The impact of a pipeline by a dropped object has been considered to consist of four distinct impact components: the dropped object, pipeline protection, the soil bed and the pipeline itself. The effect of these components as energy absorbers and the effect on system response has been investigated. Quasi-static and dynamic testing has been earned out to investigate the interaction between the various impact components. Quasi-static testing has been widely used to develop initial predictions, since closer observation of interaction is easier. The validity of applying these predictions to dynamic situations has been addressed using results from dynamic impact testing. The Dropped Object: Two areas have been investigated which address the dropped object within the impact system: the dropped object's impact profile and its deformability. Testing has been carried out to study the effect of typical loading profiles. Research has shown that the dropped object profile significantly effects the pipe response; a cone shaped indentor generates deformation with far less energy than either a wedge or a patch shape. The applicability of a method to predict the interaction between two deforming structures, using a method of shared energy, has been investigated for quasi-static and dynamic loading. During quasi-static testing it was found possible to predict a combined response using individual responses. During dynamic testing prediction was not possible, since inertia effects where found to dominate the response. The Concrete Protective Coating: A programme of work carried out has qualified the role of a pipeline protective coating and assessed the effect of four different types of concrete reinforcement. Summary Although the study has not been exhaustive, it is clear that reinforcements, which hold the concrete coating to the pipe, allow the coating to continue its protection. Fibres added to a concrete mix were found to reduce the damage to the pipe. However mesh reinforcements were found to hold the concrete together most effectively and provided the greatest added protection. The Soil Support: All foundations absorb some energy. Tests have been carried out to investigate the effect of a soil bed on the response of a laterally loaded pipeline. During dynamic tests on sand supported pipes it was noted that no energy was absorbed during the initial deformation, possibly corresponding to local indentation of the pipe wall. After this the sand was seen to react and absorbed a proportion of the energy, depending on the hammer's drop height. The energy absorbed by the soil continued to increase until an energy plateau was reached, after which the soil absorbed no further energy. It was noted that the displacement at which this energy plateau was reached increased as the drop height increased. Two possible causes of the energy plateau have been discussed. The first cause questioned an assumption that the pipe would deform as if on simple supports. The second possible cause suggested a change from dynamic to quasi-static response and investigated the relationship between acceleration, velocity and reaction force. Of the possible causes of the energy plateau, the most likely is thought to be soil related. Investigation into the Deformation of Locally Loaded Pipes: The investigation into pipeline deformation has been carried out using experimental, numerical and theoretical analysis methods. Quasi-static test results have been used to investigate four pipeline parameters and their influence on energy absorbed by the pipeline, (length, L, wall thickness, t, diameter, D and material yield stress, ay). This investigation led to an empirical equation, which brought all energy-displacement (E-8) curves on to a common curve, for a wide range of these variables. This empirical relationship has been developed to predict deformation, for the range of parameters investigated. Dynamic results obtained were normalised using these empirical equations and data was seen to fall into two broad groups, one group comprising seam welded pipe and one group comprising cold drawn pipe. Strain rate effects were proposed as the most likely cause of this bi-grouping. Limitations in the experimentally derived empirical relationship have been identified, resulting from an insufficient range of pipe samples tested.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:301215 |
| Date | January 1999 |
| Creators | Oliver, Kerry Derrick |
| Publisher | University of Aberdeen |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://digitool.abdn.ac.uk/R?func=search-advanced-go&find_code1=WSN&request1=AAIU603188 |
Page generated in 0.0017 seconds