Global torch path generation for 2-D laser cutting process using simulated annealing

Recently the major objective of laser processing in production line is the precision or shape cutting of sheet metals. This paper addresses the problem of a torch path generation for the 2D laser cutting of a stock plate nested with irregular parts. In this paper, the basic object is a polygon(a many-sided figure) with holes. Under the constraint of the relative positions of parts enforced by nesting, the developed torch path algorithm generates feasible cutting paths. The objective of this work is to traverse this cutting contours with a minimum path length. The proposed torch path algorithm is based on a simulated annealing, that is an improved version of previously suggested TSP models. Since every piercing point of parts is not fixed in advance, the algorithm solves an relaxed optimization problem with constraints. For solving the torch path optimization problem an efficient generation mechanism of neighborhood structure and annealing schedule were presented. In this way, a global solution can be obtained in a reasonable time. Several examples are represented to illustrate the method.