Due to the constant growth of global maritime shipping, the optimization of grabs used for unloading bulk carriers is more and more relevant. Especially for a coarse material like limestone, the average payload can be improved, by altering the grab design. It is suggested that the penetration and cutting depth of a grab can be improved, which will have a positive effect on the payload. This research is limited to the contact surface of the grab with the bulk material, the knife. Subsequently, the following research question is answered in this research:

How to enhance the penetration and cutting depth considering coarse cohesionless material?

This is done, by implementing a DEM model using the bulk calibration approach [1], several in situ
and laboratory experiments are performed to measure certain macro properties. The experiment is numerically replicated, and micro parameters are changed iteratively until the predicted macro response matches the desired results. A comprehensive study of similar papers considering the penetration resistance using a DEM model is executed, which resulted in selecting the penetration resistance and porosity as target variables. The particle shape of a DEM model has a significant influence on the bulk material behavior. The novelty of this research consists of doing a preliminary analysis before the final calibration process, in which several polyhedral and (multi)-spherical particle shapes are compared. For investigating the penetration resistance, a new test setup for in situ use is developed and constructed. The implementation of this setup proved to be successful, eleven data samples were collected. For the bulk density, an in situ and laboratory setup are compared, of which the laboratory setup proved to be more reliable. Several additional laboratory experiments are executed for the particle density, Particle Size Distribution (PSD) and wall friction. Using the results of the experiments, a DEM model is created, with which a preliminary analysis is done using extreme values of the calibration variables. Four spherical and four polyhedral shapes are compared, which resulted in the two-spherical shape begin the most optimal option. All the polyhedral shapes proved to be not fitting, especially due to their underestimation of the penetration resistance. Due to the high variance of the experimental penetration tests, two parameter sets with identical porosity are calibrated: one with a low resistance target, and one with a high resistance target. Using Design of Experiments (DoE) in combination with surrogate modelling with a linear regression, several optimal parameters sets are created, in which the uncertainty is considered as well. After selecting two optimal parameter sets, the simulations are verified by comparing the experiments and simulations qualitatively. Additionally, a quantitative verification using the stiffness of the force data of each time-step is performed as well. Using the calibrated models, an additional DEM model is created with a bigger particle bed, to investigate the penetration and cutting resistance. By altering the geometry of the knife, a decrease in resistance was achieved, mainly in the cutting phase. It is recommended to consider the flowing behavior of limestone as well, with which a complete material model can be calibrated and validated, and full-scale simulations can be performed.

Over 5000 offshore wind turbine generators have been installed in the past decades. 80% of these wind turbines have a monopile foundation. Monopiles are hollow steel tubular piles with a diameter of up to 10m. The first phase of the monopile installation consists of the self-weight penetration into the seabed due to the mass of the monopile. Followed by the placement of the hammer, which installs the piles to the designated depth. A scour protection is deposited near the piles to avoid erosion (scour holes) around the monopiles. The scour protection consists of rock material with different size gradations and densities. Generally, the scour protection is installed in two campaigns. The first campaign consists of the installation of smaller rocks (filter layer), followed by the installation of bigger rocks (armour layer). Between these two campaigns, the monopile is installed through the filter layer. To achieve a more time-efficient installation sequence, in some projects a full scour protection (single layer) is deposited prior to the installation of the monopile. However, the technical feasibility of monopile installation through a scour protection has not been researched. Within practice, difficulties arise during the installation of monopiles through a single layer of scour protection. This leads to timely and costly delays.
This thesis contributes by use of simulations, to study the technical feasibility of monopile installation through scour protection. With a focus on the scour protection behavior in the first phase of the installation, where the monopile penetrates the seabed due to its mass, so-called self-weight penetration.
During monopile self-weight penetration, physical processes within the subsoil, scour protection, and monopile wall create penetration resistance. The most important physical processes during monopile self-weight penetration are identified using literature:
• The development of tip resistance due to the scour protection layer.
• The development of shaft resistance due to the scour protection layer.
• The self-weight penetration due to the weight and dimensions of the monopile.
• The interaction between the monopile wall and the scour protection rock and inter-rock interaction (interlocking, rotation, breaking, and protrusion of the subsoil).
The Discrete Element Method (DEM) is considered most suitable for the representation of the scour protection behavior and is an effective method for addressing engineering problems in granular materials. The most important reason for using DEM is due to the discontinuous behavior of the scour protection material.
The Discrete Element Method (DEM) is considered most suitable for the representation of the scour protection behavior. DEM is an effective method for addressing engineering problems in granular materials. The most important reason for using DEM is due to the discontinuous behavior the scour protection materials.
To set up the simulation, calibrated DEM input parameter sets are required to recreate reliable bulk behavior. These parameter input sets are found with the use of two Key Performance Indicators (KPIs) of the bulk material: the porosity and the angle of repose. These KPIs are included because they represent the inter particle frictional resistance to movement. With the calibrated input sets combined with data from a case study, the Full-Scale Monopile Simulation is set up. Consistent with the case study, the simulation is designed using a monopile with a tip diameter of 8m and a wall thickness of 0.08m. The seabed consists of a sandy subsoil and a scour protection on top.
To achieve a realistic stable simulation with a reasonable computational time, experiments are executed. Within an experimental design, a reference simulation is developed, followed by a parameter sensitivity analysis. The most important conclusions of the parameter sensitivity analysis are:
• When using an 8m diameter monopile, a minimum simulation domain of 12*12m2 is recommended. Resulting in a minimum of 2m space around the monopile.
• A subsoil is required for successful penetration through the scour protection layer.
• The subsoil input parameters are of significant influence on the penetration behavior of the monopile, due to the rock penetration into the subsoil.
• The scour protection thickness can be included as a variable parameter.
To verify whether the reference simulation properly represents practice, a qualitative validation is executed using the case study. Successful penetration is achieved within both the reference simulation and the case study. However, the current simulation does not present a comparable depth over time correlation. An underestimation of the penetration resistance is observed. Increasing the rolling resistance and therefore limiting the angular velocity of the particles results in higher penetration resistance. Restricting the angular velocity of the scour protection completely, results in an unsuccessful penetration.
Further research should focus on representative penetration resistance of the scour protection, combined with an alternative and more representative inclusion of the subsoil. Furthermore, the validation should be elaborated using adequate data from practice.

Bulk handling grabs are often used for unloading bulk materials such as iron ore and coal from bulk carrying ships. To improve and compare virtual simulations of new grabs, accurate real life kinematic data of grabs is required. Moreover kinematic data can potentially provide insight in current use of grabs, which can be interesting information for both manufacturer and customer of the grab. The possibilities towards implementation of a sensor system on bulk handling grabs are investigated by selecting a suitable sensor and measuring kinematics and operations of imitated grab movements. This research focuses on investigating kinematics of multi rope grabs with dual scoops. A sensor selection procedure is followed resulting in the selection of an Inertial Measurement Unit (IMU) sensor with Global Navigation Satellite System (GNSS) receiver. In several experiments the sensor data is compared with reference data of imitated grab movements assumed as ground truth. When using multiple IMU’s (one on each grab scoop) providing orientation data, the opening angle can be determined. The results of the experiments indicate that the position data provided by the sensor can be used for classification of activities, but less for high precision movement trajectory reconstruction. Position and orientation data of imitated grab movements can be interpreted to generate performance indicators that describe grab operations. Experiments showed that performance indicators of imitated grab movements can be identified accurate up to an error of two seconds when comparing known performance indicators and sensor data analysis. When a high level of accuracy of the kinematic data is reached, performance indicators describing grab operations can be determined more accurately.

Bulk materials such as iron ore transported by bulk carriers are commonly unloaded by the use of grabs. The unloading process is time consuming and increasing the unloading capacity is of big importance for terminals. A validated co-simulation developed by the TU Delft in cooperation with Nemag enables to test grabs in the virtual environment. In practice a variation in unloading capacity is seen, due to dynamic operational conditions, such as different crane operators and bulk surface irregularities. To improve grab design the influence of operational conditions on the unloading capacity is investigated. Based on those results a design modification has been proposed and showed promising results.

Grabs are often used for unloading bulk carriers that transport iron ore cargoes around the globe. The unloading process is time-consuming, and terminals strive to maximise their turnover capacity. Grab performance is a combination of maximum crane capacity, grab design and bulk material behaviour. Due to bulk uncertainties occurred by varying physical bulk properties, moisture content and consolidation, grab performance is hard to predict. To incorporate the bulk variability in the design process of grabs, or other large-scale bulk handling equipment, an optimization framework is proposed in which equipment is systematically optimized. With the framework an iron ore grab is optimized for handling iron ore pellets and iron ore fines. By a minimum number of experiments grab design was improved, increasing the turnover capacity for iron ore pellets by 12% and iron ore fines with 8%, and reducing the effect of varying bulk conditions. Optimization methods such as Latin Hypercube sampling,surrogate modelling and genetic algorithms proved to be successful for grab optimization. Further research of implementing the framework in the design process, and surrogate modelling is recommended.