QD
Q. Dingeldein
info
Please Note
<p>This page displays the records of the person named above and is not linked to a unique person identifier. This record may need to be merged to a profile.</p>
2 records found
1
Dynamic Positioning systems are used to keep a vessel at a desired position and heading while it is exposed to environmental disturbances. In this thesis, the potential of adding predicted wave-induced disturbances to a Model Predictive Control framework is investigated. The main idea is that the controller can react more anticipatively to wave loads, instead of only reacting after the vessel has already moved away from the reference position.
A numerical simulation framework was developed for a model-scale vessel in 3 Degrees of Freedom. The vessel motions in surge, sway and yaw were included, together with a thrust allocation system and an MPC-based controller. Irregular waves were generated using a wave spectrum, while the hydrodynamic wave loads were calculated using Response Amplitude Operators and Quadratic Transfer Functions from Nemoh. First-order wave forces were used to describe the oscillatory wave-frequency excitation. The second-order wave loads were used to determine the mean wave drift force, which was included as predicted disturbance in the wave-aware MPC.
The wave-aware MPC was compared with a baseline MPC without wave prediction. The controller performance was evaluated using Mean Absolute Error, Root Mean Square Error, 95th percentile error and energy consumption. A seed analysis was also performed to investigate the influence of different irregular wave realizations.
The results show that the predicted mean second-order wave force can improve the mean station-keeping offset. In the seed analysis, the mean horizontal offset was reduced by approximately 26\% for the tested moderate and heavier sea states. This shows that the controller is able to use the predicted wave drift force for compensating the slowly varying drift component. However, the total station-keeping performance did not improve robustly. The total RMSE remained almost unchanged in the moderate sea state and became worse in the heavier sea state. The oscillatory part of the motion also increased on average.
It is therefore concluded that predicted wave-induced disturbances can improve Dynamic Positioning performance, but only to a limited and specific extent in the current implementation. The predicted mean wave drift force is useful for reducing the mean offset, but it is not sufficient to improve the complete vessel motion around the reference position. The developed wave-aware MPC should therefore be seen as a promising first step, but not yet as a generally better controller than the baseline MPC. ...
A numerical simulation framework was developed for a model-scale vessel in 3 Degrees of Freedom. The vessel motions in surge, sway and yaw were included, together with a thrust allocation system and an MPC-based controller. Irregular waves were generated using a wave spectrum, while the hydrodynamic wave loads were calculated using Response Amplitude Operators and Quadratic Transfer Functions from Nemoh. First-order wave forces were used to describe the oscillatory wave-frequency excitation. The second-order wave loads were used to determine the mean wave drift force, which was included as predicted disturbance in the wave-aware MPC.
The wave-aware MPC was compared with a baseline MPC without wave prediction. The controller performance was evaluated using Mean Absolute Error, Root Mean Square Error, 95th percentile error and energy consumption. A seed analysis was also performed to investigate the influence of different irregular wave realizations.
The results show that the predicted mean second-order wave force can improve the mean station-keeping offset. In the seed analysis, the mean horizontal offset was reduced by approximately 26\% for the tested moderate and heavier sea states. This shows that the controller is able to use the predicted wave drift force for compensating the slowly varying drift component. However, the total station-keeping performance did not improve robustly. The total RMSE remained almost unchanged in the moderate sea state and became worse in the heavier sea state. The oscillatory part of the motion also increased on average.
It is therefore concluded that predicted wave-induced disturbances can improve Dynamic Positioning performance, but only to a limited and specific extent in the current implementation. The predicted mean wave drift force is useful for reducing the mean offset, but it is not sufficient to improve the complete vessel motion around the reference position. The developed wave-aware MPC should therefore be seen as a promising first step, but not yet as a generally better controller than the baseline MPC. ...
Dynamic Positioning systems are used to keep a vessel at a desired position and heading while it is exposed to environmental disturbances. In this thesis, the potential of adding predicted wave-induced disturbances to a Model Predictive Control framework is investigated. The main idea is that the controller can react more anticipatively to wave loads, instead of only reacting after the vessel has already moved away from the reference position.
A numerical simulation framework was developed for a model-scale vessel in 3 Degrees of Freedom. The vessel motions in surge, sway and yaw were included, together with a thrust allocation system and an MPC-based controller. Irregular waves were generated using a wave spectrum, while the hydrodynamic wave loads were calculated using Response Amplitude Operators and Quadratic Transfer Functions from Nemoh. First-order wave forces were used to describe the oscillatory wave-frequency excitation. The second-order wave loads were used to determine the mean wave drift force, which was included as predicted disturbance in the wave-aware MPC.
The wave-aware MPC was compared with a baseline MPC without wave prediction. The controller performance was evaluated using Mean Absolute Error, Root Mean Square Error, 95th percentile error and energy consumption. A seed analysis was also performed to investigate the influence of different irregular wave realizations.
The results show that the predicted mean second-order wave force can improve the mean station-keeping offset. In the seed analysis, the mean horizontal offset was reduced by approximately 26\% for the tested moderate and heavier sea states. This shows that the controller is able to use the predicted wave drift force for compensating the slowly varying drift component. However, the total station-keeping performance did not improve robustly. The total RMSE remained almost unchanged in the moderate sea state and became worse in the heavier sea state. The oscillatory part of the motion also increased on average.
It is therefore concluded that predicted wave-induced disturbances can improve Dynamic Positioning performance, but only to a limited and specific extent in the current implementation. The predicted mean wave drift force is useful for reducing the mean offset, but it is not sufficient to improve the complete vessel motion around the reference position. The developed wave-aware MPC should therefore be seen as a promising first step, but not yet as a generally better controller than the baseline MPC.
A numerical simulation framework was developed for a model-scale vessel in 3 Degrees of Freedom. The vessel motions in surge, sway and yaw were included, together with a thrust allocation system and an MPC-based controller. Irregular waves were generated using a wave spectrum, while the hydrodynamic wave loads were calculated using Response Amplitude Operators and Quadratic Transfer Functions from Nemoh. First-order wave forces were used to describe the oscillatory wave-frequency excitation. The second-order wave loads were used to determine the mean wave drift force, which was included as predicted disturbance in the wave-aware MPC.
The wave-aware MPC was compared with a baseline MPC without wave prediction. The controller performance was evaluated using Mean Absolute Error, Root Mean Square Error, 95th percentile error and energy consumption. A seed analysis was also performed to investigate the influence of different irregular wave realizations.
The results show that the predicted mean second-order wave force can improve the mean station-keeping offset. In the seed analysis, the mean horizontal offset was reduced by approximately 26\% for the tested moderate and heavier sea states. This shows that the controller is able to use the predicted wave drift force for compensating the slowly varying drift component. However, the total station-keeping performance did not improve robustly. The total RMSE remained almost unchanged in the moderate sea state and became worse in the heavier sea state. The oscillatory part of the motion also increased on average.
It is therefore concluded that predicted wave-induced disturbances can improve Dynamic Positioning performance, but only to a limited and specific extent in the current implementation. The predicted mean wave drift force is useful for reducing the mean offset, but it is not sufficient to improve the complete vessel motion around the reference position. The developed wave-aware MPC should therefore be seen as a promising first step, but not yet as a generally better controller than the baseline MPC.
Integrated Method for Mapping Shipyard Layouts
Obtaining an Insight into the Current State Building Process and Employee Movements
Bachelor thesis
(2023)
-
S.E. Jongbloed, P.J.M. Lebbink, D.C. Pap, Q. Dingeldein, M. van der Schaaf, C.H. Thill
There exists no general method for analysing the efficiency of a shipyard layout through the mapping of building processes and employee movements. Therefore an Integrated Method for mapping shipyard layouts is developed. This is done by implementing multiple ways of analysing and optimising industrial processes, as found in the literature. Usage of the developed Integrated Method starts with the analysis of different Lean wastes. Next, it is important to get a good understanding of the shipyard’s building processes through communication with employees and Value Stream mapping. Subsequently, the found processes can be visualised on a map using Rainbow Models. Finally, these Rainbow Models are verified with data from GPS tracking. Using this combination of methods as one Integrated Method helps to get a good picture of how different shipyard processes run, where they take place and which bottlenecks are present. With this information, shipyards can be made more efficient and future-proof.
...
There exists no general method for analysing the efficiency of a shipyard layout through the mapping of building processes and employee movements. Therefore an Integrated Method for mapping shipyard layouts is developed. This is done by implementing multiple ways of analysing and optimising industrial processes, as found in the literature. Usage of the developed Integrated Method starts with the analysis of different Lean wastes. Next, it is important to get a good understanding of the shipyard’s building processes through communication with employees and Value Stream mapping. Subsequently, the found processes can be visualised on a map using Rainbow Models. Finally, these Rainbow Models are verified with data from GPS tracking. Using this combination of methods as one Integrated Method helps to get a good picture of how different shipyard processes run, where they take place and which bottlenecks are present. With this information, shipyards can be made more efficient and future-proof.