Title :
Structure design about fuzzy rule based on context sensitive constraint
Author :
Zhang Kejia ; Li Chunsheng
Author_Institution :
Modern Educ. Tech. Center, Northeast Pet. Univ., Daqing, China
Abstract :
In the process of achieving decision by intelligent method, in order to improve accuracy and reliability in using fuzzy rules. We define the C-E Model as the basis, bringing in the semantic restriction function to describe the fuzzy rules tendency. By quantitatively transformating for expert opinions, initializing semantic restriction function for fitting rules. We raise an error-correction method that can correct errors of semantic restriction function. In the process of matching rules,meanwhile, building the fuzzy rules that can be self-adjusted. To use the design idea of frame expression method for reference, in combination with the BRBC Model, the design thought of the frame-fuzzy rules group is adopted, and the design that based on the BRBC-F with semantic restriction is accomplished. In the mean time, a more scientific set of calculation method for the structure coincidence is brought out. On this account, to excavate experts´profund cognition and the potential experience. Lower the impact of specialized knowledge, improve the refined fuzzy sets computation. Finally, in combination with certain Production Plant of Daqing Oilfield, utilizing the BRBC-F to accomplish the expert knowledge design about the diagnose for the broken-down wells.
Keywords :
expert systems; fault diagnosis; fuzzy set theory; industrial plants; petroleum industry; BRBC Model; C-E model; Daqing oilfield; broken-down wells diagnosis; context sensitive constraint; error-correction method; expert opinion; frame expression method; frame-fuzzy rules group; fuzzy rule; production plant; rules matching function; semantic restriction function; Cognition; Computational modeling; Computers; BRBC-F Model; CE Model; Fuzzy Rule; Self-Adaption; Semantic Restriction; Tendency;
Conference_Titel :
Computer Science & Education (ICCSE), 2014 9th International Conference on
Conference_Location :
Vancouver, BC
Print_ISBN :
978-1-4799-2949-8
DOI :
10.1109/ICCSE.2014.6926517