Moving turbines offshore requires a top-to-toe reassessment of the equipment and its response to maritime conditions
Simon Rees, Frazer Nash Consultancy, Bristol, UK
Throughout history, the engineering challenges of the marine environment have been consistently underestimated. The ancients characterized the power of the sea in the myths of Poseiden and the fall of Atlantis, but the oceans have retained the power to surprise us into the present day, and sacrificing chickens has never been truly effective. There are plenty of more recent examples that illustrate the dangers that face us, ranging from the destruction of the Mulberry harbour off the Normandy coast in 1944 to the recent loss of the ferry Estonia.
However, the sea has many more subtle and pervasive ways to destroy than by the force of waves alone. Although there is an ongoing academic debate about the prevalence of extreme wave conditions, this problem is usually recognized by the engineer and dealt with by increasing the strength of the design. Less well appreciated are the problems brought by prolonged exposure to a cold, damp, corrosive atmosphere, with limited opportunities for maintenance. High salinity and poor weather are unfortunately more common in those areas of greatest interest to the potential harvester of marine power.
Marinization is not merely modifying a design with materials and redundancies that increase the robustness of the system. Although this improves the quality of the device, it is invariably accompanied by an increase in cost that may not be supportable by the project as a whole. A more thorough approach to marinization is to consider all those factors that would normally be taken into account – cost to manufacture, maintain, replace – and to define an optimum system specification tuned to the actual environment.
For example, consider a ship’s hull. A simple material replacement option for a large commercial steel hull, which is prone to corrosion, is to replace it with aluminium. However, in practice it is much cheaper to use sacrificial zinc anodes to protect the hull electrochemically, and accept that they will need to be replaced occasionally. That solution involves two lower specification materials and a maintenance strategy that costs less through the life of the vessel than material replacement.
North Hoyle was the UK’s first offshore wind farm. ‘Second round’ sites will be farther from the shore. Photo: Anthony Upton
This may seem an absurdly simple example, and in reality the evaluation of costs and benefits may prove to be considerably more marginal. There will also be constraints to consider when taking existing terrestrial equipment offshore, but these can also be expressed in the language of life-cycle costing. With a clear and coherent assessment strategy, a balance can still be found.
The probable situation
The probable situation is that there is a proposal to use a device developed for terrestrial use in a marine application. The device will have been proven over a suitable time by testing or service deployment, and the attraction of offshore conditions leads to a financial case for moving out to sea. Such a device may be a wind turbine, for example, whose foundations, tower, nacelle and turbines are all well understood in onshore conditions. Alternatively it may be a wave or tide device.
In order to maintain reliable performance some degree of marinization is required, but the changes must take account of the costs of modifying the existing design and tooling, and the available capital budget.
Aggressive corrosion problems must be identified and resolved immediately. However, after this essential point the implications of design changes become more complex.
Degrees of freedom
The options open to the developer depend on the degrees of freedom available in the existing design.
Given these freedoms, the options include:
- Changing the fundamental design to make it right for the environment
- Improving the design to allow maintenance with a minimum of effort and cost
- Testing to identify weaknesses, and to define an appropriate management and maintenance regime.
The balance is clear – the more fundamental changes that can be made, the lower the cost of operation and maintenance, but the greater the initial investment. A choice between design routes can only be made on the basis of a properly informed and constructed cost model of the development, and that model requires an understanding of what the design options involve.
Getting the design right
Marinizing an existing design does provide the benefit that weak components in the system should be apparent from terrestrial operations. It is reasonable to assume that exposure to the marine environment will only increase the failure rate of those components. It is also reasonable to assume that there will be some components that are perfectly adequate onshore but fail more frequently offshore. Identifying these items and predicting their performance demands a combination of different technologies.
Firstly, there must be confidence in the structural integrity of the design. The designer must know the loads to which the device will be exposed, and the strength and fatigue properties of the materials. Particularly problematic components can be subjected to structural reliability assessments, which utilize statistical assessments of load and material property variations to predict likelihood of failure. This technique is also useful for forensic studies into failed components.
Secondly, the designer must have confidence in the mechanical integrity of the components used. Again understanding physical loads and spectra are essential, but it is at this stage that a thorough assessment of the potential failure modes for each component should be undertaken. Returning to the wind turbine example, there is a move towards direct drive generators and even a return to two bladed designs to reduce the number and complexity of components. Where such radical changes are impractical, de-rating of bearings and other fundamental components can provide significant benefits for reasonably low cost.
Seals are another weak point – designs that let surplus lubricant out can also let water in. Condition monitoring is important, with surveys showing that up to 50 per cent of shutdowns of onshore turbines are caused by electrical and electronic systems failures, so the system should include a degree of self-diagnosis and remote override.
Denmark’s Horns Rev wind farm. Replacement, with as much work as possible completed on shore, was the answer to generator problems at these units. Photo: Elsam
Aggressive corrosion has been mentioned above but more subtle and long term problems are inevitable in a marine environment. A factor to take into account in material selection is electrochemical corrosion. This can be used to your advantage, but in the damp marine atmosphere apparently reasonable metal combinations become batteries and suffer incredibly rapid reactions. Even if this problem has been dealt with, marine grade coatings, sealants, lubricants and paints should be liberally applied.
Finally, it is important to think about the configuration of the system. Redundancy should be introduced where design and cost allows, but beware spurious Mean Time Between Failure (MTBF) targets – a redundant system may have twice as many components, halving its MTBF, but its availability will be greater. If an item is either known to be a reliability problem, or is in some way consumable, modularization should be used to allow for easy replacement.
Maintenance at sea
The cost of maintenance at sea is between five and ten times more than on land. That means that a fundamental of the design is that it can be maintained with a minimum of effort and cost. It is absolutely necessary to review the design to ensure that any maintenance work can be done as easily and quickly as possible.
The CEC-Joule funded Opti-OWECS study (see panel) into offshore wind-turbine operations assessed four different maintenance strategies:
- No maintenance. Neither preventative nor corrective maintenance are executed, and major overhauls are performed every five years or so. One of the few alternatives is exchanging a whole turbine if availability drops below a predefined minimum or after a certain amount of operational hours. Given the current level of turbine failure rates, this option is not presently viable.
- Corrective maintenance. Repair is carried out soon after a turbine is down, or, alternatively, when a certain number of turbines are down. No permanent maintenance crew is needed;
- Opportunity maintenance. Corrective maintenance is executed on demand and the opportunity is used to perform preventive maintenance at the same time. No permanent maintenance crew is needed;
- Periodic maintenance. There are scheduled visits performing preventative maintenance, and corrective actions are performed as necessary by a permanent dedicated maintenance crew.
The study concluded that the designs at that time could not be economically operated with reduced corrective or preventative maintenance regimes because they had originally been designed for a terrestrial environment. However, if the other maintenance strategies can be implemented there are potentially very great cost benefits. It also concluded that O&M strategy should be optimised with respect to localised energy production costs rather than pure capital or O&M costs.
Looking more widely than the power industry, there are many examples of what can be done to achieve a reduced maintenance load. The commercial aircraft industries are a particularly good example where power companies may be able to take advantage of this experience. There are also many other examples in the process and generating industries, but one of the luxuries that the offshore generating industry does not have is the ability to flood maintenance tasks with labour. Some suggestions of what should be considered are given below.
Modularization, mentioned above in the context of reliable design, will also be a boon to the maintenance technician. The size and weight of modules is a restraint as there are health and safety limits on what a single person can handle, especially as they have to position the item where accessibility is poor. Modules can be made larger as well as smaller if this reduces the amount of maintenance necessary at sea. It is much cheaper to bring something back on land for repair, even if it means towing it, than to engage in prolonged activity on-site, and this strategy is employed by Ocean Power Delivery for their Pelamis device.
Permanent handling gear, integrated with the device, simplifies known maintenance tasks and helps with the manoeuvring of heavy parts. Special multi-functional transportable tooling should be designed where the capital cost can be justified by a reduction in expensive offshore maintenance time. Another time saving feature is an on-board diagnostic system that quickly identifies where problems have occurred. Finally, the designer should ensure that the number of steps in any task is minimized, for example by removing the need to make changes to settings after installation.
All potential maintenance tasks, planned and unplanned, should be subject to a risk assessment (preferably involving someone who routinely undertakes maintenance), and any identified safety hazards designed out. Safety assessment for offshore operations is a major topic in its own right and is not discussed here, but good safety practice stems from good design and must be taken into account from the start.
Test the design
There is no way to get away without testing. It is the nature of marine engineering that something will have been underestimated, or missed entirely, such is the number of different processes of destruction the device must face. The tests that will have been done to prove the performance of existing equipment will, in all probability, be representative of land based conditions. There is a set that is required to prove acceptability for the marine environment, which includes:
- Survival in the marine extremes
- Operation in the marine extremes
- Salt spray.
These expose the device to representative levels of force, moisture, salt and temperature, and are governed by a number of national and international standards (depending on type and level of equipment). However, there is no substitute for the real thing and genuine confidence in the technologies can only be obtained from a prolonged trial at sea.
Even when a design is proven for the marine environment, installed and running, the job is not over. A marine application will require appreciably more effort to manage and maintain throughout its life than an equivalent land based item, the problems it experiences will be more acute and the frequency with which these problems arise will be greater.
Operational costs will never be as low as on land, as even routine tasks will be harder to execute. However, the environment that poses these problems is far richer in energy resource than the land, and providing the cost implications of marine design, operation, maintenance and disposal are understood there is no reason why the rewards should not be reaped.
Europe’s sea view
In 2001 an EC funded project, Concerted Action on Offshore Wind Energy in Europe, acknowledged that “Operation and maintenance of offshore wind farms is more difficult and expensive than equivalent onshore wind farms”. Onshore wind turbine availability levels are generally very high – in excess of 97 per cent with appropriate servicing and maintenance In practice, this typically equates to visiting a wind turbine four times a year, either for regular service or for repair tasks.
However, offshore conditions cause more onerous erection and commissioning operations and accessibility for routine servicing and maintenance is a major concern. During harsh winter conditions, a complete wind farm may be inaccessible for a number of days due to sea, wind and visibility conditions.
Among the other points made in the report:
- Onshore equipment can be sourced and mobilized within a short period of time, usually within hours, and is available on site within a day. Offshore lifting cranes are uncommon, and will generally have to travel a considerable distance to site;
- Access to the turbine may be difficult or impossible in harsh weather conditions due to wave heights, wind speeds and poor visi bility. There are a number of projects addressing the issue of improved access;
- Electrical and control system failures account for the highest percent age of failures. Typically, failures of this nature occur due to the number of components, poor electrical connections, corrosion, lightning strikes, etc.;
- Wind farm sites being considered in the North and Baltic Seas present harsh maritime. Cathodic protection technology of subsea structures is integral to the front end engineering design, with due consideration of modern paint systems and metal spray coatings, particularly for application within the splash zone;
- More work is needed in developing support structures which can withstand stresses caused by wind and wave loading, together with reductions in material fatigue strength caused by corrosion;
- Power for the turbine controller, electrical actuators, monitoring and communications systems are drawn from the turbines gross output, or imported from the grid system. In the event of loss of power generation or lost electrical grid connection, backup power is required.