(463f) Parallel Plant: A Smart System Approach for Ammonia Synthesis Plant

Smart plant contains high intensive information and massive operational parameters, not only extensive chemical reactions. With the development of computation capability and future 5G, the ACP theory (i.e., artificial systems, computational experiments, and parallel computing) will play a more important role in modeling and control of complex systems like ammonia synthesis plant. The necessity of making accurate predictions of production conversion rate out of a large number of operational parameters has become a crucial problem in smart plant. Previous attempts have been made to seek reactive rate based on empirical equations in ammonia synthesis. However, there are still uncertainties about parallel plant chemical reaction mechanism using a large number of operational parameters. This work proposes a novel deep learning prediction approach that utilizes long short-term memory and gated recurrent unit in conjunction with ACP theory. The proposed approach is tested and validated by real-world dataset, and the results outperformed traditional prediction models.