PROFILING RESIDENTIAL WATER USERS’ ROUTINES BY EIGENBEHAVIOR MODELLING
-
Upload
smarth2o -
Category
Engineering
-
view
38 -
download
1
Transcript of PROFILING RESIDENTIAL WATER USERS’ ROUTINES BY EIGENBEHAVIOR MODELLING
PROFILING RESIDENTIAL WATER USERS’ ROUTINES BY EIGENBEHAVIOR MODELLING
Andrea Cominola, Andrea Moro, Luca Riva, Matteo Giuliani, Andrea Castelletti
RESIDENTIAL URBAN WATER MANAGEMENT
. Demand-side management
Long term water security
Source: United Nations. Department of Economic and Social Affairs. Population Division, 2010 Leflaive, X., et al. (2012), "Water", in OECD, OECD Environmental Outlook to 2050: The Consequences of Inaction, OECD Publishing, Paris.
2000 2030 2050
+130%
Dom
estic
water
de
man
d
41 megacities worldwide
Short term costs
Interventions during low recharge periods
RESIDENTIAL URBAN WATER MANAGEMENT
. Demand-side management
. Smart metering
1990 1994
50
30
10
1995 1999
2000 2004
2005 2009
2010 2015
134 studies over the last 25 years
Benefits and challenges of using smart meters for advancingresidential water demand modeling and management: A review
A. Cominola a, M. Giuliani a, D. Piga b, A. Castelletti a, c, *, A.E. Rizzoli da Department of Electronics, Information, and Bioengineering, Politecnico di Milano, Milan, Italyb IMT Institute for Advanced Studies Lucca, Lucca, Italyc Institute of Environmental Engineering, ETH Zurich, Zurich, Switzerlandd Istituto Dalle Molle di Studi sull'Intelligenza Artificiale, SUPSI-USI, Lugano, Switzerland
a r t i c l e i n f o
Article history:Received 2 April 2015Received in revised form21 July 2015Accepted 21 July 2015Available online xxx
Keywords:Smart meterResidential water managementWater demand modelingWater conservation
a b s t r a c t
Over the last two decades, water smart metering programs have been launched in a number of mediumto large cities worldwide to nearly continuously monitor water consumption at the single householdlevel. The availability of data at such very high spatial and temporal resolution advanced the ability incharacterizing, modeling, and, ultimately, designing user-oriented residential water demand manage-ment strategies. Research to date has been focusing on one or more of these aspects but with limitedintegration between the specialized methodologies developed so far. This manuscript is the firstcomprehensive review of the literature in this quickly evolving water research domain. The papercontributes a general framework for the classification of residential water demand modeling studies,which allows revising consolidated approaches, describing emerging trends, and identifying potentialfuture developments. In particular, the future challenges posed by growing population demands, con-strained sources of water supply and climate change impacts are expected to require more and moreintegrated procedures for effectively supporting residential water demand modeling and management inseveral countries across the world.
© 2015 Elsevier Ltd. All rights reserved.
1. Introduction
World's urban population is expected to raise from current54%e66% in 2050 and to further increase as a consequence of theunlikely stabilization of human population by the end of the cen-tury (Gerland et al., 2014). By 2030 the number of mega-cities,namely cities with more than 10 million inhabitants, will growover 40 (UNDESA, 2010). This will boost residential water demand(Cosgrove and Cosgrove, 2012), which nowadays covers a largeportion of the public drinking water supply worldwide (e.g.,60e80% in Europe (Collins et al., 2009), 58% in the United States(Kenny et al., 2009)).
The concentration of the water demands of thousands or mil-lions of people into small areas will considerably raise the stress onfinite supplies of available freshwater (McDonald et al., 2011a).Besides, climate and land use change will further increase the
number of people facingwater shortage (McDonald et al., 2011b). Insuch context, water supply expansion through the construction ofnew infrastructures might be an option to escape water stress insome situations. Yet, geographical or financial limitations largelyrestrict such options in most countries (McDonald et al., 2014).Here, acting on the water demand management side through thepromotion of cost-effective water-saving technologies, revisedeconomic policies, appropriate national and local regulations, andeducation represents an alternative strategy for securing reliablewater supply and reduce water utilities' costs (Gleick et al., 2003).
In recent years, a variety of water demand management stra-tegies (WDMS) has been applied (for a review, see Inman andJeffrey, 2006, and references therein). However, the effectivenessof these WDMS is often context-specific and strongly depends onour understanding of the drivers inducing people to consume orsave water (Jorgensen et al., 2009). Models that quantitativelydescribe how water demand is influenced and varies in relation toexogenous uncontrolled drivers (e.g., seasonality, climatic condi-tions) and demand management actions (e.g., water restrictions,pricing schemes, education campaigns) are essential to explorewater users' response to alternative WDMS, ultimately supporting
* Corresponding author. Department of Electronics, Information, and Bioengi-neering, Politecnico di Milano, Milan, Italy.
E-mail address: [email protected] (A. Castelletti).
Contents lists available at ScienceDirect
Environmental Modelling & Software
journal homepage: www.elsevier .com/locate/envsoft
http://dx.doi.org/10.1016/j.envsoft.2015.07.0121364-8152/© 2015 Elsevier Ltd. All rights reserved.
Environmental Modelling & Software 72 (2015) 198e214
Benefits and challenges of using smart meters for advancingresidential water demand modeling and management: A review
A. Cominola a, M. Giuliani a, D. Piga b, A. Castelletti a, c, *, A.E. Rizzoli da Department of Electronics, Information, and Bioengineering, Politecnico di Milano, Milan, Italyb IMT Institute for Advanced Studies Lucca, Lucca, Italyc Institute of Environmental Engineering, ETH Zurich, Zurich, Switzerlandd Istituto Dalle Molle di Studi sull'Intelligenza Artificiale, SUPSI-USI, Lugano, Switzerland
a r t i c l e i n f o
Article history:Received 2 April 2015Received in revised form21 July 2015Accepted 21 July 2015Available online xxx
Keywords:Smart meterResidential water managementWater demand modelingWater conservation
a b s t r a c t
Over the last two decades, water smart metering programs have been launched in a number of mediumto large cities worldwide to nearly continuously monitor water consumption at the single householdlevel. The availability of data at such very high spatial and temporal resolution advanced the ability incharacterizing, modeling, and, ultimately, designing user-oriented residential water demand manage-ment strategies. Research to date has been focusing on one or more of these aspects but with limitedintegration between the specialized methodologies developed so far. This manuscript is the firstcomprehensive review of the literature in this quickly evolving water research domain. The papercontributes a general framework for the classification of residential water demand modeling studies,which allows revising consolidated approaches, describing emerging trends, and identifying potentialfuture developments. In particular, the future challenges posed by growing population demands, con-strained sources of water supply and climate change impacts are expected to require more and moreintegrated procedures for effectively supporting residential water demand modeling and management inseveral countries across the world.
© 2015 Elsevier Ltd. All rights reserved.
1. Introduction
World's urban population is expected to raise from current54%e66% in 2050 and to further increase as a consequence of theunlikely stabilization of human population by the end of the cen-tury (Gerland et al., 2014). By 2030 the number of mega-cities,namely cities with more than 10 million inhabitants, will growover 40 (UNDESA, 2010). This will boost residential water demand(Cosgrove and Cosgrove, 2012), which nowadays covers a largeportion of the public drinking water supply worldwide (e.g.,60e80% in Europe (Collins et al., 2009), 58% in the United States(Kenny et al., 2009)).
The concentration of the water demands of thousands or mil-lions of people into small areas will considerably raise the stress onfinite supplies of available freshwater (McDonald et al., 2011a).Besides, climate and land use change will further increase the
number of people facingwater shortage (McDonald et al., 2011b). Insuch context, water supply expansion through the construction ofnew infrastructures might be an option to escape water stress insome situations. Yet, geographical or financial limitations largelyrestrict such options in most countries (McDonald et al., 2014).Here, acting on the water demand management side through thepromotion of cost-effective water-saving technologies, revisedeconomic policies, appropriate national and local regulations, andeducation represents an alternative strategy for securing reliablewater supply and reduce water utilities' costs (Gleick et al., 2003).
In recent years, a variety of water demand management stra-tegies (WDMS) has been applied (for a review, see Inman andJeffrey, 2006, and references therein). However, the effectivenessof these WDMS is often context-specific and strongly depends onour understanding of the drivers inducing people to consume orsave water (Jorgensen et al., 2009). Models that quantitativelydescribe how water demand is influenced and varies in relation toexogenous uncontrolled drivers (e.g., seasonality, climatic condi-tions) and demand management actions (e.g., water restrictions,pricing schemes, education campaigns) are essential to explorewater users' response to alternative WDMS, ultimately supporting
* Corresponding author. Department of Electronics, Information, and Bioengi-neering, Politecnico di Milano, Milan, Italy.
E-mail address: [email protected] (A. Castelletti).
Contents lists available at ScienceDirect
Environmental Modelling & Software
journal homepage: www.elsevier .com/locate/envsoft
http://dx.doi.org/10.1016/j.envsoft.2015.07.0121364-8152/© 2015 Elsevier Ltd. All rights reserved.
Environmental Modelling & Software 72 (2015) 198e214
RESIDENTIAL URBAN WATER MANAGEMENT
. Demand-side management
. Smart metering
. Water consumers
DEMAND MANAGEMENT STRATEGIES
. Technological
. financial
. legislative
. operation and maintenance
. education
RESIDENTIAL URBAN WATER MANAGEMENT
. Demand-side management
. Smart metering
. Water consumers
CUSTOMIZED DEMAND SIDE MANAGEMENT
5% reduction through customized social norms
CHALLENGES
. Smart metered BIG DATA
. How to extract RELEVANT USERS PROFILES out of large smart
metered datasets?
. How to characterize these profiles, in order to inform
DEMAND MANAGEMENT?
CUSTOMER SEGMENTATION by eigenbehavior modelling
Eigenbehaviorextraction
Water consumers clustering
Clusters characterization
CUSTOMER SEGMENTATION by eigenbehavior modelling
Eigenbehaviorextraction
Water consumers clustering
Clusters characterization
175 water consumers
Hourly resolution water consumption data
7 months monitoring periodMarch – November 2015
CASE STUDY
LOCARNO | CH
CUSTOMER SEGMENTATION by eigenbehavior modelling
Eigenbehaviorextraction
ROUTINEof each user
HOUSEHOLD WATER CONSUMPTION
Eagle, Nathan, and Alex Sandy Pentland. "Eigenbehaviors: Identifying structure in routine."Behavioral Ecology and Sociobiology 63.7 (2009): 1057-1066.
EIGENBEHAVIOR EXTRACTION
SMART METERED DATA
H (hour of day)
DATA LABELING
D (d
ays)
H (hour of day)
D (d
ays)
Label 0Label 1Label 2Label 3
0.0
0.1
0.2
0.3
0.4
0.5
L0 L1 L2 L3
% o
f rea
ding
sLO = 0 L/hL1 = (0,12] L/hL2 = (12,100] L/hL3 > 100 L/h
EIGENBEHAVIOR EXTRACTION
SMART METERED DATA
H (hour of day)
DATA LABELING
D (d
ays)
H (hour of day)
D (d
ays)
Label 0Label 1Label 2Label 3
BINARY TRANSFORMATION
H (hour of day)
D (d
ays)
H (hour of day) H (hour of day)
Label 1 Label 2 Label 3
H (hour of day)
Label 0
EIGENBEHAVIOR EXTRACTION
SMART METERED DATA
H (hour of day)
DATA LABELING
D (d
ays)
H (hour of day)
D (d
ays)
Label 0Label 1Label 2Label 3
BINARY TRANSFORMATION
D (d
ays)
PRINCIPAL COMPONENT ANALYSIS
Dimensionality reduction
1st EIGENBEHAVIOR – USER ROUTINE
users
0
230
230
230
23
L0
L1
L2
L3
profile P1 profile P2 profile P3
0.00 0.05 0.10 0.15 0.20 0.25
users
0 23 0 23 0 23 0 23
L0 L1 L2 L3
profi
le P
1pr
ofile
P2
profi
le P
3
0.00
0.05
0.10
0.15
0.20
0.25
0
0.25
users
0
230
230
230
23
L0
L1
L2
L3
profile P1 profile P2 profile P3
0.00 0.05 0.10 0.15 0.20 0.25
users
0
230
230
230
23
L0
L1
L2
L3
profile P1 profile P2 profile P3
0.00 0.05 0.10 0.15 0.20 0.25
1st eigenbehaviorloadings
CUSTOMER SEGMENTATION by eigenbehavior modelling
Eigenbehaviorextraction
ROUTINE of each user
HOUSEHOLD WATER CONSUMPTION
Eagle, Nathan, and Alex Sandy Pentland. "Eigenbehaviors: Identifying structure in routine."Behavioral Ecology and Sociobiology 63.7 (2009): 1057-1066.
Water consumers clustering
Clusters characterization
Users community profiles
1st EIGENBEHAVIOR – USER ROUTINE
users
0
230
230
230
23
L0
L1
L2
L3
profile P1 profile P2 profile P3
0.00 0.05 0.10 0.15 0.20 0.25
users
0 23 0 23 0 23 0 23
L0 L1 L2 L3
profi
le P
1pr
ofile
P2
profi
le P
3
0.00
0.05
0.10
0.15
0.20
0.25
0
0.25
1st eigenbehaviorloadings
CHALLENGES
. Smart metered BIG DATA
-> PCA
. How to extract RELEVANT USERS PROFILES out of large smart
metered datasets?
-> Eigenbehavior extraction and clustering
. How to characterize these profiles, in order to inform
DEMAND MANAGEMENT?
-> Profiles characterization and factor mapping
thank you
August 22-25 Monte Verità, Switzerland.More Info: www2.idsia.ch/cms/smartwater/
www.smarth2o-fp7.eu
@smartH2Oproject #SmartH2O
@AndreaCominola@NRMPolimi
Andrea [email protected]
Politecnico di MilanoDepartment of Electronics, Information and Bioengineering