Performance prediction and parameters optimization of an opposed-piston free piston engine generator using response surface methodology

The opposed-piston free-piston engine generator (FPEG) has high energy conversion efficiency, but its thermodynamic performance is significantly affected by various system parameters owing to the absence of mechanical constraints. This paper introduces a novel approach to predicting the performance of opposed-piston FPEG by leveraging a comprehensive thermodynamic model with a stable operating process and a quadratic regression model based on response surface methods incorporating both design and operating parameters. The established models were verified using variance analysis, enabling the evaluation of the statistical significance of various independent parameters. Furthermore, key factors were identified, and their coupling effects on crucial indicators like indicated work, indicated thermal efficiency, and power output were thoroughly analyzed. A Multi-Objective Evolutionary Algorithm based on Decomposition was proposed to maximize the thermodynamic performance, and the optimal parameter set was obtained from the Pareto solution by means of three decision-making methods. The findings indicate that a higher power cylinder scavenging pressure and rebound cylinder base pressure, lower ignition position and rebound cylinder diameter, along with an appropriate optimal value of power cylinder diameter, effectively enhanced the overall system performance of the opposed-piston FPEG. The optimized maximum indicated work, indicated thermal efficiency, and power output were found to be 395.7 J, 64.74%, and 5.10 kW, respectively. In conclusion, the integration of response surface methodology, multi-objective optimal algorithms, and decision-making methods was demonstrated to be an effective strategy for designing the comprehensive parameters of the opposed-piston FPEG system, leading to a significant enhancement in overall system performance.