Page 10 - IJOCTA-15-2

Artificial intelligence-assisted station keeping for improved drillship operations
the vessel’s dynamics. The accelerometers mea- m-long monohull vessel, specifically designed for
sure the vessel’s acceleration, which is used to deepwater environments. To ensure precision dur-
calculate its velocity and displacement, thereby ing operations, the drillship is equipped with an
determining the vessel’s change in position. The advanced Dynamic Positioning System (DPS),
AI-controlled system operates by processing real- enabling it to maintain position and orientation.
time data from these onboard sensors, mainly the One of the notable structural features is its rect-
information from the INS. Based on this data, angular moonpool. It passes through the hull for
the AI algorithm predicts the required thruster equipment such as drilling risers and blowout pre-
force to counteract external disturbances caused venters. The moonpool in this vessel has cross-
by environmental forces such as waves, wind, and sectional dimensions of 24.8 m × 12.8 m and in-
◦
currents. Once the required force is determined, cludes a 90 cutout angle, enhancing its efficiency
the system activates specific thrusters to gen- in facilitating drilling operations while minimizing
erate the calculated force and return the vessel hydrodynamic resistance. The key geometric and
to its intended position. This integration of AI hydrodynamic parameters are detailed in Table
with the INS creates a sophisticated real-time 1. These parameters provide critical insights into
control mechanism. The system dynamically ad- the drillship’s structural design and operational
justs the thrust forces applied by the thrusters, characteristics, forming the basis for the study’s
ensuring the vessel stays within its permissi- subsequent numerical simulations and analyses.
ble position limits and achieves precise station-
The numerical model of the drillship was
keeping. The combined use of real-time sensor
developed based on key characteristics outlined
data and AI-powered control allows the system 14
by. This model was validated for accuracy, en-
to respond quickly and effectively to external dis-
suring its reliability for advanced analysis. The
turbances, significantly improving the accuracy,
problem was solved numerically using the AN-
reliability, and operational performance of the
SYS Explicit Dynamics solver, with the resulting
drillship’s station-keeping capabilities. This ad-
simulation data used to generate a database that
vanced methodology (Figure 2) ensures that the
supports the development of the AI controller’s
vessel remains stable and functional, even in chal-
numerical code. Figure 3 shows a visual repre-
lenging maritime environments.
sentation of the drillship’s working model.

Figure 3. A 3D model of the drillship

3.1. Simulation environment

The drillship’s response was analyzed for various
wave directions to understand its behavior com-
prehensively. Eight wave scenarios were consid-
ered, as shown in Figure 4:
Figure 2. Methodology
Abbreviation: AI: Artificial intelligence (i) Head sea,
(ii) Following sea,
3. Numerical analysis (iii) Beam sea (Port),
(iv) Beam sea (Starboard),
The drillship Glomar C. Luigs was chosen as the (v) Stern quarter (Port starboard),
reference vessel for this study. 14 It is a 231.5 (vi) Bow quarter (Port starboard).
205