Cutting the cost of offshore wind energy

EUROPE: Estimates of offshore operation and maintenance costs vary considerably. Although a number of turbines have been operating satisfactorily for several years, machine sizes have increased during that period and some of the more recent installations have encountered problems. That has made reliable assessment of long-term trends difficult.

Kentish Flats. Serving accounted for half of its running costs in one year

Figures reported to the International Energy Agency (IEA), comparing onshore and offshore costs, give some indication of the uncertainty that surrounds operations and maintenance (O&M) (see chart, below). The data, provided by six countries and the European utilities organisation, Eurelectric, were published in the 2010 edition of the IEA’s Projected Costs of Generating Electricity. The Netherlands reported the lowest estimate - $11/MWh (lower than its onshore equivalent of $18/MWh) - while Germany reported the highest, at $46/MWh.

Although the amount of operational experience is limited, useful data have been reported by some of the early British offshore wind farms. These were awarded capital grants by the government and, in return, were required to report on the performance of the wind farms for the first three years. These reports have included descriptions of problems that have been encountered, the electricity generated and the costs of O&M The figures (see chart, page 10) show a spread of values, between £8/MWh ($12/MWh) and £22/MWh ($33/MWh), with the majority in the range £10/MWh to £20/MWh ($15-30/MWh).

Although there is no clear consensus on O&M costs, the British results suggest that values below £10/MWh ($15/MWh) are achievable. For example, in 2008 Kentish Flats, situated on the River Thames estuary to the east of London, recorded a value of £8/MWh ($12/MWh) in a year when the availability was 89%. As some costs, such as insurance and possibly some elements of the service contract, are ‘fixed’, this suggests that a slightly lower value may be possible with higher availability. This would be on a par with the Dutch reports of $11/MWh.

Planned and unplanned maintenance

The figures from Kentish Flats suggest that servicing costs account for roughly half of all offshore O&M costs. These are the only costs over which the plant operators have full control so there is a strong incentive to keep them down. Servicing costs comprise two elements - for planned and unplanned maintenance.

Experience from onshore wind farms suggests that unscheduled maintenance costs account for between 25% and 40% of total servicing spend. If the value of the revenue lost while the turbines are down was added to this figure, the total would be higher. There are few figures available on unscheduled maintenance costs offshore.

In Europe, planned maintenance takes place, where possible, during the summer months. Wind speeds and wholesale electricity prices both tend to be lower at this time, so the loss of revenue while the turbines are switched off is also lower. As the winds are lower, rough seas are less likely to hamper access to the turbines.

The report from the Kentish Flats wind farm for 2007 shows that there were 89 days when access was not possible, but this varied from 4% of June days to 16% of December days. At the more exposed Barrow wind farm, in the Irish Sea, access was not possible for 49% of the time from July 2006 to June 2007 and this contributed to a low availability figure of 67%. Developers are looking at ways to reduce the time when access is not possible.

Condition monitoring

The high costs of repairing offshore turbines makes a strong case for the use of condition monitoring systems. During the winter months, daily load factors of 50% or more are not unknown and so the loss of revenue from a 3MW turbine that receives remuneration at Û15/MWh for its electricity is Û540/day. If just three days of downtime can be avoided in a year, that saves Û1,500 of lost revenue, quite apart from the repair costs and the hiring of specialist vessels needed to gain access to the turbine, which may not be available at short notice. These costs need to be compared with the annualised costs of purchasing and maintaining condition monitoring systems, which are in the region of €2,000 to €5,000.

The concept of condition monitoring is not new. So-called ‘hot box’ detectors have been used on railway systems for many years to give warnings of overheating bearings, and the techniques are widely used in shipping.

The principles are the same. If, for a modest outlay, sensors can monitor critical components, operators can be alerted to likely problems that could result in expensive repairs and loss of revenue. The typical parameters that are monitored are shown in the table below. The outlay can be justified provided the savings over the life of the plant exceed the initial cost of the condition monitoring equipment. In a case study presented at the European Wind Energy Association’s offshore wind conference in 2009, instrument manufacturer SPM Condition Monitoring Solutions suggested condition monitoring had brought annual savings of €34,000 per turbine.

Some parameters, such as vibration levels, can be accurately and reliably measured but others are more challenging, such as detecting structural faults in blades. A number of techniques are being explored, including greater use of fibre optic strain gauges and acoustic emission techniques. The latter ‘listen’ to the noise signals coming from rotating machinery or from within a turbine blade and compare the signals with those from ‘healthy’ components.

The future

Wind turbine technology is evolving, with machine sizes increasing and design features such as permanent magnet generators becoming more widespread. Although failure statistics from onshore turbines are a useful guide to performance offshore, repair costs for the larger items, such as generators and gearboxes, are likely to be significantly higher. This reinforces the need for reliable condition monitoring systems. The growth of offshore wind is likely to lead to further improvements in these systems so that operations and maintenance costs can be minimised.