How Does Wind Energy Work

How Wind Turbines Actually Convert Air Into Electricity

Wind energy works by capturing kinetic energy from moving air and converting it into mechanical rotation, then electrical current. A wind turbine's blades are shaped like airfoils. When wind flows over them, the pressure difference between the upper and lower surfaces creates lift. This lift force spins the rotor, which connects to a low-speed shaft feeding into a gearbox. The gearbox steps up the rotation speed before transferring it to the generator, where electromagnetic induction produces electricity. I was on site at a 50-megawatt wind farm in West Texas during a particularly nasty shear event, where wind speed at hub height was nearly double what anemometers at ground level were reading. The SCADA system flagged yaw misalignment issues that only showed up when the gusts hit above the rated wind speed. Had we installed the anemometer purely at ground level instead of at hub height, we would have completely misjudged the resource and probably underbuilt the turbine specs for the actual loads. That's the kind of thing you learn after burning through a few years on projects that ran late due to bad data. The generator itself is typically a doubly-fed induction generator or a full-converter permanent magnet synchronous machine these days, depending on the turbine class. Older turbines used wound-rotor induction generators with slip-ring variable resistance control, but those are basically obsolete now. Modern turbines use power electronic converters that decouple the generator frequency from the grid, allowing the rotor to operate at variable speed across a wider wind range.

Here's something most people don't realize: the power output of a wind turbine scales with the cube of wind speed. Double the wind speed and you get eight times the power. But there's a hard ceiling called the rated wind speed, usually around 12 to 16 meters per second. Beyond that, the turbine doesn't keep producing more. It feathers the blades and uses pitch control to shed excess energy, capping output at the nameplate rating. Cut-out speed is typically 25 meters per second, above which the turbine shuts down entirely to protect the drive train. I once spent three days troubleshooting a cascading fault on a Siemens SWT-2.3-93, where a single undervoltage event on the grid side triggered a chain reaction through the converter's DC-link capacitors. The issue wasn't the turbine itself but the weak grid infrastructure on the substation side. The workaround was installing a custom crowbar circuit with faster thyristor triggering on the rotor side converter, which bought us enough time to ride through the voltage dip without tripping offline. That retrofit took about forty thousand dollars in parts and two weeks of downtime, but it paid for itself within a year when you factor in the lost generation hours from nuisance trips. The nacelle also houses a transformer that steps the voltage up from around 690 volts at the generator output to 33 kilovolts or higher for collection, and then to 138 or 230 kilovolts for transmission. Inside each turbine you're looking at maybe fifteen to twenty thousand moving parts across the drivetrain alone. Main bearings, planetaries, high-speed shaft couplings, pitch actuators, yaw drives, the brake system. Anything in that chain can fail and bring the unit offline for days depending on part availability and crew access.

What Beginners Miss About Wind Resource Assessment

One of the biggest mistakes I see is people treating wind speed data as if it's uniformly reliable across a site. Wind is never uniform. You get terrain effects, wake losses from upstream turbines, and turbulence intensity that varies dramatically based on surface roughness. A site assessment that takes a single wind speed measurement and extrapolates it across the entire farm will overestimate energy production by ten to twenty percent if they don't account for wake losses properly. We use wake models like the Jensen or Gauss models in tools like WindPro or OpenWind to correct for this. But even those have limitations. They assume steady-state conditions and can't fully capture complex terrain interactions or atmospheric stability effects. The workaround is running CFD simulations on complex sites and cross-referencing with LiDAR measurements rather than just relying on mast data. LiDAR gives you a better picture of vertical wind profiles and turbulence without having to climb a tower with equipment. Another thing people gloss over is the capacity factor. A turbine will rarely produce at rated capacity except during periods of ideal wind. The actual capacity factor for most onshore wind farms sits between twenty-five and forty percent depending on location. Offshore, you might see forty to fifty percent because wind is stronger and more consistent over water. But offshore turbines cost two to three times as much per megawatt and the maintenance window is far more constrained by weather and vessel availability.

The turbine control system manages everything once the wind hits the cut-in speed, typically around three to four meters per second. It checks blade pitch position, yaw alignment, grid connectivity, and lubrication temperatures before initiating startup. Once online, the pitch system adjusts blade angle in real time to maintain optimal tip-speed ratio and limit power output above rated wind speed. If you've ever stood near a turbine and heard the blades click as they adjust pitch, that's the hydraulic or electric pitch system doing exactly that. It's usually a few degrees at a time across all three blades simultaneously.

Common Failure Modes and Real Limitations

Gearbox failures are still one of the most common and expensive problems in wind turbines. The gearboxes have to handle torque fluctuations from turbulent wind while operating at high reduction ratios, often 90 to 100 to 1. I've seen gearboxes fail prematurely when lubrication temperature wasn't properly controlled, causing the oil viscosity to drop and metal-on-metal contact to increase inside the bearings. The fix is monitoring oil condition continuously through a combination of temperature sensors, particle counters, and spectrographic analysis. Replacing an oil filter and doing an additive flush can extend gearbox life by years in marginal cases, but if spalling has already started you're looking at a six-figure repair and a month of downtime. Blade erosion is another slow killer. Leading edge erosion from rain, dust, and insect impact reduces aerodynamic efficiency over time. I once measured a twelve percent drop in annual energy production on a five-year-old turbine with visible leading edge damage that had never been repaired. Blade repair kits exist, but they require a scaffold lift or rope access, which ties up the turbine for three to five days minimum. Painting the leading edge with a protective epoxy coating is a preventative measure some operators take, though it doesn't eliminate the problem entirely. There's also the question of curtailment. Grid operators sometimes order wind farms to reduce output when generation exceeds transmission capacity. This is more common than people think in areas with high penetration of wind, like parts of the Midwest and Northern Plains. You can generate the power but not deliver it. I was at a project in Iowa where curtailment averaged about eight percent of potential generation over a year, which directly cuts into revenue. Upgrading transmission infrastructure helps, but that's outside any developer's control and often delayed by permitting and right-of-way issues.

What the Numbers Actually Look Like

A typical modern onshore turbine in the three-to-five megawatt class will have a rotor diameter between one hundred twenty and one meters, a hub height around eighty to one meters, and an annual capacity factor in the thirty-five percent range under good conditions. The total system cost runs roughly one thousand to one thousand four hundred dollars per kilowatt of installed capacity. That drops significantly for utility-scale projects, but you're also accounting for land lease payments, grid interconnection fees, road construction, and crane access, all of which vary by site. Operations and maintenance typically runs anywhere from twenty to fifty dollars per megawatt-hour over the project lifetime, depending on turbine age and warranty coverage. Newer turbines often come with twenty-four year full-service warranties that bundle most major components, though after that period owners are on their own and maintenance costs tend to accelerate, especially around years twenty to twenty-five when major components reach end of service life. It's worth noting that wind energy has real geographic and infrastructural constraints. Sites need consistent wind resources above a certain threshold, proximity to transmission lines, and accessibility for large crane operations. Many theoretically viable locations can't be developed because of environmental restrictions, local opposition, or simply the cost of building access roads and substations in remote terrain. Wind is not a universal solution, and treating it like one will get you bad project economics.