SLIDE 1
Nipun Batra, Amarjeet Singh, Kamin Whitehouse
14 May 2016
Exploring The Value of Energy Disaggregation through actionable feedback
1
General eco feedback vs Actionable Feedback
Eco feedback
Misc. 22% Light 10% Fridge 11% HVAC 56%
Exploring The Value of Energy Disaggregation through actionable - - PowerPoint PPT Presentation
Exploring The Value of Energy Disaggregation through actionable - - PowerPoint PPT Presentation
Exploring The Value of Energy Disaggregation through actionable feedback Nipun Batra , Amarjeet Singh, Kamin Whitehouse 14 May 2016 1 General eco feedback vs Actionable Feedback Eco feedback Misc. 22% Light HVAC 10% 56% Fridge 11%
SLIDE 2
SLIDE 3
General eco feedback vs Actionable Feedback
Eco feedback
Misc. 22% Light 10% Fridge 11% HVAC 56%
SLIDE 4
General eco feedback vs Actionable Feedback
Eco feedback
Misc. 22% Light 10% Fridge 11% HVAC 56%
Power (W)
175 350 525 700
Power (W)
175 350 525 700 Home 1 Home 2
Actionable feedback
Fridge consumption over 24 hours
Misc. 22% Light 10% Fridge 11% HVAC 56%
SLIDE 5
General eco feedback vs Actionable Feedback
Eco feedback
Misc. 22% Light 10% Fridge 11% HVAC 56%
Power (W)
175 350 525 700
Power (W)
175 350 525 700 Home 1 Home 2
Actionable feedback
Fridge consumption over 24 hours
Misc. 22% Light 10% Fridge 11% HVAC 56%
High power state
SLIDE 6
General eco feedback vs Actionable Feedback
Eco feedback
Misc. 22% Light 10% Fridge 11% HVAC 56%
Power (W)
175 350 525 700
Power (W)
175 350 525 700 Home 1 Home 2
Actionable feedback
Fridge consumption over 24 hours
Misc. 22% Light 10% Fridge 11% HVAC 56%
High power state High power state
SLIDE 7
General eco feedback vs Actionable Feedback
Eco feedback
Misc. 22% Light 10% Fridge 11% HVAC 56%
Power (W)
175 350 525 700 Home 2
Actionable feedback
Fridge consumption over 24 hours
Your fridge defrosts too much, wasting 30% energy
Misc. 22% Light 10% Fridge 11% HVAC 56%
SLIDE 8
Approach overview- How to give feedback
Power (W)
175 350 525 700
Specific features of trace to infer why energy usage is high
Length of duty cycles
SLIDE 9
Approach overview- How to give feedback
Power (W)
175 350 525 700
Specific features of trace to infer why energy usage is high
Actual power value
SLIDE 10
Feedback methods on Fridge and HVAC
Both appliances commonly found across homes
Others 38% Fridge 8% HVAC 54%
SLIDE 11
Evaluation overview
Submetered traces
Power (W)
350 700
Submeter sensor
SLIDE 12
Can we give such feedback?
Disaggregated traces
Power (W)
350 700
NILM
Household aggregate
Submetered traces
Power (W)
350 700
Submeter sensor
2000 4000
2000 4000
Smart meter
SLIDE 13
Do disaggregated traces provide features needed for providing feedback?
Disaggregated traces
Power (W)
350 700
NILM
Household aggregate
Submetered traces
Power (W)
350 700
Submeter sensor
2000 4000
2000 4000
Smart meter
SLIDE 14
125 250 375 500
Fridge is a duty cycle based appliance; compressor turns ON and OFF periodically
SLIDE 15
Defrost cycles occurs periodically and consume high amount of power
125 250 375 500
SLIDE 16
125 250 375 500
Defrost introduces heat increasing ON duration of next cycles
SLIDE 17
Fridge usage increases compressor ON durations (and reduce compressor OFF durations)
125 250 375 500
SLIDE 18
Night hours typically have “baseline” usage
175 350 525 700
Baseline duty % = Median duty % in the night
SLIDE 19
Defrost energy
175 350 525 700
Defrost energy = Energy consumed in defrost state + Extra energy consumed in next few compressor cycles
SLIDE 20
Defrost energy
175 350 525 700
Defrost energy = Energy consumed in defrost state + Extra energy consumed in next few compressor cycles
SLIDE 21
Usage energy
175 350 525 700
Usage energy = Extra energy consumed over baseline
SLIDE 22
Experimental setup
Wiki Energy data set
- 1. 97 fridges
- 2. 58 HVAC
SLIDE 23
13 out of 95 homes can be given feedback based on usage energy saving upto 23% fridge energy
SLIDE 24
13 out of 95 homes can be given feedback based on usage energy saving upto 23% fridge energy
NILM algorithms show poor accuracy in identifying homes which can be given feedback based on usage energy
SLIDE 25
17 out of 95 homes can be given feedback on excess defrost saving upto 25% fridge energy
SLIDE 26
Such feedback can’t be given with disaggregated traces, since these techniques fare poorly on defrost detection.
SLIDE 27
Benchmark NILM algorithms on our data set give accuracy comparable or better than state-of-the-art
Kolter 2012 Parson 2012 Parson 2014 Batra 2014 CO FHMM Hart Error Energy % 17.5 35 52.5 70
SLIDE 28
125 250 375 500
“Average” error in energy would be low even if NILM predicted this
SLIDE 29
But, we wanted to predict..
125 250 375 500
SLIDE 30
It’s the details that we care about
125 250 375 500
SLIDE 31
Like fridge, HVAC duty cycles to maintain the set temperature
1000 2000 3000 4000
SLIDE 32
As temperature increases during the day, more energy required to cool the home
1000 2000 3000 4000
SLIDE 33
1000 2000 3000 4000
People typically turn up the temperatures when they leave home
SLIDE 34
Recommended
- Min. Temp (F)
77 79 81 83 85 2 4 6 8 10 12 14 16 18 20 22 24
EnergyStar.gov recommended HVAC setpoint schedule
Sleep Morning Work Evening
SLIDE 35
Recommended
- Min. Temp (F)
77 79 81 83 85 2 4 6 8 10 12 14 16 18 20 22 24
Setpoint schedule score
Sleep Morning Work Evening
SLIDE 36
Recommended
- Min. Temp (F)
77 79 81 83 85 2 4 6 8 10 12 14 16 18 20 22 24
Setpoint schedule score
Sleep
Sleep score = 1 if sleep temp. > 82, (82-temp.)/4 if 78<sleep temp. <82 0 otherwise
SLIDE 37
Learning HVAC setpoint
1000 2000 3000 4000
77 85 5 1015 20
HVAC trace Weather Learnt setpoint
SLIDE 38
Giving feedback
77 85 5 1015 20
Features from HVAC trace
69 75 5 10 15 20
1000 2000 3000 4000
77 85 5 10 15 20
Learnt setpoint Don’t need feedback Need feedback
SLIDE 39
84% accuracy on giving feedback using submetered traces
39
SLIDE 40
NILM methods give 15-30% worse accuracy for feedback
40
SLIDE 41
Benchmark NILM algorithms on our data set give accuracy comparable or better than state-of-the-art
Batra 2014 CO FHMM Hart Error Energy % 7.5 15 22.5 30
SLIDE 42
Error in prediction of minutes of HVAC usage (%)
6 12 18 24 Hart FHMM CO
Night Morning
Morning hours which have lesser NILM accuracy are important for HVAC feedback
SLIDE 43
Conclusions
Appliance level data does enable actionable energy saving feedback
SLIDE 44