Analytic versus solver-based calculated daily operations of district energy plants

Research output: Contribution to journalJournal articleResearchpeer-review

Abstract

Flexible District Energy plants providing heating and cooling to cities represent an important part of future smart renewable energy systems. Equipped with large combined heat and power units, heat pumps and thermal energy storage they have the possibility to provide flexibility – but an optimized unit commitment is required. A common conclusion has been that unit commitment based on analytic methods is not useful. However, the market-based operation of District Energy plants often being reduced to participation in one or two electricity markets, simplifies the unit commitment problem and brings analytic unit commitment methods back as potentially attractive methods for District Energy plants. This is demonstrated in this paper by establishing a complex generic District Energy plant which is yet so simplified that a solver-based Mixed Integer Linear Programming method is able to deliver optimal unit commitments. An advanced analytic unit commitment method for district energy plants is proposed and the comparison of the unit commitments made by this method with the optimal solver-based unit commitments shows that the method delivers operation income within 1% of the optimal operation income, which is fully adequate for daily operation planning, yearly budgeting and long-term investment analysis for this generic District Energy plant.
Original languageEnglish
JournalEnergy
Volume175
Pages (from-to) 333-344
Number of pages12
ISSN0360-5442
DOIs
Publication statusPublished - 15 May 2019

Fingerprint

Budget control
Thermal energy
Linear programming
Energy storage
Pumps
Cooling
Heating
Planning
Hot Temperature
Power markets

Cite this

@article{5073d041592e4ff1afdd5f7c7aef7031,
title = "Analytic versus solver-based calculated daily operations of district energy plants",
abstract = "Flexible District Energy plants providing heating and cooling to cities represent an important part of future smart renewable energy systems. Equipped with large combined heat and power units, heat pumps and thermal energy storage they have the possibility to provide flexibility – but an optimized unit commitment is required. A common conclusion has been that unit commitment based on analytic methods is not useful. However, the market-based operation of District Energy plants often being reduced to participation in one or two electricity markets, simplifies the unit commitment problem and brings analytic unit commitment methods back as potentially attractive methods for District Energy plants. This is demonstrated in this paper by establishing a complex generic District Energy plant which is yet so simplified that a solver-based Mixed Integer Linear Programming method is able to deliver optimal unit commitments. An advanced analytic unit commitment method for district energy plants is proposed and the comparison of the unit commitments made by this method with the optimal solver-based unit commitments shows that the method delivers operation income within 1{\%} of the optimal operation income, which is fully adequate for daily operation planning, yearly budgeting and long-term investment analysis for this generic District Energy plant.",
author = "Andersen, {Anders N.} and {\O}stergaard, {Poul Alberg}",
year = "2019",
month = "5",
day = "15",
doi = "10.1016/j.energy.2019.03.096Get",
language = "English",
volume = "175",
pages = "333--344",
journal = "Energy",
issn = "0360-5442",
publisher = "Pergamon Press",

}

Analytic versus solver-based calculated daily operations of district energy plants. / Andersen, Anders N.; Østergaard, Poul Alberg.

In: Energy, Vol. 175, 15.05.2019, p. 333-344.

Research output: Contribution to journalJournal articleResearchpeer-review

TY - JOUR

T1 - Analytic versus solver-based calculated daily operations of district energy plants

AU - Andersen, Anders N.

AU - Østergaard, Poul Alberg

PY - 2019/5/15

Y1 - 2019/5/15

N2 - Flexible District Energy plants providing heating and cooling to cities represent an important part of future smart renewable energy systems. Equipped with large combined heat and power units, heat pumps and thermal energy storage they have the possibility to provide flexibility – but an optimized unit commitment is required. A common conclusion has been that unit commitment based on analytic methods is not useful. However, the market-based operation of District Energy plants often being reduced to participation in one or two electricity markets, simplifies the unit commitment problem and brings analytic unit commitment methods back as potentially attractive methods for District Energy plants. This is demonstrated in this paper by establishing a complex generic District Energy plant which is yet so simplified that a solver-based Mixed Integer Linear Programming method is able to deliver optimal unit commitments. An advanced analytic unit commitment method for district energy plants is proposed and the comparison of the unit commitments made by this method with the optimal solver-based unit commitments shows that the method delivers operation income within 1% of the optimal operation income, which is fully adequate for daily operation planning, yearly budgeting and long-term investment analysis for this generic District Energy plant.

AB - Flexible District Energy plants providing heating and cooling to cities represent an important part of future smart renewable energy systems. Equipped with large combined heat and power units, heat pumps and thermal energy storage they have the possibility to provide flexibility – but an optimized unit commitment is required. A common conclusion has been that unit commitment based on analytic methods is not useful. However, the market-based operation of District Energy plants often being reduced to participation in one or two electricity markets, simplifies the unit commitment problem and brings analytic unit commitment methods back as potentially attractive methods for District Energy plants. This is demonstrated in this paper by establishing a complex generic District Energy plant which is yet so simplified that a solver-based Mixed Integer Linear Programming method is able to deliver optimal unit commitments. An advanced analytic unit commitment method for district energy plants is proposed and the comparison of the unit commitments made by this method with the optimal solver-based unit commitments shows that the method delivers operation income within 1% of the optimal operation income, which is fully adequate for daily operation planning, yearly budgeting and long-term investment analysis for this generic District Energy plant.

U2 - 10.1016/j.energy.2019.03.096Get

DO - 10.1016/j.energy.2019.03.096Get

M3 - Journal article

VL - 175

SP - 333

EP - 344

JO - Energy

JF - Energy

SN - 0360-5442

ER -