A novel system for discovery and reuse of typical process route based on information entropy and PSO-Kmeans clustering algorithm

一种新的基于信息熵和PSO-Kmeans聚类算法的典型工艺路线发现与重用体系

¹ School of Mechanical Engineering, Baoji University of Arts and Sciences, Baoji 721016, China
² Shaanxi Key Laboratory of Advanced Manufacturing and Evaluation of Robot Key Components, Baoji 721016, China
³ School of Mechanical Engineering, Northwestern Polytechnical University, Xi'an 710072, China

Received: 24 May 2022

Abstract

Manufacturing enterprises will accumulate a large number of manufacturing instances as they run and develop. Being able to excavate and reuse the instance resources reasonably is one of the most effective ways to improve manufacturing and support innovation. To determine the reuse object scientifically and raise the reuse flexibility, a novel system for discovery and reuse of typical process route based on the information entropy and PSO-Kmeans clustering algorithm is proposed in this paper. In this system, a similarity measurement method of machining process routes based on the information entropy of multistage longest common subsequence is developed. Then a discovery method of typical process route based on the spectral clustering idea and PSO-Kmeans clustering algorithm is invented, and the two reuse approaches based on the typical process route are analyzed and discussed. Finally, the three case studies are rendered and the results reveal that the proposed system can provide better support for manufacture instance reuse.

摘要

制造企业在经营和发展过程中会积累大量的制造实例, 对这些实例资源进行合理地挖掘和重用, 是提高制造效率和支持创新的最有效途径之一。为了科学确定重用对象和提高重用灵活性, 提出了一种基于信息熵和PSO-Kmeans聚类算法的典型工艺路线发现与重用体系。在该体系下, 提出了一种基于多级最长公共子序列信息熵的机加工艺路线相似度度量方法。在此基础上, 提出了一种基于谱聚类思想和PSO-Kmeans聚类算法的典型工艺路线发现方法, 并分析讨论了2种基于典型工艺路线的机加工艺重用途径。通过3个验证实例, 说明所提出的体系可以更好地支持制造实例重用。