ACM e-Energy · 2026

Energy Injection Identification enabled Disaggregation with Deep Multi-Task Learning

Xudong Wang, Guoming Tang, Junyu Xue, Srinivasan Keshav, Tongxin Li, Chris Ding

ACM International Conference on Future and Sustainable Energy Systems

Read paper Repository BibTeX
Laboratory aggregate power traces with and without energy injection above the true on and off states of six appliances.

ACM e-Energy · 2026

Seeing demand behind energy injections

Controlled energy injection changes the meter signal while appliances continue operating. These laboratory measurements show why appliance recognition and injection identification need to be considered together.

Energy Injection Identification enabled Disaggregation with Deep Multi-Task Learning Figure 1 from the paper

Can appliances be identified when solar and storage obscure meter readings?

DualNILM jointly recognizes appliance states and identifies energy injected behind the meter. Its transformer architecture combines temporal learning tasks to separate consumption from injections, with evaluation on measured and synthesized datasets.

Research themes

Connections

Recovering structure from limited measurements

Graph recovery estimates electrical network parameters. Charging curve clustering and energy disaggregation recover useful structure from observed behavior and aggregate meter readings.

Cite this paper

Xudong Wang, Guoming Tang, Junyu Xue, Srinivasan Keshav, Tongxin Li, Chris Ding. Energy Injection Identification enabled Disaggregation with Deep Multi-Task Learning. ACM International Conference on Future and Sustainable Energy Systems, 2026. https://doi.org/10.1145/3744255.3798113

BibTeX
@inproceedings{tongxin-dualnilm,
  title = {{Energy Injection Identification enabled Disaggregation with Deep Multi-Task Learning}},
  author = {Xudong Wang and Guoming Tang and Junyu Xue and Srinivasan Keshav and Tongxin Li and Chris Ding},
  year = {2026},
  booktitle = {ACM International Conference on Future and Sustainable Energy Systems},
  url = {https://doi.org/10.1145/3744255.3798113},
  doi = {10.1145/3744255.3798113}
}