The core objective of the project Hybrid20 is to establish at the center viable methods for construction of probabilistic hybrid semiparametric models of physical systems (including materials, production processes or products in operation) and to document their applicability in four Use-Cases. Hybrid semiparametric models combine knowledge (parametric sub-models) and observation (nonparametric sub-models). Probabilistic hybrid models treat some of their variables as random variables. We require their outputs to provide distributions rather than point estimates (output variables as random variables). Distributions allow for derivation of output confidence metrics, which is the key requirement to establish trust and further use of complex models in many real-world applications of condition monitoring and material design.
|Effective start/end date
|1/01/22 → 31/12/26
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.