With the ongoing digitalization of industrial production, innovative ways of creating energy transparency on the shop floor are emerging. Virtual energy metering points, which use process data to predict the energy and resource demand, enable a cost-effective increase in energy transparency on machine level. In this paper, an approach based on offline trained neural networks is presented, through which the energy and resource consumption is continuously predicted for various production systems on machine and component level with high accuracy. Also the necessary data availability and the transferability to processes that are not included in the training dataset are discussed.