MIT promises to grow wind farm returns

Topics
No items found.
Richard Heap
July 24, 2015
This content is from our archive. Some formatting or links may be broken.
This content is from our archive. Some formatting or links may be broken.
MIT promises to grow wind farm returns

Is there a way to shave months off the development time for a new wind farm? The Massachusetts Institute of Technology thinks so.

In the next week, it is set to present a new statistical technique at the International Joint Conference on Artificial Intelligence in Buenos Aires, which it says will improve the energy yield from wind farms. It also says that its method only needs three months of data, which could save several weeks on project planning.

If MIT is correct then this would save developers time and money; and therefore boost financial returns from projects. MIT says the approach would have the biggest effect on offshore wind farms, where installing and maintaining measurement stations is pricey.

And, of course, it will have applications in onshore wind too, where subsidies are often in a state of change with developers racing to finish schemes before rules change. The current building boom in Germany ahead of changes in 2017 is one example of this.

There are plenty of reasons to welcome MIT’s contribution.

Like any industry, it is easy for people working in wind to focus so much on the short-term goals of developing schemes and making money that they can lose track of the industry’s long-term aims.

But we should never forget wind is under constant pressure to show that it can compete economically with fossil fuels and other energy sources, such as nuclear.

It also shows the benefit of bringing in insight from others outside the sector. It is too easy to discount the ideas of people who do not work in wind full time, but we should remember that they are in fact perfectly placed to question established ways of working.

And it should also remind us why it is important for firms to continue to invest in research and development to improve both working practices and technology. It is all too easy to cut back on research when the industry is under pressure to deliver energy at lower costs but, in fact, it is a perfect time to invest. We must keep innovating.

So what has MIT found?

The scientist behind the study, Kalyan Veeramachaneni, said consultants typically forecast wind speeds at a given site by measuring wind speeds for a year and then correlating this with a nearby weather station. The consultant then comes up with long-term predictions.

MIT’s approach is different because it uses only three months’ of site data but correlates it with information from multiple weather stations — from 15 or more sites — which can give more accurate long term forecasts. The researchers said that, based on their tests, their method could predict wind speeds for the next two years three times more accurately than existing models. This is because its method finds non-linear links between the data-sets.

That is the theory. Now we need developers to test the model.

With the financial benefits on offer, it should be easy for MIT to find someone willing to give it a try.

Is there a way to shave months off the development time for a new wind farm? The Massachusetts Institute of Technology thinks so.

In the next week, it is set to present a new statistical technique at the International Joint Conference on Artificial Intelligence in Buenos Aires, which it says will improve the energy yield from wind farms. It also says that its method only needs three months of data, which could save several weeks on project planning.

If MIT is correct then this would save developers time and money; and therefore boost financial returns from projects. MIT says the approach would have the biggest effect on offshore wind farms, where installing and maintaining measurement stations is pricey.

And, of course, it will have applications in onshore wind too, where subsidies are often in a state of change with developers racing to finish schemes before rules change. The current building boom in Germany ahead of changes in 2017 is one example of this.

There are plenty of reasons to welcome MIT’s contribution.

Like any industry, it is easy for people working in wind to focus so much on the short-term goals of developing schemes and making money that they can lose track of the industry’s long-term aims.

But we should never forget wind is under constant pressure to show that it can compete economically with fossil fuels and other energy sources, such as nuclear.

It also shows the benefit of bringing in insight from others outside the sector. It is too easy to discount the ideas of people who do not work in wind full time, but we should remember that they are in fact perfectly placed to question established ways of working.

And it should also remind us why it is important for firms to continue to invest in research and development to improve both working practices and technology. It is all too easy to cut back on research when the industry is under pressure to deliver energy at lower costs but, in fact, it is a perfect time to invest. We must keep innovating.

So what has MIT found?

The scientist behind the study, Kalyan Veeramachaneni, said consultants typically forecast wind speeds at a given site by measuring wind speeds for a year and then correlating this with a nearby weather station. The consultant then comes up with long-term predictions.

MIT’s approach is different because it uses only three months’ of site data but correlates it with information from multiple weather stations — from 15 or more sites — which can give more accurate long term forecasts. The researchers said that, based on their tests, their method could predict wind speeds for the next two years three times more accurately than existing models. This is because its method finds non-linear links between the data-sets.

That is the theory. Now we need developers to test the model.

With the financial benefits on offer, it should be easy for MIT to find someone willing to give it a try.

Is there a way to shave months off the development time for a new wind farm? The Massachusetts Institute of Technology thinks so.

In the next week, it is set to present a new statistical technique at the International Joint Conference on Artificial Intelligence in Buenos Aires, which it says will improve the energy yield from wind farms. It also says that its method only needs three months of data, which could save several weeks on project planning.

If MIT is correct then this would save developers time and money; and therefore boost financial returns from projects. MIT says the approach would have the biggest effect on offshore wind farms, where installing and maintaining measurement stations is pricey.

And, of course, it will have applications in onshore wind too, where subsidies are often in a state of change with developers racing to finish schemes before rules change. The current building boom in Germany ahead of changes in 2017 is one example of this.

There are plenty of reasons to welcome MIT’s contribution.

Like any industry, it is easy for people working in wind to focus so much on the short-term goals of developing schemes and making money that they can lose track of the industry’s long-term aims.

But we should never forget wind is under constant pressure to show that it can compete economically with fossil fuels and other energy sources, such as nuclear.

It also shows the benefit of bringing in insight from others outside the sector. It is too easy to discount the ideas of people who do not work in wind full time, but we should remember that they are in fact perfectly placed to question established ways of working.

And it should also remind us why it is important for firms to continue to invest in research and development to improve both working practices and technology. It is all too easy to cut back on research when the industry is under pressure to deliver energy at lower costs but, in fact, it is a perfect time to invest. We must keep innovating.

So what has MIT found?

The scientist behind the study, Kalyan Veeramachaneni, said consultants typically forecast wind speeds at a given site by measuring wind speeds for a year and then correlating this with a nearby weather station. The consultant then comes up with long-term predictions.

MIT’s approach is different because it uses only three months’ of site data but correlates it with information from multiple weather stations — from 15 or more sites — which can give more accurate long term forecasts. The researchers said that, based on their tests, their method could predict wind speeds for the next two years three times more accurately than existing models. This is because its method finds non-linear links between the data-sets.

That is the theory. Now we need developers to test the model.

With the financial benefits on offer, it should be easy for MIT to find someone willing to give it a try.

Is there a way to shave months off the development time for a new wind farm? The Massachusetts Institute of Technology thinks so.

In the next week, it is set to present a new statistical technique at the International Joint Conference on Artificial Intelligence in Buenos Aires, which it says will improve the energy yield from wind farms. It also says that its method only needs three months of data, which could save several weeks on project planning.

If MIT is correct then this would save developers time and money; and therefore boost financial returns from projects. MIT says the approach would have the biggest effect on offshore wind farms, where installing and maintaining measurement stations is pricey.

And, of course, it will have applications in onshore wind too, where subsidies are often in a state of change with developers racing to finish schemes before rules change. The current building boom in Germany ahead of changes in 2017 is one example of this.

There are plenty of reasons to welcome MIT’s contribution.

Like any industry, it is easy for people working in wind to focus so much on the short-term goals of developing schemes and making money that they can lose track of the industry’s long-term aims.

But we should never forget wind is under constant pressure to show that it can compete economically with fossil fuels and other energy sources, such as nuclear.

It also shows the benefit of bringing in insight from others outside the sector. It is too easy to discount the ideas of people who do not work in wind full time, but we should remember that they are in fact perfectly placed to question established ways of working.

And it should also remind us why it is important for firms to continue to invest in research and development to improve both working practices and technology. It is all too easy to cut back on research when the industry is under pressure to deliver energy at lower costs but, in fact, it is a perfect time to invest. We must keep innovating.

So what has MIT found?

The scientist behind the study, Kalyan Veeramachaneni, said consultants typically forecast wind speeds at a given site by measuring wind speeds for a year and then correlating this with a nearby weather station. The consultant then comes up with long-term predictions.

MIT’s approach is different because it uses only three months’ of site data but correlates it with information from multiple weather stations — from 15 or more sites — which can give more accurate long term forecasts. The researchers said that, based on their tests, their method could predict wind speeds for the next two years three times more accurately than existing models. This is because its method finds non-linear links between the data-sets.

That is the theory. Now we need developers to test the model.

With the financial benefits on offer, it should be easy for MIT to find someone willing to give it a try.

Is there a way to shave months off the development time for a new wind farm? The Massachusetts Institute of Technology thinks so.

In the next week, it is set to present a new statistical technique at the International Joint Conference on Artificial Intelligence in Buenos Aires, which it says will improve the energy yield from wind farms. It also says that its method only needs three months of data, which could save several weeks on project planning.

If MIT is correct then this would save developers time and money; and therefore boost financial returns from projects. MIT says the approach would have the biggest effect on offshore wind farms, where installing and maintaining measurement stations is pricey.

And, of course, it will have applications in onshore wind too, where subsidies are often in a state of change with developers racing to finish schemes before rules change. The current building boom in Germany ahead of changes in 2017 is one example of this.

There are plenty of reasons to welcome MIT’s contribution.

Like any industry, it is easy for people working in wind to focus so much on the short-term goals of developing schemes and making money that they can lose track of the industry’s long-term aims.

But we should never forget wind is under constant pressure to show that it can compete economically with fossil fuels and other energy sources, such as nuclear.

It also shows the benefit of bringing in insight from others outside the sector. It is too easy to discount the ideas of people who do not work in wind full time, but we should remember that they are in fact perfectly placed to question established ways of working.

And it should also remind us why it is important for firms to continue to invest in research and development to improve both working practices and technology. It is all too easy to cut back on research when the industry is under pressure to deliver energy at lower costs but, in fact, it is a perfect time to invest. We must keep innovating.

So what has MIT found?

The scientist behind the study, Kalyan Veeramachaneni, said consultants typically forecast wind speeds at a given site by measuring wind speeds for a year and then correlating this with a nearby weather station. The consultant then comes up with long-term predictions.

MIT’s approach is different because it uses only three months’ of site data but correlates it with information from multiple weather stations — from 15 or more sites — which can give more accurate long term forecasts. The researchers said that, based on their tests, their method could predict wind speeds for the next two years three times more accurately than existing models. This is because its method finds non-linear links between the data-sets.

That is the theory. Now we need developers to test the model.

With the financial benefits on offer, it should be easy for MIT to find someone willing to give it a try.

Full archive access is available to members only

Not a member yet?

Become a member of the 6,500-strong A Word About Wind community today, and gain access to our premium content, exclusive lead generation and investment opportunities.

Full archive access is available to members only

Not a member yet?

Become a member of the 6,500-strong A Word About Wind community today, and gain access to our premium content, exclusive lead generation and investment opportunities.