Browsing by Author "Thompson, Christopher P."

Now showing 1 - 8 of 8
  • Thumbnail Image
    ItemOpen Access
    Improving the performance of cascade correlation neural networks on multimodal functions
    (Newswood Limited, 2010-12-31) Riley, Mike J. W.; Thompson, Christopher P.; Jenkins, Karl W.
    Intrinsic qualities of the cascade correlation algorithm make it a popular choice for many researchers wishing to utilize neural networks. Problems arise when the outputs required are highly multimodal over the input domain. The mean squared error of the approximation increases significantly as the number of modes increases. By applying ensembling and early stopping, we show that this error can be reduced by a factor of three. We also present a new technique based on subdivision that we call patchworking. When used in combination with early stopping and ensembling the mean improvement in error is over 10 in some cases.
  • Thumbnail Image
    ItemOpen Access
    An inertial two-phase model of wax transport in a pipeline during pigging operations
    (Elsevier, 2017-04-15) Boghi, Andrea; Brown, Lloyd; Sawko, Robert; Thompson, Christopher P.
    Pig in pipelines performs operations for cleaning the pipe interior and internal inspection. In the past few years many 1D models have been developed to simulate the process because of their reduced computational cost; however, they rely on simplifications which are not always valid. In this paper, the results of a three-dimensional (3D) numerical investigation of the interaction between a waxy-oil and a dynamic sealing pig in a pipeline are presented. The results are obtained at a reduced computational cost by using a moving frame of reference, and an “injection” boundary condition for the wax deposited on the wall. The effect of the temperature and the wax particles’ size has been investigated. The 3D results show the structure assumed by the debris field in front of the pig. In particular, a lubrication region at the bottom of the pipe, whose dimensions are temperature dependent, is shown. This information cannot be deduced from 1D modeling. The influence of the oil on the mixture viscosity and the internal bed dynamics are discussed. This work provides insights into the interaction between the debris field in front of the pig and pipeline hydraulics.
  • No Thumbnail Available
    ItemOpen Access
    Investigation of intelligent adaptive image enhancement to aid night time driving
    (Cranfield University, 1999-01) Rio, A.; Thompson, Christopher P.; Fish, D.
    Driving at night is a difficult task. In an attempt to ease this task, most automotive companies are developing systems that aim to increase the safety of the driver and his/her passengers at night. Jaguar Cars Ltd have been involved in such project for several years and have developed a Night Vision System (NVS) based upon the Near Infrared (NIR) and Head-Up Display (HUD) technologies. This thesis is concerned with the application of digital image enhancement algorithms to further increase the driver's visual range at night. The purpose of this research work is to provide the driver with a safe and non-disturbing, enhanced view of the road scene ahead, which is presented on a head-up display. In this automotive environment, specific requirements such as real-time processing, robustness and reliability must be kept in mind to design algorithms that will not compromise the safety of the driver, his/her passengers and other road users. To fulfill these requirements, we have developed a novel intelligent image enhancement scheme for night time driving that actively adapts to the road scene. This scheme results in the enhancement of the contrast in a portion of the projected HUD road scene as if extra headlamps were directed to the region of the image that represents where the road is going. Human Factors studies have shown that this region is where the driver is concentrating his attention when driving. The position of the region of interest is defined by the computation of an approximation of the vanishing point of the road, updated for every frame using a novel, reliable and optimised road edge detection algorithm. The enhancement of the contrast within the region of interest is obtained by applying several novel low-level algorithms based upon the grey level segmentation of the image into regions and the use of the global histogram equalisation and quantised bi-histogram equalisation algorithms. These novel algorithms have all been implemented on the Matrox Genesis board based upon the multitasking, multiprocessor and parallel DSP TMS320C80 chip from Texas Instruments. All algorithms described in this thesis are able to sustain real-time processing at the NTSC frame rate of 30 frames per second. This new concept for a night time driving aid is an attractive solution that meets the numerous requirements driven by Human Factors research in an automotive environment, in particular safety requirements.
  • Thumbnail Image
    ItemOpen Access
    A non-inertial two-phase model of wax transport in a pipeline during pigging operations
    (Elsevier, 2018-03-06) Boghi, Andrea; Brown, Lloyd; Sawko, Robert; Thompson, Christopher P.
    The removal of wax deposit from pipelines is commonly accomplished using pigs. In order to avoid the formation of wax plugs in pipes, bypass pigs, which create a liquid jet to disperse the scraped deposit, are employed. Despite many One-Dimensional (1D) models have been developed to predict the dynamics of bypass pigs, the details of the interaction between the liquid jet and the debris have not been investigated numerically yet. In this work the fluid dynamics of a wax-in-oil slurry in front of a moving bypass pig is studied by means of three-dimensional (3D) numerical simulations. A mathematical model which couples the pig and the wax-in-oil slurry dynamics, solved in the pig frame of reference, has been developed. The results show that the pig quickly reaches an equilibrium velocity, and the pig acceleration is proportional to the square of the mixture relative velocity. Comparing the present with previous sealing-pig results it appears that the bypass flow is more effective in deterring plug formation. Moreover, the 3D fields have the advantage of showing the wax distribution in each pipe section whereas the 1D model cannot distinguish between deposited and suspended wax.
  • Thumbnail Image
    ItemOpen Access
    A novel shape optimization method using knot insertion algorithm in B-spline and its application to transonic airfoil design
    (Academic Journals, 2011-11-16T00:00:00Z) Sherar, P. A.; Thompson, Christopher P.; Xu, B.; Zhong, B.
    A new method using the cubic B-spline curves with nominal uniform knot set to parameterize the geometry is proposed to deal with shape optimization problems. In the method, the control points of the B-spline curves are set to be the design variables in the optimization scheme. A knot insertion algorithm has been introduced in order to keep the geometry unchanged whilst increasing the number of control points at the final optimization stage. The super-reduced idea and the mesh refinement are also employed to deal with the equality constraint and speed up the optimization process. The method is applied to two problems. The first is a 2-dimensional Poisson problem, and the second is an airfoil design problem. In both applications, the results show that the new method is much more efficient when compared with the traditional methods. In the airfoil design problem, the drag of the airfoil has been reduced significantly with much less function calls.
  • No Thumbnail Available
    ItemOpen Access
    Numerical computation and software design
    (Cranfield University, 1999-12) Lesage, P-Y; Thompson, Christopher P.
    The development of simulation tools is becoming an important area in industry, recently fostered by the tremendous improvements in computer hardware. Many physical problems can be simulated by being modelled by mathematical equations which can then be solved numerically. This thesis is concerned with the development of a Finite Difference solver for time dependent partial differential equations. The development involves a number of challenging requirements that the solver must meet: to have the capacity of solving conservation and non-conservation laws (using several numerical techniques), to be robust, efficient and to have a modular and extendible design. Firstly, we focus on the architecture of the program and how an original design approach was used in order to carry out its development. A combination of Object- Oriented Design and Structured Design was adopted.
  • Thumbnail Image
    ItemOpen Access
    Numerical simulation of two-phase gas and non-Newtonian shear-thinning fluid flows in pipelines
    (Cranfield University, 2011-08) Jia, Ninghong; Thompson, Christopher P.; Gourma, Mustapha
    The one goal of this research is to present the adaptive mesh refinement (AMR) technique for one dimensional two-phase slug flows. Uniform fine meshes for these long devices are costly and, in general situations, the optimum space discretisation could not be determined a priori. The adaptive mesh refinement (AMR) procedure permits this problem to be remedied by refining the mesh locally, within regions where sharp discontinuities and steep gradients are present. With the appropriate algorithm and data organisation, it helps to reduce CPU time and speed up simulations of flows in long pipes, while preserving accuracy and acceptable execution times. The main objective of this research is to investigate the behaviour of the gas and non-Newtonian shear-thinning fluids in horizontal pipes. Predictions of drag reduction ratio and holdup are presented for the stratified flow of gas and non-Newtonian Ostwald-deWaele liquid. For slug flow regimes, the mechanistic slug unit model is adopted in order to estimate the pressure gradients along the slug unit. The slug unit model is rearranged and reinterpreted as inviscid Burgers’s equation for incompressible phases. For both stratified and slug flow regimes, three dimensional CFD (computational fluid dynamics) simulations were performed in order to compare the drag reduction ratio and pressure gradients. In stratified flows, CFD is also used in an attempt to evaluate the liquid wall friction factor and to compare the obtained values with those given by empirical standard correlations.The estimated pressure gradient and drag reductions are compared with experimental data. Calculations showed an excellent agreement between the simulation and experimental data. Shear thinning effects are also correctly modelled in this work.
  • No Thumbnail Available
    ItemOpen Access
    Water flow on accreting ice surfaces
    (Cranfield University, 2002-09) Charpin, Jean P. F.; Myers, T. G.; Thompson, Christopher P.
    Ice growth may rapidly degrade the aerodynamic performance of an aircraft. It can also severely damage structures such as communication towers or power lines. Subsequently, de-icing and anti-icing systems have been developed and a number of codes designed to predict ice shapes. When ice accretion starts, two different types of ice can appear, depending on the temperature and conditions. All of the incoming fluid may freeze almost instantaneously and turn into rime ice. Alternatively, a fraction of the incoming fluid may freeze and turn into glaze ice while the other part remains liquid and may flow over the ice. Previous work on ice accretion has mainly targeted the ice shape and neglected the owing water layer. The present study focuses on this. A set of governing equations is derived for both rime ice growth and coupled ice growth and water flow. When rime ice accretes, a mass balance is used to calculate the shape. In the presence of both ice and water, the ice growth is governed by an energy balance and the water flow by a mass balance. These equations are solved numerically for the water flow alone and the coupled ice growth and water flow for two- and three-dimensional at inclined planes. The behaviour of both ice and water is studied. The model is then extended to deal with arbitrary substrates and solutions are sought for industrially important applications such as ice accretion on power lines or aerofoils. This research work forms part of the ICECREMO project. ICECREMO is a three-dimensional ice accretion and water flow code developed collaboratively by DERA, British Aerospace, Rolls Royce, GKN Westlands Helicopters and Cranfield University under the auspices of the UK department of Trade and Industry.