Research of improved particle swarm algorithm based on apparel intelligent marker making

Abstract

The intelligent algorithm of apparel marker making is the core technology of apparel marker making, this paper brings forward an improved particle swarm algorithm at the base of analyzing the merit and demerit of genetic algorithm, simulated annealing algorithm and particle swarm optimization algorithm, furthermore, a demonstration test is done by C++ under the environment of VS2005. The result indicates the improved particle swarm algorithm can achieve ideal material utilization. Keywords-apparel intelligent marker making; genetic algorithm; simulated annealing algorithm; improved particle swarm algorithm I. INTRODUCION Garment marker making technology is one of the important technologies of clothing enterprises, the operation time of marker process and the utilization ratio of maker making directly affect the enterprise's cost, profit, product competition ability and economic benefits. Marker making technology has been used as the most important way to raise the material utilization ratio, reduce the material loss and production cost for garment manufacturing enterprises in the production process .Therefore, a set of intelligent and efficient marker making technology can not only reduce the staff's working strength, but also greatly improve the producing speed and the utilization ratio of material, and but also bring considerable economic benefit to enterprises. With the development and popularization of garment CAD software, the core technology of garment CAD, the intelligent marker algorithm has been improved also. The optimization algorithms which can solve the maker problem conclude artificial neural network, genetic algorithm, heuristic algorithm and simulated annealing algorithm. Therein, the artificial neural network is more difficult to solve optimal marker problem, as it's not suitable for non rectangular pieces; and as genetic algorithm has no special restrictions to its required problem, and more facilitate for organically integration of rectangular parts graphic arithmetic and genetic algorithm, so there are many solutions can solve the optimal marker problem. But the genetic algorithm has limitation which we call the premature phenomenon, and its local searching ability is poor; simulated annealing algorithm has strong overall search ability. In a word, marker making problem is a typical NP complete problem, namely the time complexity of finding the solution is in exponential order. In the relatively large size of the problem, the loss of viability often occurs because of the increasing computing time and, so it is necessary to find other feasible scheme. This paper introduces a method that based on particle swarm algorithm, and does experiment at the same time analysis its effectiveness.