A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Supervised Machine Learning Based Fundamental Period Prediction of Historical Ottoman Minarets
https://orcid.org/http://orcid.org/0009-0002-5447-0799
https://orcid.org/http://orcid.org/0009-0003-1199-2244
https://orcid.org/http://orcid.org/0009-0003-6262-9955
https://orcid.org/http://orcid.org/0009-0000-8205-929X
https://orcid.org/http://orcid.org/0009-0005-6831-7770
https://orcid.org/http://orcid.org/0009-0003-8679-2917
https://orcid.org/http://orcid.org/0009-0003-5174-443X
https://orcid.org/http://orcid.org/0000-0003-3436-8551
https://orcid.org/http://orcid.org/0000-0001-8484-6027
Ottoman masonry minarets in Islamic regions have long been a symbol of enduring architectural tradition, particularly in their resilience to seismic activities. However, with proper care and maintenance, these minarets can continue to stand tall for generations to come. Ottoman masonry minarets are a prime example of heritage, despite being at risk from lateral excitations due to their brittle materials, slenderness, and unique shapes. Efforts are imperative to preserve the surviving minarets in earthquake-prone areas. The focus should be on identifying and reinforcing the most vulnerable ones. By assessing the earthquake spectrum of a region, the seismic vulnerabilities of minarets can be effectively identified through the periods of their lowest modes. In this study, a solution that is proposed by harnessing the power of Supervised Machine Learning (SML), combines in-situ experimental techniques with numerical modeling and approaches that have been shown to be effective in achieving accuracy when aligned with empirical formulas established within the last decade. This groundbreaking solution addresses the practical hurdles faced by conventional methods, particularly in obtaining authorization to install measurement devices on heritage structures and concerns about potential disruptions to the integrity of historical sites. The SML-based approach confidently predicts the fundamental period of minarets with error levels of <30% compared with the experimental data, considering only limited information about geometries and materials and no need for any invasive test setups or operational modal analyses.
Supervised Machine Learning Based Fundamental Period Prediction of Historical Ottoman Minarets
https://orcid.org/http://orcid.org/0009-0002-5447-0799
https://orcid.org/http://orcid.org/0009-0003-1199-2244
https://orcid.org/http://orcid.org/0009-0003-6262-9955
https://orcid.org/http://orcid.org/0009-0000-8205-929X
https://orcid.org/http://orcid.org/0009-0005-6831-7770
https://orcid.org/http://orcid.org/0009-0003-8679-2917
https://orcid.org/http://orcid.org/0009-0003-5174-443X
https://orcid.org/http://orcid.org/0000-0003-3436-8551
https://orcid.org/http://orcid.org/0000-0001-8484-6027
Ottoman masonry minarets in Islamic regions have long been a symbol of enduring architectural tradition, particularly in their resilience to seismic activities. However, with proper care and maintenance, these minarets can continue to stand tall for generations to come. Ottoman masonry minarets are a prime example of heritage, despite being at risk from lateral excitations due to their brittle materials, slenderness, and unique shapes. Efforts are imperative to preserve the surviving minarets in earthquake-prone areas. The focus should be on identifying and reinforcing the most vulnerable ones. By assessing the earthquake spectrum of a region, the seismic vulnerabilities of minarets can be effectively identified through the periods of their lowest modes. In this study, a solution that is proposed by harnessing the power of Supervised Machine Learning (SML), combines in-situ experimental techniques with numerical modeling and approaches that have been shown to be effective in achieving accuracy when aligned with empirical formulas established within the last decade. This groundbreaking solution addresses the practical hurdles faced by conventional methods, particularly in obtaining authorization to install measurement devices on heritage structures and concerns about potential disruptions to the integrity of historical sites. The SML-based approach confidently predicts the fundamental period of minarets with error levels of <30% compared with the experimental data, considering only limited information about geometries and materials and no need for any invasive test setups or operational modal analyses.
Supervised Machine Learning Based Fundamental Period Prediction of Historical Ottoman Minarets
https://orcid.org/http://orcid.org/0009-0002-5447-0799
https://orcid.org/http://orcid.org/0009-0003-1199-2244
https://orcid.org/http://orcid.org/0009-0003-6262-9955
https://orcid.org/http://orcid.org/0009-0000-8205-929X
https://orcid.org/http://orcid.org/0009-0005-6831-7770
https://orcid.org/http://orcid.org/0009-0003-8679-2917
https://orcid.org/http://orcid.org/0009-0003-5174-443X
https://orcid.org/http://orcid.org/0000-0003-3436-8551
https://orcid.org/http://orcid.org/0000-0001-8484-6027
Ottoman masonry minarets in Islamic regions have long been a symbol of enduring architectural tradition, particularly in their resilience to seismic activities. However, with proper care and maintenance, these minarets can continue to stand tall for generations to come. Ottoman masonry minarets are a prime example of heritage, despite being at risk from lateral excitations due to their brittle materials, slenderness, and unique shapes. Efforts are imperative to preserve the surviving minarets in earthquake-prone areas. The focus should be on identifying and reinforcing the most vulnerable ones. By assessing the earthquake spectrum of a region, the seismic vulnerabilities of minarets can be effectively identified through the periods of their lowest modes. In this study, a solution that is proposed by harnessing the power of Supervised Machine Learning (SML), combines in-situ experimental techniques with numerical modeling and approaches that have been shown to be effective in achieving accuracy when aligned with empirical formulas established within the last decade. This groundbreaking solution addresses the practical hurdles faced by conventional methods, particularly in obtaining authorization to install measurement devices on heritage structures and concerns about potential disruptions to the integrity of historical sites. The SML-based approach confidently predicts the fundamental period of minarets with error levels of <30% compared with the experimental data, considering only limited information about geometries and materials and no need for any invasive test setups or operational modal analyses.
Supervised Machine Learning Based Fundamental Period Prediction of Historical Ottoman Minarets
Lecture Notes in Civil Engineering
Milani, Gabriele (editor) / Ghiassi, Bahman (editor) / Tran, Chon (author) / Tran, Day Ngoc (author) / Pham, Thinh Minh (author) / Tran, Thai Minh (author) / Le, Duong Thai (author) / Quang Truong, Y (author) / Nguyen, Trung Quoc (author) / Nguyen, Quy Thue (author)
International Brick and Block Masonry Conference ; 2024 ; Birmingham, United Kingdom
2024-12-13
10 pages
Article/Chapter (Book)
Electronic Resource
English
A fast and practical approximations for fundamental period of historical Ottoman minarets
| British Library Online Contents | 2019
Revisiting the Frequency Laws for Ottoman Minarets. Analysis of Uncertainties
| Taylor & Francis Verlag | 2023
From One Polygon to Another: A Distinctive Feature of Some Ottoman Minarets
| Springer Verlag | 2011