Modelling of load spectra containing clusters of less probable load cycles

When the load is stochastic and the service life is a consequence of the fatigue process, the loading histories are transferred into load spectra using rainflow counting. Once the empirical densities of load cycles are known, their probability density is estimated. Here, different probability density mixtures and algorithms for their estimation have been compared according to their ability to model rainflow matrices. The estimated probability density should also consider loading cycles with a small probability. The main scientific contribution of this article is that procedures were identified, which enable more thorough consideration of those clusters with less probable load cycles.

Automated summary

This study compared different probability density mixtures and estimation algorithms for modeling rainflow matrices. Procedures were identified to enable more thorough consideration of clusters with less probable load cycles.