基于人工神经网络的海洋平台振动主动控制

Active Control of Offshore Platforms Using Neural Networks

将经典最优控制算法与人工神经网络相结合,采用BP神经网络模型,实现了受随机波浪力作用下的海洋平台的振动主动控制。由于神经网络的鲁棒性、容错性和很好的泛化能力,仿真结果表明控制是可行、有效的,并且克服了传统算法本身的时滞问题,为海洋平台的振动控制提供了一条新的思路。

With the combination of classical optimal control and artificial neural networks, active structural control of offshore platforms under random waves was accomplished using the BP neural networks mode. The obtained results show that the active control is feasible and effective, and finally overcomes the time delay of the classical algorithms, according to the robustness, fault tolerance and generalized capability of artificial neural network.