Please use this identifier to cite or link to this item: https://scholarbank.nus.edu.sg/handle/10635/166995
DC FieldValue
dc.titleDEVELOPMENT OF AN EXPERT SYSTEM FOR PIPELINE ROUTING
dc.contributor.authorONG CHENG HOCK
dc.date.accessioned2020-04-23T01:08:42Z
dc.date.available2020-04-23T01:08:42Z
dc.date.issued1990
dc.identifier.citationONG CHENG HOCK (1990). DEVELOPMENT OF AN EXPERT SYSTEM FOR PIPELINE ROUTING. ScholarBank@NUS Repository.
dc.identifier.urihttps://scholarbank.nus.edu.sg/handle/10635/166995
dc.description.abstractA semantic network expert system is developed for the simulation of the management of a hypothetical pipeline network. The expert system is implemented in a microcomputer with a graphic display system, a sampling/sending system, a parser and querying system and a searching system. Set of rules or clauses are constructed to develop these systems. The database of the expert system mainly contains the knowledge of the connectivity of piping network, the status of piping items and the evaluation values used during the search process for finding a shortest flow path for a particular fluid transfer. The problems of finding pipeline routes are formulated into state descriptions; and an effective heuristic search method is then used to search for the goal state in the problem state space. The search method used in this state-space approach employs forward deduction approach and uses evaluation values derived from the knowledge of experienced piping control operators to expand intermediate piping items between the source item and the destination item in order to search an optimal path. It is found that the blind search process is ineffective and not realistic in problem solving as they have to search for too many solution paths between two terminal items. The process requires longer computation time and large memory storage. On the other hands, the heuristic search process guided by evaluation values derived from knowledge of experienced operators could find an optimal flow path within a short computation time and less memory storage required. However the optimality of the path found depends on the evaluation values, which indirectly depends on the knowledge acquisition process. Thus it is found that the path found may not always be optimal but usually is near optimal.
dc.sourceCCK BATCHLOAD 20200406
dc.typeThesis
dc.contributor.departmentCHEMICAL ENGINEERING
dc.contributor.supervisorY. K. HUEN
dc.description.degreeMaster's
dc.description.degreeconferredMASTER OF ENGINEERING
Appears in Collections:Master's Theses (Restricted)

Show simple item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
b1728188x.pdf13.87 MBAdobe PDF

RESTRICTED

NoneLog In

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.