Multi-agent cooperative structural vibration control of three coupled flexiblebeams based on value decomposition network

作者：Qiu, Zhi-cheng^*; He, Cheng-hu; Zhang, Xian-min

来源：ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE, 2022, 114: 105002.

DOI：10.1016/j.engappai.2022.105002

• Three-coupled flexible beam
• Active vibration control
• Wavelet analysis
• Reinforcement learning
• Value decomposition network
• Deterministic policy gradient

Multi-coupled flexible body (MCFB) structure has the characteristics of small damping, close modes andcoupled vibration, and the structural vibration is difficult to freely attenuate, which will lead to rapid fatiguefailure of the structure. In order to suppress the vibration of the MCFB system quickly, a system identificationscheme and a multi-agent reinforcement learning (MARL) scheme are proposed to obtain the cooperativecontroller. The mathematical model of the three-coupled flexible beam (TCFB) system is identified by waveletanalysis method, particle swarm optimization (PSO) and sinusoidal excitation response method. Based on valuedecomposition network (VDN) and deterministic policy gradient (DPG), a cooperative MARL framework isconstructed to obtain the controller (VDN-DPG controller). The cooperative VDN-DPG controller is appliedto the vibration control simulation and experiments of the TCFB system, and compared with proportional-derivative (PD) control without cooperation. The results show that the performance of the VDN-DPG controlleris better than that of the PD controller, which verifies the effectiveness of the multi-agent cooperative controlscheme based on VDN-DPG