Hybrid Models Combining Machine Learning and Physical Knowledge

1. Hybrid Modeling and Uncertainty Quantification in Transfer Learning

In the energy sector, wind turbines, compressors or other production facilities are typically numerically simulated. While those simulations are precise, they are too computationally expensive to use for inference about unknown parameters. The idea is to use machine learning as a surrogate model and explore the uncertainty of the trained model, both on simulated and real data.

The first task will be to systematically investigate the model performance of a model before and after transfer between data as a function of difference between the data sets (K-S test), sample sizes and noise. Next step is to explore how the model transfers when including domain knowledge (hybrid AI). The end goal of the project is to use an AI model to mimic the simulation and transfer the model to real data (transfer learning). The focus is on uncertainty, both how well the model can learn the simulation as well as how it performs on the real data. The project can be adapted to students from several departments such as Computer Science, Mathematical Sciences or Engineering Cybernetics.

2. Predictive maintenance on windmill data – in collaboration with Trønder Energi

• Handling of erroneous input data in predictive maintenance of wind turbines with explanatory variables, e.g. vibration
• Exploring hybrid AI or transfer learning for predictive maintenance data-driven models (e.g. predicting the expected life time of gearboxes and bushings of wind turbines)
• Making explainability and relationship of the prediction results and causes

The projects can be adapted to students from several departments such as Computer Science, Mathematical Sciences or Engineering Cybernetics.