Research On Dynamic Positioning System Of Deep Water Semi-submersible Drilling Rig

Since the last century, the oil and gas resource on land and inshore has been explored andalmost run out. Now the offshore oil and gas exploration is focused on deep water ocean. Asthe water is going deeper and deeper, the moored semi-submersible platform and jack upplatform can’t meet the operation requirement anymore from whatever the economic or safetyaspect. Big variable load, precise positioning, not affected by water depth and large deck areaetc, all those advantages make the DP based drilling rig become the best choice for deep waterocean exploration. DP was used in ocean resource exploration since it was born. Therefore,research on DP of mooring system has significant value. Due to the complication of the realsea environment and the variable load, to make a precise model and design a good controlleralways is the core work for DP research.The semi-submersible drilling rig is set as research object in this paper. First, amathematical model was established according to the certain ocean environment. The Kalmanfilter was employed to separate the low-rfequency and high-frequency component rfom themotion state. Then, the DP controller is designed by implementing optimal control theory onthe low-frequency signal. The self-tuning filter is used to improve the Kalman filterperformance base on the ifxed model. It includes the following jobs.1.A mathematical model of the rig dynamic positioning system is studied. Firstly, theconcept of rig DP system is introduced in this paper. Then, the description of general solutionand research recently for DP control system is presented. The rig’s dynamic model,hydrodynamic model and the environment force (wind, wave, current) model was discussed.The dynamics of the rig is separated into a low-frequency model and high-frequency model.Base on the studies an integrated mathematical model for the rig dynamic positioning systemwas provided.2.Design the Kalman filter. Establish the Kalman state observer according to thedynamic model, adopt Kalman filter to estimate the high frequency and low frequencymovement statement of the rig base on the measured signal, the low-frequency part to be usedfor rig motion controller design.3.Design a DP controller. Combining with forward feed control of the wind, design theLQG dynamic positioning controller based on the estimated low-rfequency signal fromKalman filter. The mathematic model for thruster was established, and the thrust forceallocation plan was presented according to the dynamic model.4.The self-tuning iflter was introduced to improve the Kalman ifltering. The self-tuning filtering was used to on-line identify the parameter of the high-frequency motion model, andcorrect the low-frequency motion signal. Combining with the high-rfequency self-tuning filterand low-frequency Kalman filter we get an integrated filter. The low-frequency signal fromthe integrated filter output was used as the input of the control algorithm. Simulation wasdone to proof the controller’s performance.