Optimization design of moving-bed thermochemical heat storage reactor based on neural network fitting and multi-objective genetic algorithm

Hydrated salt thermochemical heat storage technology plays an important role in solar building heating systems. To overcome the unstable energy input and output caused by traditional fixed-bed reactors, this paper proposes a continuous thermochemical heat storage and supply integrated system based on a moving-bed reactor using magnesium chloride hexahydrate. A numerical model for the coupled heat and mass transfer and thermochemical heat storage process of moving-bed reactor is developed. The effects of fluid and solid inlet velocities, and reaction-zone width on reaction conversion, heat exchange efficiency, and energy storage rate are analyzed. It is found that these parameters have significant and complicated effects on the overall performance. To obtain the optimal overall performance, parameter optimization is further performed based on neural-network fitting and a multi-objective genetic algorithm. The results show that when the fluid velocity is 1.367 m s⁻¹, solid velocity is 0.000098 m s⁻¹, and reaction-zone width is 0.065 m, the reactor achieves the optimal overall performance with reaction conversion of 0.875, heat exchange efficiency of 0.682, and heat storage rate of 192.37 W. The heat storage rate is increased by 30.5 % while the reaction conversion and heat exchange efficiency remain nearly unchanged compared with the base case.