Delving into data collected from an operator's turbines can lead to a series of tweaks to the technology and operation and maintenance (O&M) that boost the output of a wind farm or even an entire asset portfolio.
When Australian developer and operator Infigen Energy brought in consulting firm AWS Truepower to do just that, the resulting benefits included a drop in unscheduled maintenance from 2.9% to 0.6%, for example. The data analysis identified a generator issue during operation at high speeds, prompting a retrofit that eliminated the problem. Clues found in the data also led to the realignment of a number of control settings to optimise the orientation and mechanical operation of individual turbines.
To achieve this kind of analysis involves taking a deep dive into the operational information generated by a wind plant's supervisory control and data acquisition (Scada) system, and combining that with meteorological data and an understanding of the specific turbine technology to figure out how it might be underperforming, how much energy is being lost and how it can be recovered.
"We are able to go in and identify what is normal operation for your turbine and your facility, and use that to help you decide how you are going to approach your O&M strategy," said Whitney Wilson, performance engineering manager at AWS Truepower at the recent American Wind Energy Association annual conference.
The work done by AWS Truepower and Infigen is part of a broader trend in the wind energy industry to find ways to use data and analytics to improve the reliability, availability and performance of existing wind farms. With a worldwide installed base of over 300GW of wind power, the potential is huge and the need to maintain and enhance plant economics increasingly critical.
"The portfolio is getting so large that there is real cashflow coming off these assets now, and the O&M performance is going to be a real driver for these businesses," says Jamie Rafferty, vice-president of operations and asset management at Invenergy, a Chicago-based power producer which has more than 3.8GW online in the US, Canada and Europe.
That is particularly true in the face of uncertainty over the market for new wind energy capacity, adds Dan Shreve, a partner at Make Consulting. "It is becoming ever more important, especially as new unit demand becomes volatile," he says.
With 300-500 data points collected at every turbine, amounting to tens of thousands at a typical wind farm, there is no shortage of information to exploit in an effort to lower costs. "Think about the amount of data that each of us as OEMs collect and that project owners and operators collect. The analysis we do with it is minuscule," says Anne McEntee, GE's vice-president for renewables.
"You can start unlocking the value in additional revenue streams like reducing wake losses and enabling units to be powered up. You can also reduce O&M expenses by better predictive maintenance. Those are things we need to be thinking about because it lowers the overall cost of energy and ownership. That's what's going to help to transform this industry," adds McEntee.
Both the OEMs and third-party suppliers are bringing a lot of creative solutions to the table, says Chris Shugart, senior vice-president of operations at California-based Pattern Energy, which owns 1.4GW of wind projects in the US, Canada and Chile: "I'm excited. There's a lot of great things going on in the industry on this."
A key trend has been taking the focus off individual turbines and looking at how to maximise efficiencies across the entire wind farm. "We have to stop thinking about just the wind turbine itself," says Mark Albenze, CEO of Siemens Energy's wind business in the Americas. "It's an energy delivery system that starts with the wind and ends at the house. So how do you get that whole system integration to be more productive and cost-efficient?"
Keith Longtin, general manager of GE's wind product line, agrees. "It is about the whole wind plant and how you optimise that. You are not necessarily optimising on power. You could be optimising on noise. There could be other parameters out there as well."
Turbine availability is being viewed in a different manner too, with the traditional measure of the percentage of time per year that a turbine can produce giving way to a formula that factors in whether it is actually producing during high wind periods.
"For us, production-based availability is more important," says Rafferty. This lets operators schedule turbine maintenance during lower wind speed times. "That way you lose less energy when you have to be down," says Wilson. "This is something we have been working on for years as an industry. It is getting better and better, and we continue to make improvements."
The industry is also getting better at developing strategies for repairing major components in the field instead of the more costly and time-consuming process of bringing cranes in to replace them. Rafferty estimates that the uptower repair capability available in the market today could have avoided about half of the gearbox replacements that have been undertaken at Invenergy sites. "That has been a very big game changer when it comes to the impact on O&M costs," he says.
Power upgrades to turbines through things like blade extensions or software programs designed to increase turbine output are also becoming more common. "The industry for a while has been actively discussing the performance of turbines and the fact that, in general, they are performing below their sales power curve across North America," says Anne Marie Graves, senior manager at DNV GL. Some of the new products being released by turbine makers can help provide a fix, she says. "We've reviewed the performance from some of these upgrades and the results have been significant."
One of the challenges when considering any of the new products and services on the market is how to get more out of wind facilities without spending too much more. "It is about balancing the cost-benefit equation so you are getting more performance and not going bankrupt in terms of achieving it," says Shreve.
A robust analysis is essential to ensuring the investment actually creates value and the answer is almost invariably site-specific, says Justin Johnson, director of performance management for EDP Renewables North America.
Given the variability and seasonality of the wind resource, says Rafferty, it can be a challenge to accurately measure the increase in annual energy production a potential technology enhancement might bring. "You need to make sure the business case has a lot of margin around it," he says.
With data and analytics taking centre stage in the drive to optimise wind asset management, the industry will also need to reassess its needs in the workforce. "We need to bring in the mathematicians and the operations research people to help us create more of those predictive models to avoid costs," says McEntee. "This is a great opportunity to bring in some new and invigorating talent in order to look at things differently."