The main purpose of fatigue engineering is to design components such that the relevant operational loads are born safely. At the same time, further requirements such as cost efficiency, weight reduction, ease of manufacturing and assembly, variant reduction etc. must be fulfilled in an optimal way. Component dimensioning depends primarily on parameters defining the external loads like driver behaviour, operation purpose of the vehicle and environmental factors as well as on parameters regarding its load bearing ability such as material properties and production processes. All these parameters are subjected to scatter, which thus has to be taken into account in component design as well as in a fatigue/reliability validation procedure, explaining the need for statistical methods. In the automotive industry, fatigue and reliability validation needs to be based on statistical methods due to scatter in customer loads and component properties. The common procedures lead to unnecessary testing expense and potentially to unnecessary design changes. In this paper, current developments towards more efficient statistical methods are presented. Nowadays, fatigue and reliability demonstration and estimation is often based on the success run method. However, this procedure is often inefficient, for instance when specimens have to be suspended prematurely due to failure of third components in system tests. In this paper, state of the art statistical methods are devised that allow to, correct test designs if failures of third components have changed the test durations, determine supplementary test designs if outliers inhibited homologation in the initial test utilize all information from censored items calculate confidence bounds for quantiles, with respect to small sample sizes take into account the exact lifetime of failures (partially passed component counting thus making fatigue and reliability testing more efficient.


    Zugriff

    Zugriff über TIB

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    New approaches for efficient statistical fatigue validation


    Beteiligte:
    Weigel, Nicolas (Autor:in) / Weihe, Stefan (Autor:in) / Speckert, Michael (Autor:in) / Feth, Sascha (Autor:in)


    Erscheinungsdatum :

    2010


    Format / Umfang :

    9 Seiten, 4 Bilder, 6 Quellen



    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Print


    Sprache :

    Englisch




    Statistical Approaches Applied to Fatigue Test Data Analysis

    Engler-Pinto, Carlos C. / Allison, John E. / Lasecki, John V. et al. | SAE Technical Papers | 2005


    Statistical approaches applied to fatigue test data analysis

    Engler-Pinto,C.C. / Lasecki,J.V. / Frisch,R.J. et al. | Kraftfahrwesen | 2005


    A verified and efficient approach towards fatigue validation of safety parts

    Weihe,S. / Weigel,N. / Dressler,K. et al. | Kraftfahrwesen | 2011


    A verified and efficient approach towards fatigue validation of safety parts

    Weihe, Stefan / Weigel, Nicolas / Dreßler, Klaus et al. | Tema Archiv | 2011


    Efficient statistical validation with edge cases to evaluate Highly Automated Vehicles

    Karunakaran, Dhanoop / Worrall, Stewart / Nebot, Eduardo | IEEE | 2020