A Comparison of Speed-Feed Fuzzy Intelligent System and ANN for Machinability Data Selection of CNC Machines

The machining process exhibits piecewise behaviour and cannot be linearly extrapolated in a wide range. It cannot be modelled effectively using theories and equations. Thus, expert systems have emerged as a major tool for decision-making in such complicated situations (Singh & Raman, 1992). The conventional method for selecting machining parameters such as cutting speed and feed rate is based on data from machining hand books and/ or on the experience and knowledge of the operator or CNC programmer. The parameters chosen in most cases are extremely conservative to protect overmatching errors from tool failures such as deflection, wear, breakage, etc. Accordingly, the metal removal rate is low due to the use of such conservative machining parameters (Park & Kim, 1998). Guidelines on machinability data selection is normally made on the basis of the manufacturer’s machinability hand book (Hashmi et al., 2003). Using machining data handbook for the choice of cutting conditions for material hardness that lies in the middle of a group is simple and straight forward. But there exists a degree of vagueness in boundary cases, where two choices of cutting speeds are applicable for one choice of material hardness. In this situation, the skilled operator makes a decision on the appropriate cutting speed, based on his experience. However, this method of data selection by individual operators is not very desirable, because it may vary from operator to operator. Therefore, it is desirable to have an operator independent data selection system for choosing machining operation (Hashmi et al., 1998). While the output variables of the machining process depend on the cutting conditions, the decision concerning the selection of the cutting parameters have an important influence on the extent, cost and quality of the production. Due to the increased use of CNC machines and severe competition between the makers, the importance of precise optimization cutting conditions has increased (Cus & Zuperl, 2006).