Purdue's approach looks at the entire system:
- Our goal is to impact both reliability & performance to reduce the cost of energy.
- We take an interdisciplinary systems-oriented approach that addresses every level from individual gears to the entire grid.
- We have a comprehensive infrastructure that includes wind energy test beds, device fabrication facilities, and partnerships.
- We support research with short-term, mid-term, and long-term impacts.
Better sensors, better control under variable wind loads
When turbines point in the wrong direction by even a few degrees, they produce less energy and break down more frequently. A 10° yaw error reduces the energy captured by 5% and produces damaging loads.
Purdue constructed a wind turbine dynamics & control test bed to enable experiments using inertial sensors inside the rotor. These sensors allow us to detect yaw errors as small as 1°, improving our ability to point wind turbines in the best direction.
Better measurements, better reliability
When a wind turbine fails unexpectedly, it is expensive to fix, which drives up the cost of wind energy relative to other sources of power. In some turbines, gearboxes that are supposed to survive for 25 years are failing in 6 months for unknown reasons.
A collaboration between Birck Nanotechnology Center and the Mechanical Engineering Tribology Laboratory has allowed us to build micro-sensors as small as the diameter of a human hair to sense defects long before bearings in the turbine are going to fail.
Additionally, in tests conducted with the National Renewable Energy Laboratory on a 750 kW gearbox, we showed that one reason for these failures is that gearboxes can resonate with the wind. Using a novel measurement procedure, we demonstrated that one likely source of damage to high-speed shaft bearings in gearboxes is the factor of 10 increase in torsional interactions between the gearbox, rotor, & generator.
Better models, better power prediction
Wind farms can experience up to a 25% loss due to rotor wakes passed from upwind turbines to those downwind of them.
3D Navier-Stokes and Large Eddy Current computational fluid dynamics simulations lead to better predictions of wind power outputs to improve the layout of wind farms.
Purdue will be able to test those predictions at a micro wind farm test bed that's under construction next to the Purdue airport. The data from our tests will let us develop models and methods for optimizing the performance of interacting wind turbines.