Energy Disaggregation via Learning Powerlets and Sparse Coding

Authors: Ehsan Elhamifar, Shankar Sastry

AAAI 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Real experiments on a publicly available energy dataset demonstrate that our proposed algorithm achieves promising results for energy disaggregation.
Researcher Affiliation Academia Ehsan Elhamifar and Shankar Sastry Electrical Engineering and Computer Sciences Department University of California, Berkeley
Pseudocode No The paper describes the proposed algorithms and frameworks using mathematical equations and textual explanations, but it does not provide any structured pseudocode or algorithm blocks.
Open Source Code No The paper does not contain any statement about making its source code publicly available, nor does it provide a link to a code repository.
Open Datasets Yes We evaluate our proposed energy disaggregation framework on the real-world REDD dataset (Kolter and Johnson 2011), a large publicly available dataset for electricity disaggregation.
Dataset Splits No The paper states 'we use the first week of recorded electricity signals for training and use the rest of data for testing', but does not mention a distinct validation set or specific split percentages for training, validation, and test.
Hardware Specification No The paper does not provide specific hardware details such as CPU/GPU models, memory, or cloud instance types used for experiments.
Software Dependencies No The paper mentions using 'the standard integer programming solver of MOSEK' but does not specify its version number or any other software dependencies with versions.
Experiment Setup Yes We choose a window size of w = 15. We model each device as a mixture of dynamical systems of the form (8), where we set the model order to be m = 3. We use our subset selection scheme in (7) to extract about 20 powerlets for each device. In order to perform disaggregation using the optimization program (19), we set λ = 30 and η = η = 1 and use the prior that kitchen appliances typically work together.