A Global Approach to the Optimal Trajectory Based on an Improved Ant Colony Algorithm for Cold Spray
- PDF / 1,819,442 Bytes
- 7 Pages / 593.972 x 792 pts Page_size
- 15 Downloads / 197 Views
Zhenhua Cai, Tingyang Chen, Chunnian Zeng, Xueping Guo, Huijuan Lian, You Zheng, and Xiaoxu Wei (Submitted June 19, 2016; in revised form September 19, 2016) This paper is concerned with finding a global approach to obtain the shortest complete coverage trajectory on complex surfaces for cold spray applications. A slicing algorithm is employed to decompose the free-form complex surface into several small pieces of simple topological type. The problem of finding the optimal arrangement of the pieces is translated into a generalized traveling salesman problem (GTSP). Owing to its high searching capability and convergence performance, an improved ant colony algorithm is then used to solve the GTSP. Through off-line simulation, a robot trajectory is generated based on the optimized result. The approach is applied to coat real components with a complex surface by using the cold spray system with copper as the spraying material.
Keywords
cold spray, intelligent algorithm, robot trajectory, traveling salesman problem
1. Introduction The cold spray process is based on a supersonic gas pistol that utilizes kinetic energy to accelerate solid particles (e.g., copper, aluminum, titanium) onto a prepared surface. This process causes the plastic deformation of the particles upon impact to create dense coatings and to provide specific properties for industrial applications (Ref 1). The cold spray process is also used to apply functional coatings on complex surfaces that have steep curvature variations. Because of the increased demand for reproducible processes, an optimal robot trajectory may enable the cold spray process to reduce both production cycles and material consumption. Therefore, an optimized robot trajectory should not only ensure a uniform full-coverage film but also guarantee the shortest trajectory length when covering the complex coating surface. Obtaining robot motion that results in a uniformly thick coating on a component with complex geometry by trial and error is time-consuming, expensive, and often very difficult (Ref 2). Off-line programming (OLP) enables us to reduce the robot downtime because the robot program can be preZhenhua Cai, Tingyang Chen, Chunnian Zeng, and Xiaoxu Wei, Automation School, Wuhan University of Technology, Wuhan 430000, China; Huijuan Lian, Information School, Wuhan University of Technology, Wuhan 430000, China; Xueping Guo, Marine Engineering College, Jimei University, Xiamen 361021, China; and You Zheng, Ningbo University of Technology, Ningbo 315000, China. Contact e-mail: [email protected].
Journal of Thermal Spray Technology
pared from a graphic environment. The OLP software normally integrates with add-in functionality, and this provides an effective and versatile approach to add optimal trajectory generation strategy in the OLP software. Publications dealing with this topic are considered in this research. Sheng et al. formalized the surface partitioning problem as an algebraic integer programming problem (Ref 3). They also suggested a stitching algorithm to decrease th
Data Loading...
