A Reduced Model for Energy Consumption Analysis in Milling

Abstract General awareness on energy consumption is globally growing, becoming a significant performance parameter also in the manufacturing field. The current work outlines a reduced model for the analysis of the energy consumption of a machine tool during face milling operations. The model is characterized by a minimum set of significant parameters describing the product, the process and the machine. The influence of these parameters on energy consumption is represented by a feed forward neural network with 20 inputs, two hidden layers and one output. The input parameters are identified a priori by means of simplifications based on physical and technological considerations, while their relevance is evaluated by a sensitivity analysis using the neural network. The network has been trained and validated on 800 experiments consisting of runs of a continuous-time simulator that estimates the energy consumption of a machine tool during part program execution.