Speaker: Roger French, Case Western Reserve University

Speaker: Dr. Roger H. French

Affiliation: Case Western Reserve University

ABSTRACT: Petabyte-scale datasets can be collected using recent advances in computing and communication. Results from data-driven modeling of these datasets can challenge how things have been understood and affect society, industry, and academia’s choices. Materials science datasets amassed in the laboratory have typically been small and sparse, but our ability to use petabyte datasets from materials systems in their real-world applications enables us to shift from observational studies to predictive and inferential science. We utilize these data science and big-data analytics approaches to address critical problems in energy science. As solar power grows beyond the current 800 GW installed base, we need to fully understand and predict the power output of photovoltaic (PV) modules over their entire > 30 year lifetimes. Degradation science 1 combines data-driven statistical and machine learning modeling with physical and chemical science to infer degradation mechanisms in order to improve PV materials’ lifetime performance and reduce system failures. We use distributed and high performance computing, based on Hadoop2, the NoSQL Hbase, and Spark to ingest, analyze, and model large volumes of time-series datasets from 3.4 GW of PV power plants 2 . This permits epidemiological studies of the performance loss rate (PLR) as a function of PV module types and climate zones, and with the addition of time-series I-V curve tracers, enable remote diagnosis of PLR mechanisms activated for different climatic exposure conditions 3 . We have developed an automated image processing and deep learning pipeline applied to electroluminescent (EL) images of PV modules to identify and further understand degradation mechanisms and predict their associated power losses 4 . Data- driven analytics using data science and machine learning represent a new front in our research studies of critically important and complex systems.

Biography:

Roger H. French is the Kyocera Professor in the Case School of Engineering, Case Western Reserve University, Cleveland, Ohio. His primary appointment is in Materials Science and Engineering, and he has secondary appointments in Macromolecular Science, Electrical Engineering & Computer Science, Biomedical Engineering, and Physics. He is the faculty director of the CWRU Applied Data Science program which offers graduate classes and an undergraduate minor university-wide. He is the director of the SDLE Research Center at CWRU, an Ohio Third Frontier, Wright Project center focused on lifetime and degradation science of long lived technologies and data science and analytics. He is a participant in the Performance and Reliability of Photovoltaic Systems task 13 of the International Energy Agency, Photovoltaic Power Systems Programme. He is a member of IEC TC82-WG2,WG3 “Solar Photovoltaic Energy Systems, Modules & Systems” US-Technical Advisory Group; the Standards Technical Panel (STP) of UL, “Flat-Plate Photovoltaic Modules and Panels”; Intl. PV Module QA Task Force; and ASTM Committees on Weathering and Durability: Service Life Prediction and Photovoltaic Electric Power Conversion. He was a recent member of the U. S. Department of Energy, Basic Energy Science Advisory Committee.Prior to joining Case Western Reserve in 2010, French was a Research Fellow in Central Research and Development, DuPont Co. (starting in 1985) and Adjunct Professor of Materials Science, University of Pennsylvania (from 1994). He received his B. S. from Cornell University and his Ph.D. from Massachusetts Institute of Technology, both in Materials Science. He has published 140 journal articles, 111 proceedings papers, 13 book chapters and given 148 invited talks and 283 conference presentations. He has 28 issued U. S. patents and 4 active filings.