Using grammatical evolution for modelling energy consumption on a computer numerical control machine

Discrete manufacturing is known to be a high consumer of energy and much work has been done in continuous improvement and energy saving methods addressing this issue. Computer Numerical Control (CNC) machines, commonly used in the manufacturing of metal parts, are highly energy-demanding because of many required sub-systems, such as cooling, lubrication, logical interfaces and electric motors. For this reason, there is a large body of work focusing on modelling the energy needs of this class of machine. This paper applies Grammatical Evolution (GE) for developing auto-regressive models for the energy consumption of a CNC machine. Empirical data from three 24-hour work shifts comprising three different types of products are used as inputs. We also introduce an autocorrelation-informed approach for the grammar, which benefits from a prior analysis of the training data for better capturing periodic or close to periodic behaviour. Finally, we compare the outcomes from real and predicted energy profiles through the use of an existing analysis tool, which is capable of extracting production-related information such as total and average KW consumption, number of parts produced and breakdown of production and idle hours. Results show that GE yields accurate and explainable models for the analysed scenario.