D ivision and C alibration Nipun Batra Haimonti Dutta Amarjeet Singh 11202013 Non Intrusive Load Monitoring NILM 2 Breaking down aggregate power observed at meter into different appliances ID: 1034580
Download Presentation The PPT/PDF document "INDiC : I mproved N on-Intrusive load ..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
1. INDiC: Improved Non-Intrusive load monitoring using load Division and CalibrationNipun BatraHaimonti DuttaAmarjeet Singh11/20/2013
2. Non Intrusive Load Monitoring (NILM)2Breaking down aggregate power observed at meter into different appliances
3. Why India?3Different socio-economic settingsHow is a residential deployment in India different?
4. Deployment OverviewFamily characteristics:Single family3 membersMedium IncomeHome characteristics:3 storey720 sq. feet4
5. Deployment Overview: Sensing5Multiple sensing modalities: Electricity, Water, AmbientWater Energy nexus provides interesting insights
6. Electricity monitoring6Smart Meter Circuit BreakerAppliance LevelMeasuring electricity consumption at Supply, MCB, ApplianceResearch questions:Value of additional information (and associated cost)?What level of invasiveness?
7. Water monitoring7Water supply available only for 2 hours in a dayPumps used to store water in tanks- Water has EMBEDDED EnergyInstrument the demand and the supply using Pulse based metersPulse based water meter
8. Ambient sensing8Energy consumption correlated with ambient settingsMeasure following ambient parametersLightTemperatureMotionSound levelBluetooth, WiFiZWave Multisensor +Android
9. Unique Features in IndiaUnreliable Grid9Voltage fluctuationQuartile Spread: Less than 1%Quartile Spread: 9%
10. Unique Features in IndiaUnreliable Grid10Voltage fluctuationLowest voltage typically seen around midnight- ACs in most home are ONHighest voltage typically seen early morning
11. Unique Features in IndiaUnreliable Grid112. BlackoutsObserved power outages upto 12 hours a day!Single power outages of upto 9 hrs observed!
12. Unique Features in IndiaUnreliable Grid3. LearningSystem Design: System should be capable of resuming in same state as it was before outage (Batteries way too difficult to manage )Inferences: Need to measure voltage in addition to current for NILM approaches!12
13. Unique Features in IndiaUnreliable network- Internet13Observed up to 1/4th packet loss on some daysLearningNeed to account for unreliable internetNeed to do local storage of dataWe followed Sense- Local store- Upload
14. Unique Features in IndiaLoad specificsBathroom level water heating- Runs off electricity as opposed to gas Contributes ~60% of total energy in wintersRoom level air conditioningUsed only in summersControl is de-centralizedThese two loads are fairly easy to disaggregate- Easy to act upon to reduce energy footprint14
15. iAWE: Indian Dataset for Ambient, Water and Electricity sensing2+ months of data 1 day fully labeled dataRest semi-labeledElectricity, Water, Ambient conditions at different granularitiesDataset released for public use15
16. iAWE: Indian Dataset for Ambient, Water and Electricity sensingDataset explorer16
17. iAWE: Indian Dataset for Ambient, Water and Electricity sensingSample IPython notebooks- Code to interact with data and view results17
18. iAWE: Indian Dataset for Ambient, Water and Electricity sensingEntire project maintained as open source on Githubhttps://github.com/nipunreddevil/Home_Deployment/https://github.com/nipunreddevil/iawe-website18
19. Demo http://energy.iiitd.edu.in:5000/19
20. ConclusionsDeveloping countries provide unique challenges for residential deploymentsUnreliable gridUnreliable networkLoad specificsValidated previous work in residential sensingRelease iAWE dataset“Behind every successful residential deployment, there is a very cooperative (and angry ) family”20
21. AcknowledgementsNSF- DEITy for fundingMilan Jain (MS IIITD) and Shailja Thakur (MS IIITD) for deployment supportTCS Research and Development for PhD fellowship21