Provide a Descriptor Model to Improve the Effect of FeO in Raw Pellets with Data Mining Approach
Data mining is extraction the knowledge from a wide range of data. In this study, data mining is used to analyze a production system in the Golgohar Sirjan Company. The collected data is saved in the Excel file and then data cleaning and data preparation operations were done in order to use in the IBM SPSS Modeler software. FeO is one of the controlling parametersduring the production of the pellet. In this study, the classification method of C & R Tree is used to predict the effect of FeO in the raw pellet against 10 input variables included SiO2, Fetot, CaO, S, Al2O3, Mgo, P, Fe2O3, C.C.S, Temp (dry temperature). The variables that create the most sensitivity on the FeO in the raw pellet are evaluated and compared according to the accuracy of the models and also, practical suggestions are presented for the directors of industry to improve the quality of pellet.
Chan, C & Lewis, B (2002), A basic primer on data mining, information systems management, vol.19, no.4, pp.56-60.
Dalkilic, T (2009), neural networks approach for determining total claim amounts in insurance,insurance:mathematics and economics.
Das,s (2009), utilization of self-organizing map and fuzzy clustering for site characterization using piezocone data, computers and geotechnics.
Eisele T.C. & Kawatra S. K (2003): Miner. Process. Extra. Metall. Rev., 24 (2003), 5.
Nezhad, (2009). Seminar “data mining technology application’. January 2009.
Perzyk, M., Kochansk, J. Kozlowski, A. Soroczynski & Biernadck R (2013).
Tan, P & Steinbach, M (2006), introduction to data mining,United States of America,Addison wesly.
V. Niiniskorpi: (2002) 61st Iron Making Conf. Proc., Iron and Steel Society, Warrendale, PA (USA).