Visual Blade Asset Integrity | The future of asset health management
“Cost-efficient and sustainable energy is of utmost importance to us, so we are keen to partner with Siemens Gamesa in order to accelerate digital transformation even further.
Leveraging AI analysis with automated stitching and failure detection of the whole turbine is done in less than 15 minutes instead of 3 hours by manual work.”
Industrial Asset Management
.NET Core, React, Angular, AWS, Azure, Python, TensorFlow, Kubernetes, Elasticsearch
Blade maintenance strategies are hard to collect, store and analyze. We provide the tools for optimizing them and secure wind farm performance and asset integrity over a lifetime.
By using artificial intelligence technology, engineering assessment, and expert knowledge to precisely determine the health of the blades and predict actionable advice.
Visual-Based Asset Integrity is a web-based platform that provides a complete overview of blade conditions. It is a blade care solution for making more efficient management decisions.
On the Hermes VBAI project, our partnership goes beyond traditional IT collaboration. We have our software engineers and machine learning experts working together with Siemens Gamesa industry specialists on new algorithms to stitch images to recognize damages and to classify them properly and correctly in order to kick-off and optimize the service process.
VBAI platform has been designed as a microservice-oriented system where microservices communicate either directly (using REST) or using event-sourcing mechanisms. Services are packaged in Docker containers which are scalable and orchestrated using container-orchestration systems such as Kubernetes. VBAI provides APIs for the 3rd party data vendors which are able to supply drone-captured data and integrate with the system.
AI part which has been built by leveraging machine learning methods based on artificial neural networks represents an important part of the system which facilitates data analysis especially in:
- Images stitching
- This way, we can precisely pinpoint where on the blade the fault is. Some of the blades are up to 80 meters long, so it’s not trivial to know that
- Images segmentation
- Separation of the blade versus background. In offshore you often have the blue ocean or the sky and in the image, we need to know what’s blade and what’s ocean
- Fault detection
- Scan the blade surface, to identify and measure faults
- Fault classification
- Identify and classify faults from one to five where five is the most severe and requires urgent attention
Cost-efficient and sustainable energy should be of utmost importance to all of us. Partnering with Siemens Gamesa, Codetiq accelerates digital and industrial transformation. Our AI analysis and automated stitching detect failure and faults in less than 15 minutes, opposed to a 3-hour manual work.
Results and Reporting
The customer portal gives access to insights of inspection data, offering different advanced filter criteria, customizable reports, and dashboards. In this way, customers have easy 24/7 access to the blade conditions at the wind farm.
Future of Hermes VBAI: introduces condition-based maintenance for blades
Over time and with enough data, Hermes will enable predictive maintenance which will significantly reduce maintenance costs.