When a grounding incident occurs, the officer of the watch is usually unable to assess the severity of the damage. Most ships do have damage response procedures in place. In case of military ships these are well developed, where a well-trained crew, in the form of damage parties, is available to respond adequately. For non-navy ships, this is hardly possible, mostly due to crew restrictions. But even for navy vessels, underwater damage is hard to assess because of accessibility issues of flooded compartments. This is undesirable because it needs to be known which compartments have flooded in order to assess the situation with respect to residual buoyancy and stability. Based on the outcome, the officer in charge can decide whether to evacuate or to stay on board and control the situation. For establishing which compartments will flood the extent of the raking damage must be known. It is proposed that the accelerations (or rather, decelerations) measured on board during a grounding incident can be used to make a prediction of the extent of raking damage.

A three degree of freedom external dynamics calculation model for ship grounding has been made, which is used to predict the extent of bottom raking damage and the grounding force with associated decelerations of the ship in terms of surge, sway and yaw. When the ship decelerates, caused by a grounding event, these motions can be logged through continuous ship motion measurements. From these deceleration time traces it is possible to calculate the bottom damage path through a double integration with respect to time. Moreover, with these decelerations, the location of the raking damage along the ship’s bottom can be calculated as well. A straight forward method is proposed, which is validated against data on ship grounding external dynamics from earlier research and data from large-scale grounding experiments, both carried out in the nineties of the previous century.

In the past decades the understanding of fracture of ductile material has increased substantially. Research has shown that the onset of fracture highly depends on the full state of stress inside the material. Better understanding of fracture resulted in more advanced and complex fracture models being conceived, allowing researchers to predict fracture in ductile solids with improved accuracy.


Nonlinear finite element analysis is a powerful tool at the disposal of researchers to predict the response of ship and offshore structures. When it comes to simulating accidental loads, such as collisions, often basic criteria are applied to include fracture in a finite element model. However, accurate fracture prediction is of great importance to determine the ice resilience of vessels, or to obtain reliable estimates of the sustained damage due to maritime collision. This research focuses on the latter.


This thesis is concerned with bridging the gap between the recent developments in fracture prediction and the application of failure criteria in finite element analysis of ship collision. A selection of recently published fracture models has been made and experiments have been conducted on S235 structural steel for calibration and validation of these models. Four small scale experiments have been conducted. These experiments serve a dual purpose: first, to gather information on the material behaviour during deformation. Second, to obtain information on the effect of different stress conditions on fracture. An iterative method has been employed to accurately model the material behaviour. For the calibration of the fracture models a method has been conceived and applied that takes into account the full histories of stress and strain during the deformation process up to fracture.

Before application as failure criteria for finite elements, the calibrated fracture models require a correction based on the size of the elements: a modification to an already existing theoretical framework has been proposed and applied to obtain element-size dependent failure criteria.


A large scale drop tower experiment has been designed to simulate a so called raking damage scenario. This experiment has been conducted on the same material as the small scale experiments and serves as a validation for a finite element model that has been created using the information on the material behaviour obtained from the small scale experiments. The different failure criteria have been implemented into the commercial finite element package LS-DYNA and have been applied to the finite element model. The results have been compared to the results of the raking damage experiment.


It was concluded that the application of complex multi-parameter failure models in analysis of maritime collision does not necessarily provide an improvement over conventional fracture prediction methods. The inability of shell elements to accurately describe strain concentrations and the effect of the element size introduce uncertainties that overrule the benefits of stress-state dependent prediction of element failure.

In order to monitor elliptical fatigue crack growth in ferromagnetic steel using magnetic methods, analternative to the self magnetic flux leakage method must be derived as elliptical cracks can grow to significant sizes before they reach through the thickness of the plate material. An approach is sought by translating subtle changes in magnetisation back to the Villari effect, a phenomenon which depicts how applied stress induces changes in magnetisation in ferromagnetic objects. Since the magnitude of these changes in magnetisation is small, other nonlinear effects of similar order such as magnetic relaxation and hysteresis are identified, measured and quantified preliminarily.

The magnetic behaviour in this project is assumed to be quasi-static, and derivations of the expressions for the magnetic field around simple geometric shapes are provided in order to understand magnetic behaviour and verify the outcome of the numerical simulations. It is shown that the numerical simulations
produce identical magnetostatic induction fields as the analytically derived expressions when using a sufficiently refined mesh.

An attempt is made to measure long-term magnetic relaxation by subjecting a solid prolate spheroid to a continuous uniform background field for periods of an hour while trying to measure differences in the induction field at a fixed distance. Short-term relaxation, the time it takes for an object to reach a certain
magnetisation when the background field is abruptly changed, is also investigated. It is concluded that both effects could not be successfully measured using the current setup. In order to draw proper conclusions, further research into this topic should be conducted using more accurate equipment for extended periods of time.

Upon investigation it is discovered that it can not be assumed that the steel specimens exhibit a uniform permanent magnetisation. A self-developed method is introduced through which non-uniform magnetisation in three directions can be calculated by means of inversion using a set of magnetic induction field measurements in a plane below the specimen when the background field is zero. These measurements are translated to magnetisation using a set of higher order square Gaussian distribution functions that are spaced in a grid over the domain of the test specimen in order to vary the magnetisation locally.

Literature that shows comparable results regarding description of non-uniform permanent magnetisation using an array of induction field measurements has not been found. The concept of hysteresis is introduced and a method is presented through which the parameters of the Jiles-Atherton hysteresis model can be determined using parameter fitting in combination with a forward numerical model created in COMSOL. Closure of minor loops require modifications to the original JA equations which are implemented in the forward model. The numerical model is encapsulated within the Shuffled Leaping Frog parameter optimisation algorithm in order to compute the correct hysteresis
parameters. It is found that it is possible to successfully determine the parameters of multiple specimens using weak magnetic fields, and therefore minor loops, which is unparalleled in literature.

Eventually, the Villari effect is introduced and an attempt is made to measure and model the effect usingan extension of the Jiles-Atherton model proposed by Naus. Experiments have shown that using this methodology the magnetostriction parameters can be succesfully obtained. A recommendation is
provided into how these results can be implemented in crack-propagation models in future research.

The structures which HMC installs offshore are fabricated onshore and subsequently moved onto a barge or ship, seafastened and then transported to the offshore location. The process of moving a structure from the onshore quayside to the barge or ship is called the load-out. This load-out can be performed by lifting, skidding or using a trailer (SPMT). This thesis research focused only on a skidded load-out onto a barge.
During the load-out the weight of the jacket or topside is gradually transferred from the quay to the barge. The barge gradually takes more of the load so ballast water needs to be continuously pumped or discharged depending on the location of the structure and the location of the ballast tank concerned. Improper ballasting during this process will cause the alignment between the quay and the barge to be disrupted which in turn causes peak loads in the topside or jacket and the barge. It is questioned if there are more suitable ballasting methods or a structural solution in order to lower these peak loads? It is also modeled what the effects are of quayside stiffness and the best method to model this stiffness.
Therefore a 2-D representation of the entire load-out is made. This model will be made using the finite element method, via a numerical model, in MATLAB. A base case load-out of a topside will be applied to this model. Using this model, optimizing the ballast configuration will be researched. Several different criteria for the optimization were tested and its different effects on the forces during the load-out were researched and quantified. The structural solution of relocating the skidbeams to an area of lower deck stiffness was also tested and the results studied. The effects of the quayside stiffness and modelling methods were also quantified using the 2-D MATLAB model.
The conclusion derived from the optimizations is that there are other ballast configurations which perform better in reducing the peak forces experienced during the load-out. The key to these optimizations is that they keep the barge-quay alignment as perfect as possible. If a critical element is present in the load-out the ballast configuration can be adjusted to lower the forces in this specific element. The results of the simulation in which the skidbeams were relocated show that this approach has no beneficial effects in reducing the forces during the load-out, mainly due to the presence of the transverse bulkheads in het barge. Furthermore for the modelling of the quayside it was proven that especially when using a low stiffness quayside, modelling the quayside without taking into account the foundation layer stiffness is inaccurate and can lead to lower forces in the model than which occur in reality.

Research into multi-axial fatigue requires knowledge on the through time and thickness development of the propagating crack front. In order to provide this knowledge methods must be developed to monitor crack initiation and growth during a fatigue test. The acoustic emission method has been recognised as a possible means to this end.
Fatigue is a failure mode that is characterised by the accumulation of damage during cyclic loading. Fatigue initiates on the microscale as dislocation movement, and subsequently develops into crack growth and eventual failure. In the assessment of fatigue damage it is critical to estimate the remaining lifetime. Prediction models have been developed for uni-axial loaded joints. However in the field of multi-axial fatigue these models are inadequate. To improve upon these models knowledge is required on the development of the fracture during cyclic multi-axial loading.
The acoustic emission method is a procedure for non-destructive testing that is based on detecting elastic stress waves originating from changes in the microstructure of an object. In the industry two common applications of acoustic emission exist. One of these aims to extract insights into the severity of damage from waveform features, the other aims to localise and correlate acoustic events. Novel applications focus mainly on the improvement of these two fields. A state of the art method is proposed which aims to localise acoustic sources in small scale volumes as a means of imaging the crack front.
Localisation of the sources is performed using a time difference of arrive scheme, which is also known as multilateration. The accuracy of this scheme is highly dependent on the accuracy of the input parameters. Therefore the times of arrival, receiver positions and speed of sound must be established meticulously.
Extensive processing is required to transform the recorded acoustic emission data into a representation of the crack propagation. For this purpose a post-processing system has been developed which automates time of arrival picking, event building and event localisation. The developed system is an extension to the AMSY-6 acoustic emission measurement system developed by Vallen Systeme.
In order to validate the accuracy of the extended measurement system an experiment has been designed. In a volumetric geometry sources are simulated in order to assess the bias and the variance, additionally noise is added to the simulation to evaluate the sensitivity of the procedures.
The study has shown that, under ideal noise conditions, the precision can amount to a cluster of about a millimetre in size. Additionally the bias of this cluster can be reduced to a minimum by means of careful calibration of the receiver positions and the speed of sound. Both of these observations indicate that an accurate representation of the propagating crack front may be obtained through a regression model, if the conditions are known and within acceptable limits. Regarding non-ideal conditions the analysis of the noise sensitivity has shown that the accuracy is relatively stable as long as the signal to noise ratio is kept above SNR ≥ 30 dB.
In conclusion this has shown that monitoring the propagation of the crack front using volumetric acoustic emission source localisation may be possible if the conditions are right. This means that the signal to noise ratio must be kept above SNR ≥ 30 dB. If this is the case, and if enough events are recorded to perform regression both the bias and the variance should pose a problem to monitoring crack growth.