Structural Health Monitoring (SHM) is becoming a crucial research topic to improve the human safety and to reduce maintenance costs. However, most of the existing SHM systems face challenges performing at real-time due to environmental effects and different operational hazards. Furthermore, the remote and constant monitoring amenities are not established yet, properly. To overcome this, Internet of Things (IoT) can be used, which would provide flexibility to monitor structures (building, bridge) from anywhere. In this paper, a complete IoT SHM platform is proposed. The platform consists of a Raspberry Pi, an analog to digital converter (ADC) MCP3008, and a Wi-Fi module for wireless communication. Piezoelectric (PZT) sensors were used to collect the data from the structure. The MCP3008 is used as an interface between the PZT sensors and the Raspberry Pi. The raspberry pi performs the necessary calculations to determine the SHM status using a proposed mathematical model to determine the damage’s location and size if any. The All the data is pushed to the Internet filter using ThingWorx platform. The proposed platform is evaluated and tested successfully.

INTRODUCTION

SHM is a nondestructive evaluation technique to monitor the integrity of civil structures such as bridges, aircraft, etc. Since the gradual deterioration of structures can happen for different reasons, such as continuous exposure to the inclement weather, overloading, etc., SHM is a vital tool to be implemented in old buildings, bridges, etc., to ensure the safety of human beings. Although researchers from different discipline took different approaches for SHM, most of the works in this field were done using civil and mechanical engineers’ approach. Their works involved mostly to analyze natural frequencies of structures to make decisions. However, in this paper, the chosen approach was to develop a technique to analyze signals (electrical) and implement the proposed technique on an embedded platform. Generally, to perform SHM, firstly, data needs to be collected using sensors. Different types of sensors such as ultrasonic [1], piezoelectric [2], [3], [4], [5], etc. can be used for SHM to generate signals traveling through solid configurations. Later, data collected from the sensors needs to be analyzed by applying different signal processing techniques, because a minor variation within the system triggered by different factors such as noises, temperature changes, environmental effects, might cause significant changes in the response from the sensors, concealing the potential signal changes due to structural defects [6]. Various signal processing techniques have been used to improve the SHM performance such as Wavelet denoising, Fast Fourier Transform (FFT), Wavelet transform, Cross-Correlation (CC), Principal Component Analysis (PCA), etc. Wavelet analysis can be used to remove noise from the signal [7] and detect damage in the structure [6]. Fast Fourier transform [8], [9] and wavelet transform [3] are usually used to get the frequency spectrum of sensors’ output signal, and these spectrums also can help to design appropriate filters to remove noises. On the other hand, CC is the degree of similarity between two signals. For SHM applications, the signals to be compared are the base signal and the real-time signal. Another useful signal processing technique used in SHM is PCA, which uses orthogonal transformation to establish the linear relationship between input and output. The linear input-output relationship developed for a targetted structure can be exploited for an SHM process.