July 27, 2017
Five research teams from the University of Michigan and Shanghai Jiao Tong University in China are sharing $1 million to study data science and its impact on air quality, galaxy clusters, lightweight metals, financial trading and renewable energy.
Since 2009, the two universities have collaborated on a number of research projects that address challenges and opportunities in energy, biomedicine, nanotechnology and data science.
In the latest round of annual grants, the winning projects focus on data science and how it can be applied to chemistry and physics of the universe, as well as finance and economics.
The Michigan Institute for Data Science, U-M's interdisciplinary hub for data science research and education, helped catalyze these new collaborations and will connect them to expertise and resources across campus that can support and augment their work.
"Issues such as health and sustainability impact society on a global scale, so it is important for us to collaborate with researchers beyond our campus to explore new directions," said Volker Sick, associate vice president for research — natural sciences and engineering.
The program funds projects that have commercial potential and are likely to attract follow-on research funding from the U.S. and Chinese governments, as well as industry.
This year's projects include:
Learning Impact of Air Pollution on Kidney Disease Epidemiology in U.S. and East China Using Big Public Health Data
Principal Investigators: Peter Song, professor of biostatistics, U-M; Zhangsheng Yu, professor, Department of Bioinformatics and Biostatistics, SJTU
Summary: Incidence rate of end-stage renal disease (ESRD) and the mortality rate among ESRD patients are high in both China and the U.S. At the same time, air quality in China has been worsening and often reaches or exceeds hazardous levels in many regions. Researchers will assess and contrast the effect of the exposure level to the air pollution on the ESRD patients' mortality and morbidity between the two countries to provide data-driven analytics for public health policymaking.
A weather-process and machine learning combined approach to improve solar forecast for PV power generation
Principal Investigators: Xianglei Huang, associate professor of climate and space sciences and engineering, U-M; Ruzhu Wang, professor, School of Mechanical Engineering, SJTU
Summary: Photovoltaic (PV) energy is a major potential source of renewable energy for future massive integration with the electricity grid. Since variations of solar irradiation over time directly affect the PV output from solar panels, accurate forecasting of the solar irradiance impinging on solar panels is critical to efficient grid integration and power management. Researchers will work on a hybrid approach for the intra-day forecast for PV production using data-driven algorithms guided by known physics and checked against the observations, as well as the ensembles of simulations from state-of-the-art weather research and forecast model.
Constraining Cosmological Parameters with Galaxy Clusters: A New Meta-analysis Approach to Cosmology
Principal Investigators: Christopher Miller, associate professor of astronomy, professor of physics, U-M; Wentao Luo, postdoctoral researcher, Department of Astronomy, SJTU
Summary: Researchers have evidence that about 7 billion years after the Big Bang, the expansion of the universe stopped slowing down and then started to speed up. This team will utilize modern data science techniques to combine some of the largest astronomical sky surveys into a single novel analysis in order to shed light on what started this acceleration, how it evolved and how it will end.
Automatic construction of a causality knowledge base from large online financial text
Principal Investigators: H.V. Jagadish, professor of electrical engineering and computer science, U-M; Kenny Qili Zhu, Department of Computer Science and Engineering, SJTU
Summary: Perceptions of the relationships between causes and effects plays a critical role in people's daily behavior and decision-making, and are of great interest in many domains, including finance, where understanding causal relationships can provide significant opportunities for economic benefits. This project aims to build a financial causality knowledge base by analyzing a large corpus of online text trying to capture the causal strength of different finance-related events. The rules in this knowledge base can be used to predict financial events and generate alerts in financial trading.
Development of a Data-based Alloys Design Methodology and Application to Magnesium
Principal Investigators: Katsuyo Thornton, professor of materials science and engineering, U-M; Leyun Wang, professor, School of Materials Science and Engineering, SJTU
Summary: The goal of this project is to develop a data‐science approach for alloy design and apply it to magnesium alloys. Researchers will build an artificial neural network model to establish the relationships between material composition, microstructural features and mechanical properties. They will train the model first for commercial and other well‐characterized magnesium alloys, and then extend it to a wider range of composition and processing histories. To this end, researchers will generate physics‐based simulation data to inform the model regarding the effects of various alloying elements and processing conditions, along with their prediction uncertainties. Finally, a few new magnesium alloys with superior mechanical properties will be identified for potential commercialization using the model.