Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor | Department of Industrial and Systems Engineering | en_US |
| dc.contributor.advisor | Chung, Sai Ho Nick (ISE) | en_US |
| dc.creator | Wang, Ye | - |
| dc.identifier.uri | https://theses.lib.polyu.edu.hk/handle/200/12971 | - |
| dc.language | English | en_US |
| dc.publisher | Hong Kong Polytechnic University | en_US |
| dc.rights | All rights reserved | en_US |
| dc.title | Adoption of AI in functional safety systems – smart elevator condition monitoring by the deep learning based data-driven modelling approach | en_US |
| dcterms.abstract | As pervasively utilised in the contemporary society, the elevator has become one of the indispensable instruments in our daily life, while also being accompanied with the emerging degradation issues that impairing their normal operations to threaten the safety of passengers. With the gradual deterioration of core electrical and mechanical (E&M) components, the continued accidents of elevators over the years have exerted great challenges on the condition monitoring of their reliability status athwart the way to prevent the hazardous events from happening. Especially among the densely populated metropolitan cities, the modernization for such widespread and extensively-utilised vertical transport means with aged elevators installed in the residential buildings or skyscrapers over the decades has become one of the prior concerns for industrial practitioners in actual practice. The effective modernization of elevators is highly reliant on the advanced technological enablers and auxiliaries while helping to monitor the condition of elevators. Conventional methods as widely adopted by manufacturers or registered contractors (RCs) are focused on the knowledge-based physics of failure (PoF) approaches with statistical modelling through the complex parameter engineering. Whereas as restricted by a certain brand of elevators, such approaches are highly reliant on the prior knowledge acquired from domain experts with the developed proprietary software in compliance with the elevator’s specific intrinsic E&M architecture, thus lacking the generalisation capability. Moreover, the traditional monitoring methods are mostly based on the data acquisition from embedded sensors as the electrical transducers or optical transceivers that are either equipped intrinsically during the manufacturing process or installed intrusively by the maintenance crews. Such invasive sensor installations might inevitably undermine the systematic integrity while interrupting the elevator’ existing circuitries with laborious costs that restricting their broad applications. Consequently, the prevailing condition monitoring methods have their inherent pitfalls while inevitably restraining and hindering the modernization process on the aging elevators in the contemporary society. | en_US |
| dcterms.abstract | With the thriving development of information technologies via advanced digitalisation, there are emerging trends in monitoring the safety-critical E&M systems by data-driven approaches. Nevertheless, regarding the domain of elevator condition monitoring, there lacks relevant research to explore the effective utilization methods or frameworks as customized in consideration of elevators’ unique signal patterns and characteristics as fetched from the industrial Artificial Intelligence of Things (AIoT) sensors. | en_US |
| dcterms.abstract | This research has been conducted aiming to establish the comprehensive smart elevator condition monitoring framework by the utilization of deep learning-based data-driven modelling methods to process the non-intrusive multi-variant electric current and electro-optical signals of elevators with health condition measurement. Moreover, the in-depth research dives into the novel algorithmic designs with theoretical breakthroughs to enhance the model’s feature extraction capability to accommodate the heterogeneous and extremely long-sequential signal patterns, while tackling the imbalanced dataset issues with the thriving generative AI technology. After completion of the progressive stages of research, it is expected that the outcomes could facilitate the effective elevator condition monitoring with the proposed intelligent deep learning models to leverage the safety level of elevators under the “Smart City” blueprint. | en_US |
| dcterms.extent | 220 pages : color illustrations | en_US |
| dcterms.isPartOf | PolyU Electronic Theses | en_US |
| dcterms.issued | 2024 | en_US |
| dcterms.educationalLevel | Ph.D. | en_US |
| dcterms.educationalLevel | All Doctorate | en_US |
| dcterms.LCSH | Elevators -- Maintenance and repair | en_US |
| dcterms.LCSH | Elevators -- Safety measures | en_US |
| dcterms.LCSH | Hong Kong Polytechnic University -- Dissertations | en_US |
| dcterms.accessRights | open access | en_US |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/12971