Machine Learning Models in Structural Engineering Research and a Secured Framework for Structural Health Monitoring

Health inspection of public structures is intended to detect incipient damage at an initial stage in order to improve maintenance. Artificial intelligence alludes to the part of computer science that comprises various techniques for fulfilling the requirements of Structural Health Monitoring (SHM). Deep Learning (DL), and Machine Learning (ML) are often utilized. Deep Learning is an instance of Machine Learning built on deep neural networks that have demonstrated remarkable achievement in numerous applications over the years. This article deals with recent literature reviews on the advent of machine learning models in the performance monitoring of civil structures. Recently, machine learning has gained considerable attention and is being built up as another class of astute techniques for the health inspection of civil structures. The main concern of this examination is to epitomize the strategies built over the last decade for the practice of ML techniques in civil engineering. In addition, types of sensors, number of sensors, sampling frequency, types of structure, structure material, data collection time, and types of excitation in the domain are also explored. Initially, a brief summary of the ML is given, and the implications of the ML in structural/civil engineering are depicted. Afterward, applications of ML methods in the domain are presented and the potential of these approaches to overcome the deficiencies of conventional methods is addressed. The observations after researching the literature, along with research opportunities and future directions in the use of ML, are then discussed. Eventually, a novel, secured framework for Structural Health Monitoring (SHM) using the Ethereum Blockchain is proposed established on the studies.