Adaptive predictive control for peripheral equipment management to enhance energy efficiency in smart manufacturing systems

Abstract

The importance of implementing energy efficiency methodologies in industrial environments has increased considerably in the last decade given the high energy costs and environmental impact (e.g., greenhouse gas emissions). This paper proposes a methodology to improve the energy efficiency of an industrial machine, without sacrificing either production or quality, using an adaptive predictive controller based on dynamic energy models that manages peripheral devices to activate/deactivate them at the proper times. The proposed adaptive mechanism aggregates robustness to the control system in industrial environments, which experiment constantly changes related to equipment degradation and that affect their energy consumption profile over time. Thus, this novel adaptive mechanism automatically updates the energy model to minimize the error between prediction and real energy consumption, including new energy behavior resulting from machine degradation. This methodology has been validated via a testbed and its performance was compared with rule-based control, which is the most widely used control strategy in industry. The energy efficiency of both approaches was evaluated using performance indicators, which show the effectiveness of the proposed control approach, highlighting remarkable improvements in reducing both energy consumption (about 2%) and sudden power peaks (more than 11%).