Random Fatigue: From Data to Theory

Integrated production scheduling and preventive maintenance planning for a single machine under a cumulative damage failure process

“…The damages are cumulative and when the accumulated damage on the machine exceeds a random threshold value, the machine fails. This kind of failure process is commonly used in modeling fatigue failures of mechanical part [11,14,16,17]. The damage increments are assumed to be independently and identically distributed (iid) random variables and the stream of shocks is assumed to be a stationary point process.…”

Flow-induced vibration of elastically supported rectangular cylinders

“…Fatigue manifests itself in the deterioration of the material's ability to carry load (Sobczyk and Spencer, 1992). When subjected to cyclic stress, the structure will fail at a stress much lower than that required to cause failure during a single, near-static application of load.…”

“…Accordingly, crack growth equations for deterministic stress conditions include the stress range and the maximum stress values. For random loading, however, it is important to characterize the extreme values of stress (Sobczyk and Spencer, 1992).…”

“…Under this condition, the direct use of a standard S-N curve is not recommended as the curve is based on constant amplitude stress. To estimate fatigue life for variable amplitude loading, Palmgren and Miner theorized that fatigue fracture resulted from a linear accumulation of partial fatigue damage (Sobczyk and Spencer, 1992). The Palmgren and Miner rule states that the damage fraction, A,-, at any stress level, S t , is linearly proportional to the ratio of the number of cycles at this stress amplitude, to the total number of cycles that would cause failure at that stress level, Nj, For each stress level, the appropriate TV,-is found from the S-N curve and a new partial damage is calculated for the number of cycles at that particular stress level.…”

“…Accordingly, it is assumed that failure occurs when the total accumulated damage reaches 100%, that is when D > 1.0. However, due to the variability inherent in fatigue testing, typical experimental values of D at failure range from 0.5 to 2.0 (Sobczyk and Spencer, 1992).…”