• CN: 11-2187/TH
  • ISSN: 0577-6686

Journal of Mechanical Engineering ›› 2025, Vol. 61 ›› Issue (19): 43-53.doi: 10.3901/JME.2025.19.043

Previous Articles    

Improvement of Differential Evolution Algorithm for Solving the Multi-root in Forward Kinematic Equations of Parallel Mechanisms

WEN Shikun1,2, JI Aihong1, LEE Heow Pueh2, CHE Linxian3, YANG Zhikang1   

  1. 1. College of Mechanical and Electrical Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210016;
    2. Department of Mechanical Engineering, National University of Singapore, Singapore 117575, Singapore;
    3. Department of Intelligent Manufacturing and Transportation, Chongqing Vocational Institute of Engineering, Chongqing 402260
  • Received:2024-11-04 Revised:2025-04-17 Published:2025-11-24

Abstract: The forward kinematics equations of a coupled parallel mechanism can be reformulated as a system of nonlinear equations with multi-root. Solving such a system in a single operation poses a significant challenge in numerical computation. A neighborhood two-strategy differential evolutionary algorithm (NTDE) based on individual information is proposed to solve this problem. Firstly, the transformation of the forward kinematic equations of a parallel mechanism into an unconstrained optimization problem is illustrated using the Stewart parallel mechanism as an example. Secondly, to enhance the optimization efficiency of the differential evolution algorithm, an adaptive subpopulation strategy is proposed that balances global and local search capabilities. Additionally, an individual judgment criterion is introduced to differentiate between individuals in the population, while the variation strategy and control parameter selection are informed by feedback. Finally, to validate the effectiveness and generalizability of the proposed method, the forward kinematic equations of the Stewart parallel mechanism, the 4-PRPaU parallel mechanism, and the 5PSS/UPU parallel mechanism are solved using the NTDE, NCDE, CADE, and MNPSO algorithms as examples. It is demonstrated by the experimental results that the NTDE algorithm effectively identifies all the roots of the forward kinematic equations for the parallel mechanism. Moreover, the RR and SR values obtained using the NTDE algorithm are shown to outperform those of the three comparative algorithms.

Key words: parallel mechanisms, forward kinematics equations, differential evolutionary algorithms, adaptive subpopulation, individual judgement criteria

CLC Number: