一种可用于航天Boost拓扑电源参数辨识的数字孪生方法  

Digital twin method for parameter identification of aerospace Boost topological power supply

作　　者：宋相毅 黄宇超 童乔凌[2] 张侨[1] SONG Xiangyi;HUANG Yuchao;TONG Qiaoling;ZHANG Qiao(School of Automation,Wuhan University of Technology,Wuhan Hubei 430070,China;School of Integrated Circuits,Huazhong University of Science and Technology,Wuhan Hubei 430074,China)

机构地区：[1]武汉理工大学自动化学院,湖北武汉430070 [2]华中科技大学集成电路学院,湖北武汉430074

出　　处：《电源技术》2024年第3期471-476,共6页Chinese Journal of Power Sources

摘　　要：电力电子变换器中元器件随时间退化会影响到航天电源的稳定性与可靠性。为确保航天电源的可靠性,并准确识别零部件状态,数字孪生依靠其非侵入性、数字化、高性价比、低风险、可视化与实时性的特点,可以解决电力电子变换器的故障趋势判断与状态检测的问题,从而保证航天电源的稳定性与可靠性。通过验证Boost拓扑电路的数字孪生模型,同时采用禁忌搜索算法以比较结果最小二乘误差为目标函数得到全局最优解,进行Boost拓扑电路参数辨识、实时监控。并且对比了100、500、1000算法迭代次数对结果的影响,随着迭代次数的增加,最终结果与实际值仅有最大0.96%的误差。验证该项技术可应用到航天电源的实时状态监控,以保证其可靠性。The deterioration of components in power electronic converter will affect the stability and reliability of space power supply.In order to ensure the reliability of space power supply and identify its state.Digital twin with non-invasive,digital,cost-effective,low-risk,visual and real-time technology,can solve the problem of fault trend judgment and status detection of power electronic converters,to ensure the stability and reliability of space power supply.The digital twin model of the Boost topology circuit was verified by the tabu search algorithm,and the parameters of the Boost topology circuit were identified and monitored in real time.Moreover,the influence of 100,500 and 1000 algorithm iterations on the result was compared.With the increase of iterations,the error between the final result and the actual value was only 0.96%.It was proved that this technology could be applied to the real-time status monitoring of aerospace power supply to ensure its reliability.

关 键 词：航天电源 数字孪生 Boost拓扑 参数辨识 禁忌搜索算法 

分 类 号：TM919[电气工程—电力电子与电力传动]