Wind turbines are typically designed to last for 25 to 30 years. During this service life they will require regular maintenance. As subsidies and power purchase guarantees have been removed, the wind industry faces significant pressure to reduce costs. Operators have been forced to reduce maintenance services and in some cases to end full maintenance contracts.
For wind turbines, as with any machinery, condition monitoring and preventative maintenance can prevent failures that could lead to secondary damage, excessive down time and higher costs. Any site maintenance of wind turbines is complicated by their remote location and height. This means that remote condition monitoring takes on even greater importance. This article examines the methods and technologies used to minimise the maintenance cost for these valuable but inaccessible assets.
Wind turbine manufacturers normally require an annual service of their product as a condition of warranty. This involves basic machine maintenance such as cleaning, lubrication, tightening bolts, changing filters and taking oil samples. Most of the maintenance takes place in the nacelle where the generator and, if present, the gearbox are housed. However, anchor bolts on the baseplate may also need to be tightened and turbine blades inspected for damage. Large turbines may even require internal blade maintenance.
Avoiding unscheduled maintenance requires estimates for the probabilities of component failures so that decisions can be taken on when to schedule maintenance to avoid an outage. The German Wind Energy Measurement Programme tracked the performance of 1,500 wind turbines in Germany between 1997 and 2006, providing detailed information on component failure probabilities and outage durations. At that time, on average, electrical faults caused stoppages every two machine-years and resulted in about 1.5 hours of downtime. Gearbox failures happened less frequently, about once every seven machine-years, but they resulted in outages of about six days. More recent studies have found that outage durations are increasing significantly as machines get larger.
The probability of a given component failing can be combined with cost of the outage which would result from its failure. The probability multiplied by the incident cost gives an expected cost. If this expected cost is greater than the cost of the preventative maintenance, then the maintenance should be carried out.
Offshore wind turbines are typically reached by boat. The boat will push on to the transition piece, the lower part of the turbine with an external ladder. At this stage, wind turbine technicians must wear a climbing harness and life jacket, as well as an immersion suit if the sea temperature is below 10°C. After climbing to a platform on the transition piece, technicians enter the tower to climb to the nacelle.
EVEN MORE CHALLENGING
Inspecting and repairing turbine blades can be even more challenging, as the blades are suspended at significant height with no access fixtures. Manual work is carried out by a rope access technician (RAT) who moves down a blade from the nacelle. Each blade must be orientated at a slight angle from the vertical, extending downwards from the nacelle to allow access, one blade at a time. Ropes looped around the blade allow the RAT to maintain contact with the blade while working. Visual inspection may also be carried out by drone, and robotic repair solutions have been developed.
Because of the challenges to manual access to blades, other methods have been developed. For inspection, there are typically three options. First, ground-based inspection uses large telephoto lenses and ultra-high definition imaging. Special software can be used to stitch together approximately 1,100 images to give a complete view of a blade. The position and dimensions of defects can also be measured using this software. Regular ground-based inspections can be carried out to generate a record of blade condition over time. Ground-based inspection has the advantages of not requiring working at height, and not requiring favourable weather to be carried out.
Drone inspection can provide a similar service to ground-based inspection, with enhanced image clarity resulting from greater freedom in positioning the camera.
Once a defect has been detected using a remote method, manual Inspection is often required as a follow-up to assess the extent of damage. It provides the best observations of defects. Detailed photos of defects can be inspected by engineers after the rope access work. Other inspection methods also used include laser scanning to create 3D images, and conductivity to determine continuity of lightning protection systems.
Once a defect has been identified as requiring repair, the blade is normally manually accessed again to fill and/or sand down the affected area.
Robot developer Aerones has developed a number of robotic solutions for inspection, cleaning and repair to be carried out without manual access. These robots have a stabilising tripod frame suspended from a cable system which carries a robot arm. This enables the robot to access both the blades and the tower of a wind turbine. Different tools can be attached to the robotic arm (pictured above) for visual inspection, laser scanning, conductivity testing, surface cleaning, drainage hole cleaning, sanding, and filling.
“Technicians who would otherwise hang hundreds of metres in the air manually working on each turbine blade, where safety concerns and weather limit operations, can now operate these inspection and maintenance services from the safety of a van, with controls that are as easy to use as a kid’s gaming console,” says Janis Putrams, CEO of Aerones.
Access challenges make condition monitoring particularly important in wind turbines. The potential for significant downtime in wind turbine gearboxes makes these the prime candidates. Insurers often require all roller bearings in the drivetrain to be replaced after five years, unless condition monitoring is being carried out.
The specific requirements of wind turbines often mean that standard methods of condition monitoring don’t apply, such as those covered in ISO 10816-3 (Mechanical Vibration — Evaluation Of Machine Vibration By Measurements On Non-Rotating Parts — Part 3: Industrial machines with nominal power above 15 KW and nominal speeds between 120rpm-15,000rpm when measured in situ). Vibration of the tower and nacelle caused by wind flow disturbances, resonances within these structures, and potentially also sea swell for offshore wind turbines may alter the time behaviour and spectra of the vibration signals compared to those of other industrial structures. However, special standards have been created to deal with wind turbines, such as ISO 10816-21: 2015, which covers horizontal axis wind turbines with a gearbox. (A second standard is also planned, ISO 10816-22, for direct drive wind turbines).
ISO 10816-21 recommends zones for evaluating vibration at continuous load operation. For the aerodynamically-excited vibration of the nacelle and tower with frequencies between 0.1 Hz and 10 Hz, vibrations should be measured over a 10-minute period. Typical measuring positions are in the nacelle at the main frame close to the rotor main bearing, on the structure above the tower flange, and in the rear end of the nacelle on one side of the generator or main frame. Measurements of acceleration or velocity should be made along the rotor shaft axis and in the radial directions.
Gearboxes and generators, with characteristic vibration with frequencies from 10 Hz to more than 1,000 Hz, can be measured using shorter evaluation periods of around one minute. All vibration measurements should be measured in the axial and radial directions. For rotor bearings with three-point suspension, measurements should be taken on the housing of the front bearing. Where two separate rotor bearings are present, measurements should be taken on both housings. Gearbox measurements should be made on the housing close to the input shaft bearing and output shaft bearing. Generator vibration is either measured on the housing for integrated gearbox-generator designs or on the bearing housings for elastically-coupled designs.
Although access to wind turbines can make maintenance challenging, a combination of remote condition monitoring and automation can enable cost-effective operation. These trends are expected to continue, driving down the cost of renewable energy.
BOX: There’s a robot for that
One year on from the launch of MIMRee, Multi-Platform Inspection, Maintenance and Repair in Extreme Environments, project partners report breakthroughs in their quest to demonstrate an end-to-end autonomous inspection and repair mission to offshore wind farms.
Awarded a £4.2 million grant from Innovate UK, the project is one of offshore wind’s most ambitious robotics project to date. The end-game is demonstration of an autonomous system capable of planning its own operational missions to offshore wind farms, whereby a ‘mother ship’ will scan moving turbine blades on approach, then launch teams of inspection drones carrying blade crawlers for forensic inspection and repair of damaged blades.
The list of successes in solving technological challenges in the first year of the two-year project suggests that the vision might well be feasible, according to project participants.
First, the Thales imaging system has achieved blur-free images of moving wind turbine blades at the Offshore Renewable Energy (ORE) Catapult’s Levenmouth Demonstration Turbine off the coast of Fife. Scanning blades for defects, without stopping turbines for days at a time, is considered a game-changer for wind farm operations.
Second, MIMRee mission planning software developed by Prof. Sara Bernardini of Royal Holloway University of London has been integrated with the Thales vessel and the inspection drones developed by a team from Manchester and Bristol Universities. The drones have successfully coordinated launch, recovery and navigation from the vessel.
One of the aims of the project is to demonstrate an integrated inspect-and-repair system for wind turbine blades, using the BladeBUG robot (pictured), which has recently demonstrated its walking abilities on a variety of blade surfaces at ORE Catapult’s National Renewable Energy Centre. Again, multiple technologies have been shown to work. First, an autonomous repair arm developed by Dr Sina Sareh’s team at the Royal College of Art Robotics Laboratory can rapidly switch between modules for cleaning, sanding and top-coating damaged areas of blades, providing real-time feedback visualisation and human-in-the-loop remote operation of repair tasks via a user-interface system, the team says.
Second, plant Integrity has produced the blade crawler’s non-destructive testing (NDT) payload. The module uses a precision scanner for exact measurement of defects under a wide variety of ambient light conditions. Third, an electronic skin, called Wootzkin, patented by robotics company, Wootzano, will enable the robot to determine the surface conditions of the blade, helping the robot to walk in an extreme environment.
BOX: Real savings expected from virtual reality
A data-rich digital twin has been created of Offshore Renewable Energy (ORE) Catapult’s Levenmouth Demonstration Turbine. The software simulation, called R2S, enables personnel to remotely visit sites and easily access technical and operational data.
Lorna Bennet, mechanical engineer with ORE Catapult, said: “We believe that R2S’s technology could revolutionise offshore wind O&M. The digital twin has proven to be a unique solution enabling technicians and interdisciplinary teams to view the turbine’s inner workings and all pieces of equipment anywhere, at any time. This has enabled remote inspection of the turbine, boosting the planning of operations and reducing the need to go offshore. It has also changed the ways individuals are working – generating significant time and financial savings.”
Laura Fairley, market development manager at engineering consultancy project partner James Fisher Asset Information Services added: “With offshore turbines getting bigger and more powerful, operations and maintenance costs have escalated. R2S has proven to be an effective planning tool.”
JF AIS is now looking to work with Vattenfall on the European Offshore Wind Demonstration Centre in Scotland.