Prediction of diamond wire sawing machine performance: An application of mamdani type fuzzy rule based inference system

The most common production method used to extract natural stones from open pits in blocks is made by a diamond wire cutting machine. The factors affecting the cutting procedure in natural stone quarries can be listed such as the parameters related to rock, technical developments related to the machinery and equipment used, geological and geotechnical conditions in the region, preferences about application of diamond wire cutting method in the field, knowledge and experience and business skills of employed workers. Since these factors vary so much, determining the cutting speed that can be achieved in the field by a diamond wire cutting machine is a very complex procedure. In this study, fuzzy logic approach was used to determine the cutting performance of diamond wire cutting machine. The fuzzy logic concept allows users to reflect their engineering experience to the fuzzy systems by providing verbal expressions so as to be understood by computers. As a result, it is aimed to create a fuzzy estimation model by going through establishing logical relations between input and output variables which affect cutting speed of diamond wire cutting machine by considering the research practices of experienced mining engineers who works in marble quarries for years. To achieve this goal; a fuzzy estimation model whose input variables are type of rock, type of diamond wire used, the type of block cutting, type of diamond wire machine used, the distance of the machine from the face during cutting procedure, and the length of the wire used in the cutting process was developed to predict the cutting speed of diamond wire cutting machine. In order to reveal the prediction performance of the proposed fuzzy model, 400 measurement data obtained from the field were evaluated. It has been achieved as far as the correlation coefficient between the measured and predicted by developed fuzzy model is 0.87 and the constructed fuzzy estimation model can successfully predict the cutting speed of a diamond wire cutting machine.