Cepsa optimizes its chemical processes with artificial intelligence-based technology
Cepsa has successfully completed its first digital transformation project in its Chemical Unit, applying specific industry 4.0 technologies and artificial intelligence, such as machine learning, big data and advanced analytics, to the production processes.
Specifically, this first project consisted of developing and implementing two real-time optimizers in the phenol production process, which apply machine-learning techniques and predictive models allowing recommendations to be offered every 15 minutes to operators in the plant control rooms in order to maximize production.
how did we do it?
To implement this project, it was necessary to mine and analyze over 3,000 process variables from different information sources (processes, climate conditions, lab data and more), build predictive behavior models based on the relationship of these variables, and to program the optimizers to offer the best recommendations regarding optimal operating values.
This first project was developed on phenol line 3 at the Palos Chemical Plant in Huelva, where it increased production by 2.5%, which will help to produce over 5,500 tons of phenol per year.