No description available.

No description available.

No description available.

No description available.

No description available.

Choke valves are important components in oil and gas production systems that are used to control the pressure and flowrate of fluids issuing from oil and gas reservoirs. The presence of sand in the production fluids can cause considerable damage to such components, and as sand is increasingly becoming an issue in oil and gas production, valve manufacturers need to find ways of reducing their product's susceptibility to solid particle erosion. A CFD-based erosion modelling tool is achieved by first solving the fluid flow through the component of interest; tracking particles through the fluid and extracting impact data on all solid surfaces; and finally relating the particle impact data to erosive wear through a semi-empirical equation. The present study has focussed on the development and validation of a CFD-based erosion modelling method for simple and complex geometries. Erosion testing has been carried out on a range of choke valve materials to provide the fundamental data required in constructing equations that relate erosion rate to particle impact velocity and angle. These equations have in turn been implemented in a commercial CFD code to provide the complete erosion modelling solution. Validation of the method has been effected by comparing predicted results to experimental test data for both simple and complex geometries. Both single phase and abrasive flows have been considered in comparisons. For the simple geometries, reasonable agreement was obtained between predicted and measured pressure drop for the simplest cases, but predicted mass loss was considerably less than the measured amount. With the complex geometries (Multi-Orifice Sleeve choke valves), good agreement for pressure drop was obtained for some valve positions, but not so good for others. Significant differences were observed in mass loss predictions for the complex geometries, which raise questions as to the usefulness of CFD-based methods for predicting component lifetime.

Helical twin screw compressors are being increasingly used in the oil and gas process industry and for refrigeration and air conditioning duties. This machine is capable of high reliability. To achieve it both thermodynamic efficiency and rotor forces must be taken into consideration at the design stage. In this study, a computer program for analysing the forces in a twin screw compressor has been developed. It takes into account all significant factors and includes a rotor profile generation program and a geometrical characteristics calculation program. The programs developed have been integrated with an existing performance simulation program and used to investigate several design aspects of a refrigeration twin screw compressor and an unusual refrigeration system design. A comprehensive examination of lobe tip designs suggests that lobe tip design parameters must be optimised if minimum power consumption is to be achieved and the use of a sealing strip gives an advantage. A para meter study for optimum rotor geometrical parameter combinations has shown that the 4+5 and 5+6 combinations have in general high efficiencies, but less rigidity, while the 5+7 and 4+6 combinations give the opposite results. The highest wrap angle and length/diameter ratio do not always lead to the highest compressor performance. A comprehensive examination of the influence of the slide valve on compressor performance and bearing forces has been conducted. Procedures for determining the optimum slide stop and volume ratios are presented. The detrimental effect of the non-return valve in a refrigeration system driven by a twin screw compressor has been examined. A non-reversing clutch fitted to the prime mover is an alternative, but must be designed according to the shut down torque which is higher than the normal running torque. This is demonstrated. An analytical model for axial-torsional coupled vibration in an Oil-injected twin screw compressor has been created. A frequency analysis of all the excitations has been conducted by using an FFT technique. It is found that the fundamental harmonic dominates the gas torque, while the first two harmonics dominate the axial forces.

A review was carried out which revealed that no simple mathematical model was available which could be used to predict the performance of a centrifugal pump when it is operating under two phase flow conditions. Experimental analyses were carried out to aid the development of such a model. A rotating channel test rig was designed to study the structure of airwater flow through an impeller passageway. The observations generated a large amount of qualitative data. A full scale centrifugal pump was also tested which provided data that allowed the results of the rotating channel experiments to be considered in more quantitative terms. These two sets of experimental work allowed a conceptual model of the two phase flow through a centrifugal pump to be constructed. The model assumes that forces act on the bubbles as they pass through the impeller passageway because of the rotation and the curvature of the impeller passageway. These forces cause the bubbles to decelerate and coalesce creating a stationary air void which leads to a partial blockage of the channel. This causes the velocity of the fluid passing through the pump to be modified. The model uses a onedimensional velocity vector approach adjusted for hydraulic losses to calculate the head raised by the pump under such conditions. The model presented provides the basis upon which a predictive tool could be developed and used in the development engineering environment. Currently operational problems that arise because of two phase flow are often quantified using scaled hydraulic models, which are expensive, or within the actual application where costly engineering solutions may be required to give acceptable performance. A mathematical model is a much more cost effective tool and its application allows the engineer to decide whether system performance would be compromised by the two phase conditions encountered and propose possible solutions. In recommending the development of such a model it is necessary to consider where and how in the project cycle it should be used. A number of companies were canvassed and it was concluded that in many cases such models are used too late in the project cycle to provide maximum return. In the case of projects which include participants from a number of companies it is suggested that an inter-company team should be constructed if development models are to be widely used. This team should be used to plan and implement the use of development models efficiently and ensure that the data produced is communicated effectively and is of the maximum value to the participants.

This thesis investigates the soil-pipeline interactions associated with the operation of large-diameter chilled gas pipelines in Britain, these are frost/pipe heave and ground cracking. The investigation was biased towards the definition of the mechanism of ground cracking and, the parameters which influence its generation and subsequent development, especially its interaction with frost heave. The study involved a literature review, questionnaire, large-scale test and small-scale laboratory model experiments. The literature review concentrated on soil-pipeline interactions and frost action, with frost/pipe heave often reported but ground cracking was seldom reported. A questionnaire was circulated within British Gas to gain further information on these interactions. The replies indicated that if frost/pipe heave was reported, ground cracking was also likely to be observed. These soil-pipeline interactions were recorded along 19% of pipelines in the survey and were more likely along the larger diameter, higher flow pipelines. A large-scale trial along a 900 mm pipeline was undertaken to assess the soil thermal, hydraulic and stress regimes, together with pipe and ground movements. Results indicated that cracking occurred intermittently along the pipeline during periods of rapid frost/pipe heave and ground movement and, that frozen annulus growth produced a ground surface profile was approximated by a normal probability distribution curve. This curve indicates maximum tensile strain directly over the pipe centre. Finally a small-scale laboratory model was operated to further define the ground cracking mechanism. Ground cracking was observed at small upward ground surface movement, and with continued movement the ground crack increased in width and depth. At the end of the experiments internal soil failure planes slanting upwards and away from the frozen annulus were noted. The suggested mechanism for ground cracking involved frozen annulus growth producing tensile strain in the overlying unfrozen soil, which when sufficient produced a crack.

No description available.