The heat transfer performance of trapezoidal supercritical carbon dioxide printed circuit heat exchanger plates (SCO₂-PCHE) is determined by both the heat transfer coefficient and pressure drop. Traditional multi-objective structural parameter optimization of heat exchangers often encounters issues such as high computational cost and difficulty in convergence.

This study aims to effectively improve the performance of the heat exchanger by proposing a multi-objective structural optimization design method for trapezoidal SCO₂-PCHE.

First, an orthogonal experiment table was created based on the geometric structure parameters of the trapezoidal SCO₂-PCHE, and Fluent software was employed to calculate the heat transfer coefficient and pressure drop. Then, the affine power law model obtained by fitting the experimental samples was used as the surrogate model of heat transfer coefficient and pressure drop, and the Strength Pareto Evolutionary Algorithm 2 (SPEA2) was applied to multi-objective structural optimization of the trapezoidal SCO₂-PCHE, resulting in an optimized Pareto front. Finally, the performances of SPEA2, OMOPSO and NSGA-II optimization algorithms were compared and verified.

Comparison results show that the JF factor obtained by SPEA2 algorithm is 21% higher than that before optimization, better than that of OMOPSO and NSGA-II algorithms, and the optimal heat transfer coefficient and pressure drop of the heat exchanger plate obtained by the optimization method based on affine power law surrogate model and SPEA2 algorithm are 16 575.53 W·(m²·K)^-1 and 173.26 kPa, respectively. The heat transfer coefficient is increased by 9.37% and the pressure drop is reduced by 27% compared with that before optimization.

The optimization design method proposed in this paper achieves best optimization performance, providing a solution with better heat transfer performance.