Optimization data analysis of injection molding based on multi-objective improvement

Abstract With the increasing emphasis on high quality and efficiency in construction machinery, parameter optimization in injection molding has become a top priority in design. Due to the numerous and difficult-to-control parameters in injection molding design, this study proposes an optimized data hybrid model for injection molding based on an improved Multi-objective Coati Optimization Algorithm. When constructing the optimization data model, the Random Forest Enhancement Algorithm is also used for the optimization function. Experimental results show that this hybrid algorithm significantly improves convergence speed compared to the Particle Swarm Optimization, Sparrow Search Algorithm, and Adaptive Covariance Scaling Monte Carlo Optimization. In the f 1 function test, the iteration counts for these algorithms are 200, 350, 400, and 300, respectively, and the convergence curve gradually flattens. Furthermore, empirical analysis of the constructed injection molding optimization data hybrid model reveals that the mean square deviation of the mold cavity temperature is 8.10 × 10 −4 (°C) 2 , and the pressure-holding time mean square deviation is 3.847 × 10 −3 s 2 , outperforming other comparison models. These results indicate that the improved Multi-objective Coati Optimization Algorithm-based injection molding optimization data hybrid model can effectively identify parameters while minimizing errors. This is crucial for meeting the high-precision, low-defect requirements of lightweight design and helps reduce material waste caused by defects.