电工技术学报2026，Vol.41Issue(6)：1934-1947,14.DOI:10.19595/j.cnki.1000-6753.tces.250434
高速磁浮牵引供电系统馈电损耗双维度协同优化策略
Dual-Dimensional Collaborative Feeding Loss Optimization Strategy for High-Speed Maglev Traction Power Supply Systems
摘要
Abstract
High-speed maglev traction power supply systems feature a configuration where parallel inverters supply power to the segmented long-stator from both ends.As a result,circulating currents and feeding network losses adversely affect the energy efficiency of the traction system.While several solutions addressing circulating currents in parallel inverters have been proposed recently,existing research rarely analyzes or addresses this issue from the perspective of overall traction power supply system efficiency.This paper proposes a dual-dimensional collaborative optimization strategy that integrates dynamic current allocation with traction substation layout design.This approach suppresses circulating currents,reduces feeding network losses,and enhances traction efficiency for high-speed maglev traction power supply systems under both high-speed testing and normal operation scenarios. Firstly,a collaborative optimization model is developed to suppress circulating currents and minimize feeding network losses simultaneously.A dynamic current allocation strategy is proposed based on the principle of inverse proportionality to cable length.This strategy ensures that the output currents of parallel inverters are dynamically adjusted according to the length of the connected feeding cables,thus achieving minimal circulating currents and reduced feeding losses.Secondly,the study introduces differentiated optimization designs for traction substation layouts to optimize feeding network losses,considering the constraints of inverter capacity and system topology.Hardware-in-the-loop(HIL)experimental results validate the effectiveness of the proposed strategies.The dynamic current allocation strategy successfully eliminates circulating currents between parallel inverters and reduces feeding network losses by up to 50.3%.Furthermore,the optimized traction substation layouts address unbalanced power distribution among inverters caused by current allocation adjustments.For high-speed testing conditions,the dual symmetric offset configuration(DSOC)strategy achieves the lowest feeding network loss of 88.71 kW·h and a traction efficiency of 81.5%.The dual midpoint co-located configuration(DMCC)strategy achieves a higher traction efficiency of 84.4%with a feeding loss of 131.85 kW·h.Still,it requires fewer traction substations,thereby reducing infrastructure complexity and cost.In normal operation conditions,the operating conditions symmetric configuration(OCSC)strategy achieves the lowest feeding loss of 52.36 kW·h and a traction efficiency of 88.3%.In comparison,the operating conditions co-located configuration(OCCC)strategy achieves the highest traction efficiency of 90.3%with a feeding loss of 131.29 kW·h.DSOC and OCSC strategies offer superior performance in minimizing feeding losses,but necessitate the construction of two traction substations and high inverter power capacities.Conversely,DMCC and OCCC strategies provide balanced power distribution and enable single-substation deployment,reducing construction costs and simplifying system architecture. The dynamic current allocation strategy,grounded in the inverse proportionality principle,effectively balances the trade-off between feeding loss minimization and circulating current suppression.The differentiated substation layout designs provide scalable solutions for various operating scenarios,ensuring energy-efficient and cost-effective system operation.This paper offers valuable insights for the design and optimization of high-speed maglev traction power supply systems,contributing to the advancement of efficient and sustainable high-speed transportation.关键词
高速磁浮列车/双端供电模式/馈电网络/损耗优化Key words
High-speed maglev/double feed mode/feeding network/loss optimization分类
信息技术与安全科学引用本文复制引用
郑彦喜,朱进权,葛琼璇,赵牧天,高瑞..高速磁浮牵引供电系统馈电损耗双维度协同优化策略[J].电工技术学报,2026,41(6):1934-1947,14.基金项目
国家重点研发计划(2023YFB4302501-02)和国家自然科学基金(52302458)资助项目. （2023YFB4302501-02）
