Thousands of wind turbines across the United States generate an estimated 48,000 megawatts of cumulative power.
Blades on these massive systems can reach 50 yards in length and represent one of the biggest maintenance challenges to keeping turbines under power. “These turbines are interesting machines because they are so large, they are quite flexible and really are running within fairly fine tolerances,” says Engineering Physics Assistant Professor Matt Allen. “There is a range of speed that they can operate in. If you go beyond that speed then you’re likely to get failure very quickly. In a high wind, they reach overspeed and can explode.”
Operators monitor data on performance and power output and inspect turbines that underperform. This might include visual inspection with binoculars, climbing the rigs or in the worst case, using a crane to take the system apart. Testing blades with sensors involves stopping the system, climbing it, mounting a sensor, climbing down and devising a telemetry system to send data. “The problem with wind turbines is that the vibrations are so slow that you’re talking about hours for each test. It can take days or weeks to get an idea of how forces are affecting the blades,” says Allen.
He has devised a system that uses a laser to collect data over the whole length of a blade, quickly giving operators critical information— from the ground—about the stress a blade can withstand. “Rather than leave the laser fixed, we have it quickly sweeping back and forth relative to how fast the blade is vibrating, so it’s almost like a barcode scanner,” says Allen. ”We have a new theory for what to do with those measurements and how to understand it. In the end, it gives an operator an understanding of how the blade is bending in a variety of frequencies.”