A Learning-Based Decision Tool towards Smart Energy Optimization in the Manufacturing Process

We developed a self-optimizing decision system that dynamically minimizes the overall energy consumption of an industrial process. Our model is based on deep reinforcement learning (DRL) framework, adopting three methods, namely: Q-network (DQN), proximal policy optimization (PPO), and advantage actor–critic (A2C) algorithms, combined with self-predicting random forest model. This smart physics-informed DRL sets key input parameters to optimize while ensuring product quality desired output parameters. The self-improving can increase its performances without further human assistance. applied approach process heating tempered glass. Indeed, identification control glass challenging task requiring expertise. In addition, optimizing dealing this issue great value-added. evaluation under configurations has been performed consequently, outcomes conclusions have explained in paper. intelligent provides optimized set for within acceptance limits minimizing consumption. work necessary foundations address issues related parameterization from theory practice providing real application; research opens new horizon towards sustainable manufacturing.