water demand forecasting for the optimal operation of large-scale water networks

18
Water demand forecasting for the optimal operation of large-scale drinking water networks the Barcelona case study A.K. Sampathirao * , J.M. Grosso ** , P. Sopasakis * , C. Ocampo-Martinez ** , A. Bemporad * and V. Puig ** * IMT Institute for Advanced Studies Lucca, Lucca, Italy, ** Automatic Control Dept., Technical University of Catalonia (UPC), Barcelona, Spain.

Upload: pantelis-sopasakis

Post on 14-Jul-2015

178 views

Category:

Technology


0 download

TRANSCRIPT

Page 1: Water demand forecasting for the optimal operation of large-scale water networks

Water demand forecasting for the optimal operation of large-scale drinking water networks the Barcelona case study

A.K. Sampathirao*, J.M. Grosso**, P. Sopasakis*, C. Ocampo-Martinez**, A. Bemporad* and V. Puig**

* IMT Institute for Advanced Studies Lucca, Lucca, Italy, ** Automatic Control Dept., Technical University of Catalonia (UPC), Barcelona, Spain.

Page 2: Water demand forecasting for the optimal operation of large-scale water networks

DWN Control: Goals ¡  Reduce energy consumption for pumping,

¡ Meet the demand requirements,

¡ Deliver smooth control actions,

¡  Keep the storage above safety limits,

¡  Respect the technical limitations: pressure limits, overflow limits & pumping capabilities,

¡  Have foresight: predict how the water demand and energy cost will move and act accordingly.

Page 3: Water demand forecasting for the optimal operation of large-scale water networks

Outline ¡ Description of the overall control system,

¡  Hydraulic model of the DWN,

¡ Modelling of the uncertain water demand time series,

¡  Economic MPC: the control algorithm,

¡  Simulation results.

Page 4: Water demand forecasting for the optimal operation of large-scale water networks

3380 3400 3420 3440 3460 3480 3500 3520 3540 3560

0

2

4

6

8

10

12

x 10−3 Prediction Error

Past Data

Observed

Forecast

The Control Module

Energy Price

Water Demand

Drinking Water Network

Online Measurements

Flow Pressure Quality

Forecasting Module

History Data

Data Validation Module

Validated Measurements

Commands Model Predictive Controller

(Uncertain) estimates

EFFINET Deliverable report D2.1, “Control-oriented modelling for operational management of urban water networks.”

Page 5: Water demand forecasting for the optimal operation of large-scale water networks

Hydraulic model

xk+1 = Adxk +Bduk +Gddk,

0 = Euk + Eddk

¡  Based on mass balance equations,

¡  Linear time-invariant discrete time system,

¡ with input-disturbance couplings

State: Storage in tanks

Input: Pumping

Disturbance: Water demand

Constraints mandated by mass balance equations.

C. Ocampo-Martinez, V. Puig, G. Cembrano, R. Creus, and M. Minoves. Improving water management efficiency by using optimization-based control strategies: the barcelona case study. Water Sci. & Tech.: Water supply, 9(5):565–575, 2009.

Page 6: Water demand forecasting for the optimal operation of large-scale water networks

Water demand forecasting ¡  Three approaches bore fruit: SARIMA, BATS and

RBF-SVM,

¡  The predictive ability of the models was evaluated using the average PMSE-24, that is:

PMSEHp =1

THp

k0+TX

k=k0

HpX

i=1

(d̂k+i|k � dk+i)2

Page 7: Water demand forecasting for the optimal operation of large-scale water networks

Water demand forecasting

3380 3400 3420 3440 3460 3480 3500 3520 3540 3560

0

2

4

6

8

10

12

x 10−3 Prediction Error

Past Data

Observed

Forecast

SARIMA model

¡  PMSE24 = 0.0158,

¡  25 parameters (quite simple) determined up to a high statistical significance.

Page 8: Water demand forecasting for the optimal operation of large-scale water networks

Water demand forecasting

RBF-SVM model

¡  PMSE24 = 0.0065,

¡  229 parameters (complex),

¡  10-fold cross-validation gave q2 = 0.9952,

¡  Explanatory variables: 200 past demands plus a set of binary calendar variables,

¡  Stringent confidence intervals.

3250 3260 3270 3280 3290 3300 3310 3320

3

4

5

6

7

8

9

10x 10

−3

Time [hr]

Dem

and [m

3 h

r−1]

RBF−SVM Prediction

Page 9: Water demand forecasting for the optimal operation of large-scale water networks

0 20 40 60 80 100 120 140 160 180 2000

0.01

0.02

0.03

0.04

0.05

0.06

0.07

Time [h]

Wa

ter

De

ma

nd

Flo

w [

m3/h

]

Forecasting of Water Demand

Future Past

Water demand forecasting

BATS model

¡  Box-Cox transformation, ARMA errors, Trends and Seasonality,

¡  PMSE24 = 0.0043,

¡ with just 26 parameters,

¡ Very stringent confidence intervals.

Page 10: Water demand forecasting for the optimal operation of large-scale water networks

Prefer to pump when the price is low!

Stay above the safety storage volume

PAST FUTURE

Volume in tank (m3)

Time (h)

Do not overflow!

Time (h)

Pumping (m3/h)

Avoid pumping when the price is high!

Account for pumping capabilities

Why MPC:

¡  Optimal: Computes the control actions by optimizing a performance criterion,

¡  Realistic: Accounts for the operational constraints,

¡  Predictive: Has foresight; acts early before the price or the demand changes.

How MPC works

J. B. Rawlings and D. Q. Mayne. Model predictive control: theory and design. Madison: Nob Hill Publishing, 2009.

Page 11: Water demand forecasting for the optimal operation of large-scale water networks

Economic MPC for DWN

From the forecasting module: dk+j|k = d̂k+j|k + ✏k+j|k

Estimation error, essentially bounded in:

Ek+j|k = {✏ : ✏min

k+j|k ✏ ✏max

k+j|k}

xk+j|k = x̂k+j|k +jX

l=1

A

l�1Gd✏k+l|kThe state sequence will satisfy:

Nominal state sequence satisfying the dynamics:

x̂k+j+1|k = Adx̂k+j|k +Bduk+j|k +Gdd̂k+j|k

Page 12: Water demand forecasting for the optimal operation of large-scale water networks

Economic MPC for DWN

Page 13: Water demand forecasting for the optimal operation of large-scale water networks

Economic MPC for DWN

4500 4505 4510 4515 4520 4525 4530 4535 4540 45450.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8 x 104

Time [hr]

Volu

me

[m3 ]

Safety Volume

Minimum Volume

Maximum Volume

MPC Upper Bound

MPC Lower Bound

Predicted Trajectory

Closed−loop trajectory

x̂k+j|k 2 X iM

j=1

A

j�1GdEk+j|k

Bounds on the predicted state sequence calculated by:

The sparsity of Gd enables this computation!

Page 14: Water demand forecasting for the optimal operation of large-scale water networks

Economic MPC for DWN

4500 4505 4510 4515 4520 4525 4530 4535 4540 45450.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8 x 104

Time [hr]

Volu

me

[m3 ]

Safety Volume

Minimum Volume

Maximum Volume

MPC Upper Bound

MPC Lower Bound

Predicted Trajectory

Closed−loop trajectory

x̂k+j+1|k = Adx̂k+j|k +Bduk+j|k+

+Gdd̂k+j|k

Predicted state sequence according to:

Page 15: Water demand forecasting for the optimal operation of large-scale water networks

MPC: Performance

10 20 30 40 50 60 70 80 90

0.2

0.4

0.6

0.8

MPC Control Action (1~20)

Contr

ol A

ctio

n

10 20 30 40 50 60 70 80 90

0.2

0.4

0.6

0.8

MPC Control Action (21~46)

Contr

ol A

ctio

n

10 20 30 40 50 60 70 80 900

0.1

0.2

Time [hr]

Wate

r C

ost

[e.u

.]

MPC in action •  88 demand nodes •  63 tanks •  114 pumping stations •  17 flow nodes

50 100 150 200 250 300 350 400 450 500

4

5

6

7

8

Economic Cost (E.U.)

50 100 150 200 250 300 350 400 450 500

0.5

1

1.5

2

Smooth Operation Cost

0 50 100 150 200 250 300 350 400 450 5000

2

4

6Safety Storage Cost (× 107)

Low price à Pumping

The system operator has information about the current and the predicted operation cost.

Page 16: Water demand forecasting for the optimal operation of large-scale water networks

5 10 15 20 25 30 35 40 45 50 550

20

40

60

80

100

Closed−loop MPC Simulation

Time [hr]

Reple

tion [

%]

5 10 15 20 25 30 35 40 45 50 550

0.5

1

1.5

Time [hr]

Dema

nd [m

3 /s]

MPC: Performance

10 20 30 40 50 60 70 80 90

0.2

0.4

0.6

0.8

MPC Control Action (1~20)

Contr

ol A

ctio

n

10 20 30 40 50 60 70 80 90

0.2

0.4

0.6

0.8

MPC Control Action (21~46)

Contr

ol A

ctio

n

10 20 30 40 50 60 70 80 900

0.1

0.2

Time [hr]

Wate

r C

ost

[e.u

.]

Foresight: Tanks starts loading up before a DMA asks for water.

Page 17: Water demand forecasting for the optimal operation of large-scale water networks

Work in progress ¡  Formulation of the control problem as a

stochastic economic MPC problem,

¡ Algorithms for the solution of large-scale optimisation problems,

¡ GPGPU implementations for the efficient solution of such optimisation algorithms.

Page 18: Water demand forecasting for the optimal operation of large-scale water networks

Thank you for your attention.

This work was financially supported by the EU FP7 research project EFFINET “Efficient Integrated Real-time monitoring and Control of Drinking Water Networks,” grant agreement no. 318556.