A fast surrogate model tailor-made for real time control

Morten Borup; Cecilie Thrysøe; Karsten Arnbjerg-Nielsen; Francesco Righetti; Peter Steen Mikkelsen

A fast surrogate model tailor-made for real time control

Morten Borup, Cecilie Thrysøe, Karsten Arnbjerg-Nielsen, Francesco Righetti, Peter Steen Mikkelsen

Research output: Contribution to conference › Conference abstract for conference › Research › peer-review

135 Downloads (Pure)

Abstract

A surrogate model of a detailed hydraulic urban drainage model is created for supplying inflow forecasts to an MPC model for 31 separate locations. The original model is subdivided into 66 relationships extracted from the original model. The surrogate model is 9000 times faster than the original model, with just a minor deviation from the original model results.

Original language	English
Publication date	2017
Number of pages	4
Publication status	Published - 2017
Event	14th IWA/IAHR International Conference on Urban Drainage 2017 - Prague Congress Centre, Prague, Czech Republic Duration: 10 Sep 2017 → 15 Sep 2017 http://www.icud2017.org/

Conference

Conference	14th IWA/IAHR International Conference on Urban Drainage 2017
Location	Prague Congress Centre
Country	Czech Republic
City	Prague
Period	10/09/2017 → 15/09/2017
Internet address	http://www.icud2017.org/

Keywords

Surrogate models
Piecewise linear
Volume-discharge relationships
Real time control
Input models

Access to Document

SurModForRTC 2 0Final published version, 298 KB

Cite this

@conference{a63311f8664f4b1fb5964e86d786504f,

title = "A fast surrogate model tailor-made for real time control",

abstract = "A surrogate model of a detailed hydraulic urban drainage model is created for supplying inflow forecasts to an MPC model for 31 separate locations. The original model is subdivided into 66 relationships extracted from the original model. The surrogate model is 9000 times faster than the original model, with just a minor deviation from the original model results. ",

keywords = "Surrogate models, Piecewise linear, Volume-discharge relationships, Real time control, Input models",

author = "Morten Borup and Cecilie Thrys{\o}e and Karsten Arnbjerg-Nielsen and Francesco Righetti and Mikkelsen, {Peter Steen}",

year = "2017",

language = "English",

note = "14th IWA/IAHR International Conference on Urban Drainage 2017, ICUD 2017 ; Conference date: 10-09-2017 Through 15-09-2017",

url = "http://www.icud2017.org/",

}

A fast surrogate model tailor-made for real time control. / Borup, Morten ; Thrysøe, Cecilie ; Arnbjerg-Nielsen, Karsten; Righetti, Francesco; Mikkelsen, Peter Steen.

2017. Abstract from 14th IWA/IAHR International Conference on Urban Drainage 2017, Prague, Czech Republic.

Research output: Contribution to conference › Conference abstract for conference › Research › peer-review

TY - ABST

T1 - A fast surrogate model tailor-made for real time control

AU - Borup, Morten

AU - Thrysøe, Cecilie

AU - Arnbjerg-Nielsen, Karsten

AU - Righetti, Francesco

AU - Mikkelsen, Peter Steen

PY - 2017

Y1 - 2017

N2 - A surrogate model of a detailed hydraulic urban drainage model is created for supplying inflow forecasts to an MPC model for 31 separate locations. The original model is subdivided into 66 relationships extracted from the original model. The surrogate model is 9000 times faster than the original model, with just a minor deviation from the original model results.

AB - A surrogate model of a detailed hydraulic urban drainage model is created for supplying inflow forecasts to an MPC model for 31 separate locations. The original model is subdivided into 66 relationships extracted from the original model. The surrogate model is 9000 times faster than the original model, with just a minor deviation from the original model results.

KW - Surrogate models

KW - Piecewise linear

KW - Volume-discharge relationships

KW - Real time control

KW - Input models

M3 - Conference abstract for conference

T2 - 14th IWA/IAHR International Conference on Urban Drainage 2017

Y2 - 10 September 2017 through 15 September 2017

ER -