Title: Design of intelligent hybrid supervisory controller based on temporal neural networks and timing modules
Authors: Magdy M. Abdelhameed; Houshang Darabi
Design and Production Engineering Department, Faculty of Engineering, Ain Shams University, 1 Elsrayat St., Abdo Basha, Abbasiah, Cairo, Egypt
Mechanical and Industrial Engineering Department, Faculty of Engineering, University of Illinois at Chicago, 842 W. Taylor St., Chicago, IL, 60607, USA
Abstract: This paper investigates the application of temporal neural networks in designing sequence controllers for time and event driven mechatronic manufacturing systems (MMSs). A proposed design of this controller is presented. The proposed design is based on temporal neural network algorithm with timing modules. Adequate guidelines of constructing the controller are presented. These guidelines are applied to design a controller for an industrial application. Functionality of the proposed controller is tested by simulation. Learning process, functionality and main features of this controller are studied and analysed. Theoretical results are presented and discussed. These results prove the proper functionality of the controller to deal with a time and event driven discrete manufacturing system. In addition, the simulation results assure the effectiveness of the proposed controller to outperform the effect of noisy inputs.
Keywords: mechatronic manufacturing systems; MMS; recurrent neural networks; event driven discrete manufacturing; IEC 61131; programmable logic controllers; PLCs; controller design; intelligent control; hybrid control; supervisory control; temporal neural networks; timing modules; simulation.
Int. J. of Industrial and Systems Engineering, 2014 Vol.17, No.1, pp.54 - 77
Available online: 28 Apr 2014