Deep generative fuel design in low data regimes via multi-objective imitation. (15th June 2023)

• Record Type: Journal Article
• Authors: Liu, Yifan; Liu, Runze; Duan, Jinyu; Wang, Li; Zhang, Xiangwen; Li, Guozhu
• Is Part Of: Chemical engineering science. Volume 274(2023)
• Publication Date: 2023-06-15
• Subjects: Deep learning -- Generative model -- Fuel -- Variational autoencoder -- Generative adversarial network -- Ensemble learning; Chemical engineering -- Periodicals; Génie chimique -- Périodiques; Chemical engineering; Periodicals; Electronic journals; 660
• DOI: 10.1016/j.ces.2023.118686
• Languages: English
• ISSNs: 0009-2509

Abstract

Graphical abstract: Highlights: A deep generative model is established to design desired fuel molecules. Fuel-relevant chemical space is enriched automatically in low data regimes. Multi-objective imitation on target fuel is realized by in-depth optimization. Abstract: Commercial fuel discovery faces a constantly decreasing return of investment due to due to increasingly tight environmental criteria and reducing potential uses for each new fuel. In this paper, a deep generative model, termed Latent Interspace Generative Adversarial Network with a Domain of Stacking (LIGANDS), has been established to screen desired fuel molecules in the large chemical space without setting design rules manually. A variational autoencoder, a generative adversarial network and a stacking model are well integrated in LIGANDS through model convergence. Given only the structures of 255 typical high-energy–density fuels in low data regimes, LIGANDS generated 3461 new fuel molecules with similar property distribution and improved energy performance as the qualified candidates of next-generation fuels. To expand and enrich the fuel-relevant chemical space with innovative molecular entities on demand, in-depth multi-objective imitation on the key properties of target fuel is realized by LIGANDS through optimizing generative molecular structures and their distribution.