Situation
- Remote, globally installed fleet
- Turbines degrade in between services and cleaning
Task
- Monitoring and diagnosis of turbine degradation
Our Contribution
- Detection of operation patterns in hundreds of years of turbine operation data
- Development of equivalent operation hours (EOH) algorithms based on real start stop, part load and critical operation
- Investigation of various key performance indicators for efficient degradation detection, modelling and trending
- Automation of operations and degradation modelling with machine learning based on historic fleet data