Demand for wind energy is soaring. The European Union (EU) has set a target to increase the share of energy consumed from renewable sources to 20 percent by 2020. Wind energy is now the second leading form of power generation capacity in the region.

To capitalize on this opportunity, wind turbine operators must maintain a complex network of assets to keep production levels high. Waiting for components to fail before tending to them can reduce output significantly.

Spotting an opportunity, Performance for Assets (P4A) teamed up with IBM Garage to create an advanced asset management monitoring system for wind turbines that enables predictive maintenance and boosts asset performance.

Building a strong idea

Lots of energy companies have asset management monitoring systems that produce vast amounts of data, but teams often don’t know how to use it effectively. Without clear alerts about when an asset is suddenly underperforming, and information about what actions to take and what to prioritize, it is difficult for technicians to detect problems before they become serious.

To help wind turbine owners extend the lives and value of their assets, P4A set out to create a new monitoring solution. We aimed to unite sophisticated analytics with field knowledge, providing action alerts for technicians before asset failure and performance drift.

We knew we had a strong concept. The next challenge was how to bring it to market successfully.

Making development a breeze

We heard about IBM Garage, a center of innovation where we could try out the latest technology. Alongside our partner I-Pulses, we set up a session with the IBM Garage in Nice, France. The IBM team led us through a three-day Design Thinking workshop to evaluate our idea from every angle, with a strong focus on the customer.

We took what we learned from the experience and worked with technical experts from IBM and I-Pulses to create a minimum viable product (MVP) for our new solution in an IBM Cloud Foundry environment.

Using IBM Watson Internet of Things (IoT) Platform hosted in the IBM Cloud, our app gathers and processes data from wind turbine sensors and merges it with weather forecast information. We rely on IBM Informix on Cloud to provide a high-performance, scalable database.

Our data scientists built hybrid machine learning models that can be put into production at scale with the IBM Watson Machine Learning service. The resulting algorithms detect problems with wind turbine components and provide diagnosis and prognosis. This provides a full picture of asset performance and health.

With the app’s user interface, field technicians and managers can explore and visualize asset performance data. They can also interact with an integrated virtual assistant based on IBM watsonx Assistant to optimize maintenance workload and machine output.

Powering greater output at lower cost

Thanks to the support from IBM and I-Pulses, the MVP for our app was ready in eight weeks. We presented it to leading asset management owners at the Euromaintenance 4.0 event. It was received very well, and we’re seeing real interest in the solution.

Our app gives asset management owners the visibility they need to boost output and efficiency. Users can tackle underperformance of assets proactively, before they start to drag down revenues. We also help asset management owners to deploy their maintenance teams more effectively, making the best use of limited resources.

P4A plans to make the app available for commercial use within three months. When we do, we’ll be able to scale up quickly and easily using IBM Cloud solutions. With help from IBM Garage, we’ve created an application that’s going to help ignite the energy sector.

Read the case study for more details.