Machine learning optimization of a zero energy building (ZEB) by waste heat recovery in a co-generation system

Abstract

Waste heat recovery is known as one of the newest methods to increase the energy and exergy efficiencies of thermal systems. The concept has been also developed in co-generation systems where productions of electricity, cooling and heating are considered, simultaneously. This study investigates tri-generation system with a co-generation ammonia-water cycle (AWC) and an organic Rankine cycle (ORC) to supply cooling, heating, and power, simultaneously. As novelty, six different working fluids of Ammonia, R124, R134a, R152a, R245fa and R290 were investigated, as six scenarios for ORC. As result, the R124 was found as the best one and the next investigations were followed just with this fluid. Minimizing total cost and maximizing exergy efficiency at the same time are the main aim of this paper. Design Expert software was used to find out about this optimum point. Four cities in Australia were selected and the effect of annual ambient temperatures of them were studied on the outputs of the system, using TRNSYS software. Also, the optimum values of the effective parameters, including the isentropic efficiencies of pumps, compressor, and turbines, heat source temperature, ammonia concentration, utmost pressures of ORC and AWC, were obtained and reported by means of TOPSIS. As results, the monthly generations in electricity, heating and cooling were estimated. It was revealed that the lower ambient temperature, the higher electricity generation, according to the thermodynamic cycle of the proposed co-generation system. Finally, the results showed that application of the suggested co-generation system can cover the electricity demand of 294 apartments, heating for 208 apartments, and cooling load for 18,999 apartments.