Publications Details

The development of confidence limits for fatigue strength data

Sutherland, Herbert J.; Veers, Paul S.

Over the past several years, extensive databases have been developed for the S-N behavior of various mate&& used in wind turbine blades, primarily fiberglass composites. These data are typically presented both in their "raw" form and curve fit to define their average properties. For design, confidence limits must be placed on these descriptions. In particular, most designs call for the "95195" design values; namely, with a 95 percent level of confidence, thedesiguerisassuredthat95percentofthematerial will 'meet or exceed the design value. For such material properties as the ultimate streng& the procedures for estimating its value at a particular confidence level is wellffiedifthemeasured values follow a normal or a log-normal distribution. Namely, based upon the number of sample points and their standard deviation, a commonly-found table may be used to determine the survival percentagea t a particular confidencel evel with respect to its mean value. The same is true for fatigue data at a constaut stress level (the number of cycles to failure N at stress level SI). However, when the stress level is allowed to vary, as with a typical S-N fatigue curve, the proceduresf or determmingc onfidencel imits are not as well delked. This paper outlines techn.iques for determimng confklence limits of fatigue data Different approachesto estimating the 95195l evel are compared. Data from the MSUIDOE and the FACT fatigue databam are used to illustrate typical results.