Machine learning-based metaheuristic optimization of an integrated biomass gasification cycle for fuel and cooling production

Abstract

Nowadays, Combined cycles have attracted much attention for improving the efficiency of the thermodynamic cycles. In these systems, the main idea is to optimally use the waste energy from the power cycle, to run the cooling, heating, and storage cycles of hydrogen and biofuel-based ammonia. In this research study, an innovative integrated biomass gasification energy system for ammonia production is used alongside hydrogen storage. In addition, gasification is used to produce the required fuel for power generation. The gasification intake air is preheated with solar panels and then enters the gasification unit. The cooling water passing through this stage is heated up to the temperature of the electrolysis and then gets decomposed. The produced hydrogen fuel is used and stored for ammonia fuel production. The power cycle consists of a gas turbine system alongside a Rankin cycle, and then the exhaust gas is guided towards the cooling cycle of a double-effect absorption chiller. The results show that this system can generate power up to 7 megawatts with an exergy efficiency of 66%. Furthermore, ammonia production and hydrogen storage rates are 0.38 and 0.101 kg per second, respectively. Genetic algorithm optimization along with machine learning methods is used to achieve the optimization points of this cycle. At the optimized point, this cycle can produce 0.5347 kg per second of ammonia with an exergy efficiency of nearly 64%.