ANN System Identification for Rapid Battery Charger using Parallel 3 Parent Genetic Algorithm
Ashima Kalra1, Shakti Kumar2, Sukhbir Singh Walia3
1Ashima Kalra, P.H.D research scholar, Punjab Technical university, Punjab, India.
2Prof. Shakti Kumar, Director, Panipat Institute of Engineering Technology, Panipat, Haryana, India.
3Dr. Sukhbir Singh Walia, Registrar, Punjab Technical university, Punjab, India.
Manuscript received on 20 August 2019 | Revised Manuscript received on 27 August 2019 | Manuscript Published on 26 August 2019 | PP: 47-51 | Volume-8 Issue-9S August 2019 | Retrieval Number: I10790789S19/19©BEIESP | DOI: 10.35940/ijitee.I1079.0789S19
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open-access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Model identification is one of the main concerns in the field of system modeling. The complete modeling of an ANN system using input output data consists of two processes: architecture selection in which number of hidden layers and the number of neurons in each hidden layer is to be decided. This is then followed by training of system by the given training data. The problem here is formulated as search and minimization problem. This paper presents the identification of ANN system for rapid Nickel Cadmium (Ni-Cd) batteries charger by applying a new version of Genetic algorithm called as parallel 3 parent genetic algorithm (P3PGA). It is a multi population based 3 parent genetic algorithm (3PGA) .The proposed approach was implemented using MATLAB R2018A and was observed to be computationally more efficient with minimum MSE. With increase in number of iterations, system performance gets improved. We further compared results of the proposed algorithm with the results of other recent soft computing based algorithms as well classical learning based algorithms namely, big bang big crunch (BB-BC), parallel big bang big crunch (PBB-BC), Levenberg-Marquardt algorithm (LM), error back propagation (EBP), Resilent prop (RPROP), particle swarm optimization (PSO), ant colony optimization (ACO) and artificial bee colony (ABC) for ANN model identification. The proposed algorithm outperformed all of the other 8 algorithms.
Keywords: Model Identification, 3 Parent Genetic Algorithm (3PGA), Parallel 3 Parent Genetic Algorithm (P3PGA)
Scope of the Article: System Integration