Coating thickness and uniformity are critical factors in achieving high-quality thermal spray coatings. With the increasing demand for high efficiency and precision in modern manufacturing, robotic manipulators have been widely employed to control the motion of spray guns relative to workpiece surfaces. However, optimizing spray path parameters to achieve the desired coating thickness distribution remains challenging due to the complexity of the optimization problem and the risk of convergence to local optima. In this study, a coating thickness distribution model for a robotic thermal spraying system is employed to predict coating thickness under different spray paths. Based on this model, a hybrid optimization approach combining Latin hypercube sampling (LHS) and the Nelder-Mead simplex method is proposed. LHS is first employed to generate well-distributed initial solutions across the design space, followed by the Nelder-Mead simplex method to perform local optimization. A desirability-based objective function is adopted to evaluate the coating thickness distribution and guide the optimization process. By combining global exploration with local refinement, the proposed approach aims to identify a near-global optimal solution for coating thickness distribution, thereby improving the accuracy and uniformity of robotic thermal spraying.

Thermal spray,Spray path,Coating thickness,Path optimization,Nelder Mead method