Investigation on Application of Support Vector Machine Algorithm in the Experimental Simulation System for Mechanism Motion Reliability
Science Discovery
Volume 5, Issue 1, February 2017, Pages: 12-18
Received: Mar. 13, 2017; Published: Mar. 15, 2017
Views 1128      Downloads 43
Authors
Zhang Tong, School of Mechanical Engineering, Beijing Institute of Technology, Beijing, China
Li Sen, School of Mechanical Engineering, Beijing Institute of Technology, Beijing, China
Qiao Jia-dong, School of Mechanical Engineering, Beijing Institute of Technology, Beijing, China
Article Tools
Follow on us
Abstract
The complexity of the mechanism brings great difficulty to the calculation of the mechanism motion reliability, and the computer simulation algorithm based on Monte Carlo method needs a great number of simulation. The further-development of LMS Virtual. Lab was carried out, and the least square support vector machine algorithm was used to construct the response surface proxy model. The PSO-GA algorithm is used to optimize the parameters and make the least square support vector machine own a better performance. For the coupling of multiple failure modes, the multiple response surfaces are combined to obtain the reliability of the system. The reliability calculation based on support vector machine algorithm is carried out for the crank slider mechanism and the cam swing bar mechanism, and the feasibility and efficiency of the method are verified by the Monte Carlo method.
Keywords
Reliability, Simulation, Support Vector Machine, Response Surface Models
To cite this article
Zhang Tong, Li Sen, Qiao Jia-dong, Investigation on Application of Support Vector Machine Algorithm in the Experimental Simulation System for Mechanism Motion Reliability, Science Discovery. Vol. 5, No. 1, 2017, pp. 12-18. doi: 10.11648/j.sd.20170501.13
ADDRESS
Science Publishing Group
1 Rockefeller Plaza,
10th and 11th Floors,
New York, NY 10020
U.S.A.
Tel: (001)347-983-5186