Experimental Study and Artificial Neural Network Modeling of Machining with Minimum Quantity Cutting Fluid

Abstract Conventional petroleum based cutting fluids have various detrimental environmental effects. Also, these cutting fluids are hazardous to operators in prolong use. Therefore, minimum quantity cutting fluid (MQCF) technique has emerged to minimize amount of cutting fluid during metal cutting. An essential step in turning operation is the correct choice of input variables, which in turn regulate important features including tool wear and surface roughness. The present investigation involves a comparative experimental study at three different machining environments such as dry air cooling (DAC), flood cooling (FC) and minimum quantity cutting fluid (MQCF). Results illustrate that MQCF shows best workpiece surface finish at different feeds and cutting speeds. Afterwards, the experimental process is modeled using artificial neural network (ANN) and trained for acquiring the required dataset of process variables. Further, the mean absolute percentage errors (MAPE) is calculated between ANN forecasted values and experimental findings and it is observed that the error lie in the range of 1% to 4%, which validate the current ANN model.