Machining process energy consumption modelling using response surface methodology and neural network

Machining parameters influence the energy consumed during machining processes. A reliable prediction model for energy consumption will enable industry to achieving energy saving by optimizing the machining parameters during process planning stage. This paper presents a two-level optimization artificial neural network modelling method to characterizing the relationship between energy consumption and the machining parameters. On the first level, the input of the artificial neural network is optimized by the regression analysis. On the second level, the parameters of the artificial neural network are optimized by genetic-simulated annealing algorithm. The method has been tested and validated based on experimental data. The experimental results show that this modelling method can lead to a more accurate predicted of energy consumption for the given machining parameters.