Fundamental frequency estimation of masonry towers based on artificial neural networks

Abstract

Historic masonry towers are regarded as symbolic and indispensable components of churches in religious countries across the globe. Masonry towers, with their brittle materials, slenderness, and distinctive shapes, are highly susceptible to lateral excitations. In the aftermath of numerous tower collapses, the urgency of preserving surviving ones in earthquake-prone regions has become apparent. Furthermore, there is a prioritization of identifying and reinforcing the most vulnerable minarets. To address this concern, an innovative approach is proposed, employing Artificial Neural Networks (ANNs) to promptly anticipate the predominant frequency of masonry towers. Predictions are based on the earthquake spectrum specific to each region, effectively alerting to the seismic vulnerabilities of towers constructed within those areas. Rather than formulating relationships based on known geometric parameters, particularly the total height, this study relies on an ANN-based model. Measurements taken from actual masonry towers are utilized as the output database for the neural networks. Consequently, the suggested ANN tool exhibits both practicality and robustness, offering an acceptable level of accuracy in estimating the desired modal information. The ANN approach is, therefore, an alternative for the same purposes of previous studies as well as standards and becomes a quick tool for future study, especially when more desired information is taken into account.