Send your latest releases to

Reliability the key to offshore wind power success

Published:  02 July, 2014

Reliability is the top priority when selecting technology for wind power projects offshore. Compared to onshore installations, repair and maintenance offshore is far more costly. This means that operators go for tried and tested solutions, basing their decisions on historical data, which may slow down the pace of technical development in the industry. ODEE reports.

Offshore wind farms have a strong focus on reliability. Reliability is an extremely important factor on offshore installations, repair and rectification costs are significantly higher in the offshore environment due in part to the increased difficulty with access issues along with intermittent weather conditions which can extend or hamper repair activities.

Two thirds of the cost for offshore wind power is related to infrastructure, maintenance and management, with only one third related to the wind turbine itself. This means it makes financial sense to install large generators capable of capturing as much energy as possible from the installation.

But apart from the fact that generators in offshore wind farms are on average more than twice the size, 3-5 MW compared to 500 kW to 2 MW onshore, the generators themselves are the same fundamental design in both applications. Still, the choices operators make are different in offshore applications, as reduced operational expenditure is a powerful driver offshore. In onshore applications, choices tend to be more driven by control of the capital expenditure.

Availability top priority

Offshore, availability is the prime concern, not the first cost. Suppliers are more likely to select a component supplier they know to be reliable, rather than a different supplier offering a lower price. Peter Wright, motors and generators service manager, ABB Limited explains: “The perception of reliability has very little to do with any theoretical reliability data the manufacturers may provide. Instead, it tends to be based on experience and field data gathered by the operator or turbine manufacturer making the purchase.

“A generator may have a design life of 20 to 30 years and often reaches this age in an industrial environment. In a wind power application, the same generator may only last half the time, or less, as the wind power nacelle is a very harsh environment. While the operational life is long enough for manufacturers to lose track of their equipment, operators keep their records and know what technology they can trust.”

Most manufacturers and operators use more than one supplier for each type of component and make comparisons. This means they often have better reliability data than the manufacturers themselves. In particular, manufacturers of wind power turbines for offshore applications are keen to keep warranty costs to a minimum and only use equipment that in their experience performs reliably.

New technology advances slowly

But the result of this approach is that technology evolves only slowly in the industry and this may be one explanation as to why direct drive generators are still not particularly common. Wright explains: “Direct drive generators eliminate the need for a gearbox, one of the greatest problem sources in wind turbine design.

Traditionally, wind turbines use a gearbox to convert the speed, 15 to 20 rotations per minute for a large, one-megawatt turbine, to the faster 1800 rpm required by the generator to generate electric power.”

He continues: “The direct drive generator operates directly on the low speed shaft of the wind turbine. This results in a design that is simple, robust and highly efficient, but which is fairly large at just under four metres in diameter. The large diameter is needed because the low speed requires a much greater number of poles.” This makes these generators too large for most onshore applications but offshore this size is not prohibitive.

A halfway house is provided by the semi-direct drive using a simple planetary gear. This offers a compromise between direct drive and a traditional design with a gearbox. This type of gear is more reliable than a full 3-stage gearbox and requires less maintenance. The configuration brings the speed up to 146 rpm and the generator used is just under 2 metres diameter.

However, both these solutions are still in the background of the standard concept with a double-fed generator and traditional gearbox. Wright says the reason, quite simply, “is that this is the way customers want it”.

Dedicated maintenance

Maintenance of offshore wind farms is typically provided by a maintenance organisation devoted to keeping the turbines operational. Access to the site is by ship or helicopter, with some wind farms having service teams on site in offshore accommodation. Heavy service, such as gearbox replacement, requires the use of a jack-up rig.

As more and more offshore wind farms move beyond their warranty periods, we may see a shift in maintenance patterns emerge with alternative suppliers of services and equipment.

In addition to the service and installation costs, wind farm operators also need to consider the value of lost production during standstill. This can be as much as £12,000 per week for a 3 MW generator. A catastrophic failure requiring a new generator can cause huge losses, as it may take 20 weeks from order to delivery of a new generator. That is nearly six months of lost production.

Because offshore wind farms are very remote, prognosis and monitoring systems will, over time, become more necessary. This facilitates just-in-time maintenance, reducing operations and maintenance costs.

Condition monitoring

Wright comments that the main reasons why equipment such as generators and gearboxes have a shorter service life in wind power than in industrial applications are the constant load changes and vibrations. The most common types of problems are associated with the bearings, the generator rotor and the gearbox.

Preventive maintenance is based on time-based service intervals for critical parts. Predictive maintenance, on the other hand, is based on the condition of the machine. Maintenance on this basis can halve the downtime, compared to a time-based approach.

Predictive maintenance uses condition-monitoring equipment to detect emerging faults and aims to sound the alarm bells before something actually goes wrong, whilst trying to avoid giving false alarms. Condition monitoring is an approach used in many other industries and can contribute to large savings. For instance, if the failure of a gearbox can be avoided by replacing a faulty bearing in time, the operator will have saved some £150,000 on the hardware alone.

Almost all generators sold during the last 10 years, explains Wright, are fitted with some kind of condition monitoring equipment. However, he says that this equipment is often primitive, generates false alarms and frequently offers no clues as to the cause of the problem. This is a major reason why predictive maintenance has not yet taken off in the industry.

A more advanced system, according to Wright, is sold by ABB as a stand-alone product. The company’s MACHsense-R remote system is used across a range of industries and monitors vibrations as well as their waveforms, comparing these to a database linking different waveforms to their causes. This is of particular relevance in an industry where mass production of wind turbines is not yet a decade old and robust operational data, where it exists, often forms disconnected clusters of information.

New generator technology may improve the reliability of wind turbines in the future. But until such time that this technology is adopted, condition monitoring of traditional technology can help provide more economical operation. Condition monitoring is a major area of interest in all industries with large capital costs and long running hours. Wind power should be no different; in fact, it ought to be even more important given the high cost of downtime and the steep installation and service expenses. Condition monitoring is an area that looks set to develop strongly over the next decade.

For further information please visit:

Last issue

View the last issue here.

View the past issue archive here.