Scientific Machine Learning to Simulate Complex Physics Processes in Porous Materials

King Abdullah University of Science and Technology, Saudi Arabia

Updated: 6 days ago

The design of industrial membrane separation materials requires advanced computation methods, such as computational fluid dynamics and computational chemistry, to design, analyze, and predict the properties and performance of these materials. This area also requires experimental studies by providing insights from molecular levels to continuous pore scale level. The modeling of porous membrane usually requires computationally expensive modeling approaches such as molecular dynamics simulations (MD), computational fluid dynamics (CFD) to understand how porous structures and operating conditions can impact membrane performance. Therefore, this objective of this research is develop efficient algorithms and models based on deep learning to accelerate the physics simulation for membrane relevant processes, which can be based on physics-informed neural network, data-driven models, and hybrid simulation models. These developed models can ultimately be deployed for industrial applications of membrane design and manufacturing processes.

The successful candidate will be based at Inorganic Membranes Research Group at KAUST, under the leadership of Professor Zhiping Lai, and will collaborate closely with Professor Bicheng Yan (KAUST).

Benefits:

Applications are sought for a one-year postdoc position (extendable). The position will include a competitive salary based on the candidate’s qualifications; benefits include medical and dental insurance, free furnished housing on the KAUST campus, annual travel allowance to visit home country, annual paid vacation, and other generous benefits.


Similar Positions