R. Naveiro, N. Campillo, M. Bernabei, G. Revilla-Lopez
Dispersants are the main additives in oils and lubricants that help keeping engines free of deposits. An efficient dispersant design requires tailoring the nature of the chemical interactions to meet the performance characteristics of a particular engine. Despite the wealth of knowledge available, the chemistry for production of dispersants in use today remains limited. The design of dispersants is done through trial and error, but this process is expensive and time-consuming. AI has the potential to guide the design of next generation materials, allowing economic and time savings. We describe a Bayesian machine learning approach for dispersant evaluation that can be used as part of virtual screening strategies to identify promising candidates. Moreover, Bayesian methods properly account for the existing uncertainty, becoming extremely useful for active learning. We will discuss these applications paying special attention to uncertainty quantification and model explainability issues.
Keywords: Bayesian methods, materials discovery, generative models.
Scheduled
Invited Session Math-In. Industrial Applications I
June 8, 2022 12:40 PM
Auditorium