Status : Verified
| Personal Name | Samudio, Roydon Jude S. |
|---|---|
| Resource Title | Application of Machine Learning-Based Stochastic Subspace Identification for Automated Modal Analysis of Transmission Towers |
| Date Issued | 02 July 2024 |
| Abstract | Although fully automated modal analysis, a Structural Health Monitoring (SHM) technique, has recently been used to monitor the current condition of various civil structures, its application to transmission towers remains limited. Most modal analysis and dynamic characterization studies related to these towers, which are essential for first-level damage detection, still require manual selection of input parameter values. This research aims to contribute to the existing discussion by applying a machine learning algorithm to Stochastic Subspace Identification (SSI) to derive the modal parameters of a transmission tower, thereby improving similar studies. In addition to utilizing Random Forest as the core intelligence of the method, the research explores three other machine learning algorithms—XGBoost, Decision Trees, and k-Nearest Neighbors (KNN)—as alternative modal prediction models within the framework. This paper presents the results of both analytical (PV stage) and operational (FI stage) modal analyses on a 230-kV transmission tower in Orion, Bataan, Philippines. Despite the limited sensor locations, ten frequency values extracted from the actual tower closely approximated those from analytical studies and predictions from related research. Significant differences were noted for damping ratios, which typically exhibit higher estimation uncertainty. However, the values obtained from the field test, which ranged from 0.68% to 3.02%, fell within the recommendations of international structural codes, such as the 4% suggested by ASCE 74. This study also demonstrates techniques for analyzing how environmental factors such as wind speed and temperature affect the modal parameters. Random Forest stands out among the machine learning models tested, showing the fastest runtime, highest performance accuracy, and smallest Coefficient of Variation values given random datasets, closely followed by XGBoost. The results of this study can be used in model updating and s |
| Degree Course | Master of Science in Civil Engineering |
| Language | English |
| Keyword | automated modal analysis; transmission tower; machine learning; structural health monitoring; modal parameters; environmental effect |
| Material Type | Thesis/Dissertation |
Preliminary Pages
529.11 Kb
Category : F - Regular work, i.e., it has no patentable invention or creation, the author does not wish for personal publication, there is no confidential information.
Access Permission : Open Access