Multi-objective machining parameter optimisation for residual stress based on quantum cat swarm

Residual stresses greatly affect parts' performances, lives, fatigue strengths, corrosion resistance, etc. Due to the lack of analytical models, machining parameter optimisation for better residual stresses is still a problem. In this paper, a multi-objective machining parameter optimisation method is proposed. Based on the support vector machine, the machining parameters' nonlinear relationships with the surface roughness and the residual stress are investigated. The cutting time consumption, surface roughness and absolute residual stress are the objectives, while the cutting speed, feed rate, axial cutting depth and the radial cutting deep are variables. The cutting power and cutting torque are constraints. The multi-objective cat swarm optimisation is designed to solve the problem, while the quantum computation is used to improve its performance. An experimental study is presented to verify the method. Some Pareto solutions are obtained with good convergence and diversity. Compared with the empirical machining parameters, the material removal rate, surface roughness and residual stress are optimised greatly. Compared with non-dominated sorting genetic algorithm II, the algorithm's precision and effectiveness are also verified.