TRISTAN
March 24th, 2005, 09:47 AM
Hi,
(sorry for my english & question if not sufficient clear):
how would you define a simple Weibull model in order to take into account, in the case of Low Cycle Fatigue phenomenum, the fact that the "system" (f.ex. a turbine disk) has many identical serie-component(f.e. many blades)submitted to a common variable (the load) and a material resistance which has: an intra-scattering per sample (n items issued from one sample) and an inter-scattering between samples (it's to say beetween disk). My simple Weibull law for the system is based indeed on the law of the minimum of n components: so, eta(syst)= eta(component)* n^(-1/beta), but this one doesn't consider the scatter beetween the samples.
Nota: i only have some observed results of failures at system level (dammage calculation to the first failure), good hypothesis on scatter contributions (intra/inter), and dammage calculations for a blade based on the median (50%) curve of LCF resistance.
Thanks for give me some idea or indicate me some reference where this kind of problem is treated
(sorry for my english & question if not sufficient clear):
how would you define a simple Weibull model in order to take into account, in the case of Low Cycle Fatigue phenomenum, the fact that the "system" (f.ex. a turbine disk) has many identical serie-component(f.e. many blades)submitted to a common variable (the load) and a material resistance which has: an intra-scattering per sample (n items issued from one sample) and an inter-scattering between samples (it's to say beetween disk). My simple Weibull law for the system is based indeed on the law of the minimum of n components: so, eta(syst)= eta(component)* n^(-1/beta), but this one doesn't consider the scatter beetween the samples.
Nota: i only have some observed results of failures at system level (dammage calculation to the first failure), good hypothesis on scatter contributions (intra/inter), and dammage calculations for a blade based on the median (50%) curve of LCF resistance.
Thanks for give me some idea or indicate me some reference where this kind of problem is treated