Font Size: a A A

Dense Medium Separation Expert Knowledge Base Development In Taixi Washey Coal Factory

Posted on:2015-04-23Degree:MasterType:Thesis
Country:ChinaCandidate:H F ZhangFull Text:PDF
GTID:2181330422987172Subject:Mining engineering
Abstract/Summary:PDF Full Text Request
Along with the maturation of Expert System, it has been widely used in the fieldof mineral processing. The whole Expert System usually consists of six parts:human-computer interface, knowledge base, inference engine, interpreter, synthesisdatabase and knowledge acquisition. However, the merits of the knowledge basedecide the Expert System capability. Coal is the main energy and an importantindustrial raw material in China, coal processing can obtain economic benefits,environmental benefits and social benefits. In recent years, with the development ofcoal preparation technology, more and more heavy medium cyclone s have been usedin preparation plant because of its good performance. The separation process of heavymedium cyclone is in the complex flow field, there are many factors affect theseparation effect of dense medium cyclone, moreover, some of these factors aresuperposed, these factors can be divided into two aspects: structure parameters andprocess parameters, in the production, once equipment has been installed, in a certainperiod of time its structure parameters can be accredited unchanged, so the study onthe process parameters has become spotlight.In this paper, simple mathematical model between the process parameters and theseparation effect was used to store expert knowledge formally in expert knowledgebase, former research on heavy medium cyclone was summarized; the influence ofprocess parameters on the separation effect of dense medium cyclone was discussed,try to obtain the model of separation effect from a variety of ways. For Taxi Coalpreparation plant2branch, this paper used scientific experimental method, theindustrial experience was developed with different factors levels which can bemonitored, different data-processing technique was used to obtain the effecting model.Industrial test results show that,ANN model gives relative prediction error of Ep1andδp1are15.89%and1.94%. The ANN model can give the prediction heavy mediumcyclone separation effect with the currently detected process parameters, this can beused in industrial production and obtain good separation effect.
Keywords/Search Tags:expert system, knowledge base, heavy medium cyclone, separation effect, process parameter
PDF Full Text Request
Related items