storage tank of 1500 kg and a single­stage LiBr-H20 absorption chiller of 7 kW rated capacity, its yearly average daily solar cooling is under-predicted by 19% compared to the detailed simulation results. The under-prediction can be caused by the solar radiation model used in the calculation, which assumes that the daily solar radiation is smooth and symmetrical about solar noon.