Gyroscopic Motion Control During The Lowering Procedure of Monopile Installation

Abstract

Due to the ever increasing demand of energy and the urgency in which this has to be produced using renewable energy resources the request for offshore wind farms is larger than ever. As a result of the competitive nature of the energy sector it is important to keep the cost of energy low in order to keep an edge over the competition. Therefore, to keep wind energy economically viable in the future the cost of producing energy should be lowered. One way this can be done is by lowering the initial costs of installation, as this is relatively high. In this thesis an installation method of monopile foundations will be introduced, which has potential to reduce costs of the installation of wind farms. More specifically, this thesis will focus on lowering of monopiles to the seabed. Problems arise when the monopile hanging on the crane wire is excited by waves hitting the submerged part of the monopile. This can cause excessivemotion of the monopile. Particularly if resonance occurs, which is the case when the wave frequencies and the natural frequencies of the monopile coincide. This could lead to catastrophic failure if no mitigating measures are taken. The most common method to counteract this phenomenon is to wait until weather conditions allow for further continuation of operations, which can consume lots of valuable time and thus increasing costs. Therefore, this thesis will focus on optimizing the workability of offshore installations of monopiles by using gyroscopicmotion actuators. These actuators, called controlmoment gyroscopes (CMG), will reducemotion of the monopile by applying a moment counteracting the moment created by the waves. This system is modelled in MATLAB. The model consists of the monopile suspended on the crane wire, a gripper and the CMG. The equations of motion are determined using the Lagrangian method and linearized around their equilibrium position. The equations of motion are then written in state space representation in order to study the behaviour of the system and to tune the PD controller. The PD controller was tuned using the loop shaping method. The MATLAB model is used to tune the controller. However, the results will be examined using OrcaFlex. Therefore, the same system is also modelled in OrcaFlex, a commercial time domain simulations package, in which an external function is built using Python to represent the CMG and the corresponding PD controller of the CMG moment. The whole system is then subjected to a sea state with a significant wave height of 2 meters and a wave period range from 5-8 seconds. The 3 hour simulations are run for each period step of 0.5 seconds from 5 to 8 seconds. From each simulation the most probable maximum is calculated. This is done for four variants of the system…