The main objective of this research was to gain further understanding of the behaviour of oil wellbore fluid filtrates, the various factors affecting filter cake quality, and how these can be controlled to reduce the invasion of the filtrate around the wellbore region. The investigations were intended to define filtration behaviour and filter cake parameters at different fluids with various particle size distributions and various wellbore conditions that are: different filtration pressures, different filtration times, and different shear rates acting on the cake surface during dynamic filtration conditions. The results indicated that all the above mentioned filtration parameters have a various degree of influence on the filter cake and filtration parameters. The forces acting on a particle suspended in the wellbore fluids during dynamic cake particle deposition were also analyzed and a dynamic filtration model was developed. This model can be used to determine the average critical shear rate of these particles and the frictional coefficient factors of the particles with the cake surface. The model has then been modified to predict equilibrium filtration flow rate at any shear stress (above the critical value) applied on the cake surface. Investigations on the transient stage of the dynamic filtration agreed with literature and showed that the size of particles deposited onto the cake decreased with increasing cake thickness, causing the particle size distribution of the dynamic filter cake to alter with cake thickness.

No description available.

The unfortunate case of hydrocarbon reservoirs being often too large and filled with uncertain details in a large range of scales has been the main reason for developments of upscaling methods to overcome computational expenses. In this field lots of approaches have been suggested, amongst which the wavelets application has come to our attention. The wavelets have a mathematically multiscalar nature which is a desirable property for the reservoir upscaling purposes. While such a property has been previously used in permeability upscaling, a more recent approach uses the wavelets in an operator-coarsening- based upscaling approach. We are interested in enhancing the efficiency in implementation of the second approach. the performance of an wavelet-based operator coarsening is compared with several other upscaling methods such as the group renormalization, the pressure solver and local-global upscaling methods. An issue with upscaling, indifferent to the choice of the method, is encountered while the saturation is obtained at coarse scale. Due to the scale discrepancy the saturation profiles are too much averaged out, leading to unreliable production curves. An idea is to downscale the results of upscaling (that is to keep the computational benefit of the pressure equation upscaling) and solve the saturation at the original un-upscaled scale. For the saturation efficient solution on this scale, streamline method can then be used. Our contribution here is to develop a computationally advantageous downscaling procedure that saves considerable time compared to the original proposed scheme in the literature. This is achieved by designing basis functions similar to multiscale methods used to obtain a velocity distribution. Application of our upscaling-downscaling method on EOR processes and also comparing it with non-uniform quadtree gridding will be further subjects of this study.

Polymer flooding is one of the most successful techniques in Enhanced Oil Recovery. One of the obstacles to implementing the technique is the understanding of fluid flow in porous media at different length scales. Although many of the microscopic processes, in microns, in the reservoir are well understood, simulating fluid flow in the reservoir at the micron scale is completely impractical. Each project requires numerous simulations to cover a wide range of scenarios. To shorten the run time, the blocks in the reservoir model are generally coarsened from the core scale in centimetres to larger scales in metres or kilometres. This helps reduce the number of gridblocks for simulations from around 10 [to the power of 13] cells to at most 10 [to the power of 5] or 10 [to the power of 6] cells. Reservoir rock properties such as porosity and permeability are averaged from the small scales using various methods, known as upscaling. In practice, upscaled permeabilities are calculated using the techniques derived for water flooding. The same upscaled model is then used for studying a variety of fluid displacements and injection schemes. The impact of using upscaled models for simulations of non-Newtonian flow displacement, as in polymer flooding, is not well understood. This study investigates the effects of upscaling errors on production forecasts in non-Newtonian flow and recommends an approach to be applied in upscaled models for better production predictions. Two permeability distributions: a two-dimensional randomly generated lognormal permeability field and a fluvial system are investigated. These models are flooded by fluids governed by a power law rheological model that represents Newtonian, shear-thinning, and shear-thickening flow behaviour. The errors in production predictions and pressure profiles are analysed. We find considerably high errors in predictions when the properties of fluid displacement are changed. These significant errors can harm economic evaluations of projects. In addition, we prove that upscaled models manipulated for a perfect match to a fine scale model under water flooding should not be used for polymer flooded modelling. Furthermore, we discover that in addition to upscaling permeability, effective viscosity should be parameterised when injecting with non-Newtonian fluid. We recommend adjusting the power law exponent of the displacing fluid model for better results. We verify the new approach and conclude that a good agreement in predictions between fine and coarse scale models can be achieved by a single phase upscaling with an adjustment of the exponent in the power law rheological model.

In this work, we used a combination of experimental techniques in order to validate pore network modelling' at both Darcy and pore scale. The pore networks were etched on a silicon wafer, and comprised 28202 rectangular capillaries 200 ~m long and 40 ~m deep, with two different capillary width distributions.' Flow experiments were perfonned with water as a model Newtonian fluid and shear-thinning partially hydrolyzed polyacrylamide (HPAM) polymer solution as the model non-Newtonian fluid. An original method was proposed in order to implement in the simulator the Carreau mod~l as the constitutive equation of the . shear-thinning fluid. Darcy scale experiments consiste~. in measuring pressure gradient between pore network inlet and outlet at different constant flow rates. Thus, an 'apparent' rheogram of the fluid was experimentally achieved. Quantitative agreement is seen between the experiments and the network calculations for both Newtonian and shear-thjnning regimes. However, at high ...../. flow rates HPAM exhibit a shear-thickening regime due to elastic effects that are not incorporated in the network model. Microscopic flow behaviour in the capillaries was visualized by using micro particle image velocimetry (~PIV) technology. This enables us to measure the velocity profile in the capillaries ~with a one micron resolution. An original method was developed to allow us to compare the computed mean velocity and the experimentally measured maximum velocity. Good correlations wer'~ found both in Newtonian and shear-thinning regime. Once the HPAM solution enters the shear-thickening regime, there is a marked increase in the amount . , of scatter observed in individual network ducts. More detailed IlP1V measurements of the micro flows in several 'neighboring channels show that asymmetric streamline patterns have been observed indicating that the effects of 'feeder' channels are leading to a very extensive 'inlet effect' as is often observed for extensional flows. The thesis is the first to present such a detailed analysis of both the Darcy scale rheology and the detailed micro-flow field using both experiment (etched silicon micromodels along with ~PIV) and pore scale network modelling. Comparison between experiment and theory shows a very interesting level of agreement.

No description available.

Froth flotation is an established method of separating mineral from its associated ore. Although there has been many years of research and development, there remains much room for development. A 4 L bench-scale, continuously overflowing and recycling single species flotation system was developed for this work. It was used to investigate the effect of particle size and retrofit designs on air recovery, a measure of froth stability, as well as the solid and liquid flowrates of the system. Particle size had an effect on air recovery and flotation performance. An increase in solids flowrate corresponded with a decrease in the particle size in the concentrate. When air recovery was below 50 %, the peak in solids flowrate corresponded with the peak in air recovery. The addition of a retrofit design, a horizontal mesh of varying hole-size and thickness, was tested at a range of superficial gas velocities in a three-way full factorial trial. It was shown that a mesh of 30 mm hole-size and 40 mm thickness resulted in a statistically significant improvement in solids flowrate to the concentrate, when compared with the unmodified base case. Position Emission Particle Tracking (PEPT) provided an opportunity to make qualitative and quantitative assessments of particle behaviour within the pulp and froth. PEPT data indicated that in the unmodified base case, swirl from the pulp continued in the lower froth, a behaviour never observed before. Variation in mesh hole-size lead to visible changes in tracer trajectories, and an increase in mesh thickness resulted in both tracers entering the froth more frequently. This work has developed a novel experimental system capable of determining the effects of a range of operational and design parameters on air recovery and flotation performance, and has corroborated and justified the trends seen through statistical analysis and novel imaging techniques.

No description available.

No description available.

The aim of this project was to investigate the potential and suitability of the application of near Infrared spectroscopy/sensors in automatic preconcentration of complex ores. Two ore types (copper and platinum) were considered for investigation. The near infrared region of electromagnetic spectrum has been used for mineral mapping in the minerals industries. However, its application as a sensing technology in the sorting of base and precious metals is still minimal. In practice, a near infrared sensor can measure characteristic features of carbonate, hydroxyl and water groups contained in minerals and rocks. Successful sensor-based sorting requires a good understanding of the minerals and their distribution in an ore. For the copper ores, mineralogical analysis was carried out using QEMSCAN® and qualitative XRD analysis. XRF analysis was used to determine the copper concentration in the various particles. In addition to the XRF elemental analysis, copper values were calculated from copper bearing minerals in the ore. XRD analysis was performed on the platinum ore. Methods of ore sorting based on near infrared readings and near infrared active functional groups (-OH, H2O, and CO32-) were investigated and strategies developed for both ore types. In addition to external environmental influence, most minerals contain water in their chemical structure. Therefore, considering the H2O absorption feature(s) for ore sorting was not considered optimal. Strategies were developed which target the discrimination of either or both carbonate and hydroxyl bearing particles as waste. Individual particles spectra were analysed and absorption features assigned to the various chemical species and minerals responsible for the absorptions. Due to individual particle mineralogical variation, particles were classified either as products, waste or middlings. For copper ore, targeting only the calcite (carbonate) dominated particles for discrimination as waste provided a better option for preconcentration. Application for the platinum ores targeted the discrimination of chlorite, antigorite, and/or calcite dominated samples as waste. Compared with sample mineralogy, samples could be classified as product or waste using near infrared.