Supplementary

postgraduate thesis: Improving the performance of lifts using artificial intelligence techniques

TitleImproving the performance of lifts using artificial intelligence techniques
Authors
Advisors
Advisor(s):Cheung, KC
Issue Date2003
PublisherThe University of Hong Kong (Pokfulam, Hong Kong)
Citation
Wong, K. [黃敬修]. (2003). Improving the performance of lifts using artificial intelligence techniques. (Thesis). University of Hong Kong, Pokfulam, Hong Kong SAR.
Abstract(Uncorrected OCR) Abstract of thesis entitled Improving the Performance of Lifts Using Artificial Intelligence Techniques submitted by Wong King Sau for the degree of Doctor of Philosophy at the University of Hong Kong in August 2003 An elevator group control system manages multiple elevators to serve hall calls in a building. Most elevator group control systems need to recognize the traffic pattern of the building and then change their control algorithms to improve the efficiency of the elevator system. However, the traffic flow in a building is very difficult to be classified into distinct patterns. Traffic recognition systems can recognize certain traffic patterns, but mixed traffic patterns are difficult to be recognized. The aim of this study was therefore to develop improved duplex elevator group control systems that do not need to recognize the traffic pattern. A fuzzy logic. control unit and genetic algorithms control unit were used. A fuzzy logic control unit integrates with the conventional duplex elevator group control system to improve performance especially in mixed traffic patterns with intermittent heavy traffic demand. This system will send more than one elevator to a floor with heavy demand, . according to the overall passenger traffic conditions in the building. The genetic algorithms control unit divides the building into three zones and assigns an appropriate number of elevators to each zone. The floors covered by each zone are adjusted every five minutes. This control unit optimizes elevator group control by equalizing the number of hall calls in each zone, the total elevator door opening time in each zone, and the number of floors served by each elevator. Both of the control units were tested by a simulator in a computer. The performance of the elevator system is given by indices such as average waiting time, wasted man-hour, and long waiting time percentage. The new performance index "wasted man-hour" indicates the total time spent by passengers in a building waiting for the lift service. Both proposed systems perform better than the conventional duplex control system. (An abstract of 297 words.) ~ Signed _ Wong King Sau
DegreeDoctor of Philosophy
SubjectFuzzy logic.
Genetic algorithms.
Elevators, Automatic.
Dept/ProgramMechanical Engineering
Persistent Identifierhttp://hdl.handle.net/10722/31202
HKU Library Item IDb2768295

 

DC FieldValueLanguage
dc.contributor.advisorCheung, KC-
dc.contributor.authorWong, King-sau-
dc.contributor.author黃敬修zh_HK
dc.date.issued2003-
dc.identifier.citationWong, K. [黃敬修]. (2003). Improving the performance of lifts using artificial intelligence techniques. (Thesis). University of Hong Kong, Pokfulam, Hong Kong SAR.-
dc.identifier.urihttp://hdl.handle.net/10722/31202-
dc.description.abstract(Uncorrected OCR) Abstract of thesis entitled Improving the Performance of Lifts Using Artificial Intelligence Techniques submitted by Wong King Sau for the degree of Doctor of Philosophy at the University of Hong Kong in August 2003 An elevator group control system manages multiple elevators to serve hall calls in a building. Most elevator group control systems need to recognize the traffic pattern of the building and then change their control algorithms to improve the efficiency of the elevator system. However, the traffic flow in a building is very difficult to be classified into distinct patterns. Traffic recognition systems can recognize certain traffic patterns, but mixed traffic patterns are difficult to be recognized. The aim of this study was therefore to develop improved duplex elevator group control systems that do not need to recognize the traffic pattern. A fuzzy logic. control unit and genetic algorithms control unit were used. A fuzzy logic control unit integrates with the conventional duplex elevator group control system to improve performance especially in mixed traffic patterns with intermittent heavy traffic demand. This system will send more than one elevator to a floor with heavy demand, . according to the overall passenger traffic conditions in the building. The genetic algorithms control unit divides the building into three zones and assigns an appropriate number of elevators to each zone. The floors covered by each zone are adjusted every five minutes. This control unit optimizes elevator group control by equalizing the number of hall calls in each zone, the total elevator door opening time in each zone, and the number of floors served by each elevator. Both of the control units were tested by a simulator in a computer. The performance of the elevator system is given by indices such as average waiting time, wasted man-hour, and long waiting time percentage. The new performance index "wasted man-hour" indicates the total time spent by passengers in a building waiting for the lift service. Both proposed systems perform better than the conventional duplex control system. (An abstract of 297 words.) ~ Signed _ Wong King Sau-
dc.languageeng-
dc.publisherThe University of Hong Kong (Pokfulam, Hong Kong)-
dc.relation.ispartofHKU Theses Online (HKUTO)-
dc.rightsThe author retains all proprietary rights, (such as patent rights) and the right to use in future works.-
dc.source.urihttp://hub.hku.hk/bib/B2768295X-
dc.subject.lcshFuzzy logic.-
dc.subject.lcshGenetic algorithms.-
dc.subject.lcshElevators, Automatic.-
dc.titleImproving the performance of lifts using artificial intelligence techniques-
dc.typePG_Thesis-
dc.identifier.hkulb2768295-
dc.description.thesisnameDoctor of Philosophy-
dc.description.thesislevelDoctoral-
dc.description.thesisdisciplineMechanical Engineering-
dc.description.natureabstract-
dc.description.naturetoc-
dc.identifier.mmsid991021776519703414-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats