Optimization of Microjet Location Using Surrogate Model Coupled with Particle Swarm Optimization Algorithm

This study aimed to present the design methodology of microjet heat sinks with unequal jet spacing, using a machine learning technique which alleviates hot spots in non-uniform flux conditions. Latin hypercube sampling was used obtain 30 sample points on three-dimensional Computational Fluid Dynamics (CFD) solutions were calculated, train model. Radial Basis Neural Network (RBNN) as surrogate model coupled Particle Swarm Optimization (PSO) optimized location jets. The RBNN provides continuous space for searching optimum values. At predicted values from model, CFD solution calculated comparison. percentage error target function 0.56%, whereas accompanied it 1.3%. algorithm has variable inputs at user discretion, including gaussian spread, number search particles, and iterations. sensitivity each obtained. Analysis Variance (ANOVA) performed investigate effect input thermal resistance. ANOVA results revealed that spread is dominant affecting