Deploying structural control devices based on machine learning is a significant advancement for infrastructure management. Such systems, capable of combining IoT and artificial intelligence, make it possible to detect and deal with various structural incidents on time. Machine learning plays an important role, as it analyzes data from sensors placed on the structure, such as vibration, acceleration, and temperature. Its ability to continuously learn over time from the data allows the device to identify faults increasingly effectively. This enables effective predictive maintenance, which reduces costs and increases safety. Several data analysis tools exploit various machine learning techniques. Starting with data from accelerometers, the study compares different unsupervised methods and their effectiveness.
Machine Learning Techniques for Structural Health Monitoring
Casillo, Mario;Cecere, Liliana;Colace, Francesco;Lorusso, Angelo;Santaniello, Domenico;
2025
Abstract
Deploying structural control devices based on machine learning is a significant advancement for infrastructure management. Such systems, capable of combining IoT and artificial intelligence, make it possible to detect and deal with various structural incidents on time. Machine learning plays an important role, as it analyzes data from sensors placed on the structure, such as vibration, acceleration, and temperature. Its ability to continuously learn over time from the data allows the device to identify faults increasingly effectively. This enables effective predictive maintenance, which reduces costs and increases safety. Several data analysis tools exploit various machine learning techniques. Starting with data from accelerometers, the study compares different unsupervised methods and their effectiveness.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.