Optimization of GMAW Process Variables to Minimize Thermal Distortion

The optimization variables include process variables such as the number of welding segments, sequence, speed, and direction, and heat input variables such as voltage and current. Parallel computing was used to reduce the time required by the Genetic Algorithm to converge to an optimal solution. The use of parallel computing enables the analysis of multiple simulation models under different combinations of optimization variables at the same time for each generation. The framework was verified with two case studies: a stiffener plate and a u-channel welding. The results show that the proposed framework can successfully optimize an arc welding process to identify the values of a set of variables that minimize the effect of thermal distortion within reasonable computation time for simple manufacturing scenarios. The framework can be used for solving larger and more complex models, however, an alternative approach for FEA simulation would be required to reduce computational time.