坦克发动机道路模拟测试平台轨迹寻优研究

坦克发动机道路模拟测试平台采用冗余构件多分支并联运动系统结构形式，满足了大负载、高灵活度、运动复杂的实际要求．鉴于系统结构形式的特殊性，论文采用结合个体定向漂移技术的基于邻域函数的尺度参数自寻优改进模拟退火算法，对此并联运动系统进行六维轨迹寻优解算，阐述了自寻优改进模拟退火算法，并将系统六维轨迹解算仿真结果与实验数据进行对比分析，在实践中得到成功验证．该方法将模拟退火的串行搜索机制与邻域函数混沌选择的并行处理机制进行了恰当结合，使算法不仅具有模拟退火算法的计算及存储量小、处理简单、寻优准确等优点，而且具有混沌选择遍历性和全局优化特性的优点．通过个体定向漂移技术提高了搜索效率，改善了邻域函数结构．

Research on Track Optimal Searching of Tank Engine's Movement State Simulating Platform

The platform is a test-equipment which is used to simulate the tank engine's movement state.A multi-branches parallel manipulator with redundant branches driven by hydraulic power is adopted,which can meet the requirements of heavy load,high agility and complicated movement.Because of the peculiar structure of the system,the study introduces an improved simulated annealing algorithm,which involves the mechanism of self-optimization for neighbor-(function) scale parameter with individual floating directionally technique,to carry out the six-dimension track searching for this system.This paper not only expatiates on the improved simulated annealing algorithm,but also compares the simulation result with that of experiment.The analysis result validates that the simulation result well matches the experimental result.Furthermore,this method perfectly combines the series-search characteristic of simulated(annealing) algorithm with the parallel-search specialty of chaos sequences.It also possesses both the local optimal ability of simulated annealing algorithm and the global optimal characteristic of chaos sequences.By applying the technique of individual floating directionally to the improved simulated annealing algorithm,the algorithm not only improves the searching efficiency,but also makes the neighbor-function structure more reasonable.