Experimental validation of genetic programming for heat exchanger circuitry optimization

Although evolutionary circuitry optimization has been recognized as a cost-effective method in recent research, it remains to be experimentally verified. Accordingly, this study reports on experiments conducted to validate the effectiveness of genetic programming for heat exchanger circuitry optimization. The numerical optimization results of the evaporator are identified for R32 in a capacity range between 3 and 4 kW and for the given airside features, thereby suggesting optimal branching and patterning characteristics obtained by implementing genetic operators through genetic programming for 3.0, 3.5, and 4.0 kW design conditions. Airside design, including fin package, remains unchanged across all investigated capacities. The manufacturability of the optimized solutions is enhanced by introducing additional constraints. Consequently, baseline and optimized heat exchangers are manufactured and tested using dedicated equipment. The test results validate the numerical model, with 66.7 % of the temperature deviation within ±1.0 K and pressure deviations within ±5.0 % of the measured values, as well as verify the benefits achievable through optimized branching and patterning. The optimized configurations at 3- and 4-kW capacity achieve inlet saturation temperatures that are 2.7 K and 4.1 K higher, respectively. The circuitry optimized at 3 kW results in an 18.0 kPa increase in pressure drop, whereas the optimization conducted at 4 kW produces an 83.5 kPa decrease. These effects translate into corresponding improvements of 18.9 % and 12.8 % in the measured coefficient of performance. The results support the development of more accurate simulations and pave the way for the practical implementation of genetic programming for actual product development.