This work numerically investigates the turbulent partially premixed flames, which are ubiquitous in combustion powered devices. This combustion mode involves many physical complexities such as flame propagation in unevenly premixed mixture of fuel and oxidiser, turbulence/flame interaction in presence of mixture fraction gradients, triple flame configuration, etc. The fundamental mechanisms of these physical processes are yet to be further understood, posing significant modelling challenges. These issues are addressed in this thesis using a presumed joint probability density function (PDF) approach with both Reynolds Averaged Navier Stokes (RANS) and Large Eddy Simulation (LES) methodologies. This joint PDF is described by a parameter, mixture fraction, describing mixing and a reaction parameter, progress variable. The laminar flamelet concept is adopted to decouple chemistry and turbulence calculations for high computational efficiency. This modelling framework is validated using two experimental test cases in this study including a canonical lifted jet flame and a practical swirling flame, both exhibiting strong partial premixing features. The simulation results obtained for these validation cases show a robust model performance for a broad range of flow and mixing conditions with an attractive computational cost for practical interests. For the lifted jet flame case, two-dimensional (2D) axisymmetric steady RANS approach is used to compute the flame lift-off height showing very good agreement with the experimental measurements for a range of jet velocities and air-dilution levels. However, a substantial difference is found between the 2D unsteady RANS (URANS) results and experimental data for the flame transient evolution from its initial ignition to final stabilisation. The comparison for this transient evolution is improved significantly in the 3D URANS simulations suggesting that the third physical dimension, the azimuthal direction, plays an important role during the flame transient evolution. The following LES study further shows that the flame most-leading point appears to be in different azimuthal positions exhibiting a spiral -like trajectory as the flame propagates upstream towards the final lift-off height. The temporal variation of this leading point during this process is captured very well by the LES model in comparison with the experimental data. The validity of this LES model is then further assessed for a confined swirling flame with practical flow conditions. The simulation results are compared against an extensive experimental dataset including velocity, mixture fraction, temperature and major species mass fraction measurements, showing an overall good agreement at various locations inside the combustion chamber. The intermediate species mass fractions are also predicted reasonably well by the LES model despite that a general over-estimation is observed. Moreover, it is found that in the transport equation for the SGS variance of progress variable, the reaction and dissipation terms are predominant, substantially greater than the terms representing turbulent diffusion and production processes. Furthermore, the statistical correlation between mixture fraction and progress variable is found to be important for the RANS closure. However, this correlation is observed to be negligible for the sub-grid scale (SGS) of LES showing reasonably good model predictions for both simulated experimental configurations.

No description available.

As the research into the improvement of gas turbine performance progresses, the combustor-turbine interface becomes of increasing importance. In new engine designs components come closer together and the study of the combustor and turbine interactions can prove to be valuable for the improvement of the aerothermal performance of the vane. This thesis presents an experimental and numerical investigation of the aerodynamic and heat transfer aspect of the interactions between the combustor and the nozzle guide vane. In the gas turbine studied the trailing edge of the combustor transition duct wall is found upstream of every second vane. In the experimental measurements carried out in a purpose-built high speed experimental facility, the wake of this wall is shown to increase the aerodynamic loss of the vane. On the other hand, the wall alters secondary flow structures and has a protective effect on the heat transfer in the leading edge-endwall junction, a region that has proven to be detrimental to component life. The effect of different clocking positions of the vane relative to the combustor wall are tested experimentally and shown to alter the aerodynamic field and the heat transfer to the vane. The experimental methods and processing techniques adopted in this work are utilized to highlight the differences between the different cases studied. A new concept of using the combustor wall to shield the nozzle guide vane leading edge is introduced, followed by a proposed design that is numerically analysed, including a new cooling system. This uses continuous cooling slots on the upstream combustor wall to cool the vane leading edge. Coolant to the endwalls is provided from continuous slots on the combustor-turbine interface. The reduction of secondary flow through the removal of the horseshoe vortex in the new design results in improved cooling of the endwalls, with a higher average adiabatic effectiveness than in the original case, using the same coolant mass flow rate. The vane surface and suction side are also successfully cooled using less air than that required for a showerhead. The new vane is tested in the experimental facility. The improved aerodynamic and thermal performance of the shielded vane is demonstrated under engine-representative inlet conditions. The new design is shown to have a lower average total pressure loss than the original vane for all inlet conditions. The heat transfer on the vane surface is overall reduced for all inlet conditions and the peak heat transfer on the vane leading edge-endwall junction is moved further upstream, to a region that can be effectively cooled from the upstream cooling slots on the combustor wall trailing edge and the endwalls.

The increasing use of biofuels to replace fossil fuels as well as more stringent emissions regulations for internal combustion engines cause a challenge for the engine manufacturer to build engines that can cope with a large range of fuel properties, but still offer low fuel consumption and very low exhaust emissions. In this work a heavy-duty diesel engine test bed has been built including the fuel and emission analysis equipment suitable for a wide range of biofuels. Also a constant volume spray vessel has been commissioned to optically investigate the macroscopic spray characteristics of different fuels. This vessel was built with the potential investigation of fuel combustion in the future. Four different biofuels, soybean oil methyl ester (SME), palm oil methyl ester (PME), used cooking oil methyl ester (UCOME) and hydrotreated vegetable oil (HVO) in blends of B10, B20, B50 and B100 have been tested, as they are potential candidates to replace mineral diesel in larger scales. The main aim of this project was to investigate the spray and combustion characteristics of various biofuels, their impact on exhaust emissions and performance and the potential optimisation of the control strategy in a heavy-duty Euro V diesel engine. The engine tests revealed that for all biofuels the nitric oxide (NOx) emissions increased compared to mineral diesel (B0), while particulate number (PN), carbon monoxide (CO) and total hydrocarbon (THC) were significantly reduced. The fuel consumption changed according to the heating value and with the three fatty acid methyl esters (FAME) full power was not reached. The macroscopic spray tests showed that lower density fuels, such as HVO, have slower penetration speeds, but wider spray cone angle resulting in better fuel-air mixing conditions. As the engine fuel injection is based on a volumetric injection the heating value and fuel density are mainly influencing the spray characteristics on the engine. In the last part the engine power has been successfully restored for all biofuels and the exhaust emissions have been reduced below the B0 benchmark limits by applying a new engine control strategy showing that the use of neat biofuels can be used on heavy-duty diesel engines without any modifications to the engine hardware and still passing the current emission regulations.

In this research, the performance analysis and control strategy of a spark-ignited free-piston engine generator were presented. A literature review of the free-piston engine fundamental information and the recent research development on the free-piston engine generator (FPEG) was provided, mainly focussing on previous work on numerical modelling, prototype design as well as the control strategy. The design and simulation of a dual-piston spark-ignited FPEG suitable for operation using either a two-stroke or four-stroke thermodynamic cycle were presented. Model validation and the general engine performance of the system were discussed. For the first time, this research demonstrated the potential advantages and disadvantages of the FPEG on using different thermodynamic gas-exchange cycles. A fast response real time model of the FPEG was designed and validated. The simplicity and flexibility of the proposed model make it feasible to be implemented and coupled with real-time hardware in the loop control system development. In addition, since it revealed how an FPEG operates according to a resonant principle, the model is useful for parameter selection in the design process. For the first time, cascade control was proposed and investigated for the piston stable operation control, using both the measured piston top dead centre of the previous stroke and the measured piston velocity at the current stroke as feedbacks, with the injected fuel mass as the control variable. The system performance was improved by implementing the cascade control compared with single loop control in terms of the controller response time, peak error and settling time.

This thesis investigated the best strategy to calculate the unit cost of isothermally forged discs used within large civil gas turbine engines. Research has shown that ultrasonic examination constraints required during the production of forged parts strongly influence the manufacturing design. These constraints have traditionally been ignored during preliminary design analysis due to the unavailability of suitable models. Cost reduction objectives are also commonly simplified using mass as a direct surrogate. These oversights can lead to significantly inaccurate predictions of both cost and material properties. This thesis assessed the impact of satisfying the ultrasonic constraint and quantified potential cost savings through direct unit cost optimisation. This work created a fast automated method for evaluating ultrasonic examination limitations, and created an optimisation workflow where a detailed analytical cost model can be rapidly evaluated. The inclusion of ultrasonic constraint analysis within an automated design loop increased both the precision of manufacturing designs and unit cost estimates. This improvement has led to detailed manufacturing designs being assessed within the preliminary design environment enabling true concurrent engineering. These new advancements enable a disc designer to obtain manufacturing designs alongside a detailed cost breakdown of the optimised manufacturing route. This approach is now routinely used within Rolls-Royce who have also patented the modelling technique. Results show that including the ultrasonic constraint increased initial manufacturing geometry mass by at least 3.8% with costs increasing by up to 2.4%. Subsequent unit cost optimisation showed cost increases of between -1.6% and +4.3% with maximum heat treatment depth increasing up to 4.6% compared to the baseline non-ultrasonic constraint scenario. These results prove that ultrasonic constraints are a necessity to predict accurate manufacturing data. Introduction of a detailed analytical cost model also found that direct cost optimisation improved savings by up to 1.9% compared to traditional mass reduction objectives. This confirms the need for cost estimation to achieve effective cost reduction.

Gas turbine design needs of high-speed turbomachinery whose layout is organised in compressor-turbine pairs mechanically linked by concentric shafts. The mechanical failure of a shaft leads to compressor-turbine decoupling provoking the acceleration of the free-running turbine. In view of such scenario, it is of paramount importance to guaranty the mechanical integrity of the turbine, in terms of high energy debris release. Certification authorities require proof that any possible failure will be contained; admitting the reliable simulation capability of the event as certification strategy. The objectives of this research activity have aimed at the development of reliable simulation tools based on analytical and semi-empirical models. The integration of all the different models/modules together in an “all-in-one” tool provides the sponsor company with the capability to simulate and assess various shaft over-speed scenarios during the early stages of an engine's design and development program. Shaft failure event cannot be understood unless engine components interaction and fast transient effects are taken into account in a global manner. The high vibration level consequence of the breakage, or the thermodynamic mismatch due to the rapid free-running compressor deceleration, trigger the surge of the compression system which affects to the performance of every engine component. Fully-transient simulation capability to model compression system post-stall performance and secondary air system behaviour has been developed. Component map prediction tools have been created for compressor reverse flow performance and turbines affected by inlet distorted flows. The development of the so-called “all-in-one” simulation tool has been completed and it has been applied to the modelling of a real case of shaft failure. Reliable prediction of thermodynamic properties evolution and over-speeding turbine terminal speed have been shown. The robustness and flexibility of the simulation tool have been demonstrated by its application to different theoretical scenarios.

This thesis documents the characterisation of the performance of a range of Diesel combustion systems derived from two conventional and two unconventional piston bowl designs used in combination with production and prototype piezoelectric fuel injection equipment (FIE).

With the increasing evidence of the potential disastrous consequences arising from global warming, there is a need to reduce greenhouse gas emissions, and gas-fired power generation represents an attractive option due to its low carbon intensity. Nevertheless, gas is not a zero-emission fuel and therefore it is necessary to control the emissions associated with its usage. Among the carbon capture techniques suitable for gas-fired generation, postcombustion is regarded as the most feasible in the short-term. The additional costs associated with the CO2 capture process can be reduced by employing modified cycle concepts such as EGR and STIG, which are characterised by a diluted combustion environment. The development of accurate numerical models for the combustion process in industrial devices under diluted conditions can be very useful in assessing the impact of dilution on the combustion process, and represents the main goal of the present work. Firstly, the impact of carbon dioxide and steam dilution on natural gas combustion has been assessed by means of detailed simulations of simple unidimensional laminar ames. It has been found that even the relatively low dilutions levels typical of EGR and STIG cycles have a significant impact on the combustion process. Also, the diluting species participate directly in the combustion chemistry, and therefore there is a need to include detailed chemistry and finite rate-effects in a CFD model for realistic configurations. In this respect, the suitability of the RANS and LES FGM/presumed-PDF approach for the modelling of swirling partially-premixed flames has been assessed. The performance of different turbulence models with different levels of mesh refinement have been assessed against in-ame measurements in a lab-scale burner and guidelines for the CFD modelling of industrial devices have been inferred. Finally, the previous findings have been employed to develop a complete CFD model for an industrial MGT combustor, which has been investigated under both air-fired and diluted operation. The numerical results have compared with the available experimental data. It has been concluded that the model is able to predict the impact of dilution on the heat release, flame stabilisation, flow-field and pollutant emissions.

Radial swirler were investigated for gas turbine combustor applications. with low NOx emissions as the main aim of the project. The flow regime of the combustor which was imposed by the radial swirler flow was shown by flow visualisation to feature a conical shaped swirling shear layer boundary and a comer recirculation zone. The flow patterns was independent of the swirl-vane angle but was a function of the swirler passage depth. A minimum swirler expansion ratio of 1.8 was required to achieve an adequate combustion efficiency. A high efficiency was not achieved in the weak region until there was a significant outer expansion and associated recirculation zone. However, there was a little influence of the expansion ratio on the weak extinction limit. Various non-conventional fuel injection methods such as swirler vane passage. radial central and wall injection were used with gaseous propane and natural gas and liquid kerosene and gasoil. Passage injection was undertaken to exploit the twin benefits of peripheral fuel injection and partial fuel and air mixing upstream of the swirlers outlet. Generally, most of the mixing between fuel/air took place in the shear layer. However, there was a major influence of the method of fuel injection on the NOx emissions. Low NOx emissions were achieved with the radial central injection, but ultra-low NOx emissions, comparable with the premixed situation, were achieved for passage and wall injection. This was due to the dependency of the local shear layer mixing near the swirler exit on the fuel placement as shown by the radial gas analysis traverse results in the plane just downstream of the radial swirler. Staged air and fuel combustion was investigated using lean-lean combustion concept. Low NOx emissions compatible with a high combustion efficiency was demonstrated with stable switching from pilot to main stage combustion. Finally, a double radial swirler with a high air flow was investigated using co and counter swirl and demonstrated ultra low NOx with a good stability with central injection into counter rotating swirler. These systems were shown to have the potential for dry solution to the industrial gas turbine NOx emissions regulations with a very high combustion efficiency.