Reducing the costs of operations and maintenance (O&M), which in turn influences the average levelised cost of electricity (LCOE), is a great incentive to developers and operators in the offshore wind energy sector. Smart technologies, including those based upon sensors and robotics, undoubtedly have a major role to play here.

Access to offshore wind turbines is still largely dependent upon weather conditions and the height of waves, particularly now that new projects are being developed further from the shore to exploit the stronger winds that can be found there. Helicopters can be used to get technicians, tools and components on site, but these are expensive to work with and their use is also restricted in bad weather. The sector is therefore investigating developments in the field of data analysis, robotics and even artificial intelligence (AI). A report by DNV GL shows that these technologies have considerable benefits for developers and operators of wind farms, including monitoring by robots, careful inspections using AI and self-driven transport to deliver components.

Greater insight

The major challenge in today’s offshore wind energy sector is how to eliminate the need to enter the turbine. Ideally, it would be possible to repair the installation remotely and to predict and prevent defects by component analysis. AI has an important role to play in attempts to achieve this.

HOME Offshore (Holistic Operation and Maintenance for Energy from Offshore Wind Farms), a research project into how new technologies can bring down maintenance costs, is underway at the University of Manchester. The project involves the use of big data and sensors as well as advanced digital simulation (in particular in the design of floating foundations), the integration of wind by better forecasting, and the use of robots for operation and maintenance.

A number of applications are already possible, not in the least because a great deal of data from the turbine is currently not being used. It is expected that existing performance data from components will be used to analyse the condition of other turbine components. State-of-the-art modelling needs to be improved to put us in a better position to run ‘what if’ scenarios. Robotics can be used for the maintenance of wind farms, undersea cables and platforms, in addition to big data options and advanced sensors and sensor technologies.

The American company Sentient Science has developed models based upon complex materials to gain more insight into the stresses that individual components are subject to. This allows the company to detect the weakest points of a specific system, to the point where it is possible to say whether a given component may have different weaknesses in different makes of turbine. This in-depth understanding of the structure of a component allows the company to predict how a component will perform in specific circumstances or machines. It can be used to gain a clearer picture of when a turbine will fail and the cause of this failure.

One example is MHI Vestas Offshore Wind (part of the Danish wind turbine manufacturer Vestas Wind Systems A/S), which has developed a SMART Fast Data solution that collects data from around 1,000 sensors in a V164-9.5 MW turbine between 1 and 50 times per second. This provides very detailed information about the health of the turbine, improving availability and thus increasing energy production.

The use of robots

New offshore projects further from the North Sea coast will need technicians to be permanently based on floating maintenance platforms. Robots can assist them in a number of ways.

Improved sensors and modelling can give a good indication of the condition of the turbine, but closer or on-the-spot inspections will certainly still be required. Drones are already being used to monitor the rotor blades of wind turbines on land. This option usually depends on the pilot being close at hand to steer the drone visually. At sea, the situation is completely different. Ideally, it would be possible to steer the drone to the turbine, with the operator on the mother ship. However, delays in data transmission of as little as a few hundred milliseconds can cause problems.

Another field in which robotics can help, is the fitting of components at sea. Transport equipment is a field where there is a lot of room for improvement. A typical situation can be the transport of a specific component that the engineer did not have available at sea.


Despite of all this, AI will not be in a position to replace the technician in the wind industry any time soon and neither will it be possible for robots to repair defective wind turbines without any human intervention. The use of AI throws up certain challenges, one of these being the capacity to learn. Tests show that a trained system can usually provide the correct answers, but it may not be clear how it arrived at a certain result. In the field of safety, this gives rise to enormous pressure regarding the correct design of the initial training. After all, the consequences are difficult to estimate if human interpretation is a factor in the way in which the machine is programmed to train itself.

Sirris, too, is performing research into smart technologies in the offshore wind industry: 

OWI-Lab is involved in the IBN Offshore Energy and VIS-OWOME projects in this field.