Estimation of value such as thickness and quality of row material based on the location in mining is important due to providing of raw material homogeneity and determining of optimum production plans in cement production. Core boring is generally performed for information of site at any location of area, but performing in every desired location is not economic because of high cost of investment. Data of belonging to desired locations are estimated from limited numbers of drillhole data by using of estimating techniques in mining. In this study, 67 thickness values of marn formation belong to raw material field of Adana Cement were used. Thickness estimations were carried out with this data by polygon, kriging and fuzzy modeling interpolation approaches and performance comparisons were made. The obtained results show that the prediction performance of the fuzzy model is higher than polygon and kriging for this row material field.
Hammadde sahalaryndaki kalite ve kalynlyk gibi de?erlerin lokasyonlara ba?ly olarak belirlenmesi, çimento üretiminde hammadde homojenli?inin sa?lanabilmesi ve optimum üretim planlarynyn yapylabilmesinden dolayy önemlidir. Herhangi bir lokasyondaki sahaya ait bilgiler için genellikle karotlu sondaj çaly?malary yapylyr ancak ilk yatyrym maliyetlerin yüksek olmasy sebebiyle istenilen her noktada yapylmasy ekonomik de?ildir. Madencilikte istenilen noktalardaki bilgiler bilinen synyrly sayyda sondajlara ait veriler ile tahmin teknikleri kullanylarak kestirilir. Bu çaly?mada, Adana Çimento hammadde sahasyna ait 67 lokasyona ait marn formasyonu kalynlyk verileri kullanylmy?tyr. Bu verileri ile poligon, kriging ve mamdani tipi bulanyk modelleme yakla?ymlary ile kestirimler yapylmy? ve kestirim performanslary test edilmi?tir. Elde edilen sonuçlar bulanyk yakla?ym performansynyn bu hammadde sahasy için di?er iki modelle göre daha yüksek oldu?unu göstermi?tir.
Primary Language | Turkish |
---|---|
Journal Section | Computer Engineering |
Authors | |
Publication Date | February 1, 2011 |
Published in Issue | Year 2011 Volume: 6 Issue: 1 |