An application of a genetic algorithm based on Particle Swarm Optimization to a multiple responses problem arising in the Tube Hydroforming Process

Tube Hydroforming (THF) is a relatively new manufacturing process mainly used in the automotive industry from the past decades, offering potential alternatives to lightweight materials. THF can significantly govern saving energy, offering several advantages over stamping and welding processes. Automotive sectors require complex-shaped extruded hollow tubes due to free-forming and calibration. THF requires less thinning to provide improved structural strength and stiffness. Lightweight vehicle units requiring less maintenance if THF are implemented with less formable Inconel 600 tubes. The impact of Hydroforming parameters (HFP) like P (internal pressure), L (axial movement), and F (tube length) on the tube output's quality like Bulging and Thinning ratios (BR&TR) are studied. RSM (Response surface methodology) was employed to develop empirical relations between HFP and experimental outputs. Particle Swarm Optimization (PSO) algorithm is applied to obtain a large amount of optimized data set for HFPs combination while simultaneously enhancing BR and reducing TR. Genetic algorithms improve the Pareto front optimized solutions of PSO's accuracy by prolonging convergence. Increasing P and L parameters values will significantly affect the output's quality. Proposed methods have performed outstanding (they avoided tube's local necking and failures like wrinkle and bursting) and the results were not possible with other techniques.

Automated summary

Tube Hydroforming (THF) is a new manufacturing process with potential alternatives to lightweight materials in the automotive industry. It offers energy savings and advantages over traditional methods, particularly important for the automotive sector requiring complex-shaped components. The study of Hydroforming parameters and the application of optimization algorithms have shown significant improvements in output quality.