Emulation of a Detailed Urban Drainage Simulator to Be Applied for Short-Term Predictions

Mahmood Mahmoodian*, J.A. Torres-Matallana, Ulrich Leopold, Georges Schutz, Francois Clemens

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference paperAcademicpeer-review

Abstract

The challenge of this study is to investigate on applicability of a data-driven Gaussian Process Emulator (GPE) technique to develop a surrogate model for a computationally expensive and detailed urban drainage simulator. The novelty is the consideration of (short) time series for the simulation inputs and outputs. Such simulation setup is interesting in applications such as Model Predictive Control (MPC) in which numerous, fast and frequent simulation results are required. Here, an emulator is developed to predict a storage tank’s volume in a small case study in Luxembourg. Three main inputs are considered as the GPE’s parameters: Initial volume in the tank, the level in which the outlet pump of the tank must start to work, and the time series of expected rainfall in the upcoming 2 h. The output of interest is the total volume of the storage tank for the next 24 h. A dataset of 2000 input-output scenarios were produced using different possible combinations of the inputs and running the detailed simulator (InfoWorks® ICM). 80% of the dataset were applied to train the emulator and 20% to validate the results. Distributions of Nash-Sutcliffe efficiency and Volumetric Efficiency are presented as indicators for quantification of the emulation error. Based on the preliminary results, it can be concluded that the introduced technique is able to reduce the simulations runtime significantly while imposing some inevitable accuracy cost. More investigation is required to validate the more generic applicability of this technique for multiple outputs and interactions between different urban drainage components.

Original languageEnglish
Title of host publicationNew Trends in Urban Drainage Modelling - UDM 2018
EditorsGiorgio Mannina
PublisherSpringer Verlag
Pages592-596
Number of pages5
ISBN (Print)9783319998664
DOIs
Publication statusPublished - 1 Sep 2019
Event11th International Conference on Urban Drainage Modelling, UDM 2018 - Palermo, Italy
Duration: 23 Sep 201826 Sep 2018

Publication series

NameGreen Energy and Technology
ISSN (Print)1865-3529
ISSN (Electronic)1865-3537

Conference

Conference11th International Conference on Urban Drainage Modelling, UDM 2018
CountryItaly
CityPalermo
Period23/09/1826/09/18

Fingerprint

urban drainage
Drainage
simulator
Time series
Simulators
storage tank
Model predictive control
prediction
simulation
Rain
Pumps
time series
train
Costs
pump
rainfall
cost

Keywords

  • Emulator
  • Gaussian process
  • InfoWorks® ICM
  • Surrogate model
  • Urban drainage

Cite this

Mahmoodian, M., Torres-Matallana, J. A., Leopold, U., Schutz, G., & Clemens, F. (2019). Emulation of a Detailed Urban Drainage Simulator to Be Applied for Short-Term Predictions. In G. Mannina (Ed.), New Trends in Urban Drainage Modelling - UDM 2018 (pp. 592-596). (Green Energy and Technology). Springer Verlag. https://doi.org/10.1007/978-3-319-99867-1_102
Mahmoodian, Mahmood ; Torres-Matallana, J.A. ; Leopold, Ulrich ; Schutz, Georges ; Clemens, Francois. / Emulation of a Detailed Urban Drainage Simulator to Be Applied for Short-Term Predictions. New Trends in Urban Drainage Modelling - UDM 2018. editor / Giorgio Mannina. Springer Verlag, 2019. pp. 592-596 (Green Energy and Technology).
@inproceedings{5c307b33709942cd97db03305756ce5b,
title = "Emulation of a Detailed Urban Drainage Simulator to Be Applied for Short-Term Predictions",
abstract = "The challenge of this study is to investigate on applicability of a data-driven Gaussian Process Emulator (GPE) technique to develop a surrogate model for a computationally expensive and detailed urban drainage simulator. The novelty is the consideration of (short) time series for the simulation inputs and outputs. Such simulation setup is interesting in applications such as Model Predictive Control (MPC) in which numerous, fast and frequent simulation results are required. Here, an emulator is developed to predict a storage tank’s volume in a small case study in Luxembourg. Three main inputs are considered as the GPE’s parameters: Initial volume in the tank, the level in which the outlet pump of the tank must start to work, and the time series of expected rainfall in the upcoming 2 h. The output of interest is the total volume of the storage tank for the next 24 h. A dataset of 2000 input-output scenarios were produced using different possible combinations of the inputs and running the detailed simulator (InfoWorks{\circledR} ICM). 80{\%} of the dataset were applied to train the emulator and 20{\%} to validate the results. Distributions of Nash-Sutcliffe efficiency and Volumetric Efficiency are presented as indicators for quantification of the emulation error. Based on the preliminary results, it can be concluded that the introduced technique is able to reduce the simulations runtime significantly while imposing some inevitable accuracy cost. More investigation is required to validate the more generic applicability of this technique for multiple outputs and interactions between different urban drainage components.",
keywords = "Emulator, Gaussian process, InfoWorks{\circledR} ICM, Surrogate model, Urban drainage",
author = "Mahmood Mahmoodian and J.A. Torres-Matallana and Ulrich Leopold and Georges Schutz and Francois Clemens",
year = "2019",
month = "9",
day = "1",
doi = "10.1007/978-3-319-99867-1_102",
language = "English",
isbn = "9783319998664",
series = "Green Energy and Technology",
publisher = "Springer Verlag",
pages = "592--596",
editor = "Giorgio Mannina",
booktitle = "New Trends in Urban Drainage Modelling - UDM 2018",
address = "Germany",

}

Mahmoodian, M, Torres-Matallana, JA, Leopold, U, Schutz, G & Clemens, F 2019, Emulation of a Detailed Urban Drainage Simulator to Be Applied for Short-Term Predictions. in G Mannina (ed.), New Trends in Urban Drainage Modelling - UDM 2018. Green Energy and Technology, Springer Verlag, pp. 592-596, 11th International Conference on Urban Drainage Modelling, UDM 2018, Palermo, Italy, 23/09/18. https://doi.org/10.1007/978-3-319-99867-1_102

Emulation of a Detailed Urban Drainage Simulator to Be Applied for Short-Term Predictions. / Mahmoodian, Mahmood; Torres-Matallana, J.A.; Leopold, Ulrich; Schutz, Georges; Clemens, Francois.

New Trends in Urban Drainage Modelling - UDM 2018. ed. / Giorgio Mannina. Springer Verlag, 2019. p. 592-596 (Green Energy and Technology).

Research output: Chapter in Book/Report/Conference proceedingConference paperAcademicpeer-review

TY - GEN

T1 - Emulation of a Detailed Urban Drainage Simulator to Be Applied for Short-Term Predictions

AU - Mahmoodian, Mahmood

AU - Torres-Matallana, J.A.

AU - Leopold, Ulrich

AU - Schutz, Georges

AU - Clemens, Francois

PY - 2019/9/1

Y1 - 2019/9/1

N2 - The challenge of this study is to investigate on applicability of a data-driven Gaussian Process Emulator (GPE) technique to develop a surrogate model for a computationally expensive and detailed urban drainage simulator. The novelty is the consideration of (short) time series for the simulation inputs and outputs. Such simulation setup is interesting in applications such as Model Predictive Control (MPC) in which numerous, fast and frequent simulation results are required. Here, an emulator is developed to predict a storage tank’s volume in a small case study in Luxembourg. Three main inputs are considered as the GPE’s parameters: Initial volume in the tank, the level in which the outlet pump of the tank must start to work, and the time series of expected rainfall in the upcoming 2 h. The output of interest is the total volume of the storage tank for the next 24 h. A dataset of 2000 input-output scenarios were produced using different possible combinations of the inputs and running the detailed simulator (InfoWorks® ICM). 80% of the dataset were applied to train the emulator and 20% to validate the results. Distributions of Nash-Sutcliffe efficiency and Volumetric Efficiency are presented as indicators for quantification of the emulation error. Based on the preliminary results, it can be concluded that the introduced technique is able to reduce the simulations runtime significantly while imposing some inevitable accuracy cost. More investigation is required to validate the more generic applicability of this technique for multiple outputs and interactions between different urban drainage components.

AB - The challenge of this study is to investigate on applicability of a data-driven Gaussian Process Emulator (GPE) technique to develop a surrogate model for a computationally expensive and detailed urban drainage simulator. The novelty is the consideration of (short) time series for the simulation inputs and outputs. Such simulation setup is interesting in applications such as Model Predictive Control (MPC) in which numerous, fast and frequent simulation results are required. Here, an emulator is developed to predict a storage tank’s volume in a small case study in Luxembourg. Three main inputs are considered as the GPE’s parameters: Initial volume in the tank, the level in which the outlet pump of the tank must start to work, and the time series of expected rainfall in the upcoming 2 h. The output of interest is the total volume of the storage tank for the next 24 h. A dataset of 2000 input-output scenarios were produced using different possible combinations of the inputs and running the detailed simulator (InfoWorks® ICM). 80% of the dataset were applied to train the emulator and 20% to validate the results. Distributions of Nash-Sutcliffe efficiency and Volumetric Efficiency are presented as indicators for quantification of the emulation error. Based on the preliminary results, it can be concluded that the introduced technique is able to reduce the simulations runtime significantly while imposing some inevitable accuracy cost. More investigation is required to validate the more generic applicability of this technique for multiple outputs and interactions between different urban drainage components.

KW - Emulator

KW - Gaussian process

KW - InfoWorks® ICM

KW - Surrogate model

KW - Urban drainage

U2 - 10.1007/978-3-319-99867-1_102

DO - 10.1007/978-3-319-99867-1_102

M3 - Conference paper

SN - 9783319998664

T3 - Green Energy and Technology

SP - 592

EP - 596

BT - New Trends in Urban Drainage Modelling - UDM 2018

A2 - Mannina, Giorgio

PB - Springer Verlag

ER -

Mahmoodian M, Torres-Matallana JA, Leopold U, Schutz G, Clemens F. Emulation of a Detailed Urban Drainage Simulator to Be Applied for Short-Term Predictions. In Mannina G, editor, New Trends in Urban Drainage Modelling - UDM 2018. Springer Verlag. 2019. p. 592-596. (Green Energy and Technology). https://doi.org/10.1007/978-3-319-99867-1_102