With the PTC in limbo, operations and maintenance takes on new importance. A range of industry leaders recently discussed ways to keep workers safe while they keep assets working, sometimes beyond their 20 year design life.
Paul Dvorak, Editor
Windpower Engineering & Development
The wind industry is asking itself two big questions. The first is: Considering all the things that can go wrong with a wind turbine, how do we get these machines through 20 productive years? While operators and technicians are successfully answering that question, more foresighted people are asking: Why stop there? The second and better question: How can we care for these turbines so they retire closer to a 30-year mark?
After two days at the recent Wind Energy Update O&M Summit in Dallas, it’s clear that PTC or no, the wind industry has engaged many active minds churning on just about every functional aspect of a turbine. Here are a few of the ideas shared at the event trying to answer the two questions above.
Creating transparency on core processes and benchmarking OEM performance
“The turbine problem wind farm owners face now is their blades and how they might be fixed with leading-edge protection or vortex generators,” said E.ON Vice President Matt Kessel. “De-rating turbines is also a life extending strategy. More emphasis is now on third party repair and services.”
Kessel emphasized creating and maintaining safe working environments. It is possible to increase revenue and profitability by reducing risk. “One way to ensure such performance is with a well-written and detailed maintenance contract,” he said.
Operation metrics involve safety first, followed by training and worker qualifications. Kessel said that safety indicators such as TRIF and LTIF are lagging indicators – they tell what has already happened. Better safety metrics include those for performance such as availability, MTBF, mean time to repair, production figures, and lost production factors. “Regarding human factors, culture is key. Make the techs feel the equipment is theirs and QAQC improves and workforce turnover drops,” he says.
This panel discussed post-warranty options and knowledge transfer, and began with the good news that newer turbines perform better than old ones, along with a request for a feature that predicts component failure. When the topic got around to end-of-warranty (EOW) inspections, everyone had something to say.
Jason Schroeder, director of supply chain for Iberdrola Renewables said the EOW process had to be clear and detailed. Some OEMs are willing to participate in the inspection and some are not, presumably because their time is tight. And you cannot start too soon. He also suggested that the OEM can bring a lot of value to the table regarding part availability. The EOW is not a zero sum game, he concluded.
Terry Callahan, Suzlon’s VP of Operations said inspection plans should begin about six months before the EOW, although others thought that was a bit too early. The inspection needs wind and time. Kinematic data is also helpful.
David Smith, VP of Commercial for Infigen said O&M crews need strong procedures and a culture of safety. “There should be a written procedure for everything. Everybody maintaining turbines the same way leads to high availability,” he said.
It’s an amusing title but Kurt Christensen, O&M Business Development Director for EDF Renewable Services recognizes the relationship is a two way street with the owner controlling traffic. There are perspectives to consider regarding how to structure the O&M services, such as an OEM-ISP (Independent service provider) hybrid combination. To select an O&M company, Christensen provided a list of 18 considerations but boiled them down to five: financial strength, liability insurance, 24/7 monitoring, “safety safety safety”, and transparency.
The EOW inspection is critical, he said, echoing earlier comments. He recommended taking one month to prepare the report and one to review it. “Perform the inspection too soon and you miss things that should be in it. Be ready to compromise on some items, and prepare for the worst news, such as a component supplier no longer in business,” he cautions.
Taking innovation into deployment
This panel discussion prompted a comment from the Chair, Make Consulting’s Dan Shreve, suggesting impatience on his part with the normal slow progress of innovation. He suggested the title, Deploying innovation to imply a more active role for the wind industry.
Eric Schick, a panelist with Pantheon Environmental, recognized that while getting a product concept from lab to product is a big leap, getting it commercially accepted is a bigger one. Schick also introduced the LAWS concept for new developments, meaning respect for Life, Air, Water, and Soil. “We are creating a symbiosis between mankind, mother earth, and the marketplace. An example: a building that is its own energy source.” Another development he mentioned was a coating for surfaces that break hydrocarbons into simpler elements.
Chris Griffin, CEO Regenedyne, presented a slide showing the rates of patent issuance and funding in the U.S. and globally, and how they are rising fast in the last few years.
Carlos Coe, Founder of Xtreme Power, suggested that more technology could bring costs down. “More sources of power on the grid can make it more reliable. And good news for the utilities is that their large central systems will still be needed,” said Coe.
Wayne Van Dyck, CEO of RePower Capital, reminded the audience that costs are always an issue and then promoted the technology his company has developed to reduce the cost of capital by 10 to 20%. For example, boosting capacity factors from 30s to 40s leads to lower power-purchase agreements and easier financing.
Safe, reliable cost effective ops, in the second half of the year
“Train your people and train your leaders,” says Infigen VP Operations Hank Owens. “A good organization shows respect for everyone, from techs to the CEO. Integrity extends to the equipment as well, as shown in operating it as it was designed and specified.”
Continuous improvement is essential for any organization, if they want to survive. Know your risks, he adds, and work with your insurer on loss prevention. The top insurance claims for his company are 40% from mechanical problems, 25% from lightning, 9% from fire, and 23.8% from business losses. Most costs are in unplanned tasks. Planned maintenance can be done quickly. Lastly, there is a positive correlation between good safety and good performance.
“Nothing happens without financing,” says K&L Gates Partner David Hattery. “And get ready for the new wind industry if the PTC does not renew, which is not a guaranteed thing,” he adds. With regard to the detail in financing contracts he says, “I’ve seen a project fail because its 500 page contract included one paragraph of exclusions that were too much,” he said.
“Financing assumptions are made on the quality of wind data, the probability of success (P) values assigned to the project,” said Catherine Helloux, VP of Projects and Structured Finance with Societe General Americas Securities. “Degradation assumptions made for how well the equipment will run in coming years also affect financing. Even the owner’s portfolio of other wind farms is considered in financing decisions. Remember, bankers are risk adverse,” she concluded.
Navigating the decision to self-perform, renew with an OEM, or hire an ISP
Cost is the big driver in the decision to self-perform maintenance, but not the only one, said Duke Energy’s Steve Scott. Making the decision comes in part by first listing all businesses objectives and options, and then evaluating how each option scores against the objectives.
A few advantages to self-performing versus OEM include full control of the site’s success and substantial cost savings, about 25 to 35%. A few drawbacks: Relationship with the OEM might require ‘repair’ so consider terms that permit continued cooperation.
The decision to self perform also requires identifying all risks associated with each option, weighing their impact, and creating a bid specification for the OEM to quote against. This specs-out the key terms of service.
Provide a counter proposal to the OEM when continuing operation is not financially viable, or when your company wants to control its own destiny. Then notify the OEM that you will not be continuing with their O&M service, and specify the minimum services you’ll need to maintain once you begin self-operation. The important minimum services are engineering support, IT & SCADA support, documentation, parts, and training.
Teresa W. Chan, Senior VP energy Warranty, AIG said circumstances that necessitate extended-warranty insurance include the expiration of original-equipment-manufacturer warrantees, and the assumption of operations and maintenance responsibilities. A few benefits of the extended-warranty insurance include control of the overall turbine conditions and control over operations and maintenance.
Extended-warranty insurance can cover mechanical breakdown, serial defects, along with a performance guarantee. Lastly, a few underwriting considerations for risk evaluation include the turbine manufacturer and model, years in operation, wind farm location, and CMS data analysis.
Best practices for taking over existing O&M contracts
Scott McBride Manager – Wind Operations, NRG Energy Services said that as a warranty period near expiration, it is time to think about turbine service and maintenance. Options include extending contracts with the OEM, interview third-party maintenance companies, or do the work in-house. “More often, owners are taking on the responsibility of maintaining the wind turbines themselves,” said McBride.
He offered a few suggestions for successful maintenance of a wind park. For example, make good use of SCADA data. That means understanding reoccurring faults and their fixes, and plan for retro-fits now and later. Also, understand how hard the machines are working, and then maintain your current SCADA data capabilities, and try to improve them. For example, using lower cost, vibration-analysis tools.
Wind turbines risk evaluation
Wallace Ebner, Principal Technical Specialist, AIG said that evaluating risk requires answering a range of questions such as: How experienced is the OEM? How many units have they produced and placed in service, and for what length of service? Risk evaluation also considers a wind turbine’s heritage, for instance, whether or not it changed and by how much.
Older-model turbines are being discontinued faster than new models to replace them. Turbine models offered 3 to 4 years ago are no longer available. This means the supply chain for replacement parts for some models is not so good. Due diligence is in order when considering a new model or manufacturer or both. Among other things, check for certification, and include your risk management people early on.
Gamesa’s workshop: How to successfully approach maintenance of turbines from other manufacturers
Gamesa is more than an OEM, revealed Warren Wilson as he introduced several of the company’s lines of business. The latest of which is life extension plans for several particular turbines. These upgrades will focus on Gamesa G47 turbines, Vestas V47s, V80, V90, and GE 1.5-MW turbines. As an example, the G47s, initially 660 kW machines, can be upgraded to 720 kW capacity with a new generator, hardware upgrades, and advanced controls. The modifications can give the equipment a 10-year life extension. To show the company is serious, Wilson said it bought a nine-turbine wind farm to demo company expertise.
Case study – An advanced O&M approach: Turning SCADA data into knowledge
Nagore Guarretxena, the Energy Division Manager of NEM Solutions, discussed the company’s advanced turbine-monitoring techniques with examples of predicted failures. She presented a case history of their predictive maintenance equipment that uses SCADA data and the analysis of work orders. Other turbine data is collected by standard ISO communication protocol. The goal is to avoid turbine downtime and improve their rate of return.
The system has been tested on three wind farms with two turbine models, 850 kW and 2 MW. SCADA data collected 50 variables per turbine over 32 months, and 6,000 work orders with the goal of predicting failures and thereby reduce costs. No other hardware installations are necessary. The system works by analyzing the data to create behavior models that detect anomalies and predict failures. Work orders evaluate the predictions.
“You can learn a lot by listening to a working turbine,” said Reynir Hilmisson with Bruel & Kjaer Vibro. In fact, he suggested that before techs climb a turbine, they should just listen for a couple minutes to learn its sounds. In two case studies, he presented before and after sound tracks from problematic and repaired turbines. The company has developed an audio technique to analyze pitch-motor problems with a 16-channel digital sensor monitor just for condition monitoring. “CMS suffers from SCADA data that has nothing to say about vibration, the source of the sounds,” he said. The new equipment is intended to remedy the shortcomings.
End of warranty using condition monitoring
“Components only fail under warranty, and turbines last 20 years,” said the always controversial and entertaining David Clark with Bachmann Electronic. Of course, Clark was teasing the audience, because neither comment is true. He suggested that condition monitoring (CM) data is essential for post-repair validation. “What I find shocking is that CM has been accepted and used extensively in steel mills, mines, and other industrial sites but still struggles for recognition in the wind industry,” he said. And then he went on to dispel some myths that hamper wider CM applications.
TCM Retrofit – Why, when and how
“During a low wind period, be sure to install a condition monitoring system and before the end-of-warranty inspection,” recommended Gram & Juhl CEO Alex Juhl. The company manufacturers CM systems which he suggests are is also a tool to improve turbine maintenance. Juhl showed that for a 2.3 MW onshore turbine, CM can provide an ROI in 2.1 to 5.5 years, by increasing production with three to four more days each year, and that leads to more than 1% increased production over the life of the turbine and reduces operating expenses in a range of $4,500 to $96,000. For a 300 turbine fleet, the savings can total over $1 million.
“All analysis start with a wind-data stream,” began EDF Renewable Energy’s Dan Charlton. A 1% increase in farm production can account for $1 million in revenue. The hard part is sifting through the mountain of data. Consider also that one turbine can generate up to 25,000 tags/day and a 200 turbine farm generates some 276,000 tags in the same period. To make matters worse, useful data could be stored in several disconnected databases. Also, all turbines are not created equal, even those from the same manufacturer. That point comes clear when you consider that two identical turbines near each other will perform differently. The big question is why? In one such case, after much research, Charlton’s team found incorrect pitch parameters in the controller. “It’s easy to report here, but finding them took a long time.” The point is that data in one place helps identify a turbine with the greatest operating expenses so crews can focus their effort on that one unit.
Moving equipment and components from warehouse to install
“Turbines are not human friendly,” said Grayling Vander Velde, safety specialist with NextEra Resources. “To make up for that, it’s important to stretch in the office before a turbine climb and again at the base of the turbine. And take micro-breaks,” he added. For instance, if you are on your knees for 15 minutes, get off of them and sit down for one minute. If bending for 10 minutes, get out of that position for a minute micro-break.
Furthermore, companies should have procedure for moving heavy objects, such as a 200 lb pitch motor. All components should have their weight listed on them. “To keep our techs in shape, we subsidize gym memberships.”
In addition, drop zones should be identified around a turbine. Many a radio and tool have made a fast trip to the ground.
The advantage of service lifts, he says, is that it lets people not quite qualified to climb get to the nacelle to see how things work, people such as the company CEO. Sadly, some states, such as Pennsylvania, make it cost prohibitive to install labor saving service lifts, because they are considered elevators, each is required by law to receive a $200 yearly inspection, which discourages their installation.
“Many at-risk behaviors have been reinforced in the past, because they have gotten away with them,” said Joe Hecker CEO of Safe by Choice to start a panel discussion on safety. “If a tech gets away with risky action once, they may try it again and lose.” Leadership reinforces at-risk behavior with a “just get it done” attitude.
Peer-to-peer reinforcements have the most value, such as when techs remind each other of proper procedures. For further reinforcement, managers should ask a team, “How did you get that job done safely?”
There is a lot of psychology to a safe working environment. “For instance, its more effective to tell a technician to keep his partner safe than to keep himself safe,” said Mike Schmidt, Windingen, MD&A.
Suzlon’s health and safety director Todd Karasek reminded the audience that if you lead, they will follow and if you can measure something in the crews, you can improve them. If you want leadership, care about the employees health and safety issues.
“To improve safe operation, be sure to provide positive reinforcement,” says Vander Velde. “Most of the reinforcing is negative and after the fact. More positives could be simply, ‘You did a good job adjusting that harness,’” he says.
Wind O&M innovation – Trends and new opportunities
“Large replacement parts, such as gearbox and blades, are almost easy to find, but with respect to smaller components, we see significant margin inflation,” said Make Consulting’s Dan Shreve. He added that he is bullish on repair services especially by small suppliers. Printed circuit boards are getting a lot of attention.
“Wind resource optimization has become more important for owners. It’s not hard to understand that what happens at the top sweep of a 100m rotor is not the same as the bottom sweep. But the turbine-optimization market is somewhat of an ad-hoc service with lumpy revenue but high-profit potential. And cash flow remains a major sticking point for asset owners,” he said.
Validating energy gains from performance upgrades
“A 1% improvement in AEP can result in $2,500 per year for some turbines,” says GE’s Jeff Bergman. The Wind PowerUp Leader discussed results from the company’s WindBoost controls. “We’ve found that vortex generators can produce on average 1.2% improvement while the company’s improved controls improve turbines on average about 3.7%, although one chart showed improvements spanning 1.2 to 4.8%. The figures were from studying many of the company’s 1.5-77 turbines. The 8 million hours of data analyzed for the figures Bergman presented came from over 500 turbines across 13 wind farms. And no hardware upgrades were performed, he said. His closing comment was that continuous monitoring of key performance indicators helps ensure that energy gains are sustained over the life of the assets.
Plant performance and turbine to turbine interactions
“From the NREL Gearbox Reliability Collaborative, we find that 48% of a gearbox failure are attributed to high-speed-shaft bearings, 18% to helical gears, and 13% to intermediate speed shaft bearings,” said NREL’s Jon Keller. That data came from about 30% of the U.S. capacity. In fact, he added, axial cracking on HSS and IMS are most common. He also announced the close of the Collaborative and announced the Atmosphere to Electrons program, eight R&D areas such as high fidelity modeling, financial risk and portfolio analysis, and integrated wind plant controls. Public workshops will be held to consider projects the DOE’s R&D arm might work on over the next few years.
Sandia Lab’s Josh Paquette discussed the drivers to blade reliability, such as inspections, effects of defects, erosion, and repairs. Sorry, a lightning study is a future effort. However, he did show stunning video generated by lidar scans across the Labs outdoor test area in west Texas. Viewers could clearly see the several turbines on the test site and their wakes captured by the lidar. The surprise is that wakes extend farther downstream than expected.
Predicting drivetrain life
The main theme, said Dr. John Coultate, R&D and Consultancy Dept Leader, Romax Technology, is that turning the ‘textbook’ theory of predictive maintenance into practice is actually not that difficult and can be achieved using technologies such as CMS, portable vibration sweeps, and turbine de-rating. He presented an example explaining the cost benefit of predictive maintenance for main bearings with a clear advantage – over $300,000 in the example, mainly due to the reduction in large crane costs because they were spread simultaneously over multiple turbines. A second topic was de-rating to extend turbine life, which is often part of the same predictive maintenance strategy. Coultate explained some of the risks and benefits of de-rating and how this related to bearing life. Lastly, he explained how to use portable vibration systems on turbines without installed CMS and still gain much of the value of predictive maintenance.
Wind turbine lifetime extension: Wind turbine life assessment technology gap
Selecting a few turbines to serve as a sample subset of a fleet is the essence of the Flight Leader Concept presented by EPRI’s technical executive Luis Cerezo. The selected turbines can add valuable input to the life-extension decision making at minimum cost and with acceptable confidence, he says about this life extension strategy. Then an O&M crew can target audit inspections of structural components to detect incipient failure and ensure structural integrity past 20 years. Advanced controls would reduce fatigue loading and enable longer life. Lidar implementation near day one, for example, could provide an 8% mean load reduction and improve mean turbine life to 28 years.
Validating energy gains from performance upgrades
Managing the aging fleet is a Gamesa program in its third year, says the company’s Sergio Velez. The program is based on three pillars: technological fatigue analysis of structural elements, correlation with operating experience, and structural monitoring systems. The goal of the program is for long term containment of O&M costs.