Yun Wang

University of California, USA

Yun Wang received his B.S. and M.S. degrees in Mechanics and Engineering Science from Peking University in 1998 and 2001, respectively. He went to the Pennsylvania State University where he received his Ph.D degree in Mechanical Engineering in 2006. Wang joined the MAE (Mechanical and Aerospace Engineering) faculty at the University of California, Irvine in 2006. Wang has produced over 70 publications in PEM fuel cell and Li-air battery, including a book on PEM Fuel Cell Water and Thermal Management Fundamentals in 2013 and a PEM fuel cell review paper in 2011 (which has been cited over 2,150 google scholar times). He received a few awards, including the prestigious Distinguished President's Award and Outstanding Educator Award from Orange County Engineering Council, the Seasky Scholarship from Dalian University of Technology, China, and the 2011-2012 Applied Energy Certificate of Excellence: Most Downloaded Authors. Several of his seminal works are highly cited in the major fuel cell journals. His 45 journal papers published in 2005-17 have been cited over 5,100 times. Scopus (last access on 10/3/2018) shows one first-authored paper is the most cited in the history of Applied Energy since 1975 (among 12,590 papers). Wang served as Track chair/co-chair, session chair/co-chair, conference chair and committee member for many international conferences of power, thermal energy, and engineering. Wang is currently Professor at the UC Irvine and ASME fellow.

Machine Learning in Dynamics and Power Management of PEM Fuel Cell

Abstract:Polymer electrolyte membrane (PEM) fuel cell has been regarded as a potential power source for various applications due to its noteworthy features of high efficiency and zero emission. Its performance and dynamics are controlled by electrochemically coupled transport processes, including fluid flow, phase change, species transport, energy conservation, and proton/electron conduction, and are important to its practical applications. Artificial neural network (ANN), inspired by the biological neural networks that constitute animal brains, provides logical connection between input and output variables to predict output values according to given input variables. It has been successfully applied to many power and energy systems to optimize component materials, dynamic control, and system design. In this talk, I will present several fundamental aspects regarding the dynamics and power management of PEM fuel cell, including dynamic response, voltage evolution, and power management, and discuss several ANN approaches that are currently under development to advance PEM fuel cell technology and applications.

Nada Zamel

Fraunhofer Institute for Solar Energy Systems, Germany

Nada Zamel obtained her Bachelors, Masters and Doctorate degrees all in Mechanical Engineering from the University of Waterloo in 2005, 2007 and 2011, respectively. Since 2006, she has published over 40 academic articles with 30 peer-reviewed journal articles. Dr. Zamel is currently a Senior Scientist in the Fuel Cell Systems Department at Fraunhofer Institute for Solar Energy Systems, ISE and has been a member of the department since 2011. Her research interests are focused on polymer electrolyte membrane fuel cells, specifically on material development and cell characterization. She is also actively involved in the research community via conference/workshop organization and as a reviewer for many journals tailored towards renewable energy. She is also actively involved in various industrial and publicly funded projects with many being collaborative between Germany and international partners.

Production of catalyst coated membranes for low temperature PEM fuel cells

Abstract: Further advancement of polymer electrolyte membrane (PEM) fuel cells, particularly for use in the automotive industry, must be achieved as a balance between cost and functionality. The catalyst layer as the heart of the cell controls the half-cell reactions and their products. Its structure governs the various transport phenomena simultaneously taking place and affects its overall activity, stability and life time. Throughout the years, the optimization of the structure of the catalyst layer, with special attention given to the cathode, has been achieved via systematic optimization of its components. Understanding the interaction between the layer’s ingredients, its structure and performance is, thus, important to its advancement. At Fraunhofer Institute for Solar Energy Systems ISE, we have been actively working on understanding the full life cycle of the manufacturing process of a catalyst coated membrane (CCM), from the ink, to its application on the membrane, to the quality control of the entire membrane electrode assembly (MEA). This investigation is carried out using various in-situ and ex-situ analysis of every step of the manufacturing process. The question then that arises is how best to deal with the extensive data collected during all processes, and how to couple it with the corresponding process parameters. Intelligent data processing can, hence, be utilized to build this link and would help to carry out a comprehensive optimization of CCM production parameters, to improve the catalyst performance and enhance our understanding of the parameters and their interconnection.

Zhiguo Qu

Xi’an Jiaotong University

Dr. Qu is a professor of Energy & Power Engineering at Xi’an Jiaotong University. He obtained the Ph. D degree in engineering thermophysics from Xi’an Jiaotong University in 2005. He worked as a visiting scholar at Advanced Heat Transfer, LLC, USA and Pennsylvania State University in 2006 and 2013, respectively. His main research interests include multiscale transport phenomena, energy conversion and storage, micro-nanofluidics, and electronic thermal management. He has published 153 SCI indexed papers in peer-reviewed journals. These papers are cited by more than 3600 times from google scholar with h-index of 31. He serves as the editorial board member for several journals. Dr. Qu is a recipient of Young Scholars of the Yangtze River, National Young Top-notch Talent Support Program of China, China National Funds for Excellent Young Scientists, Young Scholar Fund from Fok Ying Tung Education Foundation of China, and the Chinese Ministry of Education program for New Century Excellent Talents.

Numerical and deep learning study on multi-scale problem for adsorption and diffusion processes in porous media

Abstract:The multi-scale heat and mass transfer process in porous media is a widespread phenomenon that exists pervasively in multi-scale gas adsorption for shale gas matrix and adsorbent bed. In this keynote lecture, a deep learning coupled with genetic algorithm model is established to predict the adsorption capacity based on the experimental data. The model provides more accurate prediction than those of theoretical models and BP neural network model at lower pressure range. Then, a modified lattice Boltzmann model (LBM) is developed on the pore-scale to accurately predict the effective diffusivity of heterogeneous shale matrix, where the multicomponent and irregular morphological features are fully considered. The effects of shale porosity, average gain diameter, organic matrix volume fraction and diffusivity, and irregular structures on the matrix diffusion ability are investigated. A modified empirical formula is proposed to effectively capture the heterogeneous shale matrix diffusion ability. The gas adsorption and separation on porous surface of the absorbent at different scales are solved by a multi-scale method that couples LBM with grand canonical Monte Carlo (GCMC). In interfacial boundary, saturation adsorption capacities are obtained by GCMC method to replace empirical values. Langmuir–Freundlich model and linear fitting formula are used to calculate the saturation adsorption capacities in Langmuir adsorption kinetics model and the adsorption heat in heat transfer in LBM model. Lastly, the mass transfer process of mixture gases in membranes is investigated by the above multi-scale method. The proposed coupled method can be helpful in the design of efficient membranes.

Key words: porous media, multi-scale, shale gas, deep learning, gas adsorption, GCMC, LBM