Machine-Learning Based Energy Estimation on a High-Speed Transportation System

Reducing the energy absorption of automatic machines used in industry is one of the main goals towards the reduction of the carbon footprint, as well of the economic cost, of mass-produced goods. Incorporating energy improvements to existing machines and established technological processes can however be challenging, due to the complexity of estimating with a sufficient level of detail the actual energy consumption of a machine and even more by the difficulty of guessing the required modifications that allow to reduce such energy consumption. This work explores the possibility of using machine learning as a tool that allows estimating the energy consumption of a transportation system from a reduced set of numerical data that represent the main feature of the motion profile, in order to develop a model to be used for planning energy-efficient motion profiles. The investigation is based on experimental data gathered for a high-speed transportation device.