Design of Durability Sequences Based on Rainflow Matrix Optimization

This paper describes a process of designing a durability sequence by selecting a combination of measured events that best reproduces a predefined loading environment. The selection of the measured events is done by a mathematical optimization technique that compares, for each channel in a multi-channel history, the rainflow matrix of the target with that of the combined events and minimizes their differences. These differences are based on the fatigue related damage content of the entire rainflow matrix or portions of the rainflow matrix for each channel. Other histogram types like rotating moments and PSD's could be used simultaneously for each channel. Since the target is also defined as a Rainflow matrix all methods available in the area of load data editing, like extrapolation and superposition can be used in defining the target. The results define the number of repeats of each of the events that make up the target. In addition to the design of durability sequences, this technique can be used to obtain a reduced number of measured events for analysis or test rig applications, while maintaining the relevant fatigue life characteristics of the service environment. An example, is provided showing how this method can be applied to the design of an automotive test track.