Structural health monitoring of structures and infrastructures through acoustic signature analysis

Vehicular traffic load is one of the main causes that lead to the road pavements failure, i.e. the worsening of the road pavement’s performances that are mainly caused by internal (concealed) cracks. These cracks are generated in the bottom layers of the road pavements and propagate toward the upper layer leading to the failure of road pavements. Because of their location, this type of damage is difficult to be detected. Future smart cities need smart infrastructures and Intelligent Transportation Systems (ITS) able to provide high levels of performance and comfort, and to share information with their users in a sustainable and an efficient way. Unfortunately, despite the current high number of solutions designed for the structural health monitoring (SHM), there is a lack in implementation of these solutions in real contexts or in installation on road infrastructures. Furthermore, the authorities that are responsible for the management of road pavements need more sustainable, efficient, and reliable methods and technologies to improve the current management process. In particular, Pavement Management Systems (PMSs) are currently used to carry out failure-based maintenance, by mean of destructive tests and measurements that not represent the real structural conditions of the road infrastructures. Such approach has by now become unsustainable and unreliable and an improvement is needed. For these reasons, the objectives of this thesis are: i) to design and set up a new non-destructive test (NDT)-based method for the SHM of road pavement based on the concept of vibro-acoustic signature; ii) to design an electronic system that is able to apply the new NDT-SHM method, based on the principles of simplicity, efficiency, and sustainability; iii) to estimate the effect of the proposed solutions on the current road infrastructure management process. A thorough literature review, simulations, and several in-lab and on-site experimental investigations were used to design and set up the method and the system proposed in this thesis. The proposed method is innovative because of the fact it considers the road pavement as a filter for the vibration and the sound that propagate from a mechanical source to a specially designed receiver. The microphone-based receiver is able to detect the acoustic response of a road pavement to a mechanical load (e.g., a vehicle pass-by), which is therefore considered the vibro-acoustic signature of the monitored pavement. It is expected that any variation of the structural conditions of the filter will lead to a detectable variation of the road vibro-acoustic signature. Specific algorithms were designed to detect and monitor road signature variations using Fourier Transform, Wavelet Transform, and Hierarchical Clustering. Results show that the proposed NDT-SHM method allows detecting and monitoring the presence of cracks (internal and external) into the road pavements. Meaningful features can be extracted from road pavement’s acoustic signatures by using Fourier and Wavelet Transforms. Algorithms based on hierarchical clustering recognize, from the extracted features specific variations of the signals transmitted (e.g., by using the wavelet coefficient’s energy as feature), or the worsening of the transmission capability of the transmission medium (e.g., by using the Shannon’s entropy of the wavelet coefficients as feature). The proposed method has been implemented in a prototype of Wireless Sensor Network (WSN) system. It consists of several renewable energy-powered, wireless, low-consuming sensor units (modular system), which are able to detect in real-time both the vibro-acoustic signatures of the monitored structures and the significant environmental conditions (e.g., temperature). The proposed method and system can improve the current road infrastructure management process (PMSs), because of the fact they switch from the failure-based to the predictive-based maintenance paradigm using innovative smart, sustainable, efficient, reliable, and properly designed technologies.