Offshore wind turbine maintenance planning application integrates AI model and various data sources into a useful tool for the planner.

Decision tool for maintenance offshore wind turbines

Automation of the planning process

Recently, Otary developed an AI model that predicts the expected energy production per wind turbine at a particular time based on the weather forecast. Production actually varies for each wind turbine depending on its positioning and wind direction. The AI model generates this data 4 times a day based on the latest weather data for the next few days. Depending on the type of maintenance, the wind turbine has to be shut down for a short or longer period of time. By tailoring a planning tool, this parameter can be taken into account to integrate the loss of electricity production in finding the balance between the various factors.

Since weather forecasts are constantly changing, an optimized plan is not necessarily the most optimal at the next forecast update. Automation allows more parameters to be taken into account and still have ‘up to date’ plans at all times. Such optimization work cannot be sustained manually.

The aim of this project is to develop a smoothly working dynamic tool for the planner that combines all data, automates calculations and updates as soon as new data are available.

Digital twin

Initially, a proof of concept will be set up in a relatively short period of time (3 to 4 months). That way, the user experience can be evaluated in order to work out the next steps.

We set up a web application that collects data from different sources and presents them in a user-friendly interface that allows the planner to simulate different scenarios or start an optimization process. The heart of the application is a digital twin of the wind farm that simulates the output of the farm per wind turbine based on all the parameters set by the planner.

Based on the meteorological forecasts and the predicted production, different parameters are used to determine which wind turbines can undergo the specific maintenance task. For this purpose, the three offshore wind farms Mermaid, Seastar and Rentel are visualized with their respective wind turbines. The tool takes into account factors such as the type of maintenance work (e.g. painting or working on the turbine), the time of day and the production capacity at that time. On this basis, the tool visually shows the wind turbines that can be worked on and those that cannot. The reasons for rejection can be consulted. The tool is fed with accurate info 4 times a day (every 6 hours).

Smart tool for planner

The tool has been enthusiastically received by the planners. The combination of several factors in the same tool provides great user convenience. At a glance, he can consult clear and accurate information to make decisions. As a result, he saves significant time and human error can be reduced. The measurable results of the tool will be announced in about six months' time.

‘This tool, developed by Coteng, has the potential to become the ideal co-pilot in offshore planning’ - Katja D'haeyer, data scientist at Otary

Optimisation tool for maintenance offshore wind turbines

Priority versus regular maintenance work

Otary wants to maintain its offshore wind turbines according to an optimally well-organized system. There are, in fact, many different maintenance activities required on a wind turbine. What is possible on a given day also depends on the skills available on a ship. Some work is done only yearly or occasionally, while other tasks are high priorities. A working day is composed of priority tasks and completed by other tasks. Otary requires a planning tool that allocates these two types of work to a given team according to a well-defined logic and optimized to the requested balance between the various parameters.Optimizing maintenance tasks with digital toolCoteng is building a tool to plan maintenance tasks. The tool defines the available team with the right profile on a given ship. They are linked to a certain urgent task (= MUST task). The working day is completed with less urgent tasks (= MAY task). This optimization tool uses the data generated by the decision tool (see above). This means that the maintenance works are carried out only on those wind turbines that fit the balance between the different parameters requested by the planner.

Furthermore, the planning takes into account, among other things, the travel time to the wind farm, the expected duration of the maintenance works, the time needed to load equipment, and so on.

Increasing efficiency

The combination of several factors in the same tool ensures great convenience. The planner can consult clear and accurate information from a quick scan to make decisions. When new data become available, plans and calculations are automatically updated. The results of the tool are measured and evaluated after a 6-month implementation period.

These tools were fully developed in 6 months, from draft to running model in production. After evaluating the user experience over a longer period, the tools can be further expanded with additional functionalities.

This project is part of the ELBE EUROCLUSTER project (GA- 101074230), which received funding from EISMEA's COSME program (call SMP-COSME-2021-CLUSTER).Funded by the European Union. However, views and opinions are solely those of the author(s) and do not necessarily reflect those of the European Union or EISMEA. Neither the European Union nor the funding body can be held responsible for them.

Get in contact

Do you want to make your company futureproof, speed up processes and gain insights by taking the correct steps? Talk to our team of experts. They are ready to find the best solution that works and pays off. No fuss, just results.

I want to reach out