Multi-objective optimization of aluminum alloy control arm manufacturing process parameters based on the Transformer-NSGA-II model

To address the issues of unstable forming quality in aluminum alloy control arm forgings and low die service life while minimizing forging costs, this study studies the influence of process parameters (blank temperature, die temperature, upper die velocity, and friction coefficient) on target parameters (maximum forming force, maximum die wear depth, and maximum damage value of forging) through numerical simulation. A multi-objective process parameter optimization approach based on the Transformer-NSGA-II model is proposed. First, the Box-Behnken experimental design was adopted to obtain data from different process parameter combinations via finite element analysis (FEA). Then, the transformer model with self-attention was utilized to establish the regression model between the process parameters and target parameters. Finally, the NSGA-II algorithm was applied to identify the optimal parameter combination. To validate the efficiency of the proposed method, a comparative analysis of variance was conducted against commonly used optimization methods. The top 10 solutions in the Pareto solution set of each method were selected and ranked via the entropy-weighted TOPSIS method. The results demonstrate that the proposed model has a lower variance and provides superior process parameter combinations. The FEA verified the optimization results, showing a reduction in the maximum forming force from 20,251 to 18,829 KN, a decrease of 7.02%; the maximum die wear depth decreased from 0.0905 to 0.0784 μm, a reduction of 13.37%; and the maximum damage value of forging decreased from 0.413 to 0.322, a decline of 29.47%. Furthermore, actual trial production confirmed that the forgings produced under the optimal parameter combination meet the quality requirements. This approach achieves synergistic optimization of the forming efficiency, forming quality and die life, demonstrating significant engineering value in enhancing the intelligence and productivity of aluminum alloy control arm forging processes. It also provides a practical process parameter optimization model for digital manufacturing and intelligent manufacturing.