Multiobjective optimization of a reciprocating magnetic refrigerator using a genetic algorithm.

This paper is intended to analyze the overall energetic and exergetic performance of an active magnetic regenerative refrigerator (AMRR). A 1.5-D numerical model has been developed to simulate a system composed of a parallel-plate regenerator, a magnetic source, a pump, heat exchangers, and control valves. The behaviour in steady state is described and the basic thermodynamic equations are depicted. The effects of several parameters like the magnetic field, the geometry, the mass, and the blowing time are studied. Various types of liquid refrigerants are also tested thanks to CoolProp library. The main goals are to maximize the COP, the exergetic efficiency, and the cooling power while respecting a precise set of constraints. A multiobjective optimization based on a genetic algorithm (GA) is used to enhance the performance regarding the Pareto efficiency. Finally, the tuning of the algorithm is discussed in order to achieve optimal results in fair computational time.