Robust design optimisation for inductive power transfer systems from topology collection based on an evolutionary multi-objective algorithm
Robust design optimisation for inductive power transfer systems from topology collection based on an evolutionary multi-objective algorithm
复制标题
DOI：
10.1049/iet-pel.2014.0468
复制
发表时间：
2015-08
影响因子：
2
通讯作者：
Yang Zou;X. Dai;Weiyi Li;Yue Sun
中科院分区：
文献类型：
--
作者：
Yang Zou;X. Dai;Weiyi Li;Yue Sun
考虑到实际应用中不可避免的不确定性，基于进化多目标鲁棒优化方法，提出了一种感应电能传输（IPT）系统的分层综合方法，该方法能够自动获得包括最优补偿方案在内的鲁棒Pareto最优设计。首先，对于几种补偿电路，给出了实用的公式，以保证容抗和感抗不仅在原边而且在副边相互抵消。还给出了一些鲁棒设计考虑。然后，针对给定的补偿量，提出了一种两级进化多目标鲁棒优化方法，得到了相应的鲁棒Pareto最优设计，使系统具有期望的输出电压、最优效率、无分岔运行、大的一次侧和二次侧距离以及对参数不确定性的可接受的鲁棒性能。然后，在各个方案的所有鲁棒帕累托前沿中，提取出拓扑集合的鲁棒帕累托前沿，设计者可以在此基础上选择最终设计。最后，将该综合方法应用于实际IPT系统的设计。标称设计得到使用传统的优化与鲁棒解决方案的比较，以证明所提出的方法的优越性。实验结果验证了所设计的系统的最佳和鲁棒性能。
Considering the inevitable uncertainty in practice, based on an evolutionary multi-objective robust optimisation, this study presents a hierarchical synthesis method for inductive power transfer (IPT) systems to automatically obtain the robust Pareto-optimal designs including an optimal compensation schematic selected from a topology collection which combines typical IPT compensation types. First, for several compensation circuits, useful formulae are presented to ensure that capacitive and inductive reactance cancels each other in not only primary but also secondary side. Some robust design considerations are also given. Then for a given compensation, a bi-level evolutionary multi-objective robust optimisation is proposed to obtain the robust Pareto-optimal designs that provide corresponding systems with desired output voltage, optimal efficiency, bifurcation-free operation, large distance between the primary and secondary side and acceptable robust performance against parameter uncertainty. Thereafter, among all robust Pareto fronts of individual schemes, robust Pareto front of topology collection is extracted out on which designer can choose the final design. Finally, this synthesis method is used to design practical IPT systems. The nominal designs are obtained using traditional optimisation in comparison with robust solutions to demonstrate the superiority of proposed approach. Experimental results verify the optimal and robust performance of the designed systems.