عنوان مقاله [English]
Failure of steel lattice towers as one of the most important structural units in high-pressure power transmission lines due to environmental factors causes many problems for power transmission by these important urban infrastructures every year. In recent years, online monitoring of the performance of structures by monitoring their dynamic response changes through sensor installation has attracted much attention. However, identifying the location of damage in structures such as lattice metal towers with several components has problems that necessitate the use of techniques with high processing volume. One of the solutions is the use of machine learning-based methods that have attracted the attention of various researchers. In this paper, we try to identify the type and extent of damage in bolted connections in these structures by using one of the classification algorithms in machine learning. Therefore, first, one of the common towers of transmission lines in the country is modeled and dynamically analyzed under wind load. Then, its different connections are modeled at the base level of the tower using software. Then, a set of different forms of damage in simulation connections and mode shapes and natural frequencies in these conditions are extracted to learn the classification algorithm. The obtained database is used to identify the damage in different scenarios. The results of this study emphasize on the effectiveness of the selected method in identifying the location of different types of damages in the leg connections of the towers and its extent.