跨声速抖振锁频问题的自适应控制研究

Transonic buffet caused by shock boundary layer interaction will lead to fatigue loads on the structure. In addition, transonic buffet flow can also induce frequency lock-in phenomenon due to the effect of fluid-structure interaction, which may further lead to large-amplitude structural vibration and catastrophic accident, with the harm being far beyond the non-lock-in state. At present, active/passive control is the main method to suppress transonic buffet flow based on rigid models. With these control, it is difficult to completely suppress buffet loads. In addition, few studies have been done on the control of frequency lock-in phenomenon, and the existing control methods designed for buffet flow are not suitable for the lock-in case. In this project, with the framework of numerical simulation, the model-free adaptive control of transonic buffet and frequency lock-in problem will be studied by using a data-driven approach, RBF neural network method, on the basis of revealing the dynamic instability mechanism and characteristics of them. The research contents include adaptive control of transonic buffet flow (or non-lock-in) and frequency lock-in phenomenon when the parameters change, as well as the adaptive control on the process of mode conversion between non-lock-in and lock-in. This study will be a foundation for the study of adaptive control of complex unstable flow, and provide guidance for controlling the destructive vibration problems caused by buffet flows in aeronautical engineering and other fields.

激波附面层干扰引起的跨声速抖振除了造成飞行器结构的疲劳损伤外，还会因流固耦合作用而诱发结构大幅振动的锁频现象，其危害程度远超非锁频状态，有可能造成灾难性事故。目前针对跨声速抖振的控制研究以刚性模型下抖振流动的主/被动控制为主，很难实现抖振载荷的完全抑制，而对危害更大的结构锁频现象的控制研究极少，并且现有的针对抖振流动设计的控制方式并不适用于锁频状态。本项目拟通过数值仿真手段，在对跨声速抖振流动和结构锁频现象动力学失稳机制及特征揭示的基础上，采用RBF神经网络方法开展基于数据驱动的跨声速抖振锁频问题的无模型自适应控制研究。具体研究内容包括，参数变化时跨声速抖振（非锁频）和锁频响应的参数自适应控制以及非锁频和锁频不同动力学特征转换过程的模式自适应控制。该研究将为复杂不稳定流动及其诱导锁频现象的自适应控制研究奠定基础，进而为航空工程等领域因抖振产生的破坏性振动问题的控制提供指导。

激波附面层干扰引起的跨声速抖振除了造成飞行器结构的疲劳损伤外，还会因流固耦合作用而诱发结构大幅振动的锁频现象，其危害程度远超非锁频状态，有可能造成灾难性事故。目前针对跨声速抖振的控制研究以刚性模型下抖振流动的主/被动控制为主，很难实现抖振载荷的完全抑制，而对危害更大的结构锁频现象的控制研究极少，并且现有的针对抖振流动设计的控制方式并不适用于锁频状态。本项目通过数值仿真手段，在对跨声速抖振流动和结构锁频现象动力学失稳机制及特征揭示的基础上，采用RBF神经网络方法开展基于数据驱动的跨声速抖振锁频问题的控制研究。实现了参数变化时跨声速抖振（非锁频）和锁频响应的参数自适应控制以及非锁频和锁频不同动力学特征转换过程的模式自适应控制。通过项目研究发表SCI论文6篇，指导研究生3人，研究成果可以为航空工程等领域因抖振产生的破坏性振动问题的控制提供指导。