Development of piezoelectric impedance-based technique for rapid compressive strength measurement of mortar specimen using artificial neural network

Abstract

Measurement of concrete strength is crucial for both the construction and maintenance stages of infrastructures. This importance is emphasized every year in Korea, where aging facilities require reliable and accurate assessment to prevent accidents. In general, three types of non-destructive testing methods are used for concrete maintenance, including the ultrasonic and rebound hardness methods, with a combination method also used. However, despite the popularity of these methods, there can be reliability issues as these methods estimate the strength of concrete through equations, rather than direct measurement. In this study, a non-destructive testing method known as the electromechanical impedance technique is used to predict the compressive strength of two different mortar specimens of 10 MPa and 30 MPa with the use of artificial neural network. In addition, the conventional method of permanently attaching the piezoelectric transducer was altered in a way to achieve temporary attachment to the host structure to make the proposed idea more useful for real field. The result from the experiments proves the possibility of predicting the difference in the compressive strength of the two mortar specimens.