Optimal tidal power generation
Please note that this project is self-funded.
The aim of this project is to develop a statistical model for the error between the slow but accurate numerical model for water height, and the fast but inaccurate model.
New developments in turbine design and construction methods for tidal barriers mean that tidal power generation is becoming more cost effective. Increasing the efficiency of tidal generation through optimal generation strategies will also help make the technology more cost effective. Cost effective tidal generation is important, because it has the potential to provide reliable base-load renewable energy.
Tidal power generation uses turbines to generate electricity from the movement of tides in and out of estuaries. Water is held behind a tidal barrier, then allowed to pass through via a turbine. The potential to generate electricity is determined by the difference in water height across the barrier. A consequence of this is that near the high/low tide mark, you can increase the total generated electricity by opening sluice gates to allow water up/down stream unimpeded.
To calculate the optimal generation strategy---that is when to open and close the sluice gates---you need to be able to calculate the level of water behind the tidal barrier based on how much water has passed through, and at what velocity. Unfortunately, although tides are very predictable, calculating the level of water behind a tidal barrier is a difficult problem, because at any given time the level of water is not constant throughout the estuary. Numerical schemes for calculating the water level beyond a tidal barrier are complicated and slow, but to numerically search for an optimal generation strategy we need to be able to do this calculation quickly. Quick approximations exist, but have errors of the order 10-15%.
Project aims and methods
You will develop a statistical error model that will improve the approximation, and thus enable improved generation strategies. The idea is to generate accurate estimates at a well-chosen set of parameter values, and then interpolate between them with a suitable model.
Working out how to choose well is an interesting and topical experimental design problem, with applications well beyond tidal power generation. You can also make the choice of design points dynamic, allowing savings by sampling only where your error model is inaccurate.
I gael gwybodaeth am strwythur y rhaglen, gofynion mynediad a sut i wneud cais ewch i’r rhaglen Mathemateg.Gweld y Rhaglen