Deep Learning based Nonlinear Active Suspension Control
T.Rajesh1, S. Arun Jayakar2, G.M.Tamilselvan3
1T.Rajesh, Assistant Professor, Department of Electronics and Instrumentation Engineering, Bannari Amman Institute of Technology Sathyamangalam, Erode (Tamil Nadu), India.
2S.Arun Jayakar, Assistant Professor, Department of Electronics and Instrumentation Engineering, Bannari Amman Institute of Technology Sathyamangalam, Erode (Tamil Nadu), India.
3G.M.Tamilselvan, Professor, Department of Electronics and Instrumentation Engineering, Bannari Amman Institute of Technology Sathyamangalam, Erode (Tamil Nadu), India.
Manuscript received on 13 December 2018 | Revised Manuscript received on 22 December 2018 | Manuscript Published on 30 December 2018 | PP: 114-118 | Volume-8 Issue-2S, December 2018 | Retrieval Number: 100.1/ijeat.B10321282S18/18©BEIESP
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: Design of comfortable and effective active-suspension system of the vehicle has been enthralling and tough control-engineering workbench-problem. Passive and active suspension system model has been outlined as Quarter model (1/4th wheels) with spring-damper arrangement. The proposed model is for a execution of active-suspension system with actuator (final control element) is incorporated that can create the control output, ‘Uc (t)’ to control the movement of the vehicle. This paper proposes Deep learning based Modified PSO (DMPSO) for effective nonlinear active suspension system. In the General PSO, the development of a molecule is represented by three practices to be specific latency, intellectual and social. The subjective conduct helps the molecule to recall its past went to best position. This proposed PSO splits the psychological conduct into two segments like previous (past) went by finest (best) position and furthermore past went to most perceptibly appalling position. This change causes the molecule to look through the objective exceptionally successfully. DMPSO approach is proposed to increasing ride comfort results in slighter damping and superior suspension strokes in the vehicle.
Keywords: Dynamic System, Suspension, Modeling, PID, Optimization, DMPSO.
Scope of the Article: Deep Learning