Model-Based Evaluation of Methods for the Determination of the Onset of Yielding by Temperature Measurement

The thermo-elastic effect describes the cooling of a classical material under tensile load. In the tensile test, this cooling is superimposed with a temperature increase due to dissipation when plasticity sets in. This can be exploited from a measurement point of view and can be used for a physically motivated identification of the onset of yielding. In this study, different evaluation routines for the identification of the onset of yielding from the temperature signal are analyzed systematically. For this purpose, an analytical model is first used to generate evaluation data that provide a unique evaluation reference. To mimic sensor signals from real experiments, the artificially generated data is manipulated applying typical measurement noise, artifacts and other disturbance variables. Based on this data set, five approaches for identifying the onset of yielding are contrasted. The methods investigated are one based on the temperature minimum itself, a line-fit-method applied to the first derivative of the temperature signal, the second derivative of the temperature signal by means of finite differences, the second derivative of the temperature signal by means of regularization, and the solution of an inverse problem.