DSD-INT 2014 - OpenMI Symposium - An OpenMI composition with channel flow and real-time control modelling for a Dutch polder system, N de Hulster (Arcadis), S Bosch (Hydroconsult)

OpenMI An OpenMI composition

with channel flow and

real-time control

modelling for a Dutch

polder system

Niels de Hulster

Klaas-Jan van Heeringen

OpenMI An OpenMI composition

with channel flow and

real-time control

modelling for a Dutch

polder system

Imagine the result

Main contractors

c

Client

Subcontractor

Niels de Hulster

Klaas-Jan van Heeringen

| 31 oktober 2014 | © ARCADIS 2014 Dia 3

Agenda

• Problem, solution and results

• Contribution OpenMi

• Probabilistic modelling

• Joint Probabilities

• Conclusion and next steps

| 31 oktober 2014 | © ARCADIS 2014 Dia 4

The problem

• Waterboard Fryslân needs measures

against flooding due to climate change.

2050

Extreme rainfall intensity + 12%

Sea level rise 15-40 cm

| 31 oktober 2014 | © ARCADIS 2014 Dia 5

Solution

• Define preferred measures supported by

hydraulic/ hydrological modelling, OpenMI

and RTC.

| 31 oktober 2014 | © ARCADIS 2014 Dia 6

Results

• Chosen combination of measures against

flooding

• Joint stochastic variables

• Effects of climate change

• Deeper insides into functioning of water

system

• Insides in the RTC and forecasting

| 31 oktober 2014 | © ARCADIS 2014 Dia 7

Contribution OpenMI

• Connection between Sobek and RTC-Tools

• Stochastic modelling

• Joint probabilities study

• No license fees

• Transparent control

| 31 oktober 2014 | © ARCADIS 2014 Dia 8

How does it work?

OpenMi

SOBEK RTC-Tools

| 31 oktober 2014 | © ARCADIS 2014 Dia 9

• Different type of events (wind, rainfall

intensity, sea level)

• Statistical analyses

• Return periods

• Flood risk?

• But: the correlation between parameters not

included, therefore: study joint probabilities

Stochastic modelling

| 31 oktober 2014 | © ARCADIS 2014 Dia 10

Joint extremes

All rainfall events

highest10%

A

ll s

torm

surg

e e

vents

Joint probability : 1%

| 31 oktober 2014 | © ARCADIS 2014 Dia 11

Joint extremes

All rainfall events

Highest 10%

A

ll s

torm

surg

e e

vents

Joint probability: >1%

| 31 oktober 2014 | © ARCADIS 2014 Dia 12

Expanding possibilities

• Hydrology and hydrodynamics

• Real Time Control

• Real Time Predictive capabilities

Forecast

OpenMi

SOBEK RTC-Tools SOBEK Rain

| 31 oktober 2014 | © ARCADIS 2014 Dia 13

Expanding possibilities

• Hydrology and hydrodynamics

• Real Time Control

• Real Time Predictive capabilities

• Extremely large datasets (800 years)

Delft FEWS

Forecast

OpenMi

SOBEK RTC-Tools SOBEK Rain

| 31 oktober 2014 | © ARCADIS 2014 Dia 14

Technical

implementation

SOBEK Model Adapter

Delft FEWS

•Boundaries •Precipitation •Evaporation •Initial states

•Predictive run RR and Flow

•His2XML converts load to input for RTC Tools

OpenMI

SOBEK

RTC Tools

Get output results

1 2 3 4 5

800 x

| 31 oktober 2014 | © ARCADIS 2014 Dia 15

Results

-0.50

-0.45

-0.40

-0.35

-0.30

-0.25

-0.20

-0.15

-0.10

-0.05

0.00

0.05

0.10

0.15

0.20

1 10 100 1000

Wat

erh

oo

gte

(m

NA

P)

Herhalingstijd (jaren)

Extremen gemiddeld boezempeil onder verschillende methodes

Tijdreeksmethode

Stochastenmethode

Stoch. result

800 year time series result

| 31 oktober 2014 | © ARCADIS 2014 Dia 16

Conclusions

• Stochastic analysis and time series analysis

leads into deeper insight of flood risk

Next?

• Flood early warning system

• Operational system for RTC

• Optimization on power usage and CO2

reduction

| 31 oktober 2014 | © ARCADIS 2014 Dia 17

Recommendations

• Faster OpenMI

• Advanced knowledge Sobek & OpenMI

required