基于改进逼近理想解的排序技术方法自发电缓冲背架性能优化

Optimization of spontaneous electrical buffer back frame based on improved TOPSIS method

针对自发电缓冲背架结构特点和应用场景,在给定的步态条件下,以背架刚度和阻尼比为优化变量,以背架发电功率和人体所受冲击载荷为优化目标,提出了基于加权广义马氏距离逼近理想解的排序技术(Technique for order preference by similarity to ideal solution,TOPSIS)的性能优化方法。采用带有精英策略的非支配排序遗传算法(Non-dominated sorting genetic algorithm-Ⅱ,NSGA-Ⅱ),获得了背架发电功率和人体所冲击载荷的Pareto最优解及决策矩阵;利用熵权法计算出各指标的权重;使用改进后的基于加权广义(Weighted Moore-Penrose,WMP)马氏距离的TOPSIS法对Pareto最优解排序,并对3种不同优化算法进行了比较。分析了不同步态条件对冲击载荷和发电功率的影响规律。结果表明:基于改进TOPSIS法的决策结果更加符合实际需求,背架整体综合性能得到改善,对不同步态条件具有良好的适应性。

According to the structure characteristic and application scene of spontaneous electrical buffer back frame,under given gait conditions,a performance optimization technique for order preference by similarity to ideal solution(TOPSIS)method based on weighted generalized Mahalanobis distance was presented with the back frame of stiffness and damping ratios as optimization variables and generating power and impact load on human body as optimization objective.First,generating power and impact load on human body of Pareto-optimal solutions’decision matrix was given by the non-dominated sorting genetic algorithm(NSGA-Ⅱ),and then the weight of each index was calculated by using entropy weight method,last the Pareto-optimal solutions are ranked through technique for order preference by TOPSIS method based on weighted Moore-Penrose(WMP)Mahalanobis distance.Three different optimization algorithms were compared.The effect of different gait conditions on impact load and power generation was analyzed.The result shows that decision based on the improved TOPSIS method satisfies the actual demand more and makes the overall performance of the back frame better.It also has good adaptability to different gait conditions.