Abstract:
Intelligent Neural Network (NN) based control schemes have surmounted many of the limitations found in the conventional control approaches such as Proportional Integral Derivative (PID) control. Nevertheless, these modern control techniques have only recently been introduced for use on industrial computational platforms such as the Programmable Logic Controller (PLC). Intelligent control on PLCs thus remains an area that is open to further research and development. In this paper, a strongly non-linear mechatronic type system, namely the Ball-on-Wheel balancing system, is developed using a PLC as its control platform. The research details the implementation of an intelligent controller on a standard, medium specification PLC. The results from the intelligent controller are then compared to those produced by a variety of conventional controllers as physical parameters are varied. Finally, the system is presented as a stimulating educational tool that addresses the knowledge gap that exists in industry pertaining to the implementation of these intelligent control algorithms on PLCs.
