Intelligent Active Control Of Structural Vibration For Jacket Offshore Platforms

With the development of ocean technology, more and more extremely large and long flexible offshore platforms used for various purposes are built in hostile ocean environments. In general, this kind of platform is a nonlinear distributed parameter system and its natural frequency fall closer to the dominant wave frequencies with the increase of water depth. What is more, its structure is very complex and the external wave force on the platform is uncertain. Thus, these platforms are prone to excessive wave-induced oscillations under both operating and extreme conditions. It results in injury of human health, structural fatigue and failure, reducing the accuracy and reliability of the equipments on the platforms, lowering the serviceability and survival of the platforms. Hence, from the view of comfort for crew, the durability of the equipments and the safety for the structures, it is essential to bring the response under control effectively. However, it is uneconomical by stiffening the strength of the platforms to suppress the dynamic response passively. Moreover, it won't meet the prospective purposes due to the complexities of the structures and the uncertainties of the environmental loads, etc.Soil-structure interaction (SSI) will strongly affect the dynamic characteristics of the offshore platforms, which in turn affects the performance of the control system. The natural characteristics of the offshore platforms considering the soil-structure interaction are studied. The piles of the platforms are modeled as dynamic Winkler beams embedded in a deep homogeneous stratum through continuously distributed linear springs and frequency-dependent dashpots. The dynamic analysis is accomplished with the FEM package ANSYS. The methods of simplifying for jacket platform are discussed for the purpose of structural control. The research has laid a solid foundation for the further research of the structural control for offshore platforms.Depending on the control methods, two main methodologies related to conventional control are commonly used in depressing the dynamic response of offshore platforms, namely, passive control and active control. Although passive control offers some benefits, its effectiveness is limited. With the developments of modern control theory and techniques, extensive investigations have shown that the use of active control can reduce the structuralvibration further over that of the passive control.During the past three decades, extensive research efforts have been made on the control of structural vibration induced by severe environmental loads and a number of achievements can be available in the literature of developing control strategies for offshore platforms. For complex, nonlinear structures, such as offshore platforms, it is impossible to get the accurate models of the structures to be controlled and the external loadings. Although traditional control methods have many attractive features, they may run into problems when controlling systems that are difficult to model. This dissertation intends to introduce an intelligent active structural control strategy for offshore platforms.The neural network based active control of structural vibration for offshore platforms is accomplished mainly in two ways, so called supervised control and predictive control. The neural network is used as a controller or an emulator to get the active control forces of the next time steps, thus realizing the active control for offshore platforms. The results obtained have demonstrated that the proposed neural network based active control algorithm is effective and feasible, thus improving both serviceability and survival.Fuzzy logic theory was developed by Lotfi A. Zadeh in 1965 to deal with the nonlinearity, uncertainty and imprecision that is often present in real world applications. The fuzzy control needs only a few simple inference rules based on the engineering experience instead of the accurate mathematical models of the controlled structures and the environmental loads. Fuzzy logic control system has been utilized fo...