چكيده به لاتين
Moving towards a sustainable society through solar energy is one of the most pivotal topics in engineering science. Due to its fluctuating nature, solar energy is intermittent and unavailable during the night. The current study proposes and evaluates an innovative solar-driven multigeneration system with a thermal energy storage unit to provide useful products such as electricity, hydrogen, fresh water, and cooling load throughout the day. In this regard, three operation modes have been introduced for the system to use different intensities of solar radiation during the day. To analyze the performance of the system, the impact of critical parameters has been evaluated through a parametric study from energy and exergy points of view. The energy efficiency of the proposed system for solar, solar-storage, and storage modes is calculated as 45.31, 25.16, and 34.38%, respectively. Moreover, the exergy efficiency of these modes is 10.85, 4.1, and 6.628%, correspondingly. The results show that the use of a thermoelectric generator to recover waste heat increases power generation in the above-mentioned modes by 110.3, 75.48, and 79.39 kW, respectively. In addition, the feasibility of deploying this system in Shiraz city was investigated by collecting weather data. The results revealed that the annual power generation capacity of the system is 2258.3 MWh. A parametric study showed that some input parameters had a contradictory effect on the considered objective functions, including exergy efficiency, total cost rate, and freshwater production rate, so a multi-objective optimization process was needed to achieve optimal system performance. Due to the time-consuming optimization process in EES software, a data-oriented optimization method was adopted through the training of neural networks. In the following, these trained networks are introduced as objective functions in the MATLAB genetic algorithm toolbox.
