Spacecrafts usually suffer from the thermal shock during operation, leading to large control error or even failure. At the same time, spacecrafts are often rigid-flexible coupled multibody systems with large degrees of freedom and multiple varying parameters. The real-time control and condition monitoring require effective order reduction methods. In this investigation, the displacement and temperature fields are discretized by the Absolute Node Coordinate Formulation (ANCF) to achieve unified description. The coupled thermal-dynamic reduced-order model (ROM) is established based on the Proper Orthogonal Decomposition (POD) method. For the purpose of further increase efficiency, a linearization method is introduced so that the calculation times of the elastic force can be significantly reduced. In order to predict the response of the thermal-dynamic coupled system with multiple varying parameters, the interpolation on Grassmann manifold is introduced to maintain the orthogonality of the basis. As verification and validation, three numerical examples are presented to show the feasibility and efficacy of the proposed method.

中文翻译：

---

多变参数热动力耦合柔性多体系统模型降阶


航天器在运行过程中常常会受到热冲击的影响，导致控制误差较大甚至失效。同时，航天器通常是刚柔耦合的多体系统，具有大自由度和多个变化参数。实时控制和状态监测需要有效的降阶方法。在本次研究中，位移场和温度场通过绝对节点坐标公式（ANCF）进行离散化，以实现统一描述。基于本征正交分解(POD)方法建立了热动力耦合降阶模型(ROM)。为了进一步提高效率，引入了线性化方法，可以显着减少弹力的计算次数。为了预测具有多个变化参数的热动力耦合系统的响应，引入格拉斯曼流形插值以保持基的正交性。作为验证和验证，给出了三个数值算例来说明所提方法的可行性和有效性。