航空学报 > 2024, Vol. 45 Issue (15): 630203-630203   doi: 10.7527/S1000-6893.2024.30203

航天机电伺服系统非线性表征建模与动态特性精细化预示

蒋孟龙1,2,3, 付剑1, 刘轶男2,3, 孙明伟1, 黄玉平2,3, 于志远2,3()   

  1. 1.北京航空航天大学 机械工程及自动化学院,北京 100191
    2.北京精密机电控制设备研究所 航天伺服驱动与传动技术实验室,北京 100076
    3.控制执行机构技术创新中心,北京 100076
  • 收稿日期:2024-01-20 修回日期:2024-02-15 接受日期:2024-03-06 出版日期:2024-04-23 发布日期:2024-04-19
  • 通讯作者: 于志远 E-mail:zhiyuan.yu@lasat.com
  • 基金资助:
    国家自然科学基金(92371101);中央高校基本科研业务费专项资金(YWF-23-L-911)

Nonlinear characterization modeling and dynamic performance accurate prediction of aerospace electromechanical actuator

Menglong JIANG1,2,3, Jian FU1, Yinan LIU2,3, Mingwei SUN1, Yuping HUANG2,3, Zhiyuan YU2,3()   

  1. 1.School of Mechanical Engineering and Automation,Beihang University,Beijing  100191,China
    2.Laboratory of Aerospace Servo Actuation and Transmission,Beijing Institute of Precision Mechatronics and Controls,Beijing  100076,China
    3.Innovation Center for Control Actuators,Beijing  100076,China
  • Received:2024-01-20 Revised:2024-02-15 Accepted:2024-03-06 Online:2024-04-23 Published:2024-04-19
  • Contact: Zhiyuan YU E-mail:zhiyuan.yu@lasat.com
  • Supported by:
    National Natural Science Foundation of China(92371101);The Fundamental Research Funds for the Central Universities(YWF-23-L-911)

摘要:

针对机、电、磁、控多非线性强耦合作用导致航天机电伺服系统动态特性预示难的问题,基于MBSE、层次分析法、增量式建模技术分别构建了机电伺服应用于推力矢量控制的V型需求模型、多层架构模型、精细化性能预示模型。一级模型通过调配电机功率和减速比,支持位置包络预示;二级模型增量电阻-电感(RL)电磁转换架构和运动三角关系,实现位置响应预示与尺寸估计;三级模型构建多物理域耦合关系;四级模型精细化关键参数非线性,达到转速和电流动态响应准确预示效果。通过3组时域测试验证了多层级模型预示精度。设计实例表明,所建多层级模型完成多约束下机电伺服设计优化获得的可行解数量比第四级高精度模型单独优化多3.6倍、计算耗时缩减50.6%;最优方案功重比达488.95 W/kg,相频带宽达9.87 Hz,有效支撑航天机电伺服系统的快速设计和精准迭代,同时为火箭系统的数字化转型提供了可行实践方法。

关键词: 机电作动器, 多物理域耦合, 增量式非线性建模, 基于模型的系统工程, 推力矢量控制

Abstract:

Aiming at the difficulties in predicting the dynamic performance of aerospace Electro-Mechanical Actuator (EMA) due to the multiple nonlinear coupling of mechanics, electricity, magnetism and controlling, a V-shape requirement model, a multi-level architecture model and an accurate prediction model for EMA in Thrust Vector Control (TVC) are built based on Model-Based System Engineering (MBSE), Analytic Hierarchy Process (AHP) and incremental modeling respectively. The first-level model enables the position enveloping prediction by adjusting motor power and reduction ratio. The second-level model realizes position response prediction and dimension estimation with incremental RL electromagnetic framework and trigonometric kinematic functions. The third-level model constructs the coupling relations of multi-physical domain, and the fourth-level model refines the non-linearity of key parameters which enables accurate prediction of rotating speed and current dynamic response. The prediction accuracy of the multi-level model is verified by three sets of time domain tests. The design instance indicates that the multi-level models under multiple constraints obtain 3.6 times more than the individual fourth-level model with a 50.6% reduction in calculating time. The optimal power-weight ratio reaches 488.95 W/kg, and the phase frequency bandwidth reaches 9.87 Hz, which strongly supports the rapid and accurate iteration of EMA design, and provides a feasible digital transformation practice of rocket system.

Key words: electro-mechanical actuators, multi-physical domain coupling, incremental non-linear modeling, model-based system engineering, thrust vector control

中图分类号: