Title: Improved NN-PID control of MIMO systems with PSO-based initialisation of weights
Authors: Tarun Varshney; Satya Sheel
Addresses: Electrical Engineering Department, MNNIT, Allahabad 211004, UP, India ' SMIEEE, T-09 / 801 Sushant Estate, Sector 52, Gurgoan, Haryana, India
Abstract: To train the neural networks (NNs) standard back propagation (BP) algorithm and its variations are widely used where initial weights are generated as random in nature. The convergence of these algorithms is very sensitive to the initial weights. In this paper, particle swarm optimisation (PSO) algorithm has been used to initialise the weights by optimisation. Various available algorithms and the proposed one have been tested and compared for the implementation of NN-PID control of two discrete-time non-linear coupled MIMO systems. Simulation results show that the controlled performance of BP algorithm for non-linear, coupled MIMO systems can be significantly improved, with the use of PSO algorithm to initialise the weights.
Keywords: particle swarm optimisation; PSO; MIMO; back propagation; initial weights; PID control; neural networks; simulation.
DOI: 10.1504/IJAAC.2014.063362
International Journal of Automation and Control, 2014 Vol.8 No.2, pp.158 - 172
Received: 04 May 2013
Accepted: 29 Mar 2014
Published online: 30 Jul 2014 *