Machine learning based structural health monitoring of bridge using k-means clustering algorithm in python Available to Purchase

This paper illustrates the modeling of concrete micro-cracking (i.e. the first mode of damage)in Reinforced Cement Concrete (RCC) bridge using K-means clustering algorithm in Python. A three-span continuous bridge is considered and performed vibration analysis under moving vehicles for the Structural Health Monitoring (SHM). Five accelerometers were deployed at various critical locations of the bridge and the collected data was stored in a file (Accelerometer.txt). All together 2352 data points are delineated into 5 clusters using K-means. Elbow method was implemented to determine the number of optimum clusters to be provided for these data points. Five new centroids were determined for 5 clusters and also assumed that the cracking of concrete takes place through these centroids. The authors have determined: (i) Crack initiation and crack propagation in concrete, (ii) Corrosion induced concrete delamination, and (iii) Failure in reinforcement. Finally, life span of the bridge can be estimated using the present Python Framework.