Slavco Velickov 0

Improving Efficiency of Water
Systems: practical examples

May 2012
Dr.ir. Slavco Velickov

Water Industry Director EMEA
Agenda
1. Bentley at a Glance
2. Water Solutions Overview
3. Trends in the Water Industry
4. Practical Cases:
1. Active Leakage Management
2. Lifecycle Asset Management
3. Energy Efficiency Improvement
5. Contact Information & Resources
2 | WWW.BENTLEY.COM
Bentley at a glance
Revenues
Revenues by
by Region
Region
3 | WWW.BENTLEY.COM
Power
Solutions Generation
Bridges Rail and

Transit
Buildings Communications
Cadastre and Land Electric and

Development Gas Utilities
Campuses Factories Roads
Metals and Water and

Mining Wastewater
Oil and Gas
4 | WWW.BENTLEY.COM
The World of Water
Estuaries and coastal

water systems
Rural water systems Urban water systems
5 | WWW.BENTLEY.COM
Water Industry Solutions
AutoCAD ArcGIS
Industry Framework Modeling Framework
sisNET SewerGEMS / CivilStorm / WaterGEMS

sisNET
Expert Water CAD PondPack /CAD Web
WWater
(Bentley Designer (Bentley StormCAD / clients
WWater) Water) Hammer GasAnalysis
HEC-Pack
GIS products Modelling products

Web
MicroStation
Publishing
GeoSpatial Server & AssetWise

Interoperability
Interoperability Enterprise
Enterprise
Connectors
Connectors Connectors
Connectors
Data Files Spatial Web Spatial Business Ancillary Files Proprietary Enterprise SCADA &
Data Files
w/ Database Linkages Databases Services Documents Documents w/ RDBWS GIS Databases Data Stores Loggers
Water & Wastewater Industry
Challenges
• Regulatory Compliance
–Adequate Supply & Treatment capacity
–Protecting Water Quality
–Business performance
–Improving efficiency
• Reliability
–Consistently achieving target levels of services
–Maintaining aging infrastructure
–Avoiding failure
• Budget
–Reducing costs while improving services
–Asset investment planning for aging infrastructure
–Aging workforce
7 | WWW.BENTLEY.COM
The Evolution of Smart Water
Networks
8 | WWW.BENTLEY.COM
The Smart Water Network - Integration
Hydraulic Enterprise
Models GIS
system
Online SCADA
model system
Planning and CRM, WO

Demand Call Centre
forecasting
9 | WWW.BENTLEY.COM
1) Leakage Reduction Case
Studies by pressure management,
hydraulic modelling, measured
data and optimization techniques
10 | WWW.BENTLEY.COM
A Worldwide Problem: Controlling and
Remediating Water Loss Is Complex
• It’s impossible to find and fix all leaks

• Partial implementation of a water loss plan is highly likely to fail
• Coordination between all components of a water loss program
is required
“ Many practitioners make common

mistakes- they may have the false
impression that each time a leak is
repaired, physical loss is reduced by
the volume saved…”
Vermersch and Rizzo
Source: IWA’s Water21 Magazine, April 2008
(Courtesy Dr. Thomas Walski)
11 IWA = International Water Association

IWA Standard Water Balance
Billed Billed Metered Consumption Revenue

Authorized
Authorized Water
Consumption Billed Unmetered Consumption
Consumption
Unbilled Unbilled Metered Consumption
Authorized
Consumption Unbilled Unmetered Consumption
System Unauthorized Consumption
Apparent
Input Non
Losses
Customer Meter Inaccuracies
Volume Revenue
Water Leakage on Transmission & Water
Losses Distribution Mains
Real
Losses Leakage and Overflows at
Reservoirs
Leakage on Service Connections
up to metering point
Leakage Types
• Background leakage
– Small flow rates, run continuously but not economically recoverable
• Reported leaks and bursts

– High flow, reported by customers and get fixed quickly
• Unreported leaks and bursts

– Medium flow rates, longest duration and only located by active leakage
detection
Strategy: A Long-term Approach with
Immediate [short-term] Benefits
Implement IWA best / good practices
Management
Pressure
Unavoidable
Real Loss
Speed and Quality Economic Active

of Repairs Level Real Loss Leakage Control
Replacing pipes with Current Annual Detecting and fixing leaks
Real Loss Volume
least impact on Replacing/installing meters
customers (DMAs)
Infrastructure
Management
Risk-based asset management for

maximum return
14 Source: The “4 Component” diagram promoted by IWA’s Water Losses Task Force
Current Practice
1. Assessment
– water balance or water auditing
based upon water
infrastructures’ physical data and
some statistics
2. Pressure Management
– Divide the network in Pressure
Zones and DMAs
– Use hydraulic model for PRVs
– Install PRVs to manage MNF
3. Active Leakage Detection

– Sounding for leaks
– Step-testing, smart balls
– Acoustic loggers (noise correlators)
– Smart balls
– Use hydraulic model and measured
(Scada) data
Pressure Management
(Case Study in Cyprus)
Water Board of Lemesos
Established in 1951
Semi-government, non-profit organisation
Supply of potable water
Number of employees : 110
Area served : 100 km2
Population Served : 170 000
Annual water needs : 14 million m3
Number of consumers : 78 000
Length of pipework : 850 km
17
Pressure Management Considerations
Pressure range:
Between 20m and 40m
Minimum of 15 m if conditions allowed
Pressure control achieved through:

Pressure reducing valves
Pressure regulating valves
Types of pressure control:

Fixed outlet
Two point control (Time or Flow)
Flow / Time modulation
Critical point control
18
Pressure Management Objectives
Pressure reducing valve
(downstream pressure control, open/close capability)
Before pressure reduction
After pressure reduction
Pressure sensor
(downstream pressure monitoring)
Strainer
District meter (meter protection)
(mechanical “Woltman” type)
Reduction
in MNF
Reduce the flow rates (MNF) of all leaks

Reduce surges and excess pressures
Reduce burst rates and background leakage, cut repair costs
Reduce some components of consumption
Effects of change can be hydraulically modelled and predicted
19
DMAs - Pressure Management
DMA categories
Small : <1000 properties
Medium : 1000 – 3000 properties
Large : 3000 – 5000 properties
Factors considered in DMAs (re)-design

Use Hydraulic Model (WaterCAD)
Minimum variation in ground level
Single entry point into the DMA
Well defined DMA boundaries
Area meters correctly sized and located
Continuous monitoring
20
Pressure Reduction (1/3)
DMA 232
before pressure reduction
Possible pressure reduction of about 0,5 bar
21
DMA 232
after pressure reduction
Reduction in AZP from 39 m to 32 m resulted in

reduction in MNF from 5.2 m3/hr to 4.3 m3/hr
22
Before pressure reduction
After pressure reduction
Reduction
in MNF
23
Minimum Night Flow Summary Results
DMA AZNP (m) Actual MNF Background Locatable losses
(Sector 2) (m³/hr) losses(m³/hr) (m³/hr)
before after before after before after before after

220 64 32 3,92 2,16 0,63 0,24 1,88 0,51
221 63 36 5,69 3,85 3,39 1,65 0,16 0,07
222
223
REDUCTION
54
53
28
29
3,07
3,58
2,24
2,56
30 m3/hr
1,53
1,70
0,71
0,82
0,05
0,35
0,03
0,20
224
225
53
64
29
34
(Sector 2)
5,50
12,96
2,52
9,78
1,68
5,42
(25%) 0,82
2,41
2,23
4,16
0,11
3,99
226 64 34 10,04 6,84 5,62 2,55 0,37 0,24
227 59 38 15,52 10,44 5,91 3,38 5,11 2,56
Annual water saving = 220 000

228
229
43
41
39
36
7,60
4,06
7,20
3,73
3,42
1,13
3,03
0,96
0,51
2,01
m3 0,50
1,85
230
231
47
52
or € 170 000
40
42
21,80
11,01
18,00
7,92
5,57
4,63
4,60
3,54
9,37
2,17
6,54
0,18
232 39 32 5,17 4,32 1,32 1,05 2,21 1,63
233 42 33 4,45 3,96 1,48 1,10 1,48 1,37
234 48 38 3,55 2,44 0,32 0,23 2,26 1,24
Total before 117,92 43,75 34,32

Total after 87,96 27,09 21,02
24
Leakage Detection using Models
and Data
Mathematical Optimization Techniques
r
Search for: X = (LNin , Kin ); LNin ∈Jn ; n = 1,...,N; i =1,...,LKn
r
Minimize: F(X )
Subject to: 0 ≤ Kin ≤ Kmax
n
N
Leakage Nodes: nL= ∑LKn
n=1
n
Where: LN i
is the index for leakage node i in group n
n
K i is emitter coefficient at leakage node i in group n
Jn is the set of junctions in demand group n
N is the number of demand groups
nL is the total number of leakage nodes
LK n is number of leakage nodes in group n

n
K max is the max emitter coefficient for demand group n
Integrated Framework  Leakage Detection
& Model Calibration
WaterGEMS (Darwin Calibrator)
Case I: system conditions (United Utilities)
• DMA system model

• 12 km pipelines
• 1000 properties
• 5 pressure loggers and one flow meter
Case Study I: previously detection
Burst A
aA Burst B
aA
KEY
DMA Boundary
BURST A
Distance <50m
Leak located from
prediction
Mains 150mm
diam
BURST B
Distance 150m
from
prediction
Mains 8”
diam
Case I: results comparison
Burst A & B
Case I: savings
Burst A
Burst B
Saving > 210,000
Euro / year
Raw Measured Value for 08016DM01_02 WALTON SUMMIT (1 Apr 2004 - 1 Jul 2004)
240
220
200
180
cub. m/hr
160
140
120
100
30m3/hr reduction
80
60
Apr 01 Apr 15 Apr 29 May 13 May 27 Jun 10 Jun 24
Case Study I: flow comparison (after)
Case II: another DMA
Field survey
Actual
Leakage
Leakage spots
identified with
WaterGEMS
Leakage Calibrator
15 m3/hr reduction
Savings of 115,000
Euro / year
Important: Check Risk on Transients !
Essential Requirements
• Build hydraulic model

• Collect field data (e.g. flows and pressures)
• Prepare and import data
• Calibrate the model and make leakage detection
runs
• Analyze results
• Look for consistent predicted leakage hotspots
• Go to the filed: check the identified locations
and take a proper acction
2) GIS-based AM using Bentley
Water
Bentley Utility Products
• Built on Bentley Map

• Include
– Bentley Water
– Bentley WasteWater
– Bentley Gas
– Bentley Electric
– Bentley Fiber
– Bentley sisNET (multi-utility solution)
– …
• Learn one – learn them all

What Does Bentley Water Do?
• Water GIS product with CAD power/precision
• Built on Bentley Map / MicroStation
• Manages & maps water infrastructure assets for
mapping, inventory and analysis
• Customized for water networks
• Dozens of elements w/predefined properties
• Can define your own elements
• Can attach any kind of data/files
Bentley Water Features
• Customizable data model
– Water has its own data model
– Can define/remove new types of elements
– Can define/remove properties of elements
• Compatible with Oracle Spatial or DGN/RDBMS
• Ensures connectivity
• Network tracing
– Segments for shutdown
– Pressure zone management
– Define your own “stop” conditions
• Leakage and breaks analysis, thematic GIS mapping
• Native export to WaterGEMS / EPANET for hydraulic
modeling, optimization and analysis
Example of IZSU: Izmir Water
Utility in Turkey
• City with 4,000,000 population at the Aegean
Cost of Turkey
• Total water customer served by IZSU is
1,200,000
• Water/Waste Water/Storm network in 5000 km
road network
• All structures related within City Limits
• 619 Resorvoirs (Dam, Pond and Wells)
• 426 Tanks
• 135 Pumps
• 11,000 Valve
– 10% of the data was in papers
– 90% of the data was in different digital formats
(ncz, plt, dxf, dwg, dgn)
Mapping: Thin client
Mapping: Web client
Manage Leak Records:
• Most utilities keep leak records
• Many forms
– Paper records
– Databases
– Spreadsheets
– Shapefiles
– Work orders
• Import to Bentley Water

• Need x-y coordinates (georeference)
Spatially View Leak Locations:
Find Bad Pipes:
Analyze Patterns
Diameter, Breaks Break Rate,

in. break/yr/km
6 25 0.105
8 15 0.082
12 8 0.062
16 2 0.041
24 3 0.056
Look for Relationships
Circumferential Longitudinal Corrosion
breaks breaks holes
Cast Iron 73 7 4
Ductile 12 2 5
iron
PVC 23 17 0
Steel 2 1 12
Pipe Renewal Planner Workflow
Pipe Score System Inventory
Pipe Break Model Other Property

History Of Interest
Break Fire Flow Criticality Analysis

Analysis Analysis Analysis
Normalized Normalized Normalized Normalized

Break Score Fire Score Criticality Score Score
Weighting
Overall
Score
Pipe Renewal Planner Results
Pipe Investment Plan
Part of risk-based Asset Management
• Pipe break and leak history feeds into asset
management decision making
• Rational, quantifiable basis for investment
planning decisions
• Thematic graphical displays
3) Optimizing Pump Operations
for Minimum Energy Cost in Water
Distribution Systems
Typical Water System
Gravity WTW End Use

Sources Pump Distribution Pump
or Pump
STORAGE
RESERVOI
R
WTW / PUMPING
STATION /
BOOSTER
WTW
BOREHOLE
Energy Consumption
• Water is pumped throughout system
• Adequate pressure is maintained by pumping
• Pumping results in high energy consumption
Carbon Intensity (in pound)
• CO2 = ExCintensity
25 22.38
20
15 12.67
10
5
1.36
0.12
0
U.S. Natural Fuel Oil Liquefield
Electric Gas (per (per Pedro Gas
Grid (per cubic foot) gallon) (per
kWh) gallon)
National Energy Consumption Projections for Public Water Supply
Source: Water and Sustainability: U.S. Electricity Consumption for Water Supply & Treatment—The Next Half
Century, EPRI, Palo Alto, CA: 2000. 1006787.
Why Pump Scheduling
• Many benefits for saving pump energy
– Reduce cost
– Improve water service efficiency
– Reduce carbon footprint
– Improve sustainability
• Many ways to save pump energy
– Regularly maintain pumps
– Keep pump operated at best efficiency point
– Make sure electricity bills are correct
– Improve pipelines to reduce head losses
– More…
• Many pumps operated by experience
– Nothing bad or wrong
– Represent good opportunity to improve
Pump Scheduling
• What to schedule
– Which pump is on duty
– When pump is on duty
– What speed is on duty
• Goal
– Minimize energy consumption
– Minimize total energy cost
• Supply requirement
– Hydraulics
– Manage pressure (water loss)
– Water quality
Formulation (matematical optimization)
r
• Search for: H = (hi,t ) i = 1,2,..., N ps , t = 1,...,T
Np
• Minimize: C = ∑C p
p=1
• Subject to: hmin ≤ hi,t ≤ hmax

vmin ≤ v j ,t ≤ vmax
ωmin ≤ ωp ≤ ωmax
Where hi,t is the target hydraulic head of pump station i at time t
vj,t is the flow veolcity of pipe j at time t
ωp is the relative speed factor for pump p,
Nps is the number of pump stations,
Cp is the energy cost of pump p,
Np is the number of pumps,
C is the total energy cost of the pumps,
hmin and hmax are the minimum required and maximum allowed
hydraulic head,
vmin and vmax are the minimum required and maximum allowed flow
velocities
CSP Case Study (Water Utility in UK)
• DMZ system
• 57 Ml/day
• 11 pump
stations and 9
tanks
• Energy cost:
£330K/year
• Recorded daily
energy cost:
£912
• Modeled daily
energy cost:
£923
Price Pattern Factor
0
0.2
0.4
0.6
0.8
1
1.2
0:00
1:00
2:00
3:00
4:00
5:00
6:00
7:00
8:00
9:00
10:00
11:00
Time
12:00
13:00
Electricity Tariff Pattern
14:00
15:00
16:00
17:00
18:00
19:00
20:00
21:00
22:00
23:00
0:00
Pump
Characteristics
• Pump curve
• Efficiency curve
Conventional Controls
• Control rules: if…then; if…then…else...
• Pumps are triggered by clock time
• Pumps are triggered by nodal pressure or HGL
• Pumps are triggered by metered flow
• Pumps are triggered by tank level
• Conventional wisdom
– Turned ON when below a low tank level
– Turned OFF when above a high tank level
– Keep pump operation in a large range of tank levels
Pump Scheduling Optimization
• Optimization criteria
– One hour control interval
– Tank minimum level is set to 20% of depth
– Tank maximum level is set to 90% of depth
– Meet minimum pressure requirement
• Results converted to control rules, e.g.
Rule 100
IF SYSTEM CLOCKTIME <= 8:00 AM
OR SYSTEM CLOCKTIME >= 10:00 PM
AND TANK BUTa2 LEVEL BELOW 5.73
THEN PUMP PILWTHSTATUS IS OPEN
ELSE PUMP PILWTHSTATUS IS CLOSED
Energy Cost comparison
Pump Existing controls Optimized controls
Pump utilization
ID (%) Daily cost (£) Pump utilization (%) Daily cost (£)
X2420052_ 100 181.99 100 181.73
X2420014_ 40 142.11 41 120.51
X2420075_ 42 201.95 37 141.19
X2410361_ 50 31.99 42 22.65
X2419963_ 50 31.99 42 22.65
X241998C_ 26 7.92 31 5.18
X2450024_ 40 37.35 21 13.87
PILWTH 82 236.19 40 98.33
NEWMRKT 23 111.63 22 88.98
Total cost(£) 983.12 695.10
• Immediate saving is 100,000 £ (29% of original

energy cost)
• By optimizing pumping hours and better supply
from storage sources
Optimized Pump Controls
• Tank levels
• Pump flows
Models and Scada data integration in
AquaSafe real-time platform
Darwin Calibrator Darwin Scheduler

WaterGEMS
Optimized Optimized
Operational
Leakage Pumping
Modeling
Detection
Flow Analysis Operational Demand

and Leakage Data Estimation and
Alarm Management Forecasting
Aquasafe Client/Server
User Interface, Configuration, Alarms, Data
Flow Management
ScadaConnector
DB
Existing Scada Server Historic Data
Bentley Real-time Platform (ex. SABESP)
Summary
• Improving Efficiency is a part of a lifecycle asset
management practice in Water Utilities and Consulting
Ecosystem
• Real-time data streams, hydraulic modeling and
optimization technology can help:
o Reducing Water Loss and Detecting leakage
hotspots
o Pipe Renewal process as part of AM
o Pumping scheduling and optimal pressure and
energy management (including CO2 footprint)
• From ‘dull pipes’ towards Smart Water Networks for
real-time modelling, decision making, asset
management and emergency response
EXPO Stand F10
Contact Information and Resources
www.bentley.com/water
www.bentley.com/waterloss
www.bentley.com/waterstories
communities.bentley.com
Bentley office in SA
Contact: [email protected]
[email protected]
Bentley office Europe and MEA

Contact: [email protected]
Bentley Institute Press (water books)
www.bentley.com/books
Tank You for Your Attention
[email protected]
www.bentley.com

Slavco Velickov 0

Uploaded by

Copyright:

Available Formats

Slavco Velickov 0

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Slavco Velickov 0

Uploaded by

Copyright:

Available Formats

Improving Efficiency of Water

Systems: practical examples

Dr.ir. Slavco Velickov

5. Contact Information & Resources

Bridges Rail and

Cadastre and Land Electric and

Campuses Factories Roads

Metals and Water and

Oil and Gas

Estuaries and coastal

Rural water systems Urban water systems

Industry Framework Modeling Framework

sisNET SewerGEMS / CivilStorm / WaterGEMS

GIS products Modelling products

GeoSpatial Server & AssetWise

Planning and CRM, WO

• It’s impossible to find and fix all leaks

“ Many practitioners make common

(Courtesy Dr. Thomas Walski)

11 IWA = International Water Association

Billed Billed Metered Consumption Revenue

• Reported leaks and bursts

• Unreported leaks and bursts

Speed and Quality Economic Active

Risk-based asset management for

3. Active Leakage Detection

Pressure control achieved through:

Types of pressure control:

Before pressure reduction

After pressure reduction

Reduce the flow rates (MNF) of all leaks

Factors considered in DMAs (re)-design

Possible pressure reduction of about 0,5 bar

after pressure reduction

Reduction in AZP from 39 m to 32 m resulted in

Before pressure reduction

After pressure reduction

before after before after before after before after

221 63 36 5,69 3,85 3,39 1,65 0,16 0,07

226 64 34 10,04 6,84 5,62 2,55 0,37 0,24

227 59 38 15,52 10,44 5,91 3,38 5,11 2,56

Annual water saving = 220 000

232 39 32 5,17 4,32 1,32 1,05 2,21 1,63

233 42 33 4,45 3,96 1,48 1,10 1,48 1,37

234 48 38 3,55 2,44 0,32 0,23 2,26 1,24

Total before 117,92 43,75 34,32

Jn is the set of junctions in demand group n

N is the number of demand groups

nL is the total number of leakage nodes

LK n is number of leakage nodes in group n

• DMA system model

• Build hydraulic model

• Built on Bentley Map

• Learn one – learn them all

• Import to Bentley Water

Diameter, Breaks Break Rate,

Pipe Break Model Other Property

Break Fire Flow Criticality Analysis

Normalized Normalized Normalized Normalized