A New Approach to Deicing Wind Turbines from Base to Blade Tip
When icing brings down a grandmother, a power line or a plane, nobody wants to talk about it because it’s always somebody’s fault. The same is true when icing slows or shuts down a wind turbine. At least no one gets hurt physically, but it still costs lots of money. There is a cold little secret in the world of wind power; turbine blade icing is a problem. I’m new around here, so my evidence is anecdotal. His clues are described in the article in our January/February 2011 issue on page 6.
By Cliff Lyon, Director Corporate Development, IceCode LLC, USA
Clue two: Over a casual lunch not long ago, I met a man who designs turbine towers. He knew the wind industry as well as anyone, or so I assumed. Naturally, I asked him what he knew about wind blade icing. His response shocked me. He said he’d never heard about blade icing being a problem.
Clue three: Ever since IceCode won the GE Ecomagination Challenge: Powering the Grid Innovation Award for our blade deicing technology, the phone has been ringing off the hook. And they are nearly all desperate wind farm operators, some with production losses over 25%. It turns out that even a little ice (which likes to form on the leading edge of the blade), changes the aerodynamics pretty quickly. That means that for the hours, days or months before the turbine has to be shut down, the turbine is producing a lot less power.
Clue four: Outside Scandinavia, there is almost no serious discussion in the industry about production losses from blade icing and no surveys, formal or otherwise. In a feasibility study called Mapping of Icing for Wind Turbine Applications, Elforsk rapport 08:40, the authors offer this; ‘Even as far South as in Texas (US), the losses due to iced up wind gauges may cost large-scale owners M$ per day.48’ The footnote is revealing: ‘48Horn B., FPL, private communication during EWEC 2008 in Brussels + e-mail.’
From the outside, the icing conversation seems to be a reluctance to face reality; one shrouded in a pall of resignation, and for good reason: the most conservative, informal estimates of lost production due to icing in cold climates are in the 10–17% range. The resignation is over the general consensus that there are no great deicing solutions.
If secret is a mischaracterisation of the icing problem in the wind industry, I beg the reader’s forgiveness. IceCode is not a wind power company per se. We are really an ice management company with almost 200 pending or awarded patents for applications involving removing ice from windshields, refrigerator evaporators, power transmission lines, boats, bridges, roads, glass roofs and more. In almost every application, our technology uses 95–99% less energy than current methods. The same is true for preventing or removing ice from a wind blade, to the extent that anyone has really deployed the current options for deicing.
But icing is a problem and a serious one as evidenced by the plethora of studies that seek to better understand the effects of icing on power output. There is even an article ‘Experimental investigation of energy losses due to icing of a wind turbine’ in the Proceedings of the International Conference on Power Engineering 2007. The abstract reflects the general thesis of most research:
Ice accretion and irregular shedding cause many potential problems during turbine operation. For example, icing causes large load imbalances; creates excessive turbine vibration; can change the natural frequency of blades; promotes higher fatigue loads, and increases the bending moment of blades....It is therefore important to conduct research on wind turbines operating in cold climate areas. This article presents an experimental investigation of wind energy losses, when a wind turbine prototype is operating under icing conditions.
Then, there are the studies that try to map and quantify regions where icing can be a problem and finally, in the absence of good ice detection tools, there is some good research on using meteorological inputs to predict icing.
All include a section about deicing methods. It is always short and somewhat morose. Nothing seems to work well. Again, from the Elforsk rapport in 2010:
Icing of wind turbine blades poses a significant challenge in cold climate regions around the world, as it increases the economic risk, thereby reducing the profitability of wind energy projects in affected areas. The wind energy research community has addressed the issue of icing since the early 1990’s. However, no commercial anti-icing or de-icing solutions currently exist for medium and severe icing conditions.
Nonetheless, the push to develop wind energy in cold climates continues for obvious reasons (more wind), and reasons not so obvious (cold air is more dense). From an excellent and no doubt, well-read report called Wind Energy Projects in Cold Climates: ‘Since power is proportional to air density, the available power in the wind will increase similarly and require a well-adapted power control system.’
That’s good news.
Ice has some of the strongest bonds in nature. Getting it off a wing is difficult. Getting it off a wing that is moving quickly requires an enormous amount of heat. Deicing a jetliner wing at preferred altitude using the age-old bleed method, reduces fuel efficiency by 10–12%. At least, wind turbines can be shut down while the blades are heated until the ice melts. Not a solution to cheer about, but still, the only one of any significance in use today.
IceCode’s wind blade deicing solution is adapted from its airplane in-flight deicing wing, a technology developed at Dartmouth College by renowned ice physicist Dr Victor Petrenko. Both have been in testing for over two years.
IceCode’s Pulse Electro-Thermal Deicing (PETD) is simple and rather obvious once you see it.
The surface of ice has an enormously high electric charge. As a result, ice doesn’t simply cake on --it bonds in three ways: via the hydrogen atoms themselves, via an electrostatic bond caused by the current and via comparatively weak van der Waals forces. PETD works by breaking the first two bonds. An electric charge lasting a few milliseconds heats the surface buried in ice just long enough to melt about a micrometre or two of the surface of the ice. Once the ice is melted, the hydrogen and electrical bonds break. The resulting water then acts as a lubricant, allowing the mass of ice to slide away. With short pulses, the heat doesn’t have time to diffuse. It is all released at the interface.
On a per application basis and using a little maths, a ‘smart’ high power electric pulse can be transmitted across substrates like glass, concrete, wood, several polymers; ABS, polyethylene, HDPE, composite materials; epoxy-glass and carbon and laminates such as polymer films on bulk metal substrates, electrically conductive paints and much more.
The result is quick and evenly applied heating targeted at the ice interface. Depending on configuration and temperature, a wind blade can be perfectly deiced in 20–60 seconds while spinning.
It is possible to retro-fit existing turbines with external patches or electrically conductive paints. But the preferred method is to integrate it into the blade design. While the incremental cost to prepare or enable a blade for PETD in manufacturing is negligible, the implications are enormous.
So, what does it mean for the wind power industry that there is finally a good, inexpensive solution to the problem of icing? I don’t think we’ll know for years, but just thinking about it is exciting. Considerations like these will no longer need to be made:
Recent research indicates that an imbalance in power, i.e. torque, once per revolution is characteristic even for a lightly iced-up wind turbine because ice formations on the blades will vary and change blade aerodynamics. As light icing is presumably more frequent than the other natural conditions that may trigger the vibration alarms, the designer should consider the influence of fatigue caused by icing. Icing might also cause surfaces to be unserviceable, which would prevent access, and ice thrown from blade or that falls from the tower or nacelle may pose a significant safety hazard.
Wind Energy Projects in Cold Climates
It’s easy to quantify the impact of PETD for commercial refrigeration: in pure energy savings it’s approximately 33%. But for wind power in most parts of the world, a deicing solution is (I know, it’s an over-used term) a paradigm shift. Off the top of my head, I foresee implications in the following areas:
Turbine design: Turbine design engineering must carefully factor torque loads from airfoils deformed by ice. In the drive towards larger blades, predicting torque loads is challenging enough without having to worry about ice forming on the leading edge of a 90 meter blade spinning at high speeds.
Increased power in cold climates: Beyond the predictable and stronger winds, recall that cold air is denser than warm air. At –30°C, air is 26.7% denser than at 35°C. Since (I’m told) power is proportional to air density, power output increases proportionally.
Increased power production for existing wind farms: Do the maths. If you can recover a 10% loss of production due to icing, the net present value of a replacement PETD-enabled blade is at least several times the cost.
Currently planned offshore wind development in the shallow waters off northern Europe and the coast of New England anticipate severe icing conditions for several months each year. Imagine the new pro forma recalculated without downtime from icing!
And, don’t tell the Texans, but they are not the Lone Star when it comes to icing issues in otherwise balmy climates. Many areas throughout the USA considered temperate, as well as locations in southern Europe, China, Japan, southern South America, Australia and New Zealand, have microclimates that suffer the same icing conditions as cold climates, notably in the hilly areas favoured for wind farm location.
Site selection: Suddenly, we can start talking about wind farms in places heretofore untenable because they are too cold, too wet or both, where land is cheap and the tourists don’t complain. Now, the biggest consideration is too much wind. That’s what I call a high class problem.
Every now and then, a stupidly simple, inexpensive new technology comes along that changes everything. Now, you can not only manage the ice on your turbine blades, but also the turbine tower itself, the transmission lines to get your power to market and the ice cubes in your cocktail.{/access}
When icing brings down a grandmother, a power line or a plane, nobody wants to talk about it because it’s always somebody’s fault. The same is true when icing slows or shuts down a wind turbine. At least no one gets hurt physically, but it still costs lots of money. There is a cold little secret in the world of wind power; turbine blade icing is a problem. I’m new around here, so my evidence is anecdotal. His clues are described in the article in our January/February 2011 issue on page 6.By Cliff Lyon, Director Corporate Development, IceCode LLC, USA
{access view=!registered}Only logged in users can view the full text of the article.{/access}{access view=registered}Clue one: Driving cross country in the winter of 1992, I saw my first wind farm along a vast stretch of cold, barren Wyoming. It was one of those clear, dry, hyper-cold days when the snow blows like dust across the highway, melts and instantly re-freezes into perfectly formed ice rails where warm truck tyres had passed only moments before.
It was then I saw six tall, proud, white wind turbines on the horizon. As they grew nearer, I was impressed by their size and beauty. The graceful curves of their blades reminded me of the lines of a sailboat hull. Only one of the six was spinning. I remember wondering why and promptly forgot about it until recently.Clue two: Over a casual lunch not long ago, I met a man who designs turbine towers. He knew the wind industry as well as anyone, or so I assumed. Naturally, I asked him what he knew about wind blade icing. His response shocked me. He said he’d never heard about blade icing being a problem.
Clue three: Ever since IceCode won the GE Ecomagination Challenge: Powering the Grid Innovation Award for our blade deicing technology, the phone has been ringing off the hook. And they are nearly all desperate wind farm operators, some with production losses over 25%. It turns out that even a little ice (which likes to form on the leading edge of the blade), changes the aerodynamics pretty quickly. That means that for the hours, days or months before the turbine has to be shut down, the turbine is producing a lot less power.
Clue four: Outside Scandinavia, there is almost no serious discussion in the industry about production losses from blade icing and no surveys, formal or otherwise. In a feasibility study called Mapping of Icing for Wind Turbine Applications, Elforsk rapport 08:40, the authors offer this; ‘Even as far South as in Texas (US), the losses due to iced up wind gauges may cost large-scale owners M$ per day.48’ The footnote is revealing: ‘48Horn B., FPL, private communication during EWEC 2008 in Brussels + e-mail.’
From the outside, the icing conversation seems to be a reluctance to face reality; one shrouded in a pall of resignation, and for good reason: the most conservative, informal estimates of lost production due to icing in cold climates are in the 10–17% range. The resignation is over the general consensus that there are no great deicing solutions.
If secret is a mischaracterisation of the icing problem in the wind industry, I beg the reader’s forgiveness. IceCode is not a wind power company per se. We are really an ice management company with almost 200 pending or awarded patents for applications involving removing ice from windshields, refrigerator evaporators, power transmission lines, boats, bridges, roads, glass roofs and more. In almost every application, our technology uses 95–99% less energy than current methods. The same is true for preventing or removing ice from a wind blade, to the extent that anyone has really deployed the current options for deicing.
But icing is a problem and a serious one as evidenced by the plethora of studies that seek to better understand the effects of icing on power output. There is even an article ‘Experimental investigation of energy losses due to icing of a wind turbine’ in the Proceedings of the International Conference on Power Engineering 2007. The abstract reflects the general thesis of most research:
Ice accretion and irregular shedding cause many potential problems during turbine operation. For example, icing causes large load imbalances; creates excessive turbine vibration; can change the natural frequency of blades; promotes higher fatigue loads, and increases the bending moment of blades....It is therefore important to conduct research on wind turbines operating in cold climate areas. This article presents an experimental investigation of wind energy losses, when a wind turbine prototype is operating under icing conditions.
Then, there are the studies that try to map and quantify regions where icing can be a problem and finally, in the absence of good ice detection tools, there is some good research on using meteorological inputs to predict icing.
All include a section about deicing methods. It is always short and somewhat morose. Nothing seems to work well. Again, from the Elforsk rapport in 2010:
Icing of wind turbine blades poses a significant challenge in cold climate regions around the world, as it increases the economic risk, thereby reducing the profitability of wind energy projects in affected areas. The wind energy research community has addressed the issue of icing since the early 1990’s. However, no commercial anti-icing or de-icing solutions currently exist for medium and severe icing conditions.
Nonetheless, the push to develop wind energy in cold climates continues for obvious reasons (more wind), and reasons not so obvious (cold air is more dense). From an excellent and no doubt, well-read report called Wind Energy Projects in Cold Climates: ‘Since power is proportional to air density, the available power in the wind will increase similarly and require a well-adapted power control system.’
That’s good news.
Ice has some of the strongest bonds in nature. Getting it off a wing is difficult. Getting it off a wing that is moving quickly requires an enormous amount of heat. Deicing a jetliner wing at preferred altitude using the age-old bleed method, reduces fuel efficiency by 10–12%. At least, wind turbines can be shut down while the blades are heated until the ice melts. Not a solution to cheer about, but still, the only one of any significance in use today.
IceCode’s wind blade deicing solution is adapted from its airplane in-flight deicing wing, a technology developed at Dartmouth College by renowned ice physicist Dr Victor Petrenko. Both have been in testing for over two years.
IceCode’s Pulse Electro-Thermal Deicing (PETD) is simple and rather obvious once you see it.
The surface of ice has an enormously high electric charge. As a result, ice doesn’t simply cake on --it bonds in three ways: via the hydrogen atoms themselves, via an electrostatic bond caused by the current and via comparatively weak van der Waals forces. PETD works by breaking the first two bonds. An electric charge lasting a few milliseconds heats the surface buried in ice just long enough to melt about a micrometre or two of the surface of the ice. Once the ice is melted, the hydrogen and electrical bonds break. The resulting water then acts as a lubricant, allowing the mass of ice to slide away. With short pulses, the heat doesn’t have time to diffuse. It is all released at the interface.
On a per application basis and using a little maths, a ‘smart’ high power electric pulse can be transmitted across substrates like glass, concrete, wood, several polymers; ABS, polyethylene, HDPE, composite materials; epoxy-glass and carbon and laminates such as polymer films on bulk metal substrates, electrically conductive paints and much more.
The result is quick and evenly applied heating targeted at the ice interface. Depending on configuration and temperature, a wind blade can be perfectly deiced in 20–60 seconds while spinning.
It is possible to retro-fit existing turbines with external patches or electrically conductive paints. But the preferred method is to integrate it into the blade design. While the incremental cost to prepare or enable a blade for PETD in manufacturing is negligible, the implications are enormous.
So, what does it mean for the wind power industry that there is finally a good, inexpensive solution to the problem of icing? I don’t think we’ll know for years, but just thinking about it is exciting. Considerations like these will no longer need to be made:
Recent research indicates that an imbalance in power, i.e. torque, once per revolution is characteristic even for a lightly iced-up wind turbine because ice formations on the blades will vary and change blade aerodynamics. As light icing is presumably more frequent than the other natural conditions that may trigger the vibration alarms, the designer should consider the influence of fatigue caused by icing. Icing might also cause surfaces to be unserviceable, which would prevent access, and ice thrown from blade or that falls from the tower or nacelle may pose a significant safety hazard.
Wind Energy Projects in Cold Climates
It’s easy to quantify the impact of PETD for commercial refrigeration: in pure energy savings it’s approximately 33%. But for wind power in most parts of the world, a deicing solution is (I know, it’s an over-used term) a paradigm shift. Off the top of my head, I foresee implications in the following areas:
Turbine design: Turbine design engineering must carefully factor torque loads from airfoils deformed by ice. In the drive towards larger blades, predicting torque loads is challenging enough without having to worry about ice forming on the leading edge of a 90 meter blade spinning at high speeds.
Increased power in cold climates: Beyond the predictable and stronger winds, recall that cold air is denser than warm air. At –30°C, air is 26.7% denser than at 35°C. Since (I’m told) power is proportional to air density, power output increases proportionally.
Increased power production for existing wind farms: Do the maths. If you can recover a 10% loss of production due to icing, the net present value of a replacement PETD-enabled blade is at least several times the cost.
Currently planned offshore wind development in the shallow waters off northern Europe and the coast of New England anticipate severe icing conditions for several months each year. Imagine the new pro forma recalculated without downtime from icing!
And, don’t tell the Texans, but they are not the Lone Star when it comes to icing issues in otherwise balmy climates. Many areas throughout the USA considered temperate, as well as locations in southern Europe, China, Japan, southern South America, Australia and New Zealand, have microclimates that suffer the same icing conditions as cold climates, notably in the hilly areas favoured for wind farm location.
Site selection: Suddenly, we can start talking about wind farms in places heretofore untenable because they are too cold, too wet or both, where land is cheap and the tourists don’t complain. Now, the biggest consideration is too much wind. That’s what I call a high class problem.
Every now and then, a stupidly simple, inexpensive new technology comes along that changes everything. Now, you can not only manage the ice on your turbine blades, but also the turbine tower itself, the transmission lines to get your power to market and the ice cubes in your cocktail.{/access}
