Pith. sign in

REVIEW 5 major objections 6 minor 57 references

Unsupervised algorithm for disaggregating low-sampling-rate electricity consumption of households

T0 review · 5 major / 6 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read An unsupervised algorithm called device usage estimation splits a household's 15-minute power readings into the energy shares of eight appliance categories, with errors usually below 20 percent.

desk verdict Novel unsupervised NILM hybrid with an honest three-dataset benchmark, but the synthetic whole-house signal and post-hoc tuning leave the real-smart-meter claim unproven. read the letter →

arxiv 1908.10713 v1 pith:HVZ6JULI submitted 2019-08-20 cs.LG eess.SP

Jordan Holweger , Marina Dorokhova , Lionel Bloch , Christophe Ballif , Nicolas Wyrsch This is my paper · ORCID
classification cs.LGeess.SP
keywords non-intrusiveloadmonitoringenergydisaggregationunsupervisedlearningMarkovmodelsmartmetersdemand-sidemanagementlowsamplingrateprofilesimulation
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper proposes device usage estimation (DUE), an unsupervised disaggregation method that takes a whole-house power signal sampled every 15 minutes plus household metadata—ages and employment state, heating type, appliance inventory—and returns the share of energy consumed by eight appliance categories over time. It builds a Markov-chain activity sequence for each adult inhabitant from the statistics of a time-use survey, then converts those activities into candidate appliance runs using nominal powers and usage probabilities, and subtracts them from the measured load until the energy budget is matched. The claim is that this hybrid of load simulation and disaggregation reaches category-level energy-share uncertainties normally under 20 percent, without needing any labeled data and with computational cost lower than most supervised benchmarks. The purpose is demand-side management: a utility could estimate the flexibility of many households and give tailored advice from ordinary smart-meter readings and a short survey, instead of installing sub-metering equipment.

What carries the argument

The carrying mechanism is the activity chain, a Markov chain over 34 activity states whose transition matrix and initial distribution depend on the type of day, the employment state, and the age group of the inhabitant; these probabilities come from the Netherlands time-of-use survey. For each person over 10, the algorithm samples a random chain, maps activity episodes to appliance runs through a household appliance inventory with nominal powers and usage probabilities, and repeatedly re-samples until the simulated load matches the measured load within a tolerance. The category recognition is sequential—heating, lighting, cooking, housekeeping, entertainment, ICT—after standby and fridge are filtered out as the day's minimum and the periodic night-time refrigeration pattern respectively.

What would settle it

Sub-meter every appliance in a set of households for at least four weeks, downsample to 15 minutes, run DUE with complete appliance inventories, and compare predicted category energy shares with the submetered truth; regularly exceeding 20 percent mean absolute energy-share error, or failing to match activity timing against resident diaries, would falsify the central claim.

Watch

Extended reading notes

Core claim

The central claim is that disaggregating a low-sampling-rate household load into appliance categories does not require supervised training on labeled appliance data. A generative Markov model of human activity, parameterized by a generic time-use survey and by each household's composition, plus a simulation of appliance power draws, can be fitted to the measured whole-house curve well enough that the energy share per category is typically within 20 percent of the ground truth. The paper demonstrates this on three public datasets down-sampled to 15-minute resolution and benchmarks it against four standard NILM algorithms; while the supervised algorithms score higher on average, the unsupervised DUE stays in a similar uncertainty range and needs no training data.

Load-bearing premise

The load-bearing premise is that Dutch citizens' daily activity statistics transfer to central European households; if that is false, the generated activity chains that drive the entire category allocation are disconnected from the households being disaggregated.

Editorial extensions

If this is right

  • A utility could offer appliance-category-level feedback and demand-side management advice using only existing 15-minute smart meter readings plus a short household survey, without sub-metering.
  • The approach transfers to a new country or region by substituting a local time-use survey, since the activity statistics are the only behavior-specific input.
  • On the tested datasets the category energy-share uncertainty remains normally under 20 percent, the same range as the supervised benchmarks despite using no labels.
  • Execution time grows linearly with the length of the dataset, so long monitoring periods and large household pools are affordable compared with methods whose cost grows faster.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The paper's own Dutch-survey caveat implies a direct empirical test: re-estimate the Markov activity statistics from a central European time-use survey and check whether the category-share errors shrink; the paper does not run this test.
  • The current all-inhabitant occupancy rule (any peak means everyone is home) is a likely error source; using occupancy sensors or per-person diaries would probably tighten the estimates.
  • Because the benchmark datasets monitor only a subset of household appliances, some of the reported error likely reflects the mismatch between the algorithm's assumed full appliance inventory and the dataset's partial coverage, rather than the method's intrinsic accuracy.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 6 minor

Summary. The paper proposes an unsupervised, hybrid load-simulation/disaggregation algorithm called device usage estimation (DUE). The method takes as inputs a 15-minute aggregate power measurement, household characteristics (number and age of inhabitants, employment state, appliance inventory, and usage habits), and activity statistics from the 2005 Dutch time-use survey. It builds Markov activity chains for each inhabitant, simulates appliance use category by category (Cooking, Entertainment, Fridge, Heating, Housekeeping, ICT, Light, Standby), and iteratively adjusts the simulated signal to match the measured aggregate. The output is the share of energy consumed by each category over the test period. The authors benchmark DUE against four NILM algorithms (CO, FHMM, DDSC, GSP) on three public datasets (ECO, SMARTENERGY.KOM, UK-DALE), downsampled to 15 minutes, and report energy share errors, estimation accuracy, and execution times. They conclude that DUE is unsupervised, has lower computational cost than most supervised benchmarks, and achieves prediction uncertainties that are normally under 20%.

Significance. If the claims are substantiated, this is a useful contribution to low-sampling-rate NILM. The person-centric activity-chain approach is novel, the target problem (category-level energy shares for demand-side management) is practically relevant, and the use of three public datasets with several baselines is a strength. The paper is clearly structured and provides enough algorithmic detail in the appendices to be implementable. The main value would be a genuinely unsupervised method that works at a 15-minute sampling rate, where most unsupervised NILM methods require higher-frequency data. However, the current experimental design does not fully establish the headline claim: the test signal is a synthetic sum of the very appliance channels the algorithm is supposed to recover, the algorithm is stochastic but evaluated with single runs, and several parameters are tuned on the test data. These gaps currently limit the strength of the significance claim.

major comments (5)
  1. [Section 2.2.2 and Section 3.3] The experimental setup does not test the claimed use case. Section 2.2.2 states: 'The reference whole-house power measurement is defined as the sum of the sub-measurements in order to ensure energy conservation.' All three datasets are downsampled appliance-level recordings, so the 'whole-house' signal is a synthetic aggregate of the same category-level signals that DUE is meant to recover. Real 15-minute smart meter data includes unmonitored loads, always-on devices, spikes, and noise; Section 3.3 acknowledges that partial coverage 'might have a significant impact on the performance of the algorithm.' Because this mismatch is not tested, the central claim that DUE disaggregates real smart meter signals with uncertainties normally under 20% is not supported by the reported experiments. A concrete fix is to run the same benchmark on the actual whole-house channels available in UK-DALE and ECO, or to add a controlled experiment with artificial unmonitored loads.
  2. [Section 2.1.3 and Appendix A, Eq. (35)] The light-scaling factor and the fridge recalibration are post-hoc adjustments made after observing the algorithm's errors. The text in Section 2.1.3 states that the factor was added 'after noticing that our method over-estimated the light consumption,' and Appendix A, Eq. (35) recalculates the fridge nominal power from the night-time aggregate measurement (Pfridge_new = Pfridge_old * fridge_mean_power / (Pfridge_old * beta2)). These are free parameters fitted on the test data itself. Because the paper reports no sensitivity analysis and no out-of-sample validation of these choices, the reported uncertainties may be optimistic. The authors should either perform a cross-validation-style analysis of these parameters or clearly distinguish 'unsupervised' from 'parameter-free' and report the sensitivity of the results to these adjustments.
  3. [Section 3.1, Table 5] The abstract and conclusion state that prediction uncertainties 'normally remain under 20%,' but Table 5 shows several DUE energy share errors well above 20%: Entertainment errors are 33% on SMARTENERGY.KOM and 37% on UK-DALE, Light is 22% on ECO, and Housekeeping is -17% where measured. If 'uncertainty' refers to a different aggregate metric (for example, the average absolute error across categories in Figure 6), that definition should be given explicitly and the per-dataset values should be reported. As written, the claim is not supported by the data in the table.
  4. [Section 2.1 and Appendix B] The DUE algorithm is stochastic: activity chains are generated with random numbers (Appendix B), device durations are drawn as tau*2*epsilon, the device ordering is randomized (Section 2.1.2), and the activation variables gamma are random. Yet Tables 5 and 6 report single values for each metric without variance or confidence intervals. Without repeated runs, the reader cannot judge whether the reported performance is stable, and the 'uncertainty' of the method itself is not characterized. The authors should report means and standard deviations over at least 10-20 independent runs per household and dataset, and compare those distributions with the deterministic baselines.
  5. [Section 2.1] The transferability of the Dutch time-use survey to central European households is acknowledged as questionable but is not validated. The sentence 'Although it is questionable if the behavior of Dutch citizens is representative of the behavior of all central Europeans, it seems like a reasonable assumption' is a single unsupported assertion. Since the activity chains drive the entire category allocation, an incorrect activity model would compromise the method's output. A concrete test is to repeat the experiments with the UK time-use survey (which the authors themselves cite in Section 2.1) or to compare the predicted activity/occupancy patterns with the actual occupancy information available in the ECO and SMARTENERGY.KOM datasets.
minor comments (6)
  1. [Abstract and Conclusion] The phrase 'similar range of uncertainty while saving on the cost of acquiring labeled data' is stronger than the evidence: Table 5 shows several DUE errors above 20% and some negative Estimation Accuracy values. The claims should be tempered to match the reported results.
  2. [Equations (2) and (3)] The notation in the Kronecker delta conditions is confusing and may contain typos; for example, the condition 't0_i = 22:22' in Eq. (2) is likely meant to be a specific time, but it is not explained, and the use of backslashes instead of logical 'and' is nonstandard. The equations should be rewritten for clarity.
  3. [Section 2.1.2 and Appendix B] The random number generator is not seeded and the algorithm is not accompanied by release code. For reproducibility, the authors should provide the code or, at minimum, describe the random seed policy and report variability across seeds.
  4. [Table 3] The footnotes in Table 3 are difficult to parse; several footnotes are mixed in the 'Note' column and some references (e.g., 'footnote 3,' 'footnote 4') are ambiguous. The table would benefit from clear, separately numbered notes and a concise explanation of the beta and tau parameters.
  5. [Figure 6] The legend in Figure 6 appears to contain a typo (it reads 'DUP' rather than 'DUE' in the provided text), and the y-axis is cropped without indication of which values fall outside the range. Please verify the legend and consider annotating clipped bars.
  6. [References] There are several typographical errors in the references (e.g., 'Methodies' instead of 'Methodologies' in [32]) and inconsistent page ranges. A careful proofreading pass is needed.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the DUE category estimates are derived from the aggregate via transparent estimation rules, benchmarked against independent sub-meter ground truth; the synthetic whole-house input is a limitation, not a circular reduction.

full rationale

The paper's derivation chain is not circular. The algorithm uses household metadata and a Dutch time-use survey to generate activity chains, then allocates appliance power signals by checking energy and power budgets against the measured aggregate; these are estimation steps, not definitions of the target quantities. The standby estimator (P_standby = min_t P) and the fridge night-template recalibration (Eq. 15) are openly presented as preprocessing/filtering rules and are not claimed as independent predictions of those categories from first principles. The reported performance is evaluated against sub-meter ground-truth categories from ECO, SMARTENERGY.KOM, and UK-DALE, not against the aggregate used as input. The main validity caveat, also stated in Section 3.3, is that the 'whole-house' test signal is defined as the sum of monitored sub-measurements, so partial appliance coverage may not match real smart-meter signals; this weakens external validity but does not make the outputs equal to the inputs by construction. There are no load-bearing self-citations, no imported uniqueness theorems, and no fitted parameter is renamed as a prediction. Accordingly, the circularity score is 0.

Assumptions & free parameters 5 free parameters · 6 assumptions · 0 invented entities

The central contribution is an algorithm, not a mathematical theorem, and it rests on many hand-set parameters and modeling assumptions. The most consequential choices are the light-scaling factor, tuned after observing overestimation; the fridge-power recalibration, fit to night data; and the appliance-usage tables. These are not hidden, but they make the output sensitive to choices that are not derived from first principles.

free parameters (5)
  • Light per-person scaling factor = 2.25
    Added after noticing that the method over-estimated light consumption; Section 2.1.3, 'Add Lighting'.
  • Fridge nominal power recalibration = Pfridge_new = Pfridge_old * fridge_mean_power / (Pfridge_old * beta2)
    Appendix A, Equation 15; fridge scale is fit to each house's night-time load curve.
  • Appliance nominal powers = Table 3 values (e.g., TV 124 W, oven 2400 W)
    Chosen from NILM wiki and field experience; used to decide where to place appliance signals.
  • Appliance usage probabilities beta1, beta2, beta3 = Table 3 values per appliance
    Hand-set probabilities controlling whether an appliance is activated during a recognized activity.
  • Mean appliance durations tau = Table 3 values per appliance
    Hand-set average durations used to generate appliance power signals.
assumptions (6)
  • domain assumption Dutch time-of-use survey activity probabilities generalize to central European households.
    Section 2.1; if Dutch behavior is unrepresentative, all activity chains and therefore category assignments are biased.
  • domain assumption Measured whole-house power equals the sum of the monitored categories and represents full household electricity use.
    Section 3.3 and Section 2.2.2; the paper itself notes datasets are incomplete, so this assumption is violated in the benchmark.
  • domain assumption The aggregate signal at 15-minute resolution contains enough information to allocate categories via simulated activity chains.
    Core premise of the DUE design, introduced in Section 2 and tested by the benchmark.
  • domain assumption Activity chains are Markov with transition matrices that depend only on day type, employment state, and age group.
    Section 2.1; simplifies human behavior, ignores history and household interactions.
  • domain assumption Inhabitants are assumed at home if any power peak is detected; partial occupancy is not considered.
    Section 2.1.2; false for many real households and affects person-specific load allocation.
  • domain assumption Electrical space heating and cooling are absent from the target households.
    Introduction and Appendix D; the algorithm explicitly excludes these categories.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Unsupervised algorithm for disaggregating low-sampling-rate electricity consumption of households." pith.science (2026). https://pith.science/paper/HVZ6JULI

@misc{pith2026190810713,
  author       = {Pith},
  title        = {Pith review of: Unsupervised algorithm for disaggregating low-sampling-rate electricity consumption of households},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/HVZ6JULI}},
  note         = {Machine review of arXiv:1908.10713}
}
read the original abstract

Non-intrusive load monitoring (NILM) has been extensively researched over the last decade. The objective of NILM is to identify the power consumption of individual appliances and to detect when particular devices are on or off from measuring the power consumption of an entire house. This information allows households to receive customized advice on how to better manage their electrical consumption. In this paper, we present an alternative NILM method that breaks down the aggregated power signal into categories of appliances. The ultimate goal is to use this approach for demand-side management to estimate potential flexibility within the electricity consumption of households. Our method is implemented as an algorithm combining NILM and load profile simulation. This algorithm, based on a Markov model, allocates an activity chain to each inhabitant of the household, deduces from the whole-house power measurement and statistical data the appliance usage, generate the power profile accordingly and finally returns the share of energy consumed by each appliance category over time. To analyze its performance, the algorithm was benchmarked against several state-of-the-art NILM algorithms and tested on three public datasets. The proposed algorithm is unsupervised; hence it does not require any labeled data, which are expensive to acquire. Although better performance is shown for the supervised algorithms, our proposed unsupervised algorithm achieves a similar range of uncertainty while saving on the cost of acquiring labeled data. Additionally, our method requires lower computational power compared to most of the tested NILM algorithms. It was designed for low-sampling-rate power measurement (every 15 min), which corresponds to the frequency range of most common smart meters.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

57 extracted references · 55 canonical work pages

  1. [1]

    Strbac , title Demand side management: Benefits and challenges , journal Energy Policy volume 36 ( year 2008 ) pages 4419--4426

    author G. Strbac , title Demand side management: Benefits and challenges , journal Energy Policy volume 36 ( year 2008 ) pages 4419--4426

    work page 2008

  2. [2]

    author G. W. Hart , title Nonintrusive appliance load monitoring , journal Proceedings of the IEEE volume 80 ( year 1992 ) pages 1870--1891

    work page 1992

  3. [3]

    author N. F. Esa , author M. P. Abdullah , author M. Y. Hassan , title A review disaggregation method in Non-intrusive Appliance Load Monitoring , journal Renewable and Sustainable Energy Reviews volume 66 ( year 2016 ) pages 163--173

    work page 2016

  4. [4]

    Liang , author S

    author J. Liang , author S. K. K. Ng , author G. Kendall , author J. W. M. Cheng , title Load Signature Study—Part I: Basic Concept, Structure, and Methodology , journal IEEE Transactions on Power Delivery volume 25 ( year 2010 ) pages 551--560

    work page 2010

  5. [5]

    Wang , author Q

    author Y. Wang , author Q. Chen , author T. Hong , author C. Kang , title Review of Smart Meter Data Analytics: Applications, Methodologies, and Challenges , journal IEEE Transactions on Smart Grid ( year 2018 ) pages 1--1

    work page 2018

  6. [6]

    Zoha , author A

    author A. Zoha , author A. Gluhak , author M. Imran , author S. Rajasegarar , title Non-Intrusive Load Monitoring Approaches for Disaggregated Energy Sensing: A Survey , journal Sensors volume 12 ( year 2012 ) pages 16838--16866

    work page 2012

  7. [7]

    author B. J. Birt , author G. R. Newsham , author I. Beausoleil-Morrison , author M. M. Armstrong , author N. Saldanha , author I. H. Rowlands , title Disaggregating categories of electrical energy end-use from whole-house hourly data , journal Energy and Buildings volume 50 ( year 2012 ) pages 93--102

    work page 2012

  8. [8]

    Zhao , author L

    author B. Zhao , author L. Stankovic , author V. Stankovic , title Electricity usage profile disaggregation of hourly smart meter data , in: booktitle 4th International Workshop on Non-Intrusive Load Monitoring

    work page

Show all 57 references
  1. [9]

    Batra , author A

    author N. Batra , author A. Singh , author K. Whitehouse , title Gemello , in: booktitle Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD '16 , publisher ACM Press , address New York, New York, USA , year 2016 , pp. pages...

    2016

  2. [10]

    Faustine , author N

    author A. Faustine , author N. H. Mvungi , author S. Kaijage , author K. Michael , title A Survey on Non-Intrusive Load Monitoring Methodies and Techniques for Energy Disaggregation Problem , journal arXiv:1703.00785 [cs] ( year 2017 )

    2017 arXiv

  3. [11]

    Wang , title Artificial Neural Network , in: booktitle Interdisciplinary Computing in Java Programming , publisher Springer US , address Boston, MA , year 2003 , pp

    author S.-C. Wang , title Artificial Neural Network , in: booktitle Interdisciplinary Computing in Java Programming , publisher Springer US , address Boston, MA , year 2003 , pp. pages 81--100

    2003

  4. [12]

    author A. G. Ruzzelli , author C. Nicolas , author A. Schoofs , author G. M. P. O'Hare , title Real-Time Recognition and Profiling of Appliances through a Single Electricity Sensor , in: booktitle 2010 7th Annual IEEE Communications Society Conference on Sensor, Mesh and Ad Ho...

    2010

  5. [13]

    Kelly , author W

    author J. Kelly , author W. Knottenbelt , title Neural NILM : Deep neural networks applied to energy disaggregation , in: booktitle Proceedings of the 2nd ACM International Conference on Embedded Systems for Energy - Efficient Built Environments , publisher ACM , year 2015 , p...

    2015

  6. [14]

    Biansoongnern , author B

    author S. Biansoongnern , author B. Plangklang , title Nonintrusive load monitoring (NILM) using an Artificial Neural Network in embedded system with low sampling rate , in: booktitle 2016 13th International Conference on Electrical Engineering/Electronics, Computer, Telecommu...

    2016

  7. [15]

    Liao , author G

    author J. Liao , author G. Elafoudi , author L. Stankovic , author V. Stankovic , title Non-intrusive appliance load monitoring using low-resolution smart meter data , in: booktitle 2014 IEEE International Conference on Smart Grid Communications (SmartGridComm) , publisher IEE...

    2014

  8. [16]

    Stankovic , author V

    author L. Stankovic , author V. Stankovic , author J. Liao , author C. Wilson , title Measuring the energy intensity of domestic activities from smart meter data , journal Applied Energy volume 183 ( year 2016 ) pages 1565--1580

    2016

  9. [17]

    Mairal , author F

    author J. Mairal , author F. Bach , author J. Ponce , author G. Sapiro , title Online dictionary learning for sparse coding , in: booktitle Proceedings of the 26th Annual International Conference on Machine Learning , publisher ACM Press , address New York, New York, USA , yea...

    2009

  10. [18]

    author J. Z. Kolter , author S. Batra , author A. Y. Ng , title Energy disaggregation via discriminative sparse coding , in: booktitle Advances in Neural Information Processing Systems , pp. pages 1153--1161

  11. [19]

    Elhamifar , author S

    author E. Elhamifar , author S. Sastry , title Energy Disaggregation via Learning Powerlets and Sparse Coding. , in: booktitle Proceedings of the Twenty-Ninth AAAI Conference on Artificial Intelligence Pattern , pp. pages 629--635

  12. [20]

    Singh , author A

    author S. Singh , author A. Majumdar , title Deep Sparse Coding for Non-Intrusive Load Monitoring , journal IEEE Transactions on Smart Grid ( year 2017 ) pages 1--1

    2017

  13. [21]

    Rahimpour , author H

    author A. Rahimpour , author H. Qi , author D. Fugate , author T. Kuruganti , title Non-Intrusive Energy Disaggregation Using Non-negative Matrix Factorization with Sum-to-k Constraint , journal IEEE Transactions on Power Systems volume PP ( year 2017 ) pages 1--1

    2017

  14. [22]

    Bonfigli , author S

    author R. Bonfigli , author S. Squartini , author M. Fagiani , author F. Piazza , title Unsupervised algorithms for non-intrusive load monitoring: An up-to-date overview , in: booktitle 2015 IEEE 15th International Conference on Environment and Electrical Engineering (EEEIC) ,...

    2015

  15. [23]

    Zoha , author A

    author A. Zoha , author A. Gluhak , author M. Nati , author M. A. Imran , title Low-power appliance monitoring using Factorial Hidden Markov Models , in: booktitle 2013 IEEE Eighth International Conference on Intelligent Sensors, Sensor Networks and Information Processing , pu...

    2013

  16. [24]

    Kim , author M

    author H. Kim , author M. Marwah , author M. Arlitt , author G. Lyon , author J. Han , title Unsupervised Disaggregation of Low Frequency Power Measurements , in: booktitle Proceedings of the 2011 SIAM International Conference on Data Mining , publisher Society for Industrial ...

    2011

  17. [25]

    Batra , author J

    author N. Batra , author J. Kelly , author O. Parson , author H. Dutta , author W. Knottenbelt , author A. Rogers , author A. Singh , author M. Srivastava , title NILMTK: An Open Source Toolkit for Non-intrusive Load Monitoring , journal Sensors (Switzerland) volume 12 ( year ...

    2014

  18. [26]

    Bonfigli , author E

    author R. Bonfigli , author E. Principi , author M. Fagiani , author M. Severini , author S. Squartini , author F. Piazza , title Non-intrusive load monitoring by using active and reactive power in additive Factorial Hidden Markov Models , journal Applied Energy volume 208 ( y...

    2017

  19. [27]

    Kumar , author M

    author K. Kumar , author M. G. Chandra , title An intuitive explanation of graph signal processing-based electrical load disaggregation , in: booktitle 2017 IEEE 13th International Colloquium on Signal Processing & its Applications (CSPA) , publisher IEEE , year 2017 , pp. pag...

    2017

  20. [28]

    He , author L

    author K. He , author L. Stankovic , author J. Liao , author V. Stankovic , title Non-Intrusive Load Disaggregation using Graph Signal Processing , journal IEEE Transactions on Smart Grid volume 9 ( year 2016 ) pages 1739--1747

    2016

  21. [29]

    Kumar , author R

    author K. Kumar , author R. Sinha , author M. G. Chandra , author N. K. Thokala , title Data-driven electrical load disaggregation using graph signal processing , in: booktitle 2016 IEEE Annual India Conference (INDICON) , publisher IEEE , year 2016 , pp. pages 1--6

    2016

  22. [30]

    Kumar , author M

    author K. Kumar , author M. G. Chandra , title Event and feature based electrical load disaggregation using graph signal processing , in: booktitle 2017 IEEE 13th International Colloquium on Signal Processing & its Applications (CSPA) , publisher IEEE , year 2017 , pp. pages 168--172

    2017

  23. [31]

    Zhao , author L

    author B. Zhao , author L. Stankovic , author V. Stankovic , title On a Training-Less Solution for Non-Intrusive Appliance Load Monitoring Using Graph Signal Processing , journal IEEE Access volume 4 ( year 2016 ) pages 1784--1799

    2016

  24. [32]

    author Sociaal en Cultureel Planbureau , title Tijdsbestedingsonderzoek 2005 - TBO 2005 , journal DANS ( year 2005 )

    2005

  25. [33]

    Rabiner , title A tutorial on hidden Markov models and selected applications in speech recognition , journal Proceedings of the IEEE volume 77 ( year 1989 ) pages 257--286

    author L. Rabiner , title A tutorial on hidden Markov models and selected applications in speech recognition , journal Proceedings of the IEEE volume 77 ( year 1989 ) pages 257--286

    1989

  26. [34]

    Gershuny , author O

    author J. Gershuny , author O. Sullivan , title United Kingdom Time Use Survey, 2014-2015 , journal Centre for Time Use Research, University of Oxford, UK Data Service ( year 2017 )

    2014

  27. [35]

    Billauer , title peakdet: Peak detection using MATLAB (non-derivative local extremum, maximum, minimum) , year 2012

    author E. Billauer , title peakdet: Peak detection using MATLAB (non-derivative local extremum, maximum, minimum) , year 2012

    2012

  28. [36]

    author J. Z. Kolter , author M. J. Johnson , title REDD: A Public Data Set for Energy Disaggregation Research , in: booktitle Workshop on Data Mining Applications in Sustainability (SIGKDD), San Diego, CA , volume volume 25 , pp. pages 59--62

  29. [37]

    Stankovic , author J

    author V. Stankovic , author J. Liao , author L. Stankovic , title A graph-based signal processing approach for low-rate energy disaggregation , in: booktitle 2014 IEEE Symposium on Computational Intelligence for Engineering Solutions (CIES) , publisher IEEE , year 2014 , pp. ...

    2014

  30. [38]

    Perraudin , author J

    author N. Perraudin , author J. Paratte , author D. Shuman , author L. Martin , author V. Kalofolias , author P. Vandergheynst , author D. K. Hammond , title GSPBOX: A toolbox for signal processing on graphs , journal ArXiv e-prints ( year 2014 )

    2014

  31. [39]

    Leijonmarck , title Exploiting Temporal Difference for Energy Disaggregation via Discriminative Sparse Coding , Master's thesis, KTH, Mathematical Statistics, year 2015

    author E. Leijonmarck , title Exploiting Temporal Difference for Energy Disaggregation via Discriminative Sparse Coding , Master's thesis, KTH, Mathematical Statistics, year 2015

    2015

  32. [40]

    Yu , author P

    author C.-N. Yu , author P. Mirowski , author T. K. Ho , title A Sparse Coding Approach to Household Electricity Demand Forecasting in Smart Grids , journal IEEE Transactions on Smart Grid ( year 2016 ) pages 1--11

    2016

  33. [41]

    Hoyer , title Non-negative sparse coding , in: booktitle Proceedings of the 12th IEEE Workshop on Neural Networks for Signal Processing , publisher IEEE , year 2012 , pp

    author P. Hoyer , title Non-negative sparse coding , in: booktitle Proceedings of the 12th IEEE Workshop on Neural Networks for Signal Processing , publisher IEEE , year 2012 , pp. pages 557--565

    2012

  34. [42]

    title Python.org , year 2018

    2018

  35. [43]

    Beckel , author W

    author C. Beckel , author W. Kleiminger , author R. Cicchetti , author T. Staake , author S. Santini , title The ECO data set and the performance of non-intrusive load monitoring algorithms , in: booktitle Proceedings of the 1st ACM Conference on Embedded Systems for Energy-Ef...

  36. [44]

    Alhamoud , author F

    author A. Alhamoud , author F. Ruettiger , author A. Reinhardt , author F. Englert , author D. Burgstahler , author D. Bohnstedt , author C. Gottron , author R. Steinmetz , title SMARTENERGY.KOM: An intelligent system for energy saving in smart home , in: booktitle 39th Annual...

    2014

  37. [45]

    Kelly , author W

    author J. Kelly , author W. Knottenbelt , title The UK-DALE dataset, domestic appliance-level electricity demand and whole-house demand from five UK homes , journal Scientific Data volume 2 ( year 2015 ) pages 150007

    2015

  38. [46]

    Beckel , author L

    author C. Beckel , author L. Sadamori , author T. Staake , author S. Santini , title Revealing household characteristics from smart meter data , journal Energy volume 78 ( year 2014 ) pages 397--410

    2014

  39. [47]

    Alhamoud , author P

    author A. Alhamoud , author P. Xu , author F. Englert , author A. Reinhardt , author P. Scholl , author D. Boehnstedt , author R. Steinmetz , title Extracting human behavior patterns from appliance-level power consumption data , in: booktitle European Conference on Wireless Se...

    2015

  40. [48]

    Kelly , title UK Domestic Appliance Level Electricity(UK-DALE-2017)-Disaggregated appliance/whole house power , year 2017

    author J. Kelly , title UK Domestic Appliance Level Electricity(UK-DALE-2017)-Disaggregated appliance/whole house power , year 2017

    2017

  41. [49]

    Liu , author A

    author C. Liu , author A. Akintayo , author Z. Jiang , author G. P. Henze , author S. Sarkar , title Multivariate exploration of non-intrusive load monitoring via spatiotemporal pattern network , journal Applied Energy volume 211 ( year 2018 ) pages 1106--1122

    2018

  42. [50]

    Kolter , author T

    author Z. Kolter , author T. Jaakkola , author J. Z. Kolter , title Approximate Inference in Additive Factorial HMMs with Application to Energy Disaggregation , journal Proc. Int. Conf. Arti. Intell. Statist. volume XX ( year 2012 ) pages 1472--1482

    2012

  43. [51]

    Dong , author B

    author H. Dong , author B. Wang , author C. T. Lu , title Deep sparse coding based recursive disaggregation model for water conservation , journal IJCAI International Joint Conference on Artificial Intelligence ( year 2013 ) pages 2804--2810

    2013

  44. [52]

    Parson , author S

    author O. Parson , author S. Ghosh , author M. Weal , author A. Rogers , title Non-Intrusive Load Monitoring Using Prior Models of General Appliance Types , in: booktitle AAAi

  45. [53]

    Makonin , author F

    author S. Makonin , author F. Popowich , title Nonintrusive load monitoring (NILM) performance evaluation , journal Energy Efficiency volume 8 ( year 2015 ) pages 809--814

    2015

  46. [54]

    author M. J. Johnson , author A. S. Willsky , title Bayesian Nonparametric Hidden Semi-Markov Models , journal Journal of Machine Learning Research volume 14 ( year 2013 ) pages 673--701

    2013

  47. [55]

    Manivannan , author B

    author M. Manivannan , author B. Najafi , author F. Rinaldi , title Machine Learning-Based Short-Term Prediction of Air-Conditioning Load through Smart Meter Analyticsan intu , journal MDPI Energies volume 10 ( year 2017 )

    2017

  48. [56]

    Hakell , author G

    author B. Hakell , author G. Fisher , author A. Hersey , title Setting the Benchmark for Non Intrusive Load Monitoring : A Comprehensive Assessment of AMI-based Load Disaggregation , type Technical Report , Pecan Street, year 2015

    2015

  49. [57]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.