A novel approach of tri-objective optimization for a building energy system with thermal energy storage to determine the optimum size of energy suppliers

摘要

We introduced a procedure to optimize a building energy supplier system's size to provide electricity, cooling, and heating loads. This optimization includes determining the best size of system components, namely, a gas turbine, a double-effect absorption chiller, PVTs, flat plate solar collectors, and a thermal energy storage to attain the maximum exergy efficiency, minimum total cost rate, and minimum CO2 index. To reduce the runtime of the the original model, an artificial neural network (ANN) is implemented by 1000 samples to train a black-box model. The black box model is used as the fitness function of the genetic algorithm for tri-objective optimization considering the exergy efficiency, total cost rate, and CO2 index as objectives. Innovation of this research is the combination of optimization and ANN, which results in a fast and accurate optimization procedure. The 3D Pareto Frontier of optimum solutions and scatters of distribution is presented. The result of modeling shows that the total amount of cooling, heating, and electrical loads during a year are 524.2 MWh, 253.7 MWh and, 621.1 MWh, respectively. Tri-objective optimization shows the best point of Pareto Frontier has the exergy efficiency of 64.23%, total cost rate of 5.78 $/h, and CO2 index of 425.15 g/kWh.