论文部分内容阅读
导管架式海洋平台在随机波浪等外载荷作用下极易产生有害振动,且其动力响应具有极强的非线性和时变性,采用被动控制方法和基于精确数学模型的主动控制方法控制海洋平台的有害振动很难达到理想的控制效果。为此文中将灰预测和粗神经网络相结合,提出了一种基于灰预测和粗神经网络的预测逆控制方法,并将其与动态刚度阵法相结合用于导管架式海洋平台的振动主动控制中。数值算例分析表明此种控制方法可有效地控制波浪和风载荷作用而引起的导管架式平台的有害振动,并能解决由于控制信号传输等原因引起的时滞问题。
The jacket-mounted offshore platform is prone to harmful vibration under external loads such as random waves. Its dynamic response is highly nonlinear and time-varying. The passive control method and the active control method based on the precise mathematical model are used to control the offshore platform Harmful vibration is difficult to achieve the desired control effect. In this paper, a combination of gray prediction and rough neural network is proposed in this paper. A predictive inverse control method based on gray prediction and rough neural network is proposed and used in combination with dynamic stiffness matrix method Control. The numerical example shows that this control method can effectively control the harmful vibration of the jacket platform caused by wave and wind loads and can solve the time delay problem caused by the control signal transmission and other reasons.