Experimental Investigations on the Comparison of Multi-Objective Design for High Thermal Energy Applications: An Insight into Response Surface Methodology

This paper focuses on optimizing the thermal properties of aluminum, copper, and iron. Response Surface Methodology was employed to determine the optimal values for thermal conductivity and thermal expansion, minimizing experimental repetitions, and reducing costs by using pre-determined input parameters. These parameters include the volume percentages of aluminum, copper, and iron. The optimized output values for thermal conductivity and thermal expansion are measured using the two-probe method and buoyancy method, respectively, and inferred from ANOVA. The Response Surface Methodology is utilized for optimization, and various approaches analyzed include Central-Composite-Design, Box-Behnken-Design, Full-Factorial-Design, and Optimal-I method. For Central-Composite-Design, deviations are 1.07 for thermal conductivity and 0.98 for thermal expansion, with R2 values of 0.9898 and 0.9355 and F values of 96.910 and 29. For Full-Factorial-Design, deviations are 39.42 for thermal conductivity and 26.83 for thermal expansion, with R2 values of 0.86 and 0.89 and F values of 20.2936 and 17.3521. For BBD, deviations are 13.49 for thermal conductivity and 22.47 for thermal expansion, with R2 values of 0.9756 and 0.9768, and F values of 66.5 and 70.1. For Optimal-I, deviations are 12.67 for thermal conductivity and 22.56 for thermal expansion, with R2 values of 0.9937 and 0.9793, and F values of 87.38 and 63.02.