Ahut-Delta并联机构改进混沌粒子群算法尺度综合

针对Ahut-Delta并联机构,提出了一种基于改进混沌粒子群算法的尺度综合方法。首先提出一种改进混沌粒子群算法,即采用混沌立方映射初始化种群,并根据迭代状态指数性调整惯性权重因子,同时进行早熟判断和混沌扰动,迭代获得最优粒子。其次将Ahut-Delta并联机构优化参数转变为粒子维度决策变量,雅可比矩阵的全域均值条件数和全域波动量构建的全域综合性能评价指标在其几何条件约束、传动角约束条件下转换为改进混沌粒子群算法的适应度函数。最终通过改进混沌粒子群算法优化搜索,优化出适应度函数值最小的最优粒子,从而获得Ahut-Delta并联机构在全域运动性能最佳的尺度参数。仿真分析结果表明,所提尺度综合方法具有正确性和有效性。

Dimensional Synthesis of Ahut-Delta Parallel Mechanism Based on Improved Chaotic Particle Swarm Algorithm

Dimensional synthesis was the core content in the parallel mechanism design. Therefore, a dimensional synthesis method based on the improved chaotic particle swarm algorithm was proposed for the Ahut-Delta parallel mechanism. Firstly, improved chaotic particle swarm algorithm was proposed. In the algorithm, initialization of population with chaos cube map was experienced. Then inertia weight was adjusted exponentially on the basis of the algorithm iterative state. Simultaneously, early maturity judgment and chaotic disturbance were utilized to obtain the optimal particle. Secondly, the optimal parameters of Ahut-Delta were changed to the dimensional variables. The population mean condition number and the population fluctuation rate of Jacobian were synthesized to a global performance index, and then the global performance index was changed to the fitness function for the improved chaotic particle swarm algorithm under the geometric constraints and the transmission angle constraints of the Ahut-Delta. Thirdly, an optimization simulation on the scale parameters for Ahut-Delta parallel mechanism was conducted by using two optimization algorithms, i.e., basic particle group algorithm and improved chaotic particle swarm algorithm. Through the analysis of the two algorithms results, the optimal particle with the minimal fitness function value was optimized by means of improved chaotic particle swarm algorithm, and the optimal scales were obtained which remarkably improved Ahut-Delta motion performance. Finally, the results of simulation and analysis verified the correctness and effectiveness of the method.