Hello,

I'm currently working on a Structural Health monitoring approach for the foundations of an offshore wind turbine based on its resonance frequencies. On the basis of a large dataset that recovers measurements of several independent variables , I have already established a linear model in order to predict the target (here, the resonance frequency). I performed a Dominance analysis, Regressions, Features selection etc., in order to evaluate which features influences my target the most. However, I would like to improve the accuracy of my linear model by adding more features to my dataset (and then select the best features to build the most suitable model.), i.e. identify underlying mathematical expression (non-linear) between the independent variables and the target. I already performed Genetic Programming (GP) with symbolic regression (SymbolicRegressor) but didn't get consistent results. Is there a method by which I could get these underlying (non-linear) mathematical relationships ?

Thanks a lot,

Lolo_jr

More Loïc Verbeke's questions See All
Similar questions and discussions