For more than a century, engineers designing wind turbine blades have relied on aerodynamic formulas dating back to the 1800s — equations they’ve known for decades don’t hold up under real-world conditions. Rather than replace them, the industry patched the gaps with ad hoc “correction factors” stitched together from wind tunnel tests and field experience, with no solid theory underneath.
Now, MIT researchers say they’ve built something that didn’t previously exist: a physics-based model that accurately describes how air actually moves around a rotor, even when conditions push the old formulas to their breaking point.
A formula older than commercial flight
Momentum theory — the foundational model describing how rotors interact with moving fluid — dates to the late 19th century. It predates powered flight, predates commercial electricity, and has anchored wind energy textbooks ever since. “These are the first thing you would read about in a wind energy textbook,” says MIT’s Michael Howland, who teaches the subject.
In 1920, physicist Albert Betz used momentum theory to calculate the theoretical ceiling on wind energy extraction: 59.3 percent of the kinetic energy in incoming wind. That figure, known as the Betz limit, remains a standard reference point in wind engineering more than a hundred years later.
The trouble is that the theory started showing cracks almost immediately. Within a few years of Betz’s calculation, engineers found that momentum theory broke down “in a pretty dramatic way” at higher blade rotation speeds and different blade angles. The failure wasn’t minor — the theory predicted that thrust force should start declining above a certain operating point, while experiments consistently showed it continuing to rise.
“It’s not just quantitatively wrong, it’s qualitatively wrong,” Howland says. A model that gets the direction of change backward isn’t one engineers can trust.
Decades of duct tape: the ‘correction factor’ problem
With no replacement theory available, the engineering community did what it often does when a foundational tool misbehaves: it patched the gaps. Designers layered empirical “correction factors” onto the original equations — adjustments reverse-engineered from wind tunnel data and operational experience, not derived from physical principles.
Those patches worked well enough in the narrow conditions they were calibrated for, but offered no theoretical grounding whatsoever. Whenever engineers pushed into new territory — novel blade profiles, unconventional farm layouts, real-time turbine control — they were extrapolating from empiricism rather than reasoning from physics.
The problem was most acute precisely where it mattered most. Momentum theory breaks down closest to the Betz limit — the very operating point turbine designers are trying to reach. “Within 10 percent of that operational set point that we think maximizes power, the theory completely deteriorates and doesn’t work,” Howland says. Engineers were effectively flying blind at the moment of peak performance.
The theory also fails entirely when turbines aren’t perfectly aligned with incoming wind — a condition Howland describes as “ubiquitous” on real wind farms. Wind shifts constantly, turbines are always catching up, and until now operators had no principled way to predict how a change in a turbine’s angle would affect its output without falling back on empirical guesswork.
Building the unified momentum model
To build something better, MIT postdoc Jaime Liew, doctoral student Kirby Heck, and Professor Howland turned to detailed computational fluid dynamics — high-resolution simulations of how air actually moves around a spinning rotor. Rather than starting from the old equations and adjusting them, they worked from first principles.
One key flaw in the original model was an assumption about air pressure downstream of the rotor. Classical momentum theory assumed that the pressure drop immediately behind the rotor would quickly return to ambient levels a short distance away. The new analysis showed that assumption becomes increasingly inaccurate as thrust force rises — precisely the high-performance regime where the old theory was already struggling.
To handle misaligned airflow, the team incorporated three-dimensional wing-lift equations originally developed for aerospace applications. Those equations were built to describe how lift behaves across the full geometry of a wing, and they translate naturally to the problem of a rotor operating at an angle to the wind. The resulting “unified momentum model” was derived theoretically and then validated through computational fluid dynamics; further validation using wind tunnel experiments and field tests is underway but not yet published.
What changes — including the Betz limit itself
One of the more striking results is that the new model revises the Betz limit upward. The century-old ceiling on wind energy extraction turns out to be slightly conservative — the new theory shows a bit more energy can be extracted than Betz calculated. The revision is modest, on the order of a few percent, but its significance within the field is hard to overstate.
“It’s interesting that now we have a new theory, and the Betz limit that’s been the rule of thumb for a hundred years is actually modified because of the new theory,” Howland says. “And that’s immediately useful.”
More practically significant is what the model can do that the Betz limit never could: optimize turbines operating at an angle to the wind. Misalignment is the normal state of affairs on a wind farm, so this is hardly an edge case. In earlier work published in 2022, Howland’s team found that deliberately misaligning some turbines slightly could reduce wake interference and improve overall farm output. The unified momentum model now provides the theoretical foundation for that strategy, replacing an empirically observed effect with a physically grounded explanation.
The model’s applicability also extends beyond wind energy. Because the underlying fluid-flow physics are shared across systems, it applies equally to ship and aircraft propellers and to hydrokinetic turbines such as tidal and river devices. “It’s in the theoretical modeling naturally,” Howland says.
Immediate impact: no new hardware required
Perhaps the most significant near-term implication is how little the new model demands of those who might use it. Because it exists as a set of mathematical equations, wind farm operators can incorporate it directly into existing control software to optimize turbine orientation, rotation speed, and blade pitch in real time — no physical modifications needed.
That low barrier to adoption matters across the entire wind energy value chain, from turbine manufacturers designing next-generation blades to operators making second-by-second adjustments in the field. “This is what we’re so excited about, is that it has immediate and direct potential for impact across the value chain of wind power,” Howland says.
The model is available as an open-source software package on GitHub, which means researchers and industry engineers can download, test, and build on it without waiting for commercial licensing or proprietary access. The research was published in Nature Communications and was supported by the National Science Foundation and Siemens Gamesa Renewable Energy.
The next step is field validation — confirming that the model’s predictions hold up against measurements from real turbines in real wind. If that work bears out, the unified momentum model could become the new baseline: the equation wind engineers reach for first, the way they reached for momentum theory for the past hundred years. This time, at least, they’d have reason to trust it all the way to the limit.