Predict, Prevent & Plan with machine learning

Use machine learning to optimize your maintenance and planning

What if you could foresee when turbine components malfunction and need maintenance – before they actually do? How much time and money would that save you in your maintenance and planning?

On any given day, 24/7, Gram & Juhl are monitoring wind turbines globally through its state-of-the-art TCM® system, diagnosing failures, providing prognostic information and recommended actions to wind farm owners and operators. Making TCM® monitoring the weapon of choice against downtime.

It is estimated that up to 30% of the total life-cycle cost of a wind farm is due to failure and maintenance activities (N. Flaherty, 2018). According to WINDWERK, unscheduled maintenance represents about 60% of the total O&M cost (WINWERK, 2017). O&M is therefore still the major challenge facing owners and operators.

But what if you could avoid unscheduled maintenance? What if you could foresee when turbine components are malfunctioning and need maintenance – before it actually happens? How much time and money would that save you in maintenance and planning?

Think of it like a car. A car typically has icons on the dashboard that will light up if something is wrong, allowing you to take action and go to the workshop. But what if the car was so intelligent that it could also predict the failures before they occurred, what if the dashboard also showed you WHEN you should go to the workshop. What if the icon showed you how long you could drive before actual malfunction, and what if you could avoid the failures completely. How much money would you save on workshop bills?

Gain actionable insight

Gram & Juhl is watching your wind turbines so you can protect your assets and reduce your operation cost.

Gram & Juhl is predicting the future by analysing the past. Detailed knowledge of a turbine’s historical data, patterns and trends attribute to the TCM® diagnostics and prognostics analysis. Based on historical data, key figures on performance, reliability, and efficiency it is possible for Gram & Juhl data analysts to investigate and interpret the condition of the wind turbine components. This type of data gathering can be used for many turbine parameters to predict impending failures and optimize operation and maintenance.

Based on more than 20 years of experience within vibration monitoring, machine learning algorithms are developed and applied, which support the analysis and make it even more reliable.