Title: Performance improvement of software-based system using an integrated approach – a case study

Authors: R. Amuthakkannan

Addresses: Mechatronics and Virtual Instrumentation Research cell, Coimbatore Institute of Technology, Coimbatore, Tamil Nadu 641 014, India

Abstract: The modern automation system consists of software and hardware components to achieve the high quality products and processes. In such type of software-based systems, optimal design is more important to improve the system performance. The perfect parameter design problems are complex because of non-linear relationships and interactions may occur among parameters. So, a proper approach is needed for a parameter optimal design. An integrated approach of neural network with genetic algorithms is proposed to address the optimal design of software-based automation system. This article outlines neural network methodology to predict the response of the software-based automation system for various process parameters values. Then, the genetic algorithm is used to predict the quantitative value of process parameter to improve the performance of the system. In this work, a cascading electro-pneumatic kit is taken as case analysis to analyse the performance of software-based system.

Keywords: genetic algorithms; neural networks; parameter optimisation; process change; software-based automation; parameter design; optimal design; process parameters; cascading electropneumatics.

DOI: 10.1504/IJISCM.2008.026709

International Journal of Information Systems and Change Management, 2008 Vol.3 No.4, pp.327 - 343

Published online: 23 Jun 2009 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article