The latest feature added to the Clir Renewables AI platform will help wind farm operators keep their maintenance costs down using machine learning

Wind farm

(Picture: Peter Skelton/KGPhotography)

In a bid to help predict and prevent wind turbine failure ahead of time, optimisation software firm Clir Renewables has launched the latest iteration of its flagship AI platform to keep operators one step ahead of their machinery.

The Canadian company’s technology uses machine learning to identify anomalies in component temperatures and detects faults as soon as possible, arming wind farm owners with the knowledge and foresight to fix the issue in advance.

By learning the behaviour of certain turbine components at all stages of their life cycles, it is able to tell when they are operating at higher than expected temperatures under particular circumstances such as increased loads.

Clir Renewables CEO Gareth Brown said: “We really wanted to focus on building detection that has limited false positives, so the tool isn’t wasting peoples’ time while maximising the benefit of early fault detection.

“The approach maximises the use of the data to drive improved performance, and crucially it can be scaled across all turbine technologies and as components are upgraded or replaced.

“It’s exciting to see when we take deep domain expertise and apply the latest and greatest AI techniques what we achieve”.

clir renewables ai platform
Clir Renewables’ AI platform uses machine learning to detect the temperature of various components within a wind turbine (Credit: Clir Renewables)

What problems does the Clir Renewables AI platform solve?

Wind farm maintenance budgets account for the vast majority of their owners’ operating expenditure, and increase exponentially when machinery faults progress undetected and ultimately require unexpected repairs.

When a component failure occurs, it also puts the entire associated wind turbine out of operation for anywhere from a few days to a few weeks, depending on the maintenance requirements, resulting in costly downtime and lost energy generation.

Faults are difficult to detect, as component temperatures can vary drastically depending on the conditions at the time.

Clir Renewables’ AI platform helps put this data in context and accounts for a wide array of factors including ambient temperature, rotor speed and seasonal variation, allowing for accurate and timely assessment of a potential problem.

It learns the behavioural patterns of the myriad parts of a wind turbine, and when it detects an anomaly it automatically recommends actions based on the severity of the issue.

Founded in 2017, Clir Renewables has grown exponentially and now supports more than 4GW of assets across the world, claiming to deliver a 5% increase in annual power production for its clients after their first year.

The company, headquartered in Vancouver, also opened an office in Glasgow, Scotland, in 2018.