Disintegration of Energy

8/13/2019 Disintegration of Energy

1/17


2/17

2

fundamentally transform how electricity, water, and natural gas are understood, studied and, ultimately, consumed.This article is focused on electrical energy, but we have also developed systems for disaggregating water and gasusage (see sidebar). All three of our systems share a common approach: they monitor side-effects of resource usage

that are manifest throughout a homes internal electricity, plumbing, or gas infrastructure. Although our techniques should work in commercial and industrial sectors, we have so far concentrated onvalidating our methods in the residential sector, which in itself presents many challenges. In addition to thesignificant amount of energy use and CO 2 emissions in the residential sector (Swan, 2009; Weber and Matthews,2007), there is a higher degree of decentralized ownership and varying levels of self-interest and expertise inreducing energy consumption compared with the industrial and commercial sectors. Perhaps more compelling,however, is that energy consumption can vary widely from home to home simply based on differences in individualbehavior . Indeed, it has been consistently found that energy use can differ by two to three times among identicalhomes with similar appliances occupied by people with similar demographics (Socolow, 1978; Winett et al., 1979;Seryak, 2003).

We believe that disaggregated data presents an enormous opportunity for households to better understand their

consumption practices, determine easy and cost-effective measures to increase their energy efficiency, andultimately reduce their overall consumption. Even a 10-15% reduction in electricity use across US homes would besubstantial, representing nearly 200 billion kWh of electricity per year. This is equivalent to the yearly power outputof 16 nuclear power plants or 81.3 million tons of coal. Such statistics have led a growing body of scientists,utilities, and regulators to view energy efficiency as the most accessible and cost effective form of alternative energy.

The Value of Disaggregated DataBefore discussing specific techniques for sensing disaggregated energy data, it is useful to highlight why we thinkthis level of sensing is valuable to residents, utilities, policy-makers, and appliance manufacturers. With regards toresidents, research in environmental psychology has uncovered some profound misconceptions about energy usagein the home. Mettler-Meibom and Wichmann (1982) interviewed 52 households in Munich, West Germany and

asked them to estimate the proportional energy cost of specific end uses. These responses were then compared toactual usage. Consumers vastly underestimated the energy used for heating and overestimated the energy used forappliances, lighting, and cooking. Similar findings were obtained by Costanzo et al. (1986) and Kemptom andMontgomery (1982). Kempton and Montgomery also found that consumers often estimate an appliances energy use

by i ts perceptual salience (e.g., television and lighting are often overestimated) and also overestimate energy used by machines that replace manual labor tasks ( e.g., dishwasher, clothes washer). Consumer estimates of aggregateenergy usage are also poor. In multiple studies spanning a total of 700 persons, Winett et al. (see Geller et al., 1982)found that only a small percentage (1-2%) knew how many KWh they used per month or per day most did noteven know where their electricity meter was located.

These misunderstandings of how energy is used in the home are reflected in the steps consumers believe they cantake to conserve energy. People tend to overestimate the effectiveness of conservation measures that depend on

changes in short-term behavior such as turning off the lights when leaving a room and underestimate technical or building innovation solutions such as deciding to replace an inefficient appliance or upgrading a homes insulation(Kempton and Montgomery, 1984). Others have shown dramatic misunderstandings of the benefits ofweatherization, retrofits, and tax breaks (Geller et al., 1982). These results point to the need for more accurate andspecific information about how actions in the home affect energy consumption. Disaggregated data could be used byenergy eco-feedback systems to both provide pertinent information about energy usage as well as to provide tailored


3/17

3

feedback at opportune times (see Froehlich, 2009 and Froehlich et al., 2010). For example, an eco-feedbackinterface could provide information on the most convenient and cost-effective steps to take to reduce energyconsumption based on the specific appliances and devices in a home as well as the way in which those systems are

used. The feedback interface might make specific recommendations about retrofit solutions, appliance upgrades orfocus on curtailing particularly wasteful behaviors. Disaggregated data could also be used to inform residents aboutmalfunctioning equipment or inefficient settings ( e.g., the water circulation pump appears to be operatingcontinuously rather than being triggered by a thermostat ).

From a policy perspective, knowing how much energy is being consumed by each class of appliances or devices iscritical to the development and evaluation of evidence-based energy-efficiency policies and conservation programsaimed at reducing capital expenditure on additional capacity (Sidler and Waide, 1999). Disaggregated data wouldallow utilities to accurately assess and prioritize energy-saving potentials of retrofit or upgrade programs. Equipmentmanufacturers and governments could compare energy measurements made under controlled test conditions tomeasurements under actual home usage conditions. This would result in more realistic test procedures and,ultimately, more energy-efficient designs as end-use cases are better understood.

In addition to providing utility companies with an evidence-based method of evaluating their own conservation programs, disaggregated data presents opportunities for power system planning, load forecasting, new types of billing procedures and the ability to pinpoint the origins of certain customer complaints. For demand response,utilities would be able to improve the quality of demand forecasting by having better models of usage ( e.g., thenumber of households with energy efficient air conditioning). With real-time eco-feedback displays in the home,utilities could also suggest specific appliances to turn off or recommend other actions in order to conserve energyduring peak load times. Finally, although some utilities currently employ different pricing schemes depending onusage ( e.g., tiered pricing based on overall usage or time-of-use pricing based on when the energy usage occurs),future pricing models could take into account the type of usage and charge accordingly. For example, heating,ventilation and air conditioning (HVAC), refrigeration, and lighting could have different pricing models.

Existing Techniques to Measure Disaggregated Energy UsageField surveys have traditionally been the most common approach used to acquire details on occupant behaviors andappliance penetration levels ( e.g., Office of Energy Efficiency, 2003). Survey data is used for conditional demandanalysis (CDA), a modeling technique which attempts to disaggregate individual end-uses of energy (Swan et al.,2009). CDA compares the time-dependent load profiles of households with known appliances against those without,thereby allowing a statistical characterization of the consumption behavior. The accuracy of CDA is limited becauseof the diversity of devices and appliances in homes and because many end-uses share temporal load profiles, makingaggregate load profiling relatively inaccurate. Furthermore, CDA has traditionally relied on self-report surveying forload disaggregation, which provides a relatively sparse dataset containing various self-reporting biases.

In contrast to survey-based methods, automated methods for sensing disaggregation do not rely on self-report andcan potentially provide extremely rich end-use datasets; however, no commercially available disaggregation method

currently exists that is easily deployable, highly accurate, and cost-effective. There are two primary techniques forautomatically measuring disaggregated energy usage down to the individual appliance or device: distributed directsensing and single-point sensing. These approaches are differentiated by their varying degree of hardware andsoftware complexity, whether they require installation by a homeowner or a licensed professional, their calibrationor training requirements, and their potential cost.


4/17

4

Distributed Direct SensingDistributed direct sensing requires a sensor at each device or appliance in order to measure consumption. Althoughconceptually straightforward and potentially highly accurate, direct sensing is often expensive due to timeconsuming installation and the requirement for one sensor for each device/appliance. Additionally, appliances thattend to consume the most electricity are frequently hard-wired ( e.g., electric hot-water heaters or lighting) ordifficult to reach ( e.g., behind a refrigerator or dryer) making installation and maintenance difficult and costly.Finally, because these sensors are distributed around the house, a communication protocol must be devised that cancommunicate this information to a local repository using, for example, multi-hop radio or powerline communication.That said, direct sensors have the potential to both sense and control the operation of various devices and appliances

by virtue of being co-located ( e.g., turning off a light when an occupant departs a room; see, for example, iControl 1);this dual capability is not possible with single-point sensing unless the sensing system communicates with adistributed control system. An additional benefit is that the calibration process typically requires no more than

providing a label for each sensor. This label corresponds to the device or appliance that is connected to the sensor.

Single-Point SensingIn response to limitations with the direct sensing approach, researchers have explored methods to inferdisaggregated energy usage via a single sensor. Pioneering work in this area is non-intrusive load monitoring(NILM), first introduced by George Hart in the 1980s (see Hart, 1992). In contrast to the direct sensing methods,

NILM relies solely on single-point measurements of voltage and current on the power feed entering the household. NILM consists of three steps: feature extraction, event detection, and event classification. The raw current andvoltage waveforms are transformed into a feature vector a more compact and meaningful representation that mayinclude real power, reactive power, and harmonics These extracted features are monitored for changes, identified asevents ( e.g., an appliance turning on or off), and classified down to the appliance or device category level using a

pattern recognition algorithm, which compares the features to a preexisting database of signatures. In its simplestform, the NILM feature vector may only contain one value: a step change in measured power to disambiguate

between devices. More advanced NILM features may contain measures of power broken down by frequency and

temporal patterns (for example, see Laughman et al. 2003). Leveraging recent advances in device and appliance power supplies, our lab has extended the NILM approach for electrical disaggregation by using high frequencysampling of voltage noise (Patel et al., 2007; Gupta et al., 2010). Voltage noise provides an additional feature vectorthat can be used to more accurately distinguish between energy usage signatures that would otherwise appear verysimilar.

Most single-point sensing approaches rely on pattern matching, which presupposes the existence of a database ofappliance and device usage signatures. For example, Berges et al. (2008; 2010) and Roberts and Kuhn (2010) areinvestigating how to build databases of these signatures for the NILM architecture. In the ideal case, these signatureswould be similar across homes and thus could be pre-loaded on the sensing device or uploaded and examined in thecloud. In the worst case, however, a training example of each appliance or device must be produced per home ,which would greatly increase installation complexity. We return to installation and calibration requirements in the

discussion section of this article.

1 http://www.icontrol.com/


5/17

5

Extractablefeatures

Real/ReactivePower (Watts

vs. VARs)

Apparentpower from

|I AC |

Harmonicsof |I AC |

Startup ofIAC

|V AC | Transientvoltage noise

signature

Continuousvoltage noise

signature

Sensing technology smart meterscapable of mediumrate sampling

current clamps(e.g., TED) orinductive sensors(e.g., Patel et al.,

2010)

current clamps orammeters current clampsor ammeters voltmeter high samplingrate voltmeter medium sampling ratevoltmeter

Disaggregation level device category large load category(e.g., appliances)

large loadcategory

large loadcategory

large loadstartup

detection

individualdevices withmechanical

switches

individual devicesemploying SMPS, orother electronic load

controls

Example devices thatcan be disaggregated

fans, motors,HVAC systems,

forced air heaters

stove, dryer,electric heaters

fans, dryers,compressors

CFLs, motors motorappliances,

dryers, electricheaters

any switchedload ( e.g.,

incandescentlights, resistive

dimmers,heating)

continuously switcheddevices: CFLs, TVs,

DVD players, chargingunits (laptops, cell

phones, etc.)

Algorithm clustering of Wattsand VARs

step change inmagnitude

magnitude ofharmonics

pattern matchingof startup

transients

magnitude pattern matchingon transient

pulses

pattern matching onfeatures of resonant

frequencyInstallation breaker or meter:

inline ammeterwith voltmeter

breaker or meter:inline ammeter, or

affixed outside(e.g., contactless,Patel et al., 2010)

breaker or meter:in line, or affixed

outside

breaker or meter:in line, or affixed

outside

plug-inanywhere

plug-in anywhere plug-in anywhere

Ease of physicalinstallation excluding

calibration(very easy, easy,hard, very hard)

very hard current clamps:hard

inductive sensors:easy

hard hard very easy very easy very easy

Ease of calibration(very easy, easy,hard, very hard)

very easy hard hard easy very hard easy very easy

Cost(including cost of

installation)

very high low medium medium very low very high high

Advantages automaticcategorization of

certain loads,works well for

appliances

simple, enablescentral database ofsignatures, reduces

per homecalibration

discriminatesamong devices

with similarcurrent draw

discriminatesamong devices

with similarcurrent draw and

startup

simplicity andcost

nearly everydevice hasobservablesignature,

independent ofload

characteristics

stable signatures acrosshomes and devices

(i.e., enables centraldatabase of signatures,

reduces per homecalibration),

independent of loadcharacteristics

Limitations I and V must besampled

synchronously, fewdevices with

diverse powerfactor

few devices withdiverse power

draws

limited to largeinductive loadsthat distort ACline, loads must be synchronous

to 60Hz

limited to loadswith diverse, longduration startup

characteristics likemotors and some

CFLs

few devicesaffect VAC line,susceptible toline variations

requires perhome calibration,

requires fastsampling (1-100

MHz)

requires mediumsampling rate (50-500

kHz)

Table 1. This table outlines the various features, algorithms, and technologies used to disaggregate end-use energy datafrom single-point sensors. The AC

subscript

indicates that features were extracted using only the 60Hz component.


6/17

6

Intermediate Sensing MethodsSome sensing systems fall somewhere between direct and single-point sensing. "Smart Breaker" devices such as thePowerhouse Dynamics eMonitor 2, are installed by an electrician inside a homes circuit breaker panel to provide acircuit-by-circuit breakdown of energy consumption. Depending on the design of a home's circuit layout, eachcircuit may feed only one appliance so an eMonitor-like approach could be used to acquire end-use consumptiondata. If multiple devices and/or appliances share a circuit, the "Smart Breaker" approach does not offer end-usedisaggregation at the individual appliance or device level although one of the approaches outlined in the next sectioncould be used to discriminate between them. That said, these multiple-sensor approaches are often cost-prohibitive,especially when the cost of professional installation is included.

Observable Features for Single-Point Energy DisaggregationSingle-point sensing has advantages over direct sensing in terms of cost, ease of installation and in overallintrusiveness but is fundamentally limited by the amount and quality of information that can be sensed from a single

point in the homes electrica l infrastructure. This section maps out the variety of features used in electrical single- point sensing for disaggregated data and discusses the quality and type of information these features provide. Theapproaches to appliance- and device-level end use monitoring can be subdivided into three groups: approaches basedon (1) aggregate power and current consumption, (2) AC line distortion and startup characteristics, and (3) voltagesignatures. A complete summary of the approaches is given in Table 1.

Approaches Based on Aggregate Power ConsumptionPerhaps the most obvious starting point to automatically discriminate between devices is to use the total powerconsumed by each device. As different devices tend to draw different amounts of power and these power draws tendto be consistent over time, total power is a reasonable feature to use for classification ( e.g., a lamp typically uses lessthan an iron, which uses less than amicrowave). More specifically, most deviceshave predictable current consumption and allare comprised of a mixture of resistive andreactive components, which give rise to realor true power (Watts) and reactive power(volt-ampere reactive or VAR). Thus, manydevices can be categorized according to themagnitude of Watts and VARs consumed (seeFigure 1). For example, a refrigerator mightconsume 40010 W and 45020 VARs. Thisapproach, commonly referred to as LoadMonitoring, was originally used to categorizehigh power devices such as refrigerators and

HVAC systems (Drenker and Kader, 1999).However, Load Monitoring is more difficultto apply to devices that consume little

2 http://www.powerhousedynamics.com/

Figure 1. An example power-based signature space based on one home(adapted from Hart, 1992). Note that resistive appliances (iron, electric waterheater) appear on the real power axis and that motors have a reactivecomponent.


7/17

7

instantaneous power (Hart, 1992) such as radios and small fans because the overlap in the feature space isconsiderable. It is also important to note that although Load Monitoring categorizes devices, it cannot disaggregatetwo similar devices in the same home ( i.e., lights in separate rooms).

A significant downside of this method is the need to install sensing components that are capable of measuring Wattsand VARs. Measuring the real and reactive power requires knowing the phase angle between the main line ACvoltage and current. This requires a system to synchronously sample the voltage (V AC) and current (I AC) waveformseither at the meter or right before the breaker box. Most commonly ( e.g., TED 3, NILMs, etc.) I AC is measuredindirectly with clamp-type current transformers. Building codes typically require these devices to be installed by alicensed electrician. The process involves dismantling the breaker box and clamping magnetic sensors around themain power feed conductors. To gather electrical current features in a consumer-installable manner, we havedeveloped a contactless current sensor that can be mounted to the outside of the breaker box and infers current fromthe magnetic fields generated by the feed conductors inside the box (Patel et al., 2010). Although this removes theneed for professional installation and greatly reduces the complexity and safety concerns of the installation

procedure, sensor placement and calibration become critical for high accuracy. In the future electrical current

information could be provided by smart meters, but it is unclear when this data will be available and if the temporalresolution will be sufficient for disaggregation (current smart meters are typically configured to report consumptiondata with 15 minute granularity).

Approaches Based on Current Consumption and Startup CharacteristicsSynchronously sampling both the voltage and current waveforms can add considerable cost to the sensing unit andinstallation. As a result many newer sensing algorithms have attempted to classify electrical device end use usingonly the magnitude of I AC or on the magnitude of I AC over long durations (on the order of 1 second) when a device isfirst actuated (startup), which eliminates the need to sample voltage and reduces the sampling rate for the currentwaveform (most designs use ~600 Hz sampling).

Many electrical devices such as heaters, fans, and compressors exhibit current waveforms with significant energiesin the 60 Hz harmonics. Distinct features in the current waveforms can be found by isolating each harmonic andanalyzing the spectral envelope over a fixed duration during the device start-up (typically 100-500 ms). Cole et al.(1998) and Leeb et al. (1995) were able to use these start-up features together with Load Monitoring to categorizeresidential appliances. These approaches usually entail some form of template matching on a known library of start-up features to classify unknown loads. Laughman et al. (2003) were able to show that this feature space is lesssusceptible to overlapping categories and is therefore able to separate many devices with similar load characteristicscompared to approaches based on aggregate power consumption. For example, two motors with similar real andreactive power consumption can exhibit significantly different start-up features making them easy to differentiate.This approach is primarily limited to devices that consume large current loads, and thus, exert significant harmonicdistortion on the current waveform. Even so, Lee et al. (2004) showed that some switching mode power supply(SMPS) devices exhibit continuous harmonic signatures in the current waveform ( i.e., not just distortions duringstartup, but constant signatures embedded in each harmonic). These signatures, however, are better modeled using

voltage, where the features have considerably less overlap as we show in our own work below.

3 The Energy Detective, http://www.theenergydetective.com


8/17

8

Approaches Based on Voltage SignaturesUsing voltage-domain measurement for electrical device disaggregation at first seems counter-intuitive. Theincoming power feed to a home is often assumed to be a well-regulated 60 Hz pure sine-wave AC source. Of course,this may be true at the point of generation but it is not true within a home. Instead, appliances conduct a variety ofnoise voltages back onto the power wiring of the home. In an attempt to limit interference among devices, theFederal Communications Commission (FCC) Part 15 and Part 18 rules restrict the noise voltage each device mayconduct back onto the power line. Researchers have found, however, that devices that comply with the FCC's limitsstill yield measurable noise signatures that are easily detectable using appropriate hardware. These signatures occurat a broad range of frequencies, not just 60 Hz and its harmonics. Marubayashi et al. (1997) categorizes three typesof voltage noise: 1) on-off transient noise, 2) steady-state line voltage noise (generated at 60 Hz and harmonics),and 3) steady-state continuous noise (generated outside of 60 Hz). Past work on electromagnetic compatibility islargely focused on ensuring this noise does not adversely affect power distribution or interfere with radio ortelevision reception. Since 2007 we have been developing methods for characterizing these voltage noise signaturesand using them to classify the operation of electrical appliances and devices in the home.

An important advantage of voltage noise signatures is that it allows the use of any electrical outlet inside the homeas a single installation point because the line voltage is sampled in parallel with the incoming AC line. This is incontrast to electrical current sensing solutions that require professional installation of sensors around power feedconductors. A second advantage is that, unlike the above approaches, voltage noise signatures can be used to notonly categorize energy usage but also identify the specific source of usage. In other words, this approach canuncover not just that a light bulb was turned on but which light bulb

Transient Voltage Noise SignaturesAny mechanically switched device, such as a light switch, induces a short duration (transient) pulse when making or

breaking an electrical circuit. This is the result of a rapid series of openings and closings due to the physical natureof the switch itself through bouncing, sliding, rocking and surface contaminants (Howell, 1979). These "contact

bounces" create impulse noise, which typically lasts only a few microseconds and consists of a rich spectrum of

frequency components ranging from a few kHz up to 100 MHz. The transient voltage noise is affected by theelectrical infrastructure of the home ( i.e., path of transmission line) and the manufacturing details of the switch, sothe transient noise generated by each switch is unique. This enables per-device identification but usually preventsgeneralization across homes.

Figure 2 (left) shows a spectrogram of a light switch being turned on in a home. Note the rich spectrum of frequencycomponents and the relative strength of each frequency component. Through empirical observation, we have foundthat these transient signatures are relatively stable over time (Patel et al., 2007).

Steady-state AC Line Voltage SignaturesUnlike transient voltage noise, which is only present for a brief instant when a device is turned on or off, steady-

state noise exists as long as a device is powered on. Devices such as coffee grinders, fans and hair dryers containelectric motors that create continuous voltage noise synchronous to the frequency of AC power (60 Hz in the USA)

and its harmonics. The magnitude of the 60 Hz harmonics can be used to detect the presence of large motor loads.Given that the steady-state noise signal does not just occur when a device is turned on or off, systems employing thisfeature have the ability to classify the signal even if the device activation/deactivation event is missed. Thus, steady-state signals are in some ways more attractive than their transient counterparts.


9/17

9

Steady-state Voltage SignaturesDue to their higher efficiency, smaller size, and lower cost compared to traditional power supplies, an increasingnumber of devices in the home use switched mode power supplies (SMPS). These devices, which include laptops,charging units, and TVs, exhibit continuous voltage signatures at the resonant switching frequencies of the SMPShardware and are usually in the range of 10 kHz and 1 MHz with a bandwidth of a few kHz. With the wide varietyof manufacturing processes and power requirements of SMPS devices, there is a minimal amount of overlap in thefrequency spectrum making the resonant frequency an attractive feature for classification.

A similar switching mode characteristic is also seen in Compact Fluorescent Lights (CFLs) and dimmer switches. ACFL power supply employs the same fundamental switching mechanism to generate the high voltages necessary to

power the lamp. Dimmers also produce continuous signatures due to the triggering of their internal TRIAC, whichcan be used to detect and identify the incandescent loads they control. In contrast to the narrowband noise produced

by SMPS, a dimmer produces broadband noise spanning up to hundreds of kHz.Figure 2 (right) shows a frequency domain waterfall plot showing a variety of devices and appliances. When adevice or appliance is turned on we see a narrowband continuous noise signature that lasts for the duration of thedevices operation. The excitation of a switching device can be considered as a series of periodic impulses. Theelectrical lines of the household ( i.e., the houses transfer function) act as filters, affecting the magnitude and

bandwidth of the resonances and the corresponding harmonics. Thus, some features can be generalized acrosshomes, and others can be used within homes to disaggregate similar appliances in different rooms, or used to trackwhich outlet a particular mobile device is connected (see Gupta et al., 2010).

Implementing Electrical Energy Disaggregation Using Voltage NoiseIn this section, we summarize our implementation for energy disaggregation using transient voltage noise. We

include a brief overview of our sensing hardware, our disaggregation algorithms, and the results of our experimentsin 14 homes. Further details can be found in (Patel et al., 2007) and (Gupta et al., 2010).

Our prototype system consists of a single custom power line interface plug-in module that can be plugged into anyelectrical outlet in the home (Figure 3). The output of the plug-in module is connected to a high speed dataacquisition system that digitizes the analog signal and streams it over a USB connection to data collection software

Figure 2. (left) Transient voltage noise signatures of a light switch being turned on. Colors indicate amplitude at each frequency.(right) Steady state continuous voltage noise signatures of various devices during various periods of operation.


10/17

10

running on the PC. The PC then samples andconditions the incoming signal. Though we testedour system on a 120 V, 60 Hz electrical

infrastructure, our approach could easily beapplied to an electrical infrastructure utilizingdifferent frequency and voltage ratings with littlechange to the hardware and no change to thesoftware. For homes that have split phase wiring(i.e., two 120 V branches that are 180-degrees outof phase), the high frequency coupling at the

breaker box between the two phases typicallyallows us to continue to monitor at a singlelocation and capture events on both lines.

To detect and classify the transient voltage noise,

we employ a simple sliding window algorithm (1-microsecond in length) to look for substantialchanges in the input line noise (both beginning andend). A feature vector of frequency components and their associated amplitude values is created by performing aFast Fourier Transform on the sliding window sample. A Euclidean distance measure compares contiguousexamples; when the distance first exceeds a predetermined threshold, the start of the transient is marked. Thewindow continues to slide until there is another drastic change in the Euclidean distance, which indicates the end ofthe transient. The feature vectors that comprise the segmented transient are then sent to a support vector machine(SVM) for classification. Note that the SVM must be trained using 3-5 labeled transient voltage noise signaturesfrom the home in which it is to be used.

To detect and classify steady-state noise, we also employ a frequency-based analysis. The incoming time domainsignal stream from the data acquisition hardware is buffered into 4 ms windows. Using Welchs method (1967), wecreate frequency-based feature vectors, which are then fed into our event detection and extraction software. Whenthe system begins, it creates a snapshot of the baseline frequency signature. Thereafter, new vectors are subtractedfrom the baseline signature to produce a difference vector. Our feature extraction algorithm finds new resonant

peaks using the difference vector and extracts quantities related to center frequency, magnitude, and bandwidth ofthe resonances. We build templates of known devices and use Nearest Neighbor Search in Euclidean space toclassify the feature vectors into their source device or appliance.

To evaluate the feasibility and accuracy of our approach, we have conducted a number of staged experiments over atotal of 14 homes of varying styles, ages, sizes and locations. In two of these homes, we have also examined thetemporal stability of voltage noise signals by conducting longitudinal deployments. To reduce confounding factors,we have thus far performed transient and steady-state noise experiments independently of each other. For transientvoltage noise analysis we tested our system in one home for six weeks and in five homes for one week each toevaluate the system performance over time and in different types of homes. Results indicate that we can learn andclassify various electrical events with accuracies ranging from 80-90% with recalibration being unnecessary evenafter 6 weeks of installation. We were able to disaggregate events down to single light switches, and were even ableto distinguish between two different switches that had the same load characteristics ( e.g., both switches wereconnected to a 100 W light bulb).

Figure 3. Our single-point disaggregated energy sensing systemconsists of a single plug-in module, acquisition hardware and thesupporting software.


11/17

11

We tested the steady-state voltage signatures approach in one home for a period of six months, and in six otherhomes using staged experiments in a fixed setting. Results indicate that devices can be classified with accuraciesranging from 89-97% within individual homes. This included the ability to disaggregate the same model and brand

of LCD televisions appearing in the same home as well as telling apart individual CFLs. We also investigated thefeasibility of using our method to train general templates for four electrical devices at one home and then classify thedevices in six other homes. In five of the six homes we were able to classify the devices with 100% accuracy, and inthe final home we were able to classify three of the four devices with 100% accuracy (the single failure case was alaptop power adaptor whose noise characteristic is dictated by the charging state). Unlike transient voltage noise,whose characteristics come from a random distribution, SMPS steady-state noise is predictable between similarhardware designs. In a follow- up study, we acquired 10 Dell 20 inch LCD monitors and created a single noisesignature using one random monitor. We obtained near 100% classification accuracy when using a single learnedmodel to a random installation of the 9 other monitors in the various homes.

Our noise-based approaches allow us to identify when and what devices or appliances are used but they do not provide power consumption data. Thus, the last step in our approach is to sense changes in whole-home, aggregate

power consumption and map these power changes to the identified events. Any of the discussed current sensingapproaches allow for this ( e.g., TED) including those which leverage contactless sensing ( e.g., Patel et al., 2010).

DiscussionWe believe that disaggregated energy use data is considerably more valuable and actionable than solely usingaggregate data at the whole-house level. However, there are a number of open problems remaining that need to besolved in order for single-point sensing to be a viable energy disaggregation sensing method in the future. Many ofthese open problems are well suited for the pervasive computing research community to investigate including thedevelopment of probabilistic-based classification approaches and new methods for ground truth labeling of energyusage data, as well as the calibration and training of the sensing apparatus itself.

The cost and ease of installation, including any necessary training or calibration, need to be considered in terms of

their likely impact on large scale adoption of energy disaggregation solutions. Although the ease with which a givendisaggregation device can be physically installed is important, the ability for that device to either work out-of-the-

box or with a small number of calibration steps is perhaps of even greater importance. We have pursued twocomplementary disaggregation approaches based on both transient voltage noise and steady-state voltage noise.Both techniques employ a fingerprinting approach for classification. That is, a database of labeled signatures mustexist in order for the method to perform well. The calibration process requires that a user walk around his/her homeactivating and deactivating each device or appliance at least once (to create the signature) and provide some sort ofhuman-readable text label for each device. Although initially burdensome, the calibration process would only needto be done once or when a new device or appliance is installed. In either case, mobile phone software could guidethe user along this process and also serve as an interface for collecting text labels.

Of course, this level of calibration is only necessary if the disaggregation signature of the device or appliance is

different across homes. In the case of transient voltage noise, we have indeed found this to be the case. In otherwords, transient voltage noise alone is not portable enough to enable a shared database of signatures to take the place of in situ calibration. In contrast, we have found that steady state noise signals from SMPS and CFLs have alarge degree of signal similarity across homes and devices (Gupta et al., 2010). This is likely because thesesignatures are largely shaped by the particular circuit design and electronic components of the device rather than the

position of the device on the homes internal powerline infrastructure. Given this signal invariance, a crowd


12/17

12

sourcing approach for signature labeling may be employed where signatures and their labels are shared acrosshomes via a common backend database.

Future work could also look at the feasibility of automatically clustering and classifying unknown signals in anunsupervised fashion. That is, a sensing system need not require a database of signatures and labels in order to

perform. Instead it learns these signals over time and then acquires the labels at a later point. An interface could then present a list of these unknown signals to the homeowner and ask for a semantic label. For example, the eco-feedback system may state: the 2 nd most power consuming appliance in your home has yet to be labeled, we think itis a hot- water heater, is this correct? In this way, the calibration effort is amortized over a longer period of time. Roberts et al. (2010) are currently in the early stages of evaluating their sensing system, which requires no a-prioryknowledge of devices/appliances in the home nor does it require that each detected device/appliance exist as a modelin its library. Although the sensing system can learn unique signals over time, it still requires user intervention to

provide semantic labels for these signals.

In addition to unsupervised learning, we are also studying the benefits of exploiting contextual cues and temporalityof device usage (a strategy exploited by conditional demand analysis (CDA) and also investigated by Hart and

colleagues). For example, many devices have predictable usage duration or are commonly used in conjunction withother devices. Many devices also have predictable states of electrical usage such as washing machines, dryers, andHVAC systems. Dynamic Bayesian Networks (DBNs) are well suited to exploit this sort of a priori information.DBNs also have the added benefit of seamlessly integrating multiple feature streams, such as those measured fromvoltage and current essentially providing a high-level method to integrate all of the features mentioned previously.

Finally, one of the primary research challenges in investigating energy disaggregation techniques is evaluation.Although computer simulation and laboratory-based testing are often useful in evaluating the feasibility of anapproach, it is difficult to replicate the complexity and nuance of device and appliance usage in these artificialenvironments. The problem with in-home evaluation is the difficulty of establishing a method to acquire groundtruth labeling about when and which devices and appliances were used. For our own evaluations, we have largelyrelied on manual labeling of electrical device activation and deactivations using staged in-home experiments with

custom built labeling software. However, this approach is labor intensive, making it difficult to conduct experimentsacross a large set of homes. It is also unclear how well these controlled activation/deactivations mimic naturalisticenergy usage. For example, some disaggregation techniques are affected by the number of appliances or devices thatare active simultaneously such dependencies need to be reflected in the staged experiments.

The labeling of appliance and device usage could be handled by distributed direct sensing or hybrid methods thatmonitor power draw on each electrical branch of a home. Here, a sensor or a set of sensors are installed at eachdevice/appliance location in order to monitor use. These distributed sensor streams are then used to evaluate theefficacy of the single-point solution ( e.g., by comparing both outputs). Fully instrumenting a house with directsensors on each outlet and for each appliance, device, and hard wired system ( e.g., lighting) is extremely resourceintensive both from an installation and a maintenance perspective. It is also unlikely that any pre-existing directsensing system would be capable of providing per-device usage information, so new direct sensors would have to be

custom designed, which presents its own challenges. Finally, one would have to ensure that the direct sensorsthemselves do not distort the signal features used by the disaggregation algorithms ( e.g., by adding additional noiseto the power wiring). So, although a direct sensing approach would clearly have benefits ( e.g., it could be used tocollect naturalistic ground truth data over long periods of time), major challenges remain in making this practical.


13/17

13

ConclusionMost people are entirely reliant or becoming entirely reliant on utility-supplied energy, gas, and water for theirfundamental activities of daily living ( e.g., cooking, bathing, washing clothes). Yet few can accurately respond toquestions about how they use those resources e.g., what are the most energy consuming activities in one's home?When does it make economic sense to replace an old appliance with a newer, more energy-efficient model? Withoutaccurate, disaggregated consumption information, it is difficult to make informed decisions about the easiest andmost cost effective steps to reduce waste. In this article, we have argued that disaggregated data is crucial in closingthe energy-literacy gap.

The beneficiaries of disaggregated consumption information extends beyond residents to utility companies,appliance-manufacturers and policy-makers, who require knowledge about the end uses of energy in order to betterevaluate their conservation programs, forecast demand, and establish new energy efficiency policies. The keychallenges include both (1) how to cost-effectively provide accurate disaggregated consumption information and (2)how to best present this information to a homeowner in an actionable form. We believe that continued research intonew methods for disaggregation, including our efforts in contactless current sensing and automated classification ofloads based on transient and steady-state voltage noise signatures are important advancements toward achieving theultimate goal of providing pervasive disaggregated energy information in the home.

References1. Berges, M., Goldman, E., Matthews, H. Scott., Soibelman, L. Training Load Monitoring Algorithms on Highly Sub-

Metered Home Electricity Consumption Data, Tsinghua Science & Technology, Volume 13, Supplement 1, October 2008,Pages 406-411.

2. Berges, M., Soibelman, L., Matthews, H. S. Building Commissioning as an Opportunity for Training Non-Intrusive LoadMonitoring Algorithms. Proc. of the 16th International Conference on Innovation in Architecture, Engineering andConstruction (AEC), State College, PA, 2010.

3. Cole, A. I. and Albicki, A. Algorithm for non -intrusive identificatio n of residential appliances, Proc. of the 1998 IEEEInternational Symposium on Circuits and Systems (ISCAS '98), Vol. 3, May 31- June 3, 1998, pp.338 - 341

4. Costanzo, M., Archer, D., Aronson, E., & Pettigrew, T. Energy conservation behavior: The difficult path from informationto action. American Psychologist (41).1986, 521--528.

5. Drenker, S. Kader, A. "Nonintrusive Monitoring of Electric Loads:, IEEE magazine on Computer Application in Power,October 1999, pp. 47-51.

6. Froehlich, J. (2009). Promoting Energy Efficient Behaviors in the Home through Feedback: The Role of Human-ComputerInteraction. HCIC 2009 Winter Workshop, Snow Mountain Ranch, Fraser, Colorado, February 4 - 8th, 2009.

7. Froehlich, J. Findlater, L., and Landay, J. (2010) The Design of Eco-Feedback Technology. Proceedings of CHI 2010,Atlanta, Georgia, USA, April 10 - 15, 2010.

8. Geller, E. S., Winett, R. A., Everett, P. B. (1982). Preserving the Environment: New Strategies for Behavior Change. 1982Pergamon Press Inc.

9. Gupta, S., Reynolds, M.S., Patel, S.N. (2010) ElectriSense: Single-Point Sensing Using EMI for Electrical Event Detectionand Classification in the Home. Proc. of Ubiquitous Computing (UbiComp) 2010, Copenhagen, Denmark, Sep 26-29, 2010

10. Hart, G.W., Nonintrusive Appliance Load Monitoring, Proceedings of the IEEE, December 1992, pp. 1870-1891.11. Howell, E.K. How Switches Produce Electrical Noise. IEEE Transactions on Electromagnetic Compatibility 21(3), 162

170 (1979).12. Kempton, W. and Montgomery, L. Folk quantification of energy, Energy, Volume 7, Issue 10, October 1982, Pages 817-

827, ISSN 0360-5442, DOI: 10.1016/0360-5442(82)90030-5.13. Laughman, C., Lee, K., Cox, R., Shaw, S., Leeb, S., Norford, L., and Armstrong, P. Power Signature Analysis. IEEE

Power and Energy Magazine, V. 1(2), Mar-Apr 2003, Pages 56-63.


14/17

14

14. Lee, W. K., Fung, G. S. K., Lam, H. Y., Chan, F. H. Y., Lucente, M. "Exploration on Load Signatures", InternationalConference on Electrical Engineering, July 2004, Japan.

15. Leeb, S.B., Shaw, S.R., and Kirtley Jr., J.L. Transient event detection in spectral envelope estimates for nonintrusive load

monitoring, IEEE Trans. Power Delivery 10 (1995) (3), pp. 1200 1210.16. Marubayashi, G. Noise Measurements of the Residential Power Line. In the Proceedingsof International Symposium onPower Line Communications and its Applications (ISPLC), pp. 104 108 (1997)

17. Office of Energy Efficiency. 2003 Survey of household energy use detailed statistical report.18. Patel, S. N., Gupta, S., Reynolds, M. The Design and Evaluation of an End-User-Deployable Whole House, Contactless

Power Consumption Sensor. Proceedings of the SIGCHI conference on Human factors in computing systems 2010(CHI2010).

19. Patel, S. N., Robertson, T., Kientz, J., Reynolds, M., Abowd, G., At the Flick of a Switch: Detecting and ClassifyingUnique Electrical Events on the Residential Power Line. Proceedings of Ubiquitous Computing (UbiComp) 2007, pp. 271 -288.

20. Roberts M. and H. Kuhns (2010) Towards Bridging the Gap Between the Smart Grid and Smart Energy Consumption,Proceedings of the 2010 ACEEE Summer Study on Energy Efficiency in Buildings, Pacific Grove, CA, August. 12 pp.

21. Seryak J, Kissock K. Occupancy and behavioral affects on residential energy use. American Solar Energy Society, Solar

conference, Austin, Texas; 2003.22. Sidler O and Waide P (1999) Metering matters! Appliance efficiency 3 (4) 199923. Socolow, R. H., (1978). The Twin Rivers program on energy conservation in housing: Highlights and conclusions. In

Socolow, R. H. (Ed.). Saving Energy in the Home: Princetons Experiments at Twin Rive rs (pp. 2-62). Cambridge, MA:Ballinger Publishing Company.

24. Swan, Lukas G., Ugursal, V. Ismet Modeling of end-use energy consumption in the residential sector: A review ofmodeling techniques, Renewable and Sustainable Energy Reviews, Volume 13, Issue 8, October 2009, Pages 1819-1835,ISSN 1364-0321, DOI: 10.1016/j.rser.2008.09.033.

25. Weber, Christopher L., Matthews, H. Scott. Quantifying the global and distributional aspects of American householdcarbon footprint, Ecological Economics, Volume 66, Issues 2-3, 15 June 2008, Pages 379-391, ISSN 0921-8009, DOI:10.1016/j.ecolecon.2007.09.021

26. Welch, P. D. The Use of Fast Fourier Transform for the Estimation of Power Spectra: A Method Based on Time AveragingOver Short, Modified Periodograms. IEEE Trans. Audio & Electroacoust., Volume AU-15, p. 70-73. 1967

27. Winett, R. A., Neale, M. S., & Grier, H. C. (1979). The effects of self-monitoring and feedback on residential electricityconsumption: Winter. Journal of Applied Behavior Analysis, 1979, 12, 173-184.

Appendix: Note to the ReviewerAt the request of the editors, we have composed a small write-up on our water and gas sensors meant to accompanyour main article as a sidebar.

Sidebar: Water and GasIn addition to sensing disaggregated end-uses of electrical energy from a single point, our lab has also beeninvestigating the disaggregation of water and gas consumption using similar approaches. We rely on conductednoise generated as side-effects of appliance or fixture usage to identify and classify events down to their source. The

goals here are also the same: create easy-to-install sensors that provide disaggregated data to better inform residentsabout their consumption practices and enhance end-use models for utilities and policy-makers.

Like electrical energy, individual behavior also plays a significant role in water usage: residential water use accountsfor 50-80% of public water supply systems and 26% of total use in the US (Vickers, 2001). While somemunicipalities have been successful in curtailing water use, overall residential water use is still increasing in many

North American cities (Brandes, 2005). In addition, the Environmental Protection Agency estimates that over 1


15/17

15

trillion gallons of water are wasted in US households alone (EPA, 2009). To better inform residents about theirwater consumption and to provide automatic leak detection, we have developed HydroSense, a pressure-basedsensing solution capable of tracking water usage down to the fixture level from a single installation point (Froehlich

et al., 2009). HydroSense works by identifying the unique pressure wave signatures generated when fixtures areopened or closed. This pressure wave is propagated throughout the homes plumbing infrastructure, thus enablingthe single-point sensing approach. So far, we have tested HydroSense in twelve homes and three apartments by

performing upwards of 5,000 controlled experimental trials ( e.g., by repeatedly opening and closing each waterfixture in the home). Our results indicate that we can classify water usage down to the individual valve level with70-95% accuracy.

Unlike electricity and water usage, which are often the result of direct human actions such as watching TV or takinga shower, gas usage is dominated by automated systems like the furnace and hot water heater. This disconnect

between activity and consumption leads to a lack of consumer understanding about how gas is used in the home and,in particular, which appliances are most responsible for this usage (Kempton and Layne, 1994). This can lead towasteful behavior ( e.g., heating empty rooms) and using inefficient settings ( e.g., unreasonably high thermostat

settings on the hot water heater). Our gas sensing approach uses a single sensor that analyzes the acoustic responseof a home's government mandated gas regulator, which provides the unique capability of sensing both the individualappliance at which gas is currently being consumed as well as an estimate of the amount of gas flow. Our approach

provides a number of appealing features including the ability to be easily and safely installed without the need of a professional. We tested our solution in nine different homes and initial results show that GasSense has an averageaccuracy of 95% in identifying individual appliance usage.

Finally, it is interesting to note how all three resources are deeply interconnected. One of the largest uses of water isin electricity production (Thirwell et al., 2007). For example, to produce one kilowatt-hour of electricity requires140 liters of water for fossil fuels and 205 liters for nuclear power plants ( ibid ). Large amounts of energy (bothelectricity and gas) are also used to treat, pump, distribute, and heat water. Thus, our energy infrastructure isintrinsically tied to water and vice versa: water is used to produce the electricity that is used to supply consumablewater.

Sidebar References1. Brandes, O.M., At a Watershed: Ecological Governance and Sustainable Water Management in Canada. Journal of

Environmental Law and Practice, 2005. 16(1): p. 79- 97.

2. Cohn, G., Gupta, S., Froehlich, J., Larson, E., and Patel, S.N. (2010) GasSense: Appliance-Level, Single-Point Sensing ofGas Activity in the Home. In the Proceedings of Pervasive 2010, (May 17-20, Helsinki, Finland), pp. 265 282.

3. Environmental Protection Agency, WaterSense (2009). Fix a Leak. http://www.epa.gov/watersense/ pubs/fixleak.html, last accessed 03/31/2010.

4. Froehlich, J., Larson, E., Campbell, T., Haggerty, C., Fogarty, J., and Patel, S. (2009) HydroSense: Infrastructure-MediatedSingle-Point Sensing of Whole-Home Water Activity. Proceedings of UbiComp 2009, Orlando, Florida, September 30th October 3rd, 2009.

5. Kempton, W. and Layne, L .: The Consumers Energy Analysis Environment. Energy Policy, vol. 22, issue 10, pp. 857 --866 (1994)

6. Thirwell, G. M., Madramootoo, C. A., Heathcote, I. W., Energy-water Nexus: Energy Use in the Municipal, Industrial, andAgricultural Water Sectors. Canada US Water Conference, Washington D.C., October 2 nd, 2007.

7. Vickers, A. (2001). Handbook of Water Use and Conservation: Homes, Landscapes, Industries, Businesses, Farms(Hardcover), WaterPlow Press; 1 edition (August 1, 2001)


16/17

16

Author BiosJon Froehlich is a PhD candidate and Microsoft Research Graduate Fellow in HumanComputer Interaction (HCI) and Ubiquitous Computing (UbiComp) at the University ofWashington co-advised by James Landay and Shwetak Patel. He was recently selected asthe UW College of Engineering Graduate Student Innovator of the Year for 2010. Hisdissertation is on promoting sustainable behaviors through automated sensing andfeedback technology. Jon received his MS in Information and Computer Sciencespecializing in analytic visualizations from the University of California, Irvine under theadvisement of Paul Dourish. Contact him at [email protected] .

Eric Larson is a senior PhD student in the laboratory of Ubiquitous Computing at theUniversity of Washington and is advised by Shwetak Patel. His main research area issupported signal processing in ubiquitous computing applications, especially forhealthcare and environmental sustainability applications. He received his B.S. and M.S.degrees in electrical engineering from Oklahoma State University, Stillwater in 2006 and2008 respectively where he specialized in image processing and perception under theadvisement of Damon Chandler. He is active in signal processing education and is amember of IEEE and HKN. Contact him at [email protected] .

Sidhant Gupta is a PhD student at University of Washington's Computer Science &Engineering department specializing in Ubiquitous Computing. His current researchfocuses on developing novel sensing technologies for the home that are use minimalsensors and building innovative electro-mechanical haptic feedback interfaces. Sidhantgraduated with a M.Sc degree in computer science from Georgia Institute of Technologyin 2009 specializing in OS and embedded systems. Contact him at [email protected] .

Gabe Cohn is a Ph.D. student in the Ubiquitous Computing Lab at the University ofWashington. He received his B.S. degree in electrical engineering from the CaliforniaInstitute of Technology in 2009, where he specialized in embedded systems and digitalVLSI. He is currently pursuing his PhD in electrical engineering at the University ofWashington in Seattle and is advised by Shwetak Patel. His research involves designingcustomized hardware for ubiquitous computing applications. In particular, he is focused

on enabling new ultra-low-power sensing solutions for the home, and creating interestingnew human-computer interfaces. He is a member of the IEEE and ACM. Contact him [email protected] ; www.gabeacohn.com .
mailto:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]://www.gabeacohn.com/http://www.gabeacohn.com/http://www.gabeacohn.com/http://www.gabeacohn.com/mailto:[email protected]:[email protected]:[email protected]:[email protected]


17/17

17

Matthew S. Reynolds is an assistant professor at Duke University's Department ofElectrical and Computer Engineering. His research interests include the physics ofsensors and actuators, RFID, and signal processing. He received the S.B. (1998), M.Eng.(1999), and Ph.D. (2003) degrees from the Massachusetts Institute of Technology and is aco-founder of the RFID systems firm ThingMagic Inc and the energy conservation firmUSenso, Inc. He is a member of the Signal Processing and Communications andComputer Engineering Groups at Duke, as well as the IEEE Microwave Theory andTechniques Society. Contact him at [email protected] .

Shwetak N. Patel is an Assistant Professor in the departments of Computer Science andEngineering and Electrical Engineering at the University of Washington. His researchinterests are in the areas of Sensor-enabled Embedded Systems, Human-ComputerInteraction, Ubiquitous Computing, and User Interface Software and Technology. He is

particularly interested in developing easy-to-deploy sensing technologies and approachesfor location, activity recognition, and energy monitoring applications. Dr. Patel was alsothe co-founder of Usenso, Inc., a recently acquired demand side energy monitoringsolutions provider. He received his Ph.D. in Computer Science from the Georgia Institute

of Technology in 2008 and B.S. in Computer Science in 2003. He was also the Assistant Director of the AwareHome Research Initiative at Georgia Tech. Dr. Patel was a TR-35 award recipient in 2009. Contact him [email protected] .
mailto:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]:[email protected]

Disintegration of Energy

Documents

Transcript of Disintegration of Energy