Lecturer, Solar-Thermal Group, Australian National University
My PhD was related to system modelling of a solar thermal energy system under development at the University of New South Wales and the University of Sydney, named the Compact Linear Fresnel Reflector (CLFR). A prototype is currently under construction at the Liddell power station in the Hunter Valley in New South Wales, Australia.
This system uses a series of parallel mirrors to focus light onto an absorber surface positioned about 8 metres above the mirrors. The absorber surface is in fact a series of pipes with water running through them. The system therefore generates hot water or steam as the water is pumped through, depending on how fast the water is pumped. We transfer this heat into the power station's final boiler feedwater preheater, thereby reducing their need to preheat that water by other means, which in turn reduces their coal consumption.
The system is in fact modular. A full-size system will have multiple absorber surfaces. The mirrors that 'belong' to each absorber will actually be interleaved at ground level, as shown in the diagram at right. This has the effect of reducing the the required spacing between mirrors since there is less risk of mirrors falling in each other's shadows. Hence the 'compact' in CLFR.
The project was motivated by the Australian Government's Mandatory Renewable Energy Target (MRET) scheme, which requires power stations to produce 2% of their power from renewable sources. Macquarie Generation has chosen to use the CLFR technology and will produce solar-heated steam for addition to their power station at the boiler inlet stage, thereby replacing the need for the final feedwater heater.
If all prototyping goes well, the Liddell CLFR system will provide 95 MW of thermal energy to the power plant, giving an estimated electrical equivalent of 35 MWe.
Below are some photos of the first prototype system, as well as a couple of maps which show the location of Liddell between Muswellbrook and Singleton in NSW (click to enlarge):
More information on the CLFR is available from the University of Sydney. More photos are available on the photos page of Ausra, Inc.. Maps were taken from the NRMA and maporama.com. Have a look also at the satellite imagery from Google.
My work is concerned with the heat transfer and system modelling aspects of the design. My academic supervisors at the University of New South Wales are Graham Morrison and Masud Behnia.
Steady-state and transient system modelling for the CLFR are underway. We are investigating how to control the system when radiation levels change; when radiation drops, we need to ensure that steam of a suitable quality continues to flow out of the absorber, and that the flow pattern is still in a steady two-phase flow regime; when the radiation rises, we must ensure that we don't enter the superheated region because we want to avoid the thermal stresses that would result in the pipe from this.
As well as looking at how the steam flow in the absorber will occur in transient conditions, we are also concerned with sizing pumps and surge tanks in the system.
The CLFR system used forced-convection boiling inside high-pressure tubes to take thermal energy from the absorber (along the optical focus) into the working fluid. This means that water is pumped in at one end of the pipe, and steam comes out the other, with no need for heat exchanger or high-temperature heat-transfer oil such as is used in the US SEGS systems. Direct steam generation has already been implemented in other prototype systems, such as the DISS system at the Plataforma Solar de Almería.
Below is a computer program designed to perform a steady-state simulation of the direct boiling process in the CLFR system. If you're interested in using the program please let me know and I'll keep you up to date on its development.
The source code package is written to compile with the free GNU C++ compiler. It has been tested under cygwin and MinGW on Windows. For more information see INSTALL.txt inside the source distribution.
The steam tables component of my code has been released as the freesteam IAPWS steam properties library on SourceForge.
When performing modelling involving water and steam, it is important to use an agreed set of values for the properties of the fluid, so that a fair comparison of system performance can be made with other similar systems, and to help with the validation of the model. A library of code for the above project is under development which implements the IAPWS 1997 steam tables for industrial use, as defined by the following IAPWS documents:
This library is available as subdirectory of the above Direct Steam Generation source code file. Code is fully object-oriented, in C++, and includes automated self-testing routines to compare the calculated values with published expected values from the IAPWS Releases above, and also with the online data from NIST.
Thanks to Bernhard Spang for his help with this code, and his useful document Thermodynamic and Transport Properties of Water and Steam. He has produced a freeware Excel add-in with similar functionality to the above library, coded in Visual Basic.
The 'cavity' in our system is an enclosed trapezoidal shape: concentrated light enters from below, having been reflected off off the mirror array. The upper surface of the trapezoid is lined with pipes containing water which is partially boiled as it passs along the pipes. Estimating the losses involved in capturing heat from the incoming solar radiation is important in estimating the overall system performance, and in deciding the best shape for the cavity. I am performing computational modelling work of the heat loss from the proposed cavity (figure at right) of the CLFR receiver. You can download animations showing the flow.
Note, my hard disk filled up while doing these simulations, so the latter half of the longer animations is corrupted. Better versions yet to come. [August 2003]
The following is a list of my publications. For a more general bibliography for my field, please see my Bibliography Database.