Active Vibration Control Of Offshore Platforms Using Fuzzy Neural Network

With the development of ocean exploitation, more and more extremely large and long flexible offshore platforms are used for drilling operation and oil exploration. Located in hostile ocean environment, offshore platforms usually experience various environmental loads such as wind, wave and earthquake loads, which will induce continuous vibration of the platforms. Excessive vibration will result in injury of human health, structural fatigue damage, decreasing reliability of the platforms. Hence, it is important to find effective control schemes to mitigate the vibration response of offshore platforms. Ideal control effect can not be achieved if we take on traditional active and passive control method.In this paper, a new active control scheme for offshore platforms, based on fuzzy neural network (FNN) method, is presented. With the main advantages of fuzzy neural network and adaptive inverse control, this dynamic stiffness matrix (DSM) based FNN adaptive predictive inverse control (APIC) scheme is used to accomplish the active control of offshore jacket platforms subjected to random wave forces. A real time measurement of the random wave forces is made by the pressure transducers which are placed on the leg of the offshore platform. And then the responses on the top of the platform under these random wave forces are quickly and accurately calculated by the DSM method. The calculated responses are taken as the input signals of FNN adaptive predictive inverse controller, through which the control force for the imminent time of the platform is forecasted. Meanwhile, a feedback adaptive predictive inverse controller is designed for the purpose of disturbance canceling and error reduction. For demonstration of the anti-disturbance capability of the control scheme, the control efficiency of the jacket platform with external disturbances of random wind loads has also been investigated in this paper. The numerical results obtained in this paper show that the DSM based FNN APIC scheme which has excellent anti-disturbance capability is feasible and effective, and can finally overcome the time delay.