By Gareth Brown, CEO, Clir Renewables
The U.S. offshore wind industry is set to take off in the coming years, with more than 5 GW of cutting-edge turbines due to come online by 2025 and take advantage of the country’s high offshore windspeeds. However, in order to make the most of this fantastic opportunity to increase renewable energy’s share of the country’s generation mix, owners must plan for operations and maintenance (O&M) that is effective at preventing both downtime and underperformance during periods of high wind resource.
By incorporating the best practice and lessons mature markets have learned over the course of decades into O&M strategies, future offshore wind project owners and operators will be able to prevent any hold-ups in performance and ultimately maximize production from the first day of operations.
Lessons from mature markets
The “offshore” nature of offshore wind raises a number of very specific challenges for scheduling O&M. Whilst repairing a faulty onshore turbine might require a wait for specialized technicians and equipment to become available, maintaining an offshore turbine requires finding a window where technician and tool availability matches safe sea conditions and crew transfer vessel availability. This means that turbines can see extended downtime over winter when sea conditions are least favorable.
Calmer summer seas and lower winds allow for more opportunities to send the O&M team to site and minimize any lost revenue due to downtime; therefore, as a rule, offshore wind farms in Europe and Asia schedule routine on-site inspection and repair for this season. If any small errors and defects identified in the course of monitoring are fixed during this period, the turbine is less likely to fail during a peak production month and remain offline for an extended period while conditions are unsafe for on-site repair.
However, this forward-thinking strategy is often undercut by ineffective year-round monitoring of wind farm performance. Small defects, particularly those that begin at a subsurface level or due to sensor error, are easily missed if an operator relies on visual inspections or standard, top-level data analysis.
It is vitally important that they are able to tackle all issues affecting turbine performance in these summer weather windows. If anomalies or early stage faults go under the radar, the owner will be faced with long term turbine underperformance at best, or a critical failure forcing the turbine offline in times of high resource with no safe access window to rectify it, at worst.
Finding the root cause
Typically, O&M teams rely on SCADA data to diagnose turbine issues. However, if only turbine SCADA is analyzed, many small or sustained instances of underperformance (which can make a big dent on balance sheets if missed) will be lost in the data “noise” caused by fluctuation of resource.
It is essential to set the turbine’s data within the context of its environment — inclusive of geospatial factors, vessel logistics and the proximity of neighboring turbines or farms — in order to gain an accurate picture of why a turbine might be underperforming. This contextual data is crucial to understanding whether underperformance is due to times of low resource or a sign of a technical problem that can be addressed.
However, traditional methods of data analysis cannot handle the sheer volume of data needed to analyze an offshore wind turbine in the context of its environment. Often, a team can take weeks or even months to sift through turbine data and get to the root cause of the problem. To add to the complexity, systems are typically not holistic, whereby different platforms are utilized depending on the particular area of analysis.
It doesn’t need to take this long. Artificial intelligence applied to the appropriate data model can identify underperformance from the data as it arises, rather than diagnose it months down the line. By building an intelligent model of turbine performance inclusive of all atmospheric, geospatial and meteorological data, it is possible to cut the time necessary to identify a potential failure down to hours.
As the sheer scale of offshore turbines and offshore wind farms increases, we are entering unknown territory around how these new designs interact with and are impacted by the harsh offshore environment.
While these turbines have been designed to produce vast amounts of energy in harsher conditions, there is still a lack of long-term data to indicate exactly how higher hub heights and harsher environments will affect turbine components — preventing owners from having a clear indication of asset condition and adjusting their maintenance strategy.
For example, the latest designed turbines being deployed with blades over 100 m in length. However, longer blades require lighter materials and innovative internal structures, which further increases the tip speed. For onshore turbines, the industry has had time to gather data to suggest that these factors increase the risk of a blade developing leading edge erosion. For offshore turbines, there is a lack of available data to confirm whether these blades will stand up to the constant interaction of microdroplets at high wind speeds typically seen offshore.
With in-depth, AI-driven analysis of contextualized turbine performance on site from day one, owners will be able to monitor for dips in performance or signs of damage, learn how their technology is working in that environment and build a long-term O&M strategy based on real performance.
With this information, the O&M team can then watch for the first signs of erosion when on site, ultimately reducing the chances of blade erosion progressing to a crack or total blade failure.
Collecting and analyzing all available offshore wind farm data — quickly, and in context — is crucial to fully understanding turbine performance and targeting O&M to make the best use of available weather windows.
Rapid and in-depth AI-driven analysis of contextualized health and performance data has become best practice in mature markets, and by adopting leading software at the commissioning stage, the U.S. offshore wind industry has the chance to hit the ground running to achieve minimal downtime and maximal returns.
Gareth Brown is CEO of Clir Renewables, a renewable energy AI software company. He is an entrepreneur and a chartered engineer with the IMechE. Gareth has over a decade of experience in the industry which spans the life cycle of renewable energy projects from identification, development, construction, through to financing and operation. Gareth set up Clir Renewables in 2017 alongside Jake Gray, offering an innovative, AI-driven software solution to help owners and operators better understand how their assets are performing and provide actionable insights on how to optimize their output. Headquartered in Vancouver, Canada, the company opened its European office in Glasgow, Scotland, in 2018, and now supports 6 GW of assets globally.