Developing a New Decision Support Framework for Sustainable Drainage Design

Jo-fai Chow
Jo-fai ChowData Scientist at H2O.ai - Customer Success, Smart Apps, Machine Learning & Data Visualization

Poster for the 9th International Conference on Urban Drainage Modelling, September 2012, Belgrade, Serbia. (Best Young Researcher Award)

Developing a New Decision Support
Framework for Sustainable Drainage Design
Jo-fai

1,2,3,
Chow

1,
Savić

Dragan

2,
Fortune

David

Netsanet

2
Mebrate

and Zoran

1
Kapelan

Introduction

Challenges

Summary

The vulnerability of drainage systems and the
importance of flood risk management have
drawn increasing public attention following
major flood events around the globe.

Identifying the optimal combination of different
SuDS techniques with regard to performance,
social-environmental impact and cost:
• The number of possible SuDS combinations
can grow into hundreds and thousands
depending on site characteristics. The
traditional trial and error approach is not
suitable. More sophisticated search methods
based on computation intelligence such as
evolutionary optimisation is recommended.
• The proposed design will be checked and
evaluated by various parties throughout a
project cycle. It is important to provide a
framework for consistent evaluation.

The existing software tools are not sufficient for
sustainable drainage design as they lack the
emphasis on social impact and cost-benefit. We
are developing new software tools that will
allow drainage designers to determine optimal
combinations of SuDS efficiently and will enable
stakeholders to compare and evaluate best
trade-off between water quantity, quality,
amenity value and whole life costs.

Sustainable drainage systems (SuDS) have
been proposed as better alternatives to
conventional drainage systems. Compared to
traditional pipe and storage networks, SuDS
bring additional values such as treatment and
biodiversity to the development site.
Traditional Drainage Systems –
Conveyance & Storage

Sustainable Drainage Systems –
Source Control & Treatment

Figure 1 – examples of both conventional and sustainable
drainage systems.

SuDS in Drainage Models
Several drainage software packages have
already included SuDS modelling modules (e.g.
WinDes and XPSWMM). This allows users to
configure various SuDS components in their
drainage models and to run simulations in
order to determine the impact of different
SuDS techniques on flooding, water quality as
well as life cycle cost.
Porous car park

Swale

Decision Support Framework
A prototype decision support framework has
been developed to look at changes in hydraulic
performance (flow and storage), water quality
(pollutants concentration) and costs (capital
and operational expenditure) based on
different SuDS techniques and sizes. Indicators
for social impact will be implemented in the
next phase of the project. In order to search for
optimal solutions effectively, multi-objective
evolutionary optimisation functionality has
been implemented into this prototype using
GANetXL (Savić, 2011). Users can choose and
compare various drainage design options from
the Pareto front with different trade-off
between costs and system performance.
Prototype SuDS Treatment Train Optimisation Framework

Location 1

Location 2

Location 3

SuDS Technique
Physical Dimension 1
Physical Dimension 2
Physical Dimension 3
Inflitration % (Side)
Inflitration % (Base)
SuDS Technique
Physical Dimension 1
Physical Dimension 2
Physical Dimension 3
Inflitration % (Side)
Inflitration % (Base)
SuDS Technique
Physical Dimension 1
Physical Dimension 2
Physical Dimension 3
Inflitration % (Side)
Inflitration % (Base)

1 to 16
1 to 10
1 to 10
1 to 10
1 to 10
1 to 10
1 to 16
1 to 10
1 to 10
1 to 10
1 to 10
1 to 10
1 to 16
1 to 10
1 to 10
1 to 10
1 to 10
1 to 10

True Value Range
Min.
Max.

10
10
34
62
67
16
1
59
25
20
14
25
13
95
74
48
94
91

1
10
19
2
0
0
1
10
6
1
0
0
1
19
9
1
0
0

True Value

16
29
26
5
25
25
16
33
43
3
25
25
16
26
36
3
25
25

Description of KPI

Infiltration Basin
11.9
21.4
3.7
16.8%
4.0%
Pervious Pavements
23.6
15.3
1.0
3.5%
6.3%
Stormwater Wetlands
25.7
29.0
1.7
23.5%
22.8%

Optimisation Objectives and Penalty Function
Description of Optimisation Objective
Positive if the performance is better than defined targets
WLC (CAPEX, OPEX and Land Value)
Penalty as a results of contraints

Hydraulics
Costs
Penalty Cost

Type
Maximise
Minimise
Avoid

Graphical Outputs

Objective Value
-0.29%
719,093
40,951,441,427.47

Flow Rate at Various Stages of SuDS Treatment Train

Value

Peak Flow at Outlet (m3/s)
Total Storage Required (m3)
Time to Reach Peak Flow (min)
Total Nitrogen
Pollutant Total Phosphorus
Concentration Total Suspended Solids
at Outlet Hydrocarbons
(mg/L)
Heavy Metals
Faecal Coliforms
SuDS 1 WLC (£ at 2012 value)
Whole Life SuDS 2 WLC (£ at 2012 value)
SuDS 3 WLC (£ at 2012 value)
Cost
Total WLC (£ at 2012 value)

20

Inflow (Connection1)

18

2.73
6187
238
6.75
8.52
5.70
10.78
5.45
18.56
£175,620
£225,519
£317,955
£719,093

Hydraulics

Outflow (Connection 1) -> Inflow
(SuDS 1)
Outflow (SuDS1) -> Inflow
(Connection 2)
Outflow (Connection 2) -> Inflow
(SuDS 2)
Outflow (SuDS 2) -> Inflow
(Connection 3)
Outflow (Connection 3) -> Inflow
(SuDS 3)
Final Outflow at Outlet

16
14

Flow (m3/s)

Description

Pond

SuDS Key Performance Indicators

Optimsation Optimisation
Range
Value

12
10
8
6
4

2

Peak Flow Constraint

0
0

50

100

Other Measurements
Description of KPI

Value

SuDS 1 Total Surface Area (m2)
2
Surface Area SuDS 2 Total Surface Area (m2)
SuDS 3 Total Surface Area (m )
Total Surface Area (m2)
SuDS 1 Land Value (£ at 2012 value)

200

250

300

SuDS 2 Land Value (£ at 2012 value)
SuDS 3 Land Value (£ at 2012 value)
Total Land Value (£ at 2012 value)

Storage Required at Various Stages of SuDS Treatment Train

254
359
743
1,357
£31,803
£89,861
£130,084
£251,747

Land Value

150

Duration (Minutes)

35

Storage (Connection 1)

30

Storage (SuDS 1)

25

Volume (m3)

Optimisation Parameters, Range and True Values

Storage (Connection 2)

20

Storage (SuDS 2)
15

Storage (Connection 3)

10

Background Calculations - Optimisation Contraints and Penalty Costs

should be <=
should be <=
should be <=
should be <=
should be <=
should be <=
should be <=
should be <=
should be <=
is
is
is

Quality

1

Storage (TOTAL)
0

50

100

Muskingum

150

200

250

300

Duration (Minutes)

Pollutant Concentration (mg/L)

Cost Summary
£600,000

Total
Nitrogen
35
30
25
20
15
10
5
0

Heavy
Metals

Concentration
(Inlet)
Total
Phosphorus

Concentration
(Regulation
Targets)
Concentration
(Outlet)

£500,000
£400,000
£300,000
£200,000
£100,000
£0

Total
Suspended
Solids

Hydrocarbo
ns

SuDS1
CAPEX

Flow Rate at Various Stages of SuDS Treatment Train

SuDS2

OPEX (over 50 years)

Land Value

SuDS3
Whole Life Cost (over 50 years)

Inflow (Connection1)

20

18

Outflow (Connection 1) -> Inflow
(SuDS 1)
Outflow (SuDS1) -> Inflow
(Connection 2)
Outflow (Connection 2) -> Inflow
(SuDS 2)
Outflow (SuDS 2) -> Inflow
(Connection 3)
Outflow (Connection 3) -> Inflow
(SuDS 3)
Final Outflow at Outlet

18

12
10
8

6
4

Final Outflow

14
12
10
8

6
4

2

0

50

100

150

200

250

10
8

4
2

0

50

100

Duration (Minutes)

150

200

250

Peak Flow Constraint

0

300

0

50

100

Duration (Minutes)

Storage Required at Various Stages of SuDS Treatment Train
30

12

6

Peak Flow Constraint

0

300

Total Storage

30

150

200

250

300

Duration (Minutes)

Storage Required at Various Stages of SuDS Treatment Train
Storage (Connection 1)

Outflow (Connection 1) -> Inflow
(SuDS 1)
Outflow (SuDS1) -> Inflow
(Connection 2)
Outflow (Connection 2) -> Inflow
(SuDS 2)
Outflow (SuDS 2) -> Inflow
(Connection 3)
Outflow (Connection 3) -> Inflow
(SuDS 3)
Final Outflow at Outlet

Final Outflow

14

2

Peak Flow Constraint

0

Inflow (Connection1)

16

Flow (m3/s)

Final Outflow

14

16

Total Storage

Storage Required at Various Stages of SuDS Treatment Train
Storage (Connection 1)

30

Total Storage

Storage (Connection 1)

25

Storage (SuDS 1)

25

Storage (Connection 2)

20

Storage (Connection 2)

20

Storage (Connection 2)

15

Storage (SuDS 2)

15

Storage (SuDS 2)

15

Storage (SuDS 2)

10

Storage (Connection 3)

10

Storage (Connection 3)

10

Storage (Connection 3)

Storage (SuDS 3)
Storage (TOTAL)

0

0

50

100

150

200

250

5
0

300

Storage (SuDS 3)

50

100

Cost Summary

Total
Nitrogen

Heavy
Metals

£600,000

35
30
25
20
15
10
5
0

£500,000

Total
Phosphorus

Concentration
(Regulation
Targets)
Concentration
(Outlet)

Total
Suspended
Solids

200

250

0

300

Costs

Final Pollutant Conc.

Storage (TOTAL)
50

100

Heavy
Metals

£300,000
£200,000

£600,000

35
30
25
20
15
10
5
0

£500,000

Total
Phosphorus

£100,000

SuDS1
CAPEX

Concentration
(Regulation
Targets)
Concentration
(Outlet)

£0

OPEX (over 50 years)

SuDS2
Land Value

SuDS3
Whole Life Cost (over 50 years)

Cost Summary

Total
Nitrogen

Concentration
(Inlet)

Hydrocarbo
ns

Total
Suspended
Solids

150

200

250

300

Duration (Minutes)

Pollutant Concentration (mg/L)

£400,000

Storage (SuDS 1)

Storage (SuDS 3)

0

Duration (Minutes)

Pollutant Concentration (mg/L)

Concentration
(Inlet)

150

5

Storage (TOTAL)
0

Duration (Minutes)

Final Pollutant Conc.

Volume (m3)

Storage (SuDS 1)

20

5

Costs

Final Pollutant Conc.
Pollutant Concentration (mg/L)

Heavy
Metals

£300,000
£200,000

£600,000

35
30
25
20
15
10
5
0

£500,000

Concentration
(Inlet)
Total
Phosphorus

£100,000

SuDS1
CAPEX

Concentration
(Regulation
Targets)
Concentration
(Outlet)

£0

OPEX (over 50 years)

SuDS2
Land Value

SuDS3
Whole Life Cost (over 50 years)

Somewhere in between:
stakeholders to decide
what is the best trade-off
between two objectives.

Cost Summary

Total
Nitrogen

£400,000

Hydrocarbo
ns

The following tasks have been scheduled for
the second phase of the project:
• Quantifying and including more optimisation
objectives for social impact.
• Mutli-objective optimisation engine written in
MATLAB codes with main focus on fast and
parallel execution utilising both CPU and GPU.
• Full integration with new drainage design
software suites (e.g. XPDrainage) developed
by Micro Drainage and XP Solutions.

Key References

Flow Rate at Various Stages of SuDS Treatment Train

20

Outflow (Connection 1) -> Inflow
(SuDS 1)
Outflow (SuDS1) -> Inflow
(Connection 2)
Outflow (Connection 2) -> Inflow
(SuDS 2)
Outflow (SuDS 2) -> Inflow
(Connection 3)
Outflow (Connection 3) -> Inflow
(SuDS 3)
Final Outflow at Outlet

16

Future Development

Savić, D.A., Bicik J. and Morley M.S. (2011).
GANetXL: A DSS generator for multiobjective
optimisation of spreadsheet-based models,
Environmental Modelling & Software, Vol. 26,
551-561.

Flow Rate at Various Stages of SuDS Treatment Train
Inflow (Connection1)

18

Least-cost option: runoff
satisfies minimum design
requirement.

Figure 3 – Comparison of traditional and new approach.

Settings

Flow Calculation Method (Choose)

20

Hydrocarbo
ns

Amenity

Description of KPI
Hydraulics

Performance

Volume (m3)

New, Integrated Approach –
Balanced Emphasis

Quality

Amenity

9,358,715,432.56
23,744,725,994.91
0.00
0.00
0.00
0.00
7,848,000,000.00
0.00
No Constraint
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
No Constraint
No Constraint
No Constraint
40,951,441,427.47

0

Cost

25

Quantity

50
50
50
100
100
100
40
40
40
allowed
allowed
allowed

Penalty Cost

Storage (SuDS 3)

5

Figure 4 – prototyping a new decision support framework
for SuDS using Microsoft Excel spreadsheet.

Flow (m3/s)

In order to fill this gap, we decided to develop
additional software features that will put more
emphasis on social impact and will enable
stakeholders to maximise multiple benefits.

Quantity

2.5
5000
180
10
10
10
10
10

TOTAL Penalty Cost:

Yet the existing software modules are not
sufficient for sustainable drainage design as
they mostly focus on water quantity and
quality aspect. There is not enough emphasis
on the amenity value and cost-benefit analysis.

Traditional Approach – Main
Emphasis on Water Quantity

Level

Flow (m3/s)

Towards Sustainability

Contraints
should be <=
should be <=
should be >=
should be <=
should be <=
should be <=
should be <=
should be <=

Peak Flow at Outlet (m3/s)
Total Storage Required (m3)
Time to Reach Peak Flow (min.)
Total Nitrogen
Pollutant Total Phosphorus
Concentration Total Suspended Solids
at Outlet Hydrocarbons
Heavy Metals
(mg/L)
Faecal Coliforms
SuDS 1 - Dimension 1 (m)
SuDS 1 - Dimension 2 (m)
SuDS 1 - Dimension 3 (m)
SuDS 2 - Dimension 1 (m)
Physical
SuDS 2 - Dimension 1 (m)
Restrictions
SuDS 2 - Dimension 1 (m)
SuDS 3 - Dimension 1 (m)
SuDS 3 - Dimension 1 (m)
SuDS 3 - Dimension 1 (m)
Use of infiltration at location 1
Infiltration Use of infiltration at location 2
Use of infiltration at location 3

Hydraulics

Volume (m3)

Figure 2 – using Micro Drainage’s WinDes to model SuDS
for a typical site development drainage design.

Description of KPI

Other Controls

Figure 6 – our vision: balanced emphasis on water
quantity, quality and amenity for sustainable drainage
design with whole life costing analysis.

Total
Suspended
Solids

Costs

£400,000
£300,000
£200,000

£100,000
£0

SuDS1
CAPEX

OPEX (over 50 years)

SuDS2
Land Value

SuDS3
Whole Life Cost (over 50 years)

Most expensive option:
runoff is further reduced
at higher costs.

Figure 5 – exploring and comparing different design
options from optimisation Pareto front .

Contact the Author
The work presented here is part of author’s 4year industrial PhD research project. For more
information, please contact the author:
• Jo-fai Chow, STREAM Research Engineer
• E-mail: jo-fai.chow@microdrainage.co.uk
• Software by XP Solutions:
http://www.xpsolutions.com/software/

Centre for Water Systems, University of Exeter, United Kingdom (www.exeter.ac.uk/cws)
2 Micro Drainage (an XP Solutions company), United Kingdom (www.microdrainage.co.uk)
3 STREAM Industrial Doctorate Centre for the Water Sector, United Kingdom (www.stream-idc.net)
* Note: image courtesy of Micro Drainage and XP Solutions

Recommended

H2O Machine Learning Use Cases by
H2O Machine Learning Use CasesH2O Machine Learning Use Cases
H2O Machine Learning Use CasesJo-fai Chow
1.5K views21 slides
A Systematic, Multi-Criteria Decision Support Framework for Sustainable Drain... by
A Systematic, Multi-Criteria Decision Support Framework for Sustainable Drain...A Systematic, Multi-Criteria Decision Support Framework for Sustainable Drain...
A Systematic, Multi-Criteria Decision Support Framework for Sustainable Drain...Jo-fai Chow
5.3K views62 slides
Deploying your Predictive Models as a Service via Domino by
Deploying your Predictive Models as a Service via DominoDeploying your Predictive Models as a Service via Domino
Deploying your Predictive Models as a Service via DominoJo-fai Chow
19.1K views33 slides
Project "Deep Water" by
Project "Deep Water"Project "Deep Water"
Project "Deep Water"Jo-fai Chow
1.5K views95 slides
Making Multimillion-Dollar Baseball Decisions with H2O AutoML, LIME and Shiny by
Making Multimillion-Dollar Baseball Decisions with H2O AutoML, LIME and ShinyMaking Multimillion-Dollar Baseball Decisions with H2O AutoML, LIME and Shiny
Making Multimillion-Dollar Baseball Decisions with H2O AutoML, LIME and ShinyJo-fai Chow
702 views50 slides
Automatic and Interpretable Machine Learning in R with H2O and LIME by
Automatic and Interpretable Machine Learning in R with H2O and LIMEAutomatic and Interpretable Machine Learning in R with H2O and LIME
Automatic and Interpretable Machine Learning in R with H2O and LIMEJo-fai Chow
313 views92 slides

More Related Content

More from Jo-fai Chow

H2O at BelgradeR Meetup by
H2O at BelgradeR MeetupH2O at BelgradeR Meetup
H2O at BelgradeR MeetupJo-fai Chow
473 views100 slides
Introduction to H2O and Model Stacking Use Cases by
Introduction to H2O and Model Stacking Use CasesIntroduction to H2O and Model Stacking Use Cases
Introduction to H2O and Model Stacking Use CasesJo-fai Chow
1.1K views78 slides
H2O at Poznan R Meetup by
H2O at Poznan R MeetupH2O at Poznan R Meetup
H2O at Poznan R MeetupJo-fai Chow
753 views114 slides
Introduction to Machine Learning with H2O and Python by
Introduction to Machine Learning with H2O and PythonIntroduction to Machine Learning with H2O and Python
Introduction to Machine Learning with H2O and PythonJo-fai Chow
716 views120 slides
Introduction to Machine Learning with H2O and Python by
Introduction to Machine Learning with H2O and PythonIntroduction to Machine Learning with H2O and Python
Introduction to Machine Learning with H2O and PythonJo-fai Chow
842 views108 slides
From Kaggle to H2O - The True Story of a Civil Engineer Turned Data Geek by
From Kaggle to H2O - The True Story of a Civil Engineer Turned Data GeekFrom Kaggle to H2O - The True Story of a Civil Engineer Turned Data Geek
From Kaggle to H2O - The True Story of a Civil Engineer Turned Data GeekJo-fai Chow
1.8K views45 slides

More from Jo-fai Chow(20)

H2O at BelgradeR Meetup by Jo-fai Chow
H2O at BelgradeR MeetupH2O at BelgradeR Meetup
H2O at BelgradeR Meetup
Jo-fai Chow473 views
Introduction to H2O and Model Stacking Use Cases by Jo-fai Chow
Introduction to H2O and Model Stacking Use CasesIntroduction to H2O and Model Stacking Use Cases
Introduction to H2O and Model Stacking Use Cases
Jo-fai Chow1.1K views
H2O at Poznan R Meetup by Jo-fai Chow
H2O at Poznan R MeetupH2O at Poznan R Meetup
H2O at Poznan R Meetup
Jo-fai Chow753 views
Introduction to Machine Learning with H2O and Python by Jo-fai Chow
Introduction to Machine Learning with H2O and PythonIntroduction to Machine Learning with H2O and Python
Introduction to Machine Learning with H2O and Python
Jo-fai Chow716 views
Introduction to Machine Learning with H2O and Python by Jo-fai Chow
Introduction to Machine Learning with H2O and PythonIntroduction to Machine Learning with H2O and Python
Introduction to Machine Learning with H2O and Python
Jo-fai Chow842 views
From Kaggle to H2O - The True Story of a Civil Engineer Turned Data Geek by Jo-fai Chow
From Kaggle to H2O - The True Story of a Civil Engineer Turned Data GeekFrom Kaggle to H2O - The True Story of a Civil Engineer Turned Data Geek
From Kaggle to H2O - The True Story of a Civil Engineer Turned Data Geek
Jo-fai Chow1.8K views
H2O Deep Water - Making Deep Learning Accessible to Everyone by Jo-fai Chow
H2O Deep Water - Making Deep Learning Accessible to EveryoneH2O Deep Water - Making Deep Learning Accessible to Everyone
H2O Deep Water - Making Deep Learning Accessible to Everyone
Jo-fai Chow566 views
H2O Deep Water - Making Deep Learning Accessible to Everyone by Jo-fai Chow
H2O Deep Water - Making Deep Learning Accessible to EveryoneH2O Deep Water - Making Deep Learning Accessible to Everyone
H2O Deep Water - Making Deep Learning Accessible to Everyone
Jo-fai Chow1K views
Kaggle Competitions, New Friends, New Skills and New Opportunities by Jo-fai Chow
Kaggle Competitions, New Friends, New Skills and New OpportunitiesKaggle Competitions, New Friends, New Skills and New Opportunities
Kaggle Competitions, New Friends, New Skills and New Opportunities
Jo-fai Chow363 views
Using H2O Random Grid Search for Hyper-parameters Optimization by Jo-fai Chow
Using H2O Random Grid Search for Hyper-parameters OptimizationUsing H2O Random Grid Search for Hyper-parameters Optimization
Using H2O Random Grid Search for Hyper-parameters Optimization
Jo-fai Chow2.5K views
Introduction to Generalised Low-Rank Model and Missing Values by Jo-fai Chow
Introduction to Generalised Low-Rank Model and Missing ValuesIntroduction to Generalised Low-Rank Model and Missing Values
Introduction to Generalised Low-Rank Model and Missing Values
Jo-fai Chow2.3K views
Improving Model Predictions via Stacking and Hyper-parameters Tuning by Jo-fai Chow
Improving Model Predictions via Stacking and Hyper-parameters TuningImproving Model Predictions via Stacking and Hyper-parameters Tuning
Improving Model Predictions via Stacking and Hyper-parameters Tuning
Jo-fai Chow2.1K views
Kaggle competitions, new friends, new skills and new opportunities by Jo-fai Chow
Kaggle competitions, new friends, new skills and new opportunitiesKaggle competitions, new friends, new skills and new opportunities
Kaggle competitions, new friends, new skills and new opportunities
Jo-fai Chow973 views
A Systematic, Multi-Criteria Decision Support Framework for Sustainable Drain... by Jo-fai Chow
A Systematic, Multi-Criteria Decision Support Framework for Sustainable Drain...A Systematic, Multi-Criteria Decision Support Framework for Sustainable Drain...
A Systematic, Multi-Criteria Decision Support Framework for Sustainable Drain...
Jo-fai Chow1.2K views
Designing Sustainable Drainage Systems by Jo-fai Chow
Designing Sustainable Drainage SystemsDesigning Sustainable Drainage Systems
Designing Sustainable Drainage Systems
Jo-fai Chow2.2K views
Developing a New Decision Support System for SuDS by Jo-fai Chow
Developing a New Decision Support System for SuDSDeveloping a New Decision Support System for SuDS
Developing a New Decision Support System for SuDS
Jo-fai Chow1.3K views
Udacity Statement (Introduction to Statistics, August 2012) by Jo-fai Chow
Udacity Statement (Introduction to Statistics, August 2012)Udacity Statement (Introduction to Statistics, August 2012)
Udacity Statement (Introduction to Statistics, August 2012)
Jo-fai Chow365 views
Coursera Statement (Computational Investing, Part I, by Jo-fai Chow
Coursera Statement (Computational Investing, Part I, Coursera Statement (Computational Investing, Part I,
Coursera Statement (Computational Investing, Part I,
Jo-fai Chow532 views
Coursera Statement (Computing for Data Analysis, Oct 2013) by Jo-fai Chow
Coursera Statement (Computing for Data Analysis, Oct 2013)Coursera Statement (Computing for Data Analysis, Oct 2013)
Coursera Statement (Computing for Data Analysis, Oct 2013)
Jo-fai Chow283 views
Coursera Statement (Data Analysis, Mar 2013) by Jo-fai Chow
Coursera Statement (Data Analysis, Mar 2013)Coursera Statement (Data Analysis, Mar 2013)
Coursera Statement (Data Analysis, Mar 2013)
Jo-fai Chow332 views

Recently uploaded

Lilypad @ Labweek, Istanbul, 2023.pdf by
Lilypad @ Labweek, Istanbul, 2023.pdfLilypad @ Labweek, Istanbul, 2023.pdf
Lilypad @ Labweek, Istanbul, 2023.pdfAlly339821
9 views45 slides
Report 2030 Digital Decade by
Report 2030 Digital DecadeReport 2030 Digital Decade
Report 2030 Digital DecadeMassimo Talia
14 views41 slides
Special_edition_innovator_2023.pdf by
Special_edition_innovator_2023.pdfSpecial_edition_innovator_2023.pdf
Special_edition_innovator_2023.pdfWillDavies22
16 views6 slides
Vertical User Stories by
Vertical User StoriesVertical User Stories
Vertical User StoriesMoisés Armani Ramírez
11 views16 slides
AMAZON PRODUCT RESEARCH.pdf by
AMAZON PRODUCT RESEARCH.pdfAMAZON PRODUCT RESEARCH.pdf
AMAZON PRODUCT RESEARCH.pdfJerikkLaureta
15 views13 slides
Roadmap to Become Experts.pptx by
Roadmap to Become Experts.pptxRoadmap to Become Experts.pptx
Roadmap to Become Experts.pptxdscwidyatamanew
11 views45 slides

Recently uploaded(20)

Lilypad @ Labweek, Istanbul, 2023.pdf by Ally339821
Lilypad @ Labweek, Istanbul, 2023.pdfLilypad @ Labweek, Istanbul, 2023.pdf
Lilypad @ Labweek, Istanbul, 2023.pdf
Ally3398219 views
Special_edition_innovator_2023.pdf by WillDavies22
Special_edition_innovator_2023.pdfSpecial_edition_innovator_2023.pdf
Special_edition_innovator_2023.pdf
WillDavies2216 views
AMAZON PRODUCT RESEARCH.pdf by JerikkLaureta
AMAZON PRODUCT RESEARCH.pdfAMAZON PRODUCT RESEARCH.pdf
AMAZON PRODUCT RESEARCH.pdf
JerikkLaureta15 views
Voice Logger - Telephony Integration Solution at Aegis by Nirmal Sharma
Voice Logger - Telephony Integration Solution at AegisVoice Logger - Telephony Integration Solution at Aegis
Voice Logger - Telephony Integration Solution at Aegis
Nirmal Sharma17 views
Piloting & Scaling Successfully With Microsoft Viva by Richard Harbridge
Piloting & Scaling Successfully With Microsoft VivaPiloting & Scaling Successfully With Microsoft Viva
Piloting & Scaling Successfully With Microsoft Viva
Automating a World-Class Technology Conference; Behind the Scenes of CiscoLive by Network Automation Forum
Automating a World-Class Technology Conference; Behind the Scenes of CiscoLiveAutomating a World-Class Technology Conference; Behind the Scenes of CiscoLive
Automating a World-Class Technology Conference; Behind the Scenes of CiscoLive
Business Analyst Series 2023 - Week 3 Session 5 by DianaGray10
Business Analyst Series 2023 -  Week 3 Session 5Business Analyst Series 2023 -  Week 3 Session 5
Business Analyst Series 2023 - Week 3 Session 5
DianaGray10209 views
Unit 1_Lecture 2_Physical Design of IoT.pdf by StephenTec
Unit 1_Lecture 2_Physical Design of IoT.pdfUnit 1_Lecture 2_Physical Design of IoT.pdf
Unit 1_Lecture 2_Physical Design of IoT.pdf
StephenTec11 views
Data-centric AI and the convergence of data and model engineering: opportunit... by Paolo Missier
Data-centric AI and the convergence of data and model engineering:opportunit...Data-centric AI and the convergence of data and model engineering:opportunit...
Data-centric AI and the convergence of data and model engineering: opportunit...
Paolo Missier34 views
Five Things You SHOULD Know About Postman by Postman
Five Things You SHOULD Know About PostmanFive Things You SHOULD Know About Postman
Five Things You SHOULD Know About Postman
Postman27 views
Spesifikasi Lengkap ASUS Vivobook Go 14 by Dot Semarang
Spesifikasi Lengkap ASUS Vivobook Go 14Spesifikasi Lengkap ASUS Vivobook Go 14
Spesifikasi Lengkap ASUS Vivobook Go 14
Dot Semarang35 views
DALI Basics Course 2023 by Ivory Egg
DALI Basics Course  2023DALI Basics Course  2023
DALI Basics Course 2023
Ivory Egg14 views
【USB韌體設計課程】精選講義節錄-USB的列舉過程_艾鍗學院 by IttrainingIttraining
【USB韌體設計課程】精選講義節錄-USB的列舉過程_艾鍗學院【USB韌體設計課程】精選講義節錄-USB的列舉過程_艾鍗學院
【USB韌體設計課程】精選講義節錄-USB的列舉過程_艾鍗學院
STPI OctaNE CoE Brochure.pdf by madhurjyapb
STPI OctaNE CoE Brochure.pdfSTPI OctaNE CoE Brochure.pdf
STPI OctaNE CoE Brochure.pdf
madhurjyapb12 views

Developing a New Decision Support Framework for Sustainable Drainage Design

  • 1. Developing a New Decision Support Framework for Sustainable Drainage Design Jo-fai 1,2,3, Chow 1, Savić Dragan 2, Fortune David Netsanet 2 Mebrate and Zoran 1 Kapelan Introduction Challenges Summary The vulnerability of drainage systems and the importance of flood risk management have drawn increasing public attention following major flood events around the globe. Identifying the optimal combination of different SuDS techniques with regard to performance, social-environmental impact and cost: • The number of possible SuDS combinations can grow into hundreds and thousands depending on site characteristics. The traditional trial and error approach is not suitable. More sophisticated search methods based on computation intelligence such as evolutionary optimisation is recommended. • The proposed design will be checked and evaluated by various parties throughout a project cycle. It is important to provide a framework for consistent evaluation. The existing software tools are not sufficient for sustainable drainage design as they lack the emphasis on social impact and cost-benefit. We are developing new software tools that will allow drainage designers to determine optimal combinations of SuDS efficiently and will enable stakeholders to compare and evaluate best trade-off between water quantity, quality, amenity value and whole life costs. Sustainable drainage systems (SuDS) have been proposed as better alternatives to conventional drainage systems. Compared to traditional pipe and storage networks, SuDS bring additional values such as treatment and biodiversity to the development site. Traditional Drainage Systems – Conveyance & Storage Sustainable Drainage Systems – Source Control & Treatment Figure 1 – examples of both conventional and sustainable drainage systems. SuDS in Drainage Models Several drainage software packages have already included SuDS modelling modules (e.g. WinDes and XPSWMM). This allows users to configure various SuDS components in their drainage models and to run simulations in order to determine the impact of different SuDS techniques on flooding, water quality as well as life cycle cost. Porous car park Swale Decision Support Framework A prototype decision support framework has been developed to look at changes in hydraulic performance (flow and storage), water quality (pollutants concentration) and costs (capital and operational expenditure) based on different SuDS techniques and sizes. Indicators for social impact will be implemented in the next phase of the project. In order to search for optimal solutions effectively, multi-objective evolutionary optimisation functionality has been implemented into this prototype using GANetXL (Savić, 2011). Users can choose and compare various drainage design options from the Pareto front with different trade-off between costs and system performance. Prototype SuDS Treatment Train Optimisation Framework Location 1 Location 2 Location 3 SuDS Technique Physical Dimension 1 Physical Dimension 2 Physical Dimension 3 Inflitration % (Side) Inflitration % (Base) SuDS Technique Physical Dimension 1 Physical Dimension 2 Physical Dimension 3 Inflitration % (Side) Inflitration % (Base) SuDS Technique Physical Dimension 1 Physical Dimension 2 Physical Dimension 3 Inflitration % (Side) Inflitration % (Base) 1 to 16 1 to 10 1 to 10 1 to 10 1 to 10 1 to 10 1 to 16 1 to 10 1 to 10 1 to 10 1 to 10 1 to 10 1 to 16 1 to 10 1 to 10 1 to 10 1 to 10 1 to 10 True Value Range Min. Max. 10 10 34 62 67 16 1 59 25 20 14 25 13 95 74 48 94 91 1 10 19 2 0 0 1 10 6 1 0 0 1 19 9 1 0 0 True Value 16 29 26 5 25 25 16 33 43 3 25 25 16 26 36 3 25 25 Description of KPI Infiltration Basin 11.9 21.4 3.7 16.8% 4.0% Pervious Pavements 23.6 15.3 1.0 3.5% 6.3% Stormwater Wetlands 25.7 29.0 1.7 23.5% 22.8% Optimisation Objectives and Penalty Function Description of Optimisation Objective Positive if the performance is better than defined targets WLC (CAPEX, OPEX and Land Value) Penalty as a results of contraints Hydraulics Costs Penalty Cost Type Maximise Minimise Avoid Graphical Outputs Objective Value -0.29% 719,093 40,951,441,427.47 Flow Rate at Various Stages of SuDS Treatment Train Value Peak Flow at Outlet (m3/s) Total Storage Required (m3) Time to Reach Peak Flow (min) Total Nitrogen Pollutant Total Phosphorus Concentration Total Suspended Solids at Outlet Hydrocarbons (mg/L) Heavy Metals Faecal Coliforms SuDS 1 WLC (£ at 2012 value) Whole Life SuDS 2 WLC (£ at 2012 value) SuDS 3 WLC (£ at 2012 value) Cost Total WLC (£ at 2012 value) 20 Inflow (Connection1) 18 2.73 6187 238 6.75 8.52 5.70 10.78 5.45 18.56 £175,620 £225,519 £317,955 £719,093 Hydraulics Outflow (Connection 1) -> Inflow (SuDS 1) Outflow (SuDS1) -> Inflow (Connection 2) Outflow (Connection 2) -> Inflow (SuDS 2) Outflow (SuDS 2) -> Inflow (Connection 3) Outflow (Connection 3) -> Inflow (SuDS 3) Final Outflow at Outlet 16 14 Flow (m3/s) Description Pond SuDS Key Performance Indicators Optimsation Optimisation Range Value 12 10 8 6 4 2 Peak Flow Constraint 0 0 50 100 Other Measurements Description of KPI Value SuDS 1 Total Surface Area (m2) 2 Surface Area SuDS 2 Total Surface Area (m2) SuDS 3 Total Surface Area (m ) Total Surface Area (m2) SuDS 1 Land Value (£ at 2012 value) 200 250 300 SuDS 2 Land Value (£ at 2012 value) SuDS 3 Land Value (£ at 2012 value) Total Land Value (£ at 2012 value) Storage Required at Various Stages of SuDS Treatment Train 254 359 743 1,357 £31,803 £89,861 £130,084 £251,747 Land Value 150 Duration (Minutes) 35 Storage (Connection 1) 30 Storage (SuDS 1) 25 Volume (m3) Optimisation Parameters, Range and True Values Storage (Connection 2) 20 Storage (SuDS 2) 15 Storage (Connection 3) 10 Background Calculations - Optimisation Contraints and Penalty Costs should be <= should be <= should be <= should be <= should be <= should be <= should be <= should be <= should be <= is is is Quality 1 Storage (TOTAL) 0 50 100 Muskingum 150 200 250 300 Duration (Minutes) Pollutant Concentration (mg/L) Cost Summary £600,000 Total Nitrogen 35 30 25 20 15 10 5 0 Heavy Metals Concentration (Inlet) Total Phosphorus Concentration (Regulation Targets) Concentration (Outlet) £500,000 £400,000 £300,000 £200,000 £100,000 £0 Total Suspended Solids Hydrocarbo ns SuDS1 CAPEX Flow Rate at Various Stages of SuDS Treatment Train SuDS2 OPEX (over 50 years) Land Value SuDS3 Whole Life Cost (over 50 years) Inflow (Connection1) 20 18 Outflow (Connection 1) -> Inflow (SuDS 1) Outflow (SuDS1) -> Inflow (Connection 2) Outflow (Connection 2) -> Inflow (SuDS 2) Outflow (SuDS 2) -> Inflow (Connection 3) Outflow (Connection 3) -> Inflow (SuDS 3) Final Outflow at Outlet 18 12 10 8 6 4 Final Outflow 14 12 10 8 6 4 2 0 50 100 150 200 250 10 8 4 2 0 50 100 Duration (Minutes) 150 200 250 Peak Flow Constraint 0 300 0 50 100 Duration (Minutes) Storage Required at Various Stages of SuDS Treatment Train 30 12 6 Peak Flow Constraint 0 300 Total Storage 30 150 200 250 300 Duration (Minutes) Storage Required at Various Stages of SuDS Treatment Train Storage (Connection 1) Outflow (Connection 1) -> Inflow (SuDS 1) Outflow (SuDS1) -> Inflow (Connection 2) Outflow (Connection 2) -> Inflow (SuDS 2) Outflow (SuDS 2) -> Inflow (Connection 3) Outflow (Connection 3) -> Inflow (SuDS 3) Final Outflow at Outlet Final Outflow 14 2 Peak Flow Constraint 0 Inflow (Connection1) 16 Flow (m3/s) Final Outflow 14 16 Total Storage Storage Required at Various Stages of SuDS Treatment Train Storage (Connection 1) 30 Total Storage Storage (Connection 1) 25 Storage (SuDS 1) 25 Storage (Connection 2) 20 Storage (Connection 2) 20 Storage (Connection 2) 15 Storage (SuDS 2) 15 Storage (SuDS 2) 15 Storage (SuDS 2) 10 Storage (Connection 3) 10 Storage (Connection 3) 10 Storage (Connection 3) Storage (SuDS 3) Storage (TOTAL) 0 0 50 100 150 200 250 5 0 300 Storage (SuDS 3) 50 100 Cost Summary Total Nitrogen Heavy Metals £600,000 35 30 25 20 15 10 5 0 £500,000 Total Phosphorus Concentration (Regulation Targets) Concentration (Outlet) Total Suspended Solids 200 250 0 300 Costs Final Pollutant Conc. Storage (TOTAL) 50 100 Heavy Metals £300,000 £200,000 £600,000 35 30 25 20 15 10 5 0 £500,000 Total Phosphorus £100,000 SuDS1 CAPEX Concentration (Regulation Targets) Concentration (Outlet) £0 OPEX (over 50 years) SuDS2 Land Value SuDS3 Whole Life Cost (over 50 years) Cost Summary Total Nitrogen Concentration (Inlet) Hydrocarbo ns Total Suspended Solids 150 200 250 300 Duration (Minutes) Pollutant Concentration (mg/L) £400,000 Storage (SuDS 1) Storage (SuDS 3) 0 Duration (Minutes) Pollutant Concentration (mg/L) Concentration (Inlet) 150 5 Storage (TOTAL) 0 Duration (Minutes) Final Pollutant Conc. Volume (m3) Storage (SuDS 1) 20 5 Costs Final Pollutant Conc. Pollutant Concentration (mg/L) Heavy Metals £300,000 £200,000 £600,000 35 30 25 20 15 10 5 0 £500,000 Concentration (Inlet) Total Phosphorus £100,000 SuDS1 CAPEX Concentration (Regulation Targets) Concentration (Outlet) £0 OPEX (over 50 years) SuDS2 Land Value SuDS3 Whole Life Cost (over 50 years) Somewhere in between: stakeholders to decide what is the best trade-off between two objectives. Cost Summary Total Nitrogen £400,000 Hydrocarbo ns The following tasks have been scheduled for the second phase of the project: • Quantifying and including more optimisation objectives for social impact. • Mutli-objective optimisation engine written in MATLAB codes with main focus on fast and parallel execution utilising both CPU and GPU. • Full integration with new drainage design software suites (e.g. XPDrainage) developed by Micro Drainage and XP Solutions. Key References Flow Rate at Various Stages of SuDS Treatment Train 20 Outflow (Connection 1) -> Inflow (SuDS 1) Outflow (SuDS1) -> Inflow (Connection 2) Outflow (Connection 2) -> Inflow (SuDS 2) Outflow (SuDS 2) -> Inflow (Connection 3) Outflow (Connection 3) -> Inflow (SuDS 3) Final Outflow at Outlet 16 Future Development Savić, D.A., Bicik J. and Morley M.S. (2011). GANetXL: A DSS generator for multiobjective optimisation of spreadsheet-based models, Environmental Modelling & Software, Vol. 26, 551-561. Flow Rate at Various Stages of SuDS Treatment Train Inflow (Connection1) 18 Least-cost option: runoff satisfies minimum design requirement. Figure 3 – Comparison of traditional and new approach. Settings Flow Calculation Method (Choose) 20 Hydrocarbo ns Amenity Description of KPI Hydraulics Performance Volume (m3) New, Integrated Approach – Balanced Emphasis Quality Amenity 9,358,715,432.56 23,744,725,994.91 0.00 0.00 0.00 0.00 7,848,000,000.00 0.00 No Constraint 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 No Constraint No Constraint No Constraint 40,951,441,427.47 0 Cost 25 Quantity 50 50 50 100 100 100 40 40 40 allowed allowed allowed Penalty Cost Storage (SuDS 3) 5 Figure 4 – prototyping a new decision support framework for SuDS using Microsoft Excel spreadsheet. Flow (m3/s) In order to fill this gap, we decided to develop additional software features that will put more emphasis on social impact and will enable stakeholders to maximise multiple benefits. Quantity 2.5 5000 180 10 10 10 10 10 TOTAL Penalty Cost: Yet the existing software modules are not sufficient for sustainable drainage design as they mostly focus on water quantity and quality aspect. There is not enough emphasis on the amenity value and cost-benefit analysis. Traditional Approach – Main Emphasis on Water Quantity Level Flow (m3/s) Towards Sustainability Contraints should be <= should be <= should be >= should be <= should be <= should be <= should be <= should be <= Peak Flow at Outlet (m3/s) Total Storage Required (m3) Time to Reach Peak Flow (min.) Total Nitrogen Pollutant Total Phosphorus Concentration Total Suspended Solids at Outlet Hydrocarbons Heavy Metals (mg/L) Faecal Coliforms SuDS 1 - Dimension 1 (m) SuDS 1 - Dimension 2 (m) SuDS 1 - Dimension 3 (m) SuDS 2 - Dimension 1 (m) Physical SuDS 2 - Dimension 1 (m) Restrictions SuDS 2 - Dimension 1 (m) SuDS 3 - Dimension 1 (m) SuDS 3 - Dimension 1 (m) SuDS 3 - Dimension 1 (m) Use of infiltration at location 1 Infiltration Use of infiltration at location 2 Use of infiltration at location 3 Hydraulics Volume (m3) Figure 2 – using Micro Drainage’s WinDes to model SuDS for a typical site development drainage design. Description of KPI Other Controls Figure 6 – our vision: balanced emphasis on water quantity, quality and amenity for sustainable drainage design with whole life costing analysis. Total Suspended Solids Costs £400,000 £300,000 £200,000 £100,000 £0 SuDS1 CAPEX OPEX (over 50 years) SuDS2 Land Value SuDS3 Whole Life Cost (over 50 years) Most expensive option: runoff is further reduced at higher costs. Figure 5 – exploring and comparing different design options from optimisation Pareto front . Contact the Author The work presented here is part of author’s 4year industrial PhD research project. For more information, please contact the author: • Jo-fai Chow, STREAM Research Engineer • E-mail: jo-fai.chow@microdrainage.co.uk • Software by XP Solutions: http://www.xpsolutions.com/software/ Centre for Water Systems, University of Exeter, United Kingdom (www.exeter.ac.uk/cws) 2 Micro Drainage (an XP Solutions company), United Kingdom (www.microdrainage.co.uk) 3 STREAM Industrial Doctorate Centre for the Water Sector, United Kingdom (www.stream-idc.net) * Note: image courtesy of Micro Drainage and XP Solutions