Dear NILM researchers,

On behalf of the organizing committee, we would like to invite you to participate in the 5th International Workshop on Non-Intrusive Load Monitoring (NILM). The workshop will be held on November 18 (2020), in conjunction with the ACM International Conference on Systems for Energy-Efficient Buildings, Cities and Transportation (BuildSys).

NILM (or disaggregation) is a growing research field which began in 1985 with a report written by George W. Hart (MIT) for Electric Power Research Institute (EPRI). NILM is used to discern what electrical loads (e.g., appliances) are running within a home/building using only the aggregate power meter. Why? To help occupants understand how they and their appliance use energy so that they could conserve to either save money, the environment, or both.

The mission of this workshop is to serve as a forum for bringing together all the researchers, practitioners, and students that are working on the topic of energy disaggregation around the world.

We are looking forward to welcoming you at this next gathering of our community!


  • General Co-Chairs

    David Irwin, University of Massachusetts, Amherst
    Stephen Makonin, Simon Fraser University
    Niki Davies, Lift Up Leaders

  • Organizing Committee

    Niki Davies, Lift Up Leaders
    Peter Davies, Austin Consultants
    Dimitrios Doukas, NET2GRID
    David Irwin, University of Massachusetts, Amherst
    Stephen Makonin, Simon Fraser University
    Oliver Parson, Bulb
    Lina Stankovic, University of Strathclyde

  • Program Committee Chairs

    Mario Bergés, Carnegie Mellon University
    Lina Stankovic, University of Strathclyde

  • Technical Program Committee

    Jose M. Alcala, Informetis Europe LTD.
    Kyri Baker, University of Colorado
    Peter Davies, Austin Consultants
    Dimitrios Doukas, NET2GRID
    Manoj Gulati, IIT-Delhi
    Xin Jin, National Renewable Energy Laboratory
    Oliver Parson, Bulb
    Katie Russell, OVO Energy
    Lina Stankovic, University of Strathclyde
    Nipun Batra, IIT Gandhinagar
    Sean Barker, Bowdoin College
    Kyle Bradbury, Duke University
    Wenpeng Luan, Tianjin University
    Lucas Pereira, University of Lisbon
    Christoph Klemenjak, University of Klagenfurt
    Alejandro Rodriguez-Silva, Simon Fraser University
    Angshul Majumdar, IIT-Delhi
    Srinivasan Iyengar, Microsoft Research - India


  • Abstract Registration: August 21 28, 2020 AoE
  • Paper Submission Due: August 21 28, 2020 AoE
  • Notification of Acceptance: September 18 23, 2020 AoE
  • Final Paper Due: October 2 11, 2020 AoE
  • Presentation Video Submission: November 8, 2020 5pm UTC



Applications of Non-intrusive Load Monitoring to Power Systems and New NILM-type Problems


Time (UTC)TopicPresentation Details
10:30 Stream begins
11:00Welcome Address by Oliver Parson and Lina Stankovic
11:30 Technical session 1 Annoticity: A Smart Annotation Tool and Data Browser for Electricity Datasets
by Benjamin Völker, Marc Pfeifer, Philipp M. Scholl, Bernd Becker (University of Freiburg)
On the Impact of the Sequence Length on Sequence-to-Sequence and Sequence-to-Point Learning for NILM
by Andreas Reinhardt, Mazen Bouchur (TU Clausthal)
Explainable NILM Networks
by David Murray, Lina Stankovic, Vladimir Stankovic (University of Strathclyde)
BERT4NILM: A Bidirectional Transformer Model for Non-Intrusive Load Monitoring
by Zhenrui Yue, Camilo Requena, Daniel Jorde, Hans-Arno Jacobsen (Technical University of Munich)
13:30 Keynote Applications of Non-intrusive Load Monitoring to Power Systems and New NILM-type Problems
Johanna Mathieu (University of Michigan)
14:30 Short talks Lightweight Non-Intrusive Load Monitoring Employing Pruned Sequence-to-Point Learning
by Jack Barber (University of Lincoln), Heriberto Cuayahuitl (University of Lincoln), Mingjun Zhong (University of Aberdeen), Wenpeng Luan (Tianjin University)
Edge computed NILM: a phone-based implementation using MobileNet compressed by TensorFlow Lite
by Shamim Ahmed, Marc Bons (FLUDIA)
Performance Analysis of Similar Appliances Identification using NILM Technique under Different Data Sampling Rates
by R. Gopinath, Mukesh Kumar, K. J. Lokesh, Kota Srinivas (CSIR- Central Scientific Instruments Organisation, Academy of Scientific and Innovative Research)
UNet-NILM: A Deep Neural Network for Multi-tasks Appliances state detection and power estimation in NILM
by Anthony Faustine (University College Dublin), Lucas Pereira (Tecnico Lisboa), Hafsa Bousbiat (University of Klagenfurt), Shridhar Kulkarn (Trinity College Dublin)
NILM based Energy Disaggregation Algorithm for Dairy Farms
by Akhilesh Yadav, Anuj Sinha, Abdessamad Saidi, Wilfried Zörner (Technische Hochschule Ingolstadt)
Finite Precision Analysis for an FPGA-based NILM Event-Detector
by Rubén Nieto, Laura de Diego-Otón, Álvaro Hernández, Jesús Ureña (University of Alcalá)
Energy Disaggregation for Small and Medium Businesses and their Operational Characteristics
by Abhinav Srivastava, Paras Tehria, Basant K. Pandey (Bidgely Technologies)
Solar Disaggregation: State of the Art and Open Challenges
by Xinlei Chen, Omid Ardakanian (University of Alberta)
An Open Problem: Energy Data Super-Resolution
by Rithwik Kukunuri, Nipun Batra (IIT Gandhinagar), Hongning Wang (University of Virginia)
Evaluation of low-complexity supervised and unsupervised NILM methods and pre-processing for detection of multistate white goods
by Mohammad Khazaei, Lina Stankovic, Vladimir Stankovic (University of Strathclyde)
Matrix Factorization for High Frequency Non Intrusive Load Monitoring: Definitions and Algorithms.
by Simon Henriet, Benoit Fuentes, Umut Simsekli, Gael Richard (Institut Polytechnique de Paris)
Bayesian model of electrical heating disaggregation
by Alexander Belikov, Laetitia Leduc, François Culière (Hello Watt)
On the Relationship between Seasons of the Year and Disaggregation Performance
by João Gois (LARSyS, ARDITI), Christoph Klemenjak (University of Klagenfurt)
15:15 Industry demos eco-bot
David Murray (University of Strathclyde)
Verv Connect Inline Adapter and Smart Isolator
Peter Davies (Verv)
IEC International Standard on NILM
Josh Honda (Informetis)
15:30 Technical session 2 Non-Intrusive Load Monitoring of Water Heaters Using Low-Resolution Data
by Christy Green, Srinivas Garimella (Georgia Institute of Technology)
Stop! Exploring Bayesian Surprise to Better Train NILM
by Richard Jones (Simon Fraser University), Christoph Klemenjak (University of Klagenfurt), Stephen Makonin (Simon Fraser University), Ivan V. Bajic (Simon Fraser University)
Virtual metering of heat supplied by hydronic perimeter heaters in variable air volume zones
by Darwish Darwazeh, Jean Duquette, Burak Gunay (Carleton University)
Phased: Phase-Aware Submodularity-Based Energy Disaggregation
by Faisal M. Almutairi (University of Minnesota), Aritra Konar (University of Virginia), Ahmed S. Zamzam (National Renewable Energy Laboratory), Nicholas D. Sidiropoulos (University of Virginia)
16:30Stream ends


Applications of Non-intrusive Load Monitoring to Power Systems and New NILM-type Problems

Johanna Mathieu, University of Michigan

Though power systems researchers still do not agree what the term “smart grid” really means, most would probably agree that smart grids have enhanced metering, telemetry, and control infrastructure to increase situational awareness, hopefully leading to improved operations and real-time control (though you will not find that definition on Wikipedia!). Converting a grid into a smart grid has generally entailed deployments of network sensors and advanced metering infrastructure (i.e., smart meters). However, I believe that the really exciting stuff is behind-the-meter: flexible electric loads, storage, and photovoltaic generation, all of which can be coordinated to provide a variety of services to distribution systems and power markets. Here, of course, is where the fields of power systems and non-intrusive load monitoring (NILM) intersect. How can power systems leverage NILM to “see” behind-the-meter? What specific applications exist now, or might exist in a future “smarter” grid? What power system challenges might inspire new NILM-type problems? In this talk, I will discuss how NILM can be leveraged for residential demand response – both characterizing the resource and for real-time load coordination. I will also describe our work on “feeder-level energy disaggregation,” a problem that is inherently related to NILM.

Johanna Mathieu is an associate professor of Electrical Engineering and Computer Science at the University of Michigan, Ann Arbor. Prior to joining Michigan in 2014, she was a postdoctoral researcher at ETH Zurich, Switzerland. She completed her PhD at the University of California at Berkeley in 2012. She is the recipient of an NSF CAREER Award and the Ernest and Bettine Kuh Distinguished Faculty Award. Her research focuses on ways to reduce the environmental impact, cost, and inefficiency of electric power systems via new operational and control strategies. She is particularly interested in developing new methods to actively engage distributed flexible resources such as energy storage, electric loads, and distributed renewable resources in power system operation, which are especially important in power systems with high penetrations of intermittent renewable energy resources such as wind and solar.