東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

FindBook

Google Book

Amazon

博客來

Uncertainty-Assisted Artificial Intelligence for Reliable Structural Health Monitoring.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Uncertainty-Assisted Artificial Intelligence for Reliable Structural Health Monitoring./
作者:	Sajedi, Seyedomid.
面頁冊數:	1 online resource (271 pages)
附註:	Source: Dissertations Abstracts International, Volume: 83-12, Section: B.
Contained By:	Dissertations Abstracts International83-12B.
標題:	Civil engineering. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=29210219click for full text (PQDT)
ISBN:	9798819394366

Uncertainty-Assisted Artificial Intelligence for Reliable Structural Health Monitoring.
Sajedi, Seyedomid.

Uncertainty-Assisted Artificial Intelligence for Reliable Structural Health Monitoring. - 1 online resource (271 pages)

Source: Dissertations Abstracts International, Volume: 83-12, Section: B.

Thesis (Ph.D.)--State University of New York at Buffalo, 2022.

Includes bibliographical references

Structural Health Monitoring (SHM) is the practice of implementing a damage detection strategy and provides essential information for the effective maintenance of civil infrastructure. Human inspections have served as the standard technique for structural assessments during normal operations or after extreme events. Considering restrictions in time and resources and the size of aging infrastructure across the US, processing large amounts of information is not always practical using the standard human methods. Therefore, automated inspections can help save millions of dollars by providing timely information on the presence and extent of damage and reducing the downtime of civil infrastructure. Artificial Intelligence (AI) can be a powerful tool in the hands of engineers and stakeholders to achieve this automation. This dissertation provides a series of studies addressing certain limitations in the existing frameworks on reliable information processing for civil infrastructure damage diagnosis leveraging machine learning. Integration of performance-based earthquake engineering concepts and modeling the consequences of error in AI-equipped damage diagnosis is one of the contributions of this research. Furthermore, the potential applications of damage-sensitive feature extraction techniques, including cumulative intensity features and Mel filter banks, are evaluated, and guidelines on the proper selection of hyperparameters are provided. This study also introduces a series of custom deep learning architectures and concepts for vibration-based SHM, including Semantic Damage Segmentation (SDS), Hybrid Deep learning for Rapid Assessment (HyDRA), and Trident. Several case studies on building and bridge structures are considered to evaluate the robustness of these frameworks for fine-resolution damage diagnosis in structures with dense sensor arrays. The potential applications of generative AI for engineering is another area of contribution where a novel Deep Generative Bayesian Optimization (DGBO) method is developed for the optimal sensor placement problem. A key contribution of this work is studying the reliability of AI automation in SHM applications that utilize vision and vibration data. The importance of deep learning model uncertainty quantification through deep Bayesian neural networks is investigated. The concept of surrogate models is proposed to interpret uncertainty for more robust predictions and as a means to trigger human intervention. The performance of surrogate models for applications in visual and vibration-based SHM case studies highlights their critical role in achieving reliable automation and enabling risk-informed decision making.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2023

Mode of access: World Wide Web

ISBN: 9798819394366Subjects--Topical Terms:

860360
Civil engineering.
Subjects--Index Terms:

Computer visionIndex Terms--Genre/Form:

542853
Electronic books.

Uncertainty-Assisted Artificial Intelligence for Reliable Structural Health Monitoring.
LDR:04194nmm a2200433K 4500 001 2353201
005 20221214062822.5
006 m o d
007 cr mn ---uuuuu
008 241011s2022 xx obm 000 0 eng d
020 $a 9798819394366
035 $a (MiAaPQ)AAI29210219
035 $a AAI29210219
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Sajedi, Seyedomid. $3 3693543
245 1 0 $a Uncertainty-Assisted Artificial Intelligence for Reliable Structural Health Monitoring.
264 0 $c 2022
300 $a 1 online resource (271 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 83-12, Section: B.
500 $a Advisor: Liang, Xiao.
502 $a Thesis (Ph.D.)--State University of New York at Buffalo, 2022.
504 $a Includes bibliographical references
520 $a Structural Health Monitoring (SHM) is the practice of implementing a damage detection strategy and provides essential information for the effective maintenance of civil infrastructure. Human inspections have served as the standard technique for structural assessments during normal operations or after extreme events. Considering restrictions in time and resources and the size of aging infrastructure across the US, processing large amounts of information is not always practical using the standard human methods. Therefore, automated inspections can help save millions of dollars by providing timely information on the presence and extent of damage and reducing the downtime of civil infrastructure. Artificial Intelligence (AI) can be a powerful tool in the hands of engineers and stakeholders to achieve this automation. This dissertation provides a series of studies addressing certain limitations in the existing frameworks on reliable information processing for civil infrastructure damage diagnosis leveraging machine learning. Integration of performance-based earthquake engineering concepts and modeling the consequences of error in AI-equipped damage diagnosis is one of the contributions of this research. Furthermore, the potential applications of damage-sensitive feature extraction techniques, including cumulative intensity features and Mel filter banks, are evaluated, and guidelines on the proper selection of hyperparameters are provided. This study also introduces a series of custom deep learning architectures and concepts for vibration-based SHM, including Semantic Damage Segmentation (SDS), Hybrid Deep learning for Rapid Assessment (HyDRA), and Trident. Several case studies on building and bridge structures are considered to evaluate the robustness of these frameworks for fine-resolution damage diagnosis in structures with dense sensor arrays. The potential applications of generative AI for engineering is another area of contribution where a novel Deep Generative Bayesian Optimization (DGBO) method is developed for the optimal sensor placement problem. A key contribution of this work is studying the reliability of AI automation in SHM applications that utilize vision and vibration data. The importance of deep learning model uncertainty quantification through deep Bayesian neural networks is investigated. The concept of surrogate models is proposed to interpret uncertainty for more robust predictions and as a means to trigger human intervention. The performance of surrogate models for applications in visual and vibration-based SHM case studies highlights their critical role in achieving reliable automation and enabling risk-informed decision making.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2023
538 $a Mode of access: World Wide Web
650 4 $a Civil engineering. $3 860360
650 4 $a Engineering. $3 586835
650 4 $a Computer science. $3 523869
650 4 $a Artificial intelligence. $3 516317
650 4 $a Geological engineering. $3 2122713
653 $a Computer vision
653 $a Deep learning
653 $a Earthquake engineering
653 $a Signal processing
653 $a Structural damage diagnosis
653 $a Structural health monitoring
655 7 $a Electronic books. $2 lcsh $3 542853
690 $a 0537
690 $a 0984
690 $a 0800
690 $a 0466
690 $a 0543
710 2 $a ProQuest Information and Learning Co. $3 783688
710 2 $a State University of New York at Buffalo. $b Civil, Structural and Environmental Engineering. $3 1023703
773 0 $t Dissertations Abstracts International $g 83-12B.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=29210219 $z click for full text (PQDT)