Optimal fuzzy inverse dynamics control of a parallelogram mechanism based on a new multi-objective PSO

This work presents a multi-objective optimization method based on high exploration particle swarm optimization, called MOHEPSO, for optimization problems with multiple objectives. In order to convert the single-objective (HEPSO) algorithm to the multi-objective one, its fundamentals should be change...

Full description

Saved in:
Bibliographic Details
Published in:Cogent engineering 2018-01, Vol.5 (1), p.1443675
Main Authors: Farokhi, A., Mahmoodabadi, M.J.
Format: Article
Language:English
Subjects:
Algorithms
Computer simulation
Fuzzy control
Fuzzy systems
Inverse dynamics
Linkage mechanisms
multi-objective optimization
Multiple objective analysis
optimal control
parallelogram five-bar mechanism
Particle swarm optimization
Performance evaluation
Pruning
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This work presents a multi-objective optimization method based on high exploration particle swarm optimization, called MOHEPSO, for optimization problems with multiple objectives. In order to convert the single-objective (HEPSO) algorithm to the multi-objective one, its fundamentals should be changed. The leaders' selection in the proposed algorithm is based on the neighborhood radius concept for the global best position and the Sigma method for the personal best position. Also, a fuzzy elimination technique is used for pruning the archive. The numerical results of the MOHEPSO algorithm on mathematical test functions are compared with those of other multi-objective optimization algorithms for the performance evaluation of the algorithm. Finally, the proposed algorithm is implemented to find the optimum values of controller coefficients for a parallelogram five-bar linkage mechanism. The introduced control strategy is designed based on the inverse dynamics concepts, improved by fuzzy systems and optimized by regarding two objective functions. The simulation results are presented to demonstrate the efficiency and accuracy of this approach.
ISSN:2331-1916
2331-1916
DOI:10.1080/23311916.2018.1443675