High-dimensional data analytics using low-dimensional models in power systems
Meng Wang, Rensselaer Polytechnic Institute, USA
Phasor Measurement Units and smart meters provide fine-grained measurements to enhance the system visibility to the operators and reduce blackouts. The spatial-temporal blocks of collected measurements have intrinsic low-dimensional structures due to the correlations governed by the underlying physical system. The central idea of the talk is to show that one can exploit the low-dimensional structure to develop fast model-free methods for information recovery with analytical guarantees.
One example is missing data recovery and error correction for synchrophasor data. The low data quality currently prevents the implementation of synchrophasor-data-based real-time monitoring and control. We developed model-free approaches to recover the PMU data even under the extreme scenarios of simultaneous and consecutive data losses and data errors across all channels for some time. By exploiting the low-dimensional structures, we formulated the data recovery problem as nonconvex optimization problems and developed fast algorithms to find the global minimum with a linear rate.
The second example is our proposed privacy-preserving data collection framework for smart meters. One can add noise and quantize the data significantly to hide the information in individual measurements. We developed computationally efficient load pattern extraction methods from highly noisy and quantized smart meter data such that the estimated load pattern is only accurate for the operator, and the information is obfuscated to a cyber intruder with partial measurements. This enables the data sharing among different parties without sacrificing the privacy.
Meng Wang is an Assistant Professor in the Department of Electrical, Computer and Systems Engineering at Rensselaer Polytechnic Institute. She received B.S. and M.S. degrees from Tsinghua University, China, in 2005 and 2007, respectively. She received the Ph.D. degree from Cornell University, Ithaca, NY, USA, in 2012.
Prior to joining RPI, she was a postdoc research scholar at Duke University. Her research areas involve machine learning and data analytics, energy systems, signal processing, and optimization. She is a recipient of Army Research Office Young Investigator Program (YIP) Award. She also received School of Engineering Research Excellence Award from Rensselaer. She is a guest editor of IEEE Journal of Selected Topics in Signal Processing Special Issue on Signal and Information Processing for Critical Infrastructures in 2018.
Writing Portable Building Analytics with the Brick Metadata Schema
Gabe Fierro, UC Berkeley, USA
Commercial buildings have long since been a primary target for applications from a number of areas: from cyber-physical systems to building energy use to improved human interactions in built environments. While technological advances have been made in these areas, such solutions rarely experience widespread adoption due to the lack of a common descriptive schema reducing the now-prohibitive cost of porting these applications and systems to different buildings.
Brick is a unified metadata schema for buildings that was born from the BuildSys community’s effort in advancing technologies for the built environment. Brick has two main components: 1. a class hierarchy describing the family of building subsystems and the entities, equipment and points therein 2. a minimal, principled set of relationships for describing the associations and connections between those entities A Brick model is a directed graph that represents an instance of a building. Building analytics applications can query Brick models for the data they need. This tutorial walks attendees through the design and usage of Brick as applied to authoring two portable analytics applications executed against real-world building data.
More information is available HERE.
Gabe Fierro is a PhD student in the Computer Science department at UC Berkeley, advised by Dr. David E. Culler. His research involves creating intelligent, sustainable infrastructure for smart buildings and cities, including work on the Brick metadata schema (https://brickschema.org) and the Mortar building data testbed (https://mortardata.org).
Data-driven Battery Profiling: Solutions and Application Case Study
Jiangchuan Liu, Simon Fraser University, Canada
Fangxin Wang, Simon Fraser University, Canada
Batteries play a crucial role in many industry fields, from large-scale energy storage systems to tiny Internet-of-Things devices. Given the complex physical structure and the chemical reactions inside a battery, understanding and predicting its current and future working conditions is challenging yet necessary for management and maintenance. In this tutorial, we will show how the recent advances of big data analytics and deep learning could be leveraged for battery state profiling, beyond traditional electrochemical modelling. We will also examine the joint optimization of battery reservoir and the power grid supply. Taking the backup batteries in cellular base stations as a case, we will demonstrate a unified learning-based framework to accurately predict the battery states and the remaining lifetime based on the historical onboard data and the external power outage events. Such data-driven profiling minimizes the reliance on the domain knowledge of specific batteries and supports large-scale deployment with reduced cost and service interruption.
Jiangchuan Liu is a University Professor in the Schoolof Computing Science, Simon Fraser University, British Columbia, Canada. He is an IEEE Fellow and an NSERC E.W.R. Steacie Memorial Fellow. He is an EMC-Endowed Visiting Chair Professor of Tsinghua University, Beijing, China and an Adjunct Professor of Tsinghua-Berkeley Shenzhen Institute. In the past he worked as an Assistant Professor at The Chinese University of Hong Kong and as a research fellow at Microsoft Research Asia.
He received the BEng degree (cum laude) from Tsinghua University, Beijing, China, in 1999, and the PhD degree from The Hong Kong University of Science and Technology in 2003, both in computer science. He is a co-recipient of the inaugural Test of Time Paper Award of IEEE INFOCOM (2015), ACM SIGMM TOMCCAP Nicolas D. Georganas Best Paper Award (2013), and ACM Multimedia Best Paper Award (2012).
His research interests include multimedia systems and networks, cloud computing, social networking, online gaming, big data computing, RFID, and Internet of things. He has served on the editorial boards of IEEE/ACM Transactions on Networking, IEEE Transactions on Big Data, IEEE Transactions on Multimedia, IEEE Communications Surveys and Tutorials, and IEEE Internet of Things Journal. He is a Steering Committee member of IEEE Transactions on Mobile Computing and Steering Committee Chair of IEEE/ACM IWQoS (2015-2017).
Fangxin Wang received the B.E. degree from Beijing University of Post and Telecommunication, Beijing, China in 2013 and the M.E. degree from Tsinghua University, Beijing, China in 2016. He is currently pursuing the Ph.D. degree in the School of Computing Science, Simon Fraser University, Burnaby, Canada. His research interests include Internet of Things, wireless networks, big data analytics and edge computing.
Open Data Driven Analytics and Modeling for Energy and Occupant Sensing Data
Mikkel Baun Kjærgaard, University of Southern Denmark, Denmark
Mobile, wearable and Internet of Things devices and the availability of low-cost sensors open up new possibilities for mapping human behavior and energy consumption via objective sensor data. This trend enables new data-driven analysis and modeling methods with applications in demand-side energy management and smart grid optimization. However, to gain the full potential of using resources on mapping human behavior and energy consumption requires that data is not only collected for one-time use. The open data paradigm prescribes a method for going beyond one-time data collection increasing the value of the data in individual organizations and the society in general. However, the open data paradigm also raises needs for 1) proper privacy protection of shared data and 2) tool support for doing data-driven analytics and modeling based on heterogeneous sensor data from multiple sources.
Type of tutorial: Mix of lecture and hands-on
Outline of tutorials:
- [30 minutes] Lecture on challenges and methods for open data driven analytics and modeling
- [30 minutes] Tutorial on tool support and existing open datasets
- [30 minutes] Interactive session with participants working with tools and open datasets
- [30 minutes] Discussion among participants of experiences, limitations and opportunities
Mikkel Baun Kjærgaard is a Professor in Software Engineering at the Center for Energy Informatics at University of Southern Denmark. He conducts research within the area of energy informatics with a specific focus on occupancy sensing, software support for data processing, modeling and analytics and energy applications in buildings and smart grids. He conducts his research with an experimental foundation at the intersection of ubiquitous computing, machine learning and systems research.
Student Travel Grant Support
ACM e-Energy owes its success, in great part, to the generous backing of our academic and corporate partners. If you are interested in sponsoring ACM e-Energy 2019, please contact the General co-Chairs: Xiaojun Lin and Steven Low.