Predictive maintenance and Structural Health Monitoring via IoT system

Abstract

Reinforced concrete buildings have proven the need to monitor the concrete and steel parts over time. The topic of structural monitoring of a building is becoming more topical with time, and many buildings from the 1960s and 1970s are under observation. The current challenge is to monitor structures effectively and continuously, applying the meaning of preventive maintenance, a concept well developed in engineering disciplines. New technologies allow us to assess the impact of time, wear, and tear, which in the long term can challenge the safety of buildings by monitoring the natural vibrations of a building. However, traditional monitoring systems in the civil infrastructure sector have always been expensive and undervalued. Therefore, borrowing innovations from computer science, a sensor system based on the new paradigms of the Internet of Things (IoT) was developed to provide a valuable alternative to proven vibration monitoring systems. The proposed system consists of a microprocessor (Raspberry Pi) and a low-cost accelerometer for microelectromechanical systems (MEMS), this type of lower costs sensor allows for investment in the safety of structures. The architecture of the monitoring system and the visualization of the vibrational model and its operation mechanism are presented. The performance of the monitoring system and the data collected are then integrated with Deep Learning techniques to obtain possible future scenarios and forecasts practically to perform tests that are as close as possible to reality and thus be able to act with the necessary maintenance to prevent undesired structural effects.