With almost 500 turbines now erected in waters around the UK, and the same number again currently under construction, the offshore wind sector is starting to build towards a critical mass.
But some things will never change: the environment into which the turbines are placed will always be hostile and difficult to reach – so keeping them up and running, with the minimum amount of manual intervention, remains an ongoing challenge.
According to the latest statistics from the European Wind Energy Association, operation and maintenance amount to a sizeable share of the total annual costs of a wind turbine. These costs can easily make up 30% of the total levelised cost per kilowatt-hour produced over the lifetime of a new turbine. It’s therefore crucial that equipment manufacturers continue to attempt to lower these costs significantly, by developing designs that require fewer regular service visits and less turbine downtime.
An important part of this process is the installation of condition-monitoring or predictive maintenance equipment on new turbines – and retrofitted to old ones – to give operators a better indication of when parts are about to fail. This equipment has become increasingly sophisticated over recent years, flagging up wear and tear to crucial rotating equipment that would not be visible to the naked eye.
For condition-monitoring providers such as SKF, the wind sector has developed to become an important chunk of its business. “We retrofitted condition monitoring on to more than 100 wind turbines in UK waters last year, and we are increasingly finding that such equipment is being fitted as standard on many new turbine installations,” says Paul Deighton, general manager of SKF Reliability Systems.
“Offshore windfarms are more expensive to maintain than their onshore counterparts, mainly because of the difficulty of gaining access to the machines and the requirement to mobilise vessels and cranes. The consequence of failure is therefore more severe offshore. So operators really need to know what is going on with their offshore windfarm assets on an accurate, realtime basis.”
SKF’s primary technology is WindCon, a predictive system that can be installed on all turbine sizes and types, and is capable of continuously monitoring single units or entire farms to reliably predict when maintenance activities will be necessary. The hardware comprises vibration sensors mounted on a turbine’s main shaft bearings, drivetrain gearbox, and generator, along with access to the turbine control system.
The sensors and proprietary software combine to monitor and track important operating conditions, highlighting potential problems such as unbalanced propeller blades, misalignment, shaft deflections, mechanical looseness, foundation weakness, bearing condition, gear damage and generator rotor/stator problems. The system provides a performance overview that identifies faults and predicts failures before they occur, enabling operators to consolidate maintenance activities and perform necessary inspection and repair work during planned turbine stops. This also means potentially extended maintenance intervals, less unexpected downtime, and fewer unexpected costs.
The collected data also enables root cause failure analysis, which can eliminate recurring faults. When accessed with SKF’s WebCon software, the system enables operators to monitor turbine conditions in real time via any computer or hand-held device with internet access.
“The system can be used on all forms of rotating equipment in a windfarm environment,” says Deighton. “Typically, you would take a vibration reading at the bearing because, from a rotating machinery point of view, that’s the fulcrum point – that’s where the action is happening. But if you have a fan or any kind of impeller hanging on the shaft, it is also possible to pick out problems with that.
“The system can identify misalignment and balance problems, mechanical looseness or if the holding-down bolts aren’t tight. It gives a representative view of what’s going on.”
The SKF technology is a multichannel surveillance system – it can take 16 analogue and two digital inputs, monitoring the data continually and simultaneously.
This, says Deighton, gives it an advantage over more basic multiplex condition monitoring systems, which analyse data from one input for a designated period of time before moving on to the next.
“If any of the channels goes outside its parameters, an alarm is triggered and the data is collected from that channel until it returns to normal or you tell it to stop. So if you get a peak in the middle of the night, with no one looking at the data, and it falls back to normal by the morning, the information will be collected and stored for analysis. The system has also been designed specifically for the wind environment, so it has features such as lightning protection to give it a greater level of reliability.”
Deighton says the system has been proved to save wind turbine operators money. For instance, it was recently installed on a wind turbine that had already had damage to the low-speed part of the gearbox. A gearbox replacement was being planned when the WindCon system was first installed.
The system not only registered the damage – it also determined that the damage was stable enough to postpone the gearbox replacement and keep the turbine in operation, says Deighton. After monitoring the part for almost 12 months, the system detected a rapid increase in the damage pattern, and the turbine was taken offline for the planned gearbox replacement. By postponing the replacement for that time, the windfarm was able to accrue interest on the money needed for the overhaul, which was almost enough to pay for the system installation, he says.
More importantly, in terms of cost savings, the farm was also able to efficiently plan for parts delivery, shipping, personnel and cranes for the replacement – something that would have been impossible with a rushed operation accompanied by several weeks of downtime, he says.
Deighton says that the WindCon system is already being developed for other applications. “We have started installing oil debris sensors into gearboxes and taking the information and feeding it back through WindCon, so we can start to trend oil debris particles.
“The other area we are looking at is using laser technology to monitor blade wear and ice formation. And we think the technology is applicable in other markets, such as marine energy, monitoring and predicting the performance of wave and tidal devices.”