Abonnement à la biblothèque: Guest

ISSN Online: 2377-424X

ISBN Print: 1-56032-797-9

International Heat Transfer Conference 11
August, 23-28, 1998, Kyongju, Korea

SLAGGING ASSESSMENT OF POWER PLANT BOILER WITH KNOWLEDGE BASED SYSTEM

Get access (open in a dialog) DOI: 10.1615/IHTC11.10
pages 3-8

Résumé

The paper presents the design of an expert system for slagging assessment in power plant boilers. It is an on-line expert system based on selected criteria for the slagging assessment. Using criteria for slagging assessment based on " clean" and "not-clean" radiation heat flux measurements, the diagnostic variables are defined for the boiler heat transfer surfaces. The development of a prototype knowledge-based system for slagging assessment in power plant boiler comprise integration of the elements including knowledge base, inference procedure and prototype configuration.
Diagnostic variables are organised by the knowledge representation method in cases. The cases are obtained by 3D model of the boiler furnace and representing individual situations expected in boiler. The inference procedure is the set of procedural rules to be followed in the processing of diagnostic variables for the retrieval of diagnostic cases. This procedure implies fuzzification of actual variables and their conversion into a linguistic variable with the respective membership function. The sets of linguistic variables representing all fields in the boiler furnace are selected by the minimum values of membership functions. The selection of the case with the maximum values of the membership function among the cases under consideration will lead to the diagnostic case.
Demonstration of the prototype knowledge-based system for slagging assessment was performed on the Sines power plant. It is a 300 MW coal fired power plant. 12 fields are used with 3 on each side of boiler.