LAUSR

Abstract Contact fatigue failures are becoming increasingly prominent in heavy-duty gears due to the remarkable effects on service lives and reliabilities of associated machines such as wind turbines. The contact fatigue estimation of such gears is complicated due to the coupling effect of mechanical properties, lubrication and stochastic wind loadings. A numerical model was established to evaluate the contact fatigue damage evolution of a megawatt level wind turbine carburized gear pair based on the measured wind load spectrum. The elastohydrodynamic lubrication (EHL) theory was employed to compute the gear contact pressure and subsequently, the stress responses were obtained efficiently through the discrete convolute, fast Fourier transformation (DC-FFT) algorithm. The hardness and the residual stress gradients were captured via the Vickers hardness test and X-ray diffraction method, respectively. The Dang Van multiaxial fatigue criterion and the Basquin equation were adopted to estimate the fatigue lives under stochastic loadings, based on which the contact pressure-life curve was further profiled. Finally, the gear damage accumulation under this load spectrum was evaluated via the linear Palmgren-Miner rule. Results indicate that both the subsurface and transition area failures should be considered in gear design. The fatigue life prediction should consider the load spectrum rather than solely rely on a constant amplitude loading condition.