Development of an Electric-Load Intelligence System for Component Level Disaggregation to Improve Energy Efficiency of Machine Tools

Abstract

Energy and resource efficient manufacturing has become a key priority due to higher energy cost, market competition and environmental regulations. Better transparency and higher levels of disaggregation of energy data are necessary for energy efficiency improvement of machine tools. Since the beginning of the 21st century, some attempts have been made by the researchers to quantify the energy data but only up to the operational state of the machine tool. Better accuracy and transparency require disaggregation up to the component level. This study proposes an Electric-Load Intelligence (E-LI) system for identification of machine tool operating state and disaggregation of time and energy consumed up to the component level. The energy profile is obtained at the power input of a machine tool and analyzed using a set of signal processing techniques and load-disaggregation algorithms. The proposed methodology is validated through a case study of milling process. Various classifiers used in the disaggregation algorithms are compared for their accuracies using the case study data. The results reveal that only a small portion of the total cutting energy (782.24 kJ) was used for actual material removal (40.73 kJ). The proposed study provides accurate data in user friendly format to assist designers and manufacturers for strategic and economic decision making.