The structural health monitoring (SHM) of structures is acquiring a key role in the present time. An in situ system able to assess the health state of bolted joints would save money and maintenance time, by allowing quick assessment of residual life and degradation state of structures. In the last decades, SHM systems based on linear acoustic/ultrasound methods have been investigated extensively. The scope of this study was to develop a reliable index able to assess the loosening/tightening health state of a bolted structure based on linear and nonlinear acoustic/ultrasound parameters. In particular, for the linear acoustic/ultrasound method, a tightening/loosening state index based on the first-order acoustic moment was developed. This method is based on the assumption that a change of signal energy would be recorded for different loosening/tightening states. As for the nonlinear methods, under single- and multi-frequency excitation, high-harmonics generation and sidebands modulation indices were developed. The developed tightening/loosening state index trend was very well reproduced by an analytical expression where they are expressed as function of the torque applied. In particular, the fully loosened (kissing bond) and tight state of bolted structures can be clearly identified by the measured plateau region. The proposed analytical trend approximates the experimental results with excellent correlation. By knowing this trend and measuring the proposed indices, it would be possible to know the torque applied on the bolt and therefore assess the health state of a bolted structure.

Ultrasonic waves are useful tools to characterize the contact forces between components in non-destructive and non-invasive manners. It has been shown that the transmission and reflection coefficients of the ultrasonic wave are sensitive to the contact pressure or other contact parameters. Theoretically, the normal and tangential stiffnesses of the contact interface govern the transmission/reflection coefficients and can be used as parameters to characterize the contact condition. However, weak and incomplete interfaces, formed by rough surfaces in partial contact, show a highly nonlinear behaviour also when they are excited under free vibrations. In particular, the amplitude of the second harmonic is a relevant index of the contact stiffness, and the nonlinear response is strongly influenced by the nominal contact pressure applied to the boundaries. In this study a new theoretical model of the nonlinear interface stiffness was developed where the stiffness of the contact interface was described as a function of the nominal contact pressure. The developed theoretical contact pressure function of the second harmonic generation at the contact interface was found to agree with good accuracy with the experimental data. Moreover, this paper presents also a theoretical and experimental study aimed at developing an integrity index capable of assessing the stiffness of the contact interface between structures when excited by free vibration or under controlled vibration excitation.

Prediction of crack growth and fatigue life estimation of metals using linear/nonlinear acousto-ultrasound methods is an ongoing issue. It is known that by measuring nonlinear parameters, the relative accumulated fatigue damage can be evaluated. However, there is still a need to measure two crack propagation states to assess the absolute residual fatigue life. A procedure based on the measurement of a third-order acoustic nonlinear parameter is presented to assess the residual fatigue life of a metallic component without the need of a baseline. The analytical evaluation of how the cubic nonlinear-parameter evolves during crack propagation is presented by combining the Paris law to the Nazarov-Sutin crack equation. Unlike other developed models, the proposed model assumes a crack surface topology with variable geometrical parameters. Measurements of the cubic nonlinearity parameter on AA2024-T351 specimens demonstrated high sensitivity to crack propagation and excellent agreement with the predicted theoretical behavior. The advantages of using the cubic nonlinearity parameter for fatigue cracks on metals are discussed by comparing the relevant results of a quadratic nonlinear parameter. Then the methodology to estimate crack size and residual fatigue life without the need of a baseline is presented, and advantages and limitations are discussed.