A Fuzzy Pooling of Coal Dust Knowledge

conference paper

English

Relations among different characteristics (e.g. coal dust concentrations and explicability index) of coal dust are complex, integrated, difficult to quantify and therefore relatively ill known. Conventional experimental records (usually tables, graphs) are not suitable for efficient uncertainty reasoning. A fuzzy knowledge base is a flexible framework for acquisition of vague, sparse and inconsistent knowledge. The fuzzy reasoning is to some extent equivalent to a conventional interpolation algorithm capable of interpolating among variously accurate non-equidistant points in a multi-dimensional space. The pooling of the experimental records, industrial observations and retrieved literature knowledge items represent a trade off between minimal modifi-cation of the original data and elimination of obvious inconsistencies among all sets of data. This is an ad hoc and time consuming process that requires highly qualified field experts and knowledge engineers. A sequence of several modifications is needed to de-crease the expected inconsistencies. A general theoretical background for the methodol-ogy of pooling/upgrading does not exist. The coal dust pool has more than 3000 en-tries. Therefore the paper presents a detailed description of sub set of data based on a relatively homogeneous set of measurements (11 variables e.g. Ash, Moisture, Mini-mum Ignition Temperature and 55 fuzzy conditional statements). No a priori knowl-edge of fuzzy mathematics is assumed.

Coal Dust, Fuzzy Logic, Foüzzy Pooling, Knowledge

Krejčí, A. - Dohnal, M.

25. 8. 2002

Česká společnost chemického inženýrství

Praha

80-86059-33-2

CHISA 2002

160

168

9