Direct waste heat recovery from a solid oxide fuel cell through Kalina cycle, two-bed adsorption chiller, thermoelectric generator, reverse osmosis, and PEM electrolyzer: 4E analysis and ANN-assisted optimization

Abstract

The present study investigates an innovative model for the utilization of redundant heat from a solid oxide fuel cell (SOFC). The proposed system comprises a Kalina cycle (KC), a two-bed adsorption chiller (AC), a thermoelectric generator (TEG), reverse osmosis (RO), and a proton exchange membrane electrolyzer (PEME). A parametric analysis is conducted to evaluate the impact of key variables on the performance of the system. Various metrics are employed, including energy, exergy, economic, and environmental (4E). The environmental analysis reveals that waste heat recovery from SOFC decreases CO2 emissions by 36.41 kg per megawatt output at the base condition. At a fuel (methane) consumption rate of 0.0743 kg/s, the system produces 2206.49 kW of power, 166.09 kW of cooling load, 404.68 m3/day of fresh water, and 18.04 kg/day of hydrogen with reported energy efficiency and exergy efficiency values of 63.91% and 51%, respectively. Besides around 60.55% of the overall exergy destruction in the entire system is attributed to the air preheater, afterburner, and water preheater of the SOFC. The system is optimized using a genetic algorithm (GA) in three separate optimization issues, aiming to minimize the total cost rate while maximizing other objective functions, such as exergy efficiency, freshwater production, and cooling load. A modern optimization model using artificial neural networks (ANNs) is employed to reduce optimization time. By considering the overall cost rate and exergy efficiency as the objective functions, the optimal point is achieved at 58.03 $/h and 59.92%, respectively. The optimum values of 120.99 $/h and 1346.30 m3/day for the objective functions of total cost rate and freshwater production, respectively, are obtained in the second issue. In the third optimization problem, the evaluated values at the optimal point are 272.6 kW and 70.69 $/h for cooling load and total cost rate, respectively. Sensitivity analysis indicates that the changes in current density significantly affect the exergy efficiency, cost rate, net power, and cooling load, while variations in the fuel utilization factor coefficient considerably influence freshwater production.