Artificial intelligence‐based tri‐objective optimization of different demand load patterns on the optimal sizing of a smart educational buildings

Abstract

In this research, an integrated energy system for providing triple loads of an educational building has been modeled, analyzed, and optimized using artificial intelligence The use of different load supply patterns for the integrated system was optimized and evaluated from the point of view of exergy, economics, and environment. SketchUp, OpenStudio, and EnergyPlus are used to obtain the building loads, and the integrated system is dynamically modeled by the MATLAB software. The integrated system has been optimized based on different patterns, including monthly, seasonal, and two-state constant loads compared with hourly dynamic loads using an artificial intelligence genetic algorithm. The results show that the dynamic load pattern gives the highest exergy efficiency and the lowest total cost rate. The exergy efficiency and total cost rate in this scenario are 58.78% and 5.022 $/h, respectively. Also, the CO2 emission index has reached 363.1 g/kWh. Due to the subsidized price of grid electricity, there is a tendency to use low-capacity gas turbines and buy electricity from the grid for the dynamic model. The gas turbine capacity is 35.56 kW, and the system has purchased 242.4 MWh from the grid. Due to high heating loads, the capacity of the gas turbine significantly increases, and no electricity is purchased from the grid in the fixed-load pattern. The highest capacity of the gas turbine is 161.6 kW, which is obtained by a seasonal fixed-load pattern. Also, 994.9 MWh of electricity is sold to the grid in this mode.