Title: Intelligent monitoring and diagnosis of manufacturing process using an integrated approach of neural network ensemble and genetic algorithm
Authors: Jianbo Yu, Lifeng Xi
Addresses: Department of Industrial Engineering and Management, Shanghai Jiaotong University, Shanghai 200240, PR China. ' Department of Industrial Engineering and Management, Shanghai Jiaotong University, Shanghai 200240, PR China
Abstract: In this paper, a hybrid learning-based model is developed for online intelligent monitoring and diagnosis of manufacturing processes. In this model, a Genetic Algorithm (GA)-based selective Neural Network (NN) ensemble (GASENN) is developed for monitoring the manufacturing process and recognising faulty quality categories of products being produced. In addition, a GA-based Rule (GARule) extraction algorithm is developed to discover the causal relationship between manufacturing parameters and product quality. These extracted rules are applied for fault diagnosis of the manufacturing process to reveal why this has occurred and how to recover from the abnormal condition with the specific guidelines on process parameter settings. Therefore, the seamless integration of GASENN and GARule provides abnormal warning, reveals assignable cause(s) and helps operators optimally set the process parameters. This model is conducted in an Ethernet network environment with various sensors, PLCs, computers, etc. The whole system is successfully applied into a japanning-line, which improves the product quality and saves manufacturing cost.
Keywords: intelligent monitoring; fault diagnosis; neural networks; GAs; genetic algorithms; data mining; intelligent diagnosis; process monitoring; manufacturing parameters; product quality.
International Journal of Computer Applications in Technology, 2008 Vol.33 No.2/3, pp.109 - 119
Published online: 10 Dec 2008 *Full-text access for editors Access for subscribers Purchase this article Comment on this article