Growth through Research, development & demonstration in Offshore Wind

Project:

ReliaBlade2-NL

Improving the reliability of wind turbine blades by understanding failure mechanisms and remaining life to enable better condition-based maintenance decisions.

WHY

A recent study by DNV indicated that the failure rate of wind turbine blades has not decreased in recent years. Despite efforts to improve control over the manufacturing process, defects still occur during certification or operation that may lead to blade failures. This can significantly affect the power output of wind farms. Moreover, as the size of turbines increases rapidly, larger blade models are needed which increases uncertainty about their reliability.

In current practice, a blade is designed based on design loads and material properties derived from many coupon tests. A blade is designed, manufactured and tested for certification, based on key load factors and practical experience of the blade design team. However, recently, it has been shown that developing and implementing condition-based maintenance for wind turbine blades is more cost-effective. By tracking damage throughout a blade's life, from production to end of use, maintenance can be more proactive, reducing the need for complete replacements.

The ReliaBlade2-NL project builds on earlier ReliaBlade 1 project in Germany and Denmark. Besides collaborating with these countries, this project also joins forces with the United Kingdom. Close cooperation will occur among these countries during the project.

WHAT

The main goal of this project is to lower the operational costs of wind turbine blades. To do this, we will focus on making the blades more reliable by monitoring their condition throughout their lifetime. The project will use digital twin technology to predict how long the blades will last and help with operation and maintenance decisions.

The project will focus on three key areas:

1. Blade modelling

We will develop an improved method for modelling wind turbine blades to predict structural changes over time based on its updated condition. This method will focus on operational blades by connecting detailed damage prediction models to a model of the entire blade using a sub-structuring approach. Probabilistic analyses – i.e. a method to predict the behaviour of a system that is inherently uncertain - will be used to characterise the stochastic behaviour of the blade materials and defects. This analysis will make use of test results, and the obtained information will be included in the predictions.

In addition, for accurate blade lifetime predictions, the project will develop a method that improves estimates of material defects and damage over time. This will be done by using periodic monitoring data from the blade's operation. As more data is collected, uncertainties about the condition of the blades will decrease, and the models will become more specific to each blade. This allows for more accurate lifetime predictions, taking into account the blade's past performance. To verify these methods and models, tests will be done at different levels, including experiments on material samples, blade subcomponents, full-scale blades, and operational turbines.

2. Structural health monitoring for blade characterisation and damage diagnosis

Continuous monitoring of the blades is essential to gather data for diagnosing damage in the most stressed areas, known as "hot spots." To achieve this, we will develop algorithms to combine, analyse, and interpret data from various sensors, such as strain, acceleration, and acoustic emission sensors, to identify the presence, location, type, and size of any damage.

In addition, monitoring will also provide data for blade characterisation. This data helps create structural models for blades in operation and detects any unusual changes during operation, even in areas away from the monitored hot spots. The data will be regularly analysed and used to update the models, improving the accuracy of predictions for the blade's remaining lifespan.

3. Decision support for operation and maintenance

We will develop a decision support tool using data from the digital twin. This tool will consider many factors that affect offshore maintenance planning, such as weather conditions (impacting operational limits and power production), resource costs and availability (e.g., vessel hire, technician teams), and specific maintenance needs for each asset.

Since the predictions of a blade's remaining lifespan are probabilistic, the tool will allow operators to assess the risk of delaying maintenance actions (e.g. waiting until the next scheduled inspection). It will help set tolerance limits for these risks. This decision support tool will assist asset owners in choosing the best time for maintenance, balancing condition-based tasks with scheduled maintenance while keeping track of the risks and benefits of different choices.

EXPECTED RESULTS

In short, ReliaBlade2-NL advances current technology by integrating all key aspects into a single framework that supports predictive maintenance decisions for turbine blades. This framework uses a digital twin that combines structural health monitoring, modelling, and probabilistic analysis to predict the remaining lifetime of the blades. By enabling timely inspections and preventive repairs, it reduces the risk of major blade failures.

The main outcome of this project will be a digital twin framework for the lifetime prediction of wind turbine blades. The results will include:

Contact Details

TNO

Michiel Hagenbeek
+31 6 2933 5590

Technology Readiness Level

Maturity level: 4.
        5,
        6,
        7,
        4 5 6 7

Project duration

Theme

O&M Turbine
Figure 1. Flowchart of the research & development concept proposed in ReliaBlade2-NL. magnify Figure 1. Flowchart of the research & development concept proposed in ReliaBlade2-NL.

Figure 1. Flowchart of the research & development concept proposed in ReliaBlade2-NL.

close

Figure 1. Flowchart of the research & development concept proposed in ReliaBlade2-NL.

Figure 2. Major failure modes in a wind turbine rotor blade. In: McGugan et al. (2015), Damage tolerance and structural monitoring for wind turbine blades). magnify Figure 2. Major failure modes in a wind turbine rotor blade. In: McGugan et al. (2015), Damage tolerance and structural monitoring for wind turbine blades).

Figure 2. Major failure modes in a wind turbine rotor blade. In: McGugan et al. (2015), Damage tolerance and structural monitoring for wind turbine blades).

close

Figure 2. Major failure modes in a wind turbine rotor blade. In: McGugan et al. (2015), Damage tolerance and structural monitoring for wind turbine blades).

Figure 3. Envisioned digital twin framework in ReliaBlade2-NL. magnify Figure 3. Envisioned digital twin framework in ReliaBlade2-NL.

Figure 3. Envisioned digital twin framework in ReliaBlade2-NL.

close

Figure 3. Envisioned digital twin framework in ReliaBlade2-NL.

Other information

This project is supported with a subsidy by the Dutch Ministry of Economic Affairs and Green Growth. Find more project information at the TKI Offshore Energy website.

Partners

Associate project partners