Configuration Optimization of a Plate Fin Precooler Based on Multi-Objective Grey Wolf Optimizer

The method of effectiveness-number of heat transfer units (ε-NTU) is adopted to establish a design indicator prediction model for plate fin precooler (PFP), and experimental verification is conducted. The average error between the experimental heat transfer capacity and the calculated heat transfer capacity is 4.65%, and the predicted mass matches the mass computed via the commercial software SolidWorks 2020. This outcome confirms the model’s reliability. An investigation is conducted into the influences of parametric factors, including hot stream flow length, cold stream flow length, hot side number of layers, and hot side fin pitch on the heat transfer capacity and mass of the PFP. To realize the maximization of heat transfer capacity and the minimization of mass, optimization is performed on the four sensitive configuration parameters by leveraging the multi-objective grey wolf optimizer (MOGWO). This optimization can significantly reduce the mass while ensuring the stability of the heat transfer capacity. Three classes of optimal configurations were derived from Pareto optimal points. Compared to the original structure, the selected schemes exhibit an average 2.95% rise in heat transfer capacity and a 10.7% reduction in mass. These findings show that the optimization method proposed in this study is effective and provides valuable guidance for precooler design.