Damage identification using piezoelectric electromechanical Impedance: A brief review from a numerical framework perspective

High-frequency electromechanical impedance measured from the piezoelectric transducer has been recognized as an effective indicator to infer minor damage occurrence. Over the past decades, much research has focused on developing tailored numerical frameworks to fully utilize the electromechanical impedance for damage identification of various engineering structures. In terms of the implementation architecture, the numerical frameworks generally can be classified into two categories, i.e., inverse model updating and forward damage prediction. The former is conducted through formulating an inverse problem based upon the response difference between the model prediction and corresponding impedance measurement under the same operating condition. Such inverse problem can be solved by means of minimizing the above difference in an iterative manner, which can be facilitated by incorporating the optimization method. The latter, on the other hand, can directly predict the damage using the machine learning model established by the known input-output relationships. As its architecture appears to be opposed to that of inverse model updating, it can be tentatively referred to as the forward framework. This article intends to provide a brief review of the state-of-the-art studies in terms of these two numerical frameworks. Different variants of methods developed and integrated into the frameworks for performance improvement and their limitations are discussed. The remaining challenges and future direction for electromechanical impedance-based damage identification are also pointed out.