Integrated Ensemble Framework for Real-Time Building Stability Monitoring using Accelerometers and Strain Gauges

Structural health monitoring is critical for ensuring the safety and integrity of buildings, especially in seismic zones or environments subject to high loads. This paper presents the "StrucSense Ensembler," an advanced ensemble learning model designed to monitor building stability by inte...

Full description

Saved in:
Bibliographic Details
Main Authors: Nithila, Ezhil E., Sundararajan, S., Haribabu, K., Krishnan, R. Santhana, Raj, J. Relin Francis, Sankar, G. Ram
Format: Conference Proceeding
Language:English
Subjects:
Accelerometers
Accuracy
Boosting
Buildings
Deformation Detection
Load modeling
Monitoring
Random forests
Real-time systems
Safety
Spatial Data Analysis
Strain measurement
StrucSense Ensembler
Structural Health Monitoring
Time-Series Data
Vibration Classification
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Structural health monitoring is critical for ensuring the safety and integrity of buildings, especially in seismic zones or environments subject to high loads. This paper presents the "StrucSense Ensembler," an advanced ensemble learning model designed to monitor building stability by integrating time-series data from BMA280 accelerometers and spatial data from strain gauges. The StrucSense Ensembler combines the strengths of Long Short-Term Memory (LSTM) networks for analyzing temporal accelerometer data and Convolutional Neural Networks (CNNs) for interpreting spatial strain data. By leveraging this hybrid approach, the Model delivers accurate predictions of structural stability, enabling proactive decision-making. The ensemble is evaluated using various metrics, including accuracy, precision, recall, and F1 score, and is demonstrated to outperform traditional models like Random Forest, Gradient Boosting Machines, and AdaBoost. The proposed StrucSense Ensembler effectively identifies potential risks such as deformation, cracks, and load imbalances, offering actionable insights for timely intervention and maintenance. This innovative approach provides a robust framework for real-time structural monitoring, significantly enhancing the safety and reliability of buildings.
ISSN:2768-0673
DOI:10.1109/I-SMAC61858.2024.10714827