No abstract

No abstract

Fax (505)845-9500 ABSTRACTThe cost of repairing or replacing failed components depends on the number and timing of failures.Although the total probability of individual component failure is sometimes interpreted as the percentage of components likely to fail, this perception is often far from correct. Different amounts of common versus independent uncertainty can cause different numbers of components to be at risk of fkilure. The FAROW tool for fatigue and reli,ability analysis of wind turbines makes it possible for the first'time to conduct a detailed economic analysis of the effects of uncertainty on fleet costs. By dividing the uncertainty into common and independent parts, the percentage of components expected to fail in each year of operation is down enables the calculation of costs associated with various probability of occurrence levels, illustrating the variability in projected costs. Estimating how the numbers of components expected to fail evolves over time is also useful in calculating the present value of projected costs and in understanding the nature of the financial risk.

No abstract

Load models for fatigue analysis and testing are tailored to the level of complexity required for the application. Random variable models are developed and applied to analyses in which load sequence effects can be neglected. Conventional narrow-band load peak and range distributions are applied to crack initiation and growth. It is shown that narrow-band load models provide useful, conservative life estimates for general Gaussian loadings. Distributions of significant peaks and ranges for wide-band loadings are developed empirically through simulations with racetrack filtering. An efficient “sequential” simulation technique is introduced for continuous generation of both narrow- and wide-band random loads. Based on a simplified crack closure model, simulations of crack growth suggest that sequence effects are most influential when any or a combination of the following are present: larger ratios of crack opening stress to maximum applied stress, lower values of applied tensile mean stress, smaller values of yield stress and crack growth coefficient. When sequence effects are present, the regularity of the spacing between tensile overloads can be important. In particular, assumption of regularly spaced overloads can be nonconservative.