|
|
Combustion Technology:
Research, Development & Training
Transnational Access to Major Research Infrastructures
|
Project listing - Cardiff University,
Cardiff
Professor
Nick Syred
- Burner, gasifiers, combustors and gas turbine combustors modelling
- Predictions of oscillations in combustors, both stationary and gas turbine, using a variety of different fuels.
- Prediction of large frequency jumps and the effect of system geometry change, as well as the effect of scale in the area of conventional burners and gas turbines.
- Prediction of engines and explosions.
Modelling Activities at Cardiff using CFD, Chemkin and Analytical Models.
There are a range of projects available here involving modelling, which are all based on a wealth of experimental data which has been measured over many years. Specifically:
Burner, gasifiers, combustors and gas turbine combustors modelling
We are extensive users of Fluent and CFD codes in the context of burners, gasifiers, combustors, gas turbine combustors etc and are looking long term to try to couple aspects of Chemkin predictions (this is a commercial software package with very detailed chemistry) to the Fluent Pre PDF package as well as identifying critical reactions in given situations so that reduced reactions schemes can be developed for application via Magnussen type reaction rate schemes in CFD codes, where non- equilibrium chemistry is important.
In particular we have CFD projects in the following areas:
- Cyclonic biomass gasifiers for sawdust and similar fuels to drive small gas turbines. Here we gasify sawdust under pressure to give a medium calorific value gas. Some ash is collected in the gasifier. CFD modelling is critical here, especially in the drive to retain more ash in the system to avoid contamination of the gas turbine blades. The generated fuel gas then passes into a highly efficient cyclone ash separator, then to a modified gas turbine combustor, where the gasified biomass fuel gas is burnt out in excess air from the gas turbine compressor. CFD studies are important in the cyclonic ash separator (very fine ash needs to be separated at low pressure drop) and in the modified gas turbine combustor, where the effects of secondary support fuel for start up/shut down must be modelled.
- Pre-calciners in cement plant. Cement manufacture is a very competitive business these days and operators seek to reduce costs by using alternative fuels such as tyre chips, packaging wastes etc., especially in the pre-calciners at the end of the kiln, in which much of the carbon dioxide in the raw cement meal is driven off. Normally pulverised coal is used here and work is needed to ensure that system changes, when these new fuels are used, can be adequately modelled. Both combustion of the alternate fuel and driving off of the carbon dioxide from the raw cement meal must be
modelled.
- Simulation rigs for fouling in large power station boilers. We are involved in a major European project, called Powerflam, run by a consortium of European Utility Operators and the object is to evolve systems which can predict the effect of new fuel mixes on heat transfer surfaces in large boilers, namely slagging, fouling and erosion effects. We have evolved a very novel rig which at small scale can simulate the temperature/time/gas atmosphere history of large boilers and CFD studies are an essential element of the system development. Slagging/fouling and erosion effects are evaluated via a series of targets located in the simulation rig. Particular concern is expressed with the use of sewage sludge and coal blends, even at substitution rates of only 3%. Extensive CFD work is needed to compliment the experimental work.
- Modelling of small gas turbine combustors for use with aviation kerosene and similar fuels. There is much interest in small gas turbines in a number of areas, both for unmanned aircraft and for small scale power generation; the problem is that conventional designs of gas turbine combustor do not scale very well, basically the residence time of the fuel is much reduced, causing problems with obtaining good burnout and low emissions.
- We have been also involved in work to improve the yield of natural gas fired reactors to produce high quality carbon black. Very detailed experimental analysis has been carried out and complimentary work is required to model the system and investigate ways in which the yield and quality of the carbon black produced can be improved, again this will involve extensive CFD and related Chemkin software use.
Predictions of oscillations in combustors, both stationary and gas turbine, using a variety of different fuels.
We have developed a phenomenological model for pulse and related combustors which describes well their behaviour in terms of pressure pulsations, temperatures etc with a global model for methane reaction. We have now managed to substitute the global methane reaction rate mechanism by coupling our model to Chemkin software and an available detailed methane reaction scheme. There have been some inevitable adjustments to rate constants but we are now getting this coupled model to give fairly good predictions of CO, CO2, H2O and NOx over most of the equivalence ratio range for which data is available.
We wish to extend this work to encompass predictions of oscillations in combustors, both stationary and gas turbine, using a variety of different fuels.
Prediction of large frequency jumps and the effect of system geometry change, as well as the effect of scale in the area of conventional burners and gas turbines.
Another approach to model combustion oscillations uses a neural network approach, based on an initial simple sinusoidal model, coupled with a self learning neural network model, which is taught via input from a number of measured conditions. We are getting successful results both for frequency, small frequency jumps and rms pressure levels, and are applying the approach to extend the range of operation beyond the normal learning limits (dictated by experimental results). In particular we wish to use the model to predict large frequency jumps and the effect of system geometry change, as well as the effect of scale. Applications again are in the area of conventional burners and gas turbines.
Prediction of engines and explosions.
In the context of engines and explosions we are using Ricardo Wave Chemkin and various CFD codes software to look at the effects of pressure, temperature, flame stretch on flame propagation, development, blow off, ignition etc. The Ricardo wave software is used to obtain a one dimensional thermofluid analysis of the engine from inlet to exhaust over the complete operating cycle. We then use CFD to focus in on critical areas of the engine cycle and look in detail at thermofluid and combustion sensitive areas for optimisation purposes and minimisation of emissions. The Chemkin software allows us to obtain fine chemical detail in critical areas of the combustion process. In this context we are now using a heptane scheme to extend the range of the work, having initially started with methane. Essentially Chemkin can give rise to families of correlation’s to match against experimental data, normally generated here experimentally at Cardiff. This then allows improvement to be made to CFD codes and inputs to the Wave software.
- General description
- Modelling of a jet flow using FLUENT.
- Turbulent structures associated with jet flows
- Modelling of novel heat transfer surfaces using Fluent
- Experimental analysis of novel heat transfer surfaces
- Turbulent structure of swirl flows
- Large-scale Combustion Hazards
- Transient Combustion Modelling
- Liquid-Fuel Combustion
- Biofuel
- Combustion Particulates
- Combustion and Laser Diagnostics
- Jet Breakup and Novel Atomisation Research
- Thermal stress modelling of heat exchanger surfaces
- Droplet Coalescence and Break-Up
|
|
For further information on the EuroFlam programme, mail to: info@euroflam.net
Page designed and executed by "Aristide Of St. Michael" in co-operation with Peter Roberts at IFRF NET,
Tony Griffiths at UWCC and Mario Graziadio at ENEL.
