Analysis and prediction of the impact of technological parameters on cutting force components in rough milling of AZ31 magnesium alloy

Abstract This paper presents the results of experimental study AZ31 magnesium alloy milling process. Dry was carried out under high-speed machining conditions. First, a stability lobe diagram determined using CutPro software. Next, studies were to verify diagram. The tests for different feed per tooth and cutting speed values two types tools. During investigations, forces in three directions recorded. obtained time series subjected general analysis composite multiscale entropy. Modelling prediction performed Statistica Neural Network software, which neural networks applied: multi-layered perceptron radial basis function. It observed that with high allows component force be lowered as result transition into conditions range. In most cases, highest analysed parameters recorded F x , whereas lowest y . Additionally, shows (with use artificial networks) components can made, both amplitudes amp root mean square rms