Signal-based nonlinear modelling for damage assessment under variable temperature conditions by means of acousto-ultrasonics

Damage assessment can be considered as the main task within the context of structural health monitoring (SHM) systems. This task is not only confined to the detection of damages in its basic algorithms but also in the generation of early warnings to prevent possible catastrophes in the daily use of the structures ensuring their proper functioning. Changes in environmental and operational conditions (EOC), in particularly temperature, affect the performance of SHM systems that constitutes a great limitation for their implementation in real world applications. This paper describes a health monitoring methodology combining the advantages of guided ultrasonic waves together with the compensation for temperature effects and the extraction of defect-sensitive features for the purpose of carrying out a nonlinear multivariate diagnosis of damage. Two well-known methods to compensate the temperature effects, namely optimal baseline selection and optimal signal stretch, are investigated within the proposed methodology where the performance is assessed using receiver operating characteristic curves. The methodology is experimentally tested in a pipeline. Results show that the methodology is a robust practical solution to compensate the temperature effects for the damage detection task.