Publications

Detailed Information

An Interval-Based Two-Stage Stochastic Optimization Model for the Unit Commitment Problem Under Demand Uncertainty : 수요의 불확실성 하에서의 발전구간 기반 2단계 추계적 발전계획 모형

DC Field Value Language
dc.contributor.advisor이경식-
dc.contributor.author정호진-
dc.date.accessioned2023-06-29T01:52:07Z-
dc.date.available2023-06-29T01:52:07Z-
dc.date.issued2023-
dc.identifier.other000000174246-
dc.identifier.urihttps://hdl.handle.net/10371/193128-
dc.identifier.urihttps://dcollection.snu.ac.kr/common/orgView/000000174246ko_KR
dc.description학위논문(석사) -- 서울대학교대학원 : 공과대학 산업공학과, 2023. 2. 이경식.-
dc.description.abstractThe unit commitment problem aims to find a minimum-cost on/off status and amount of generation for each generator while satisfying the electricity demand and operational requirements. To efficiently deal with demand uncertainty, the two-stage stochastic optimization models have been widely used in the literature, where the on/off status is decided in the first stage and the amount of generation is in the second stage. However, they often suffer from excessive computational burden as the number of demand scenarios increases. In this thesis, we propose an interval-based two-stage stochastic optimization model to mitigate the drawback under the period-wise independent demand assumption. In the model, an interval of a generator, which is a range of the amount of generation, is determined along with its on/off status for each period. It enables the second-stage problem to be decomposed in a period-wise manner, which reduces the need for a large number of scenarios. We also propose a compact Benders reformulation by exploiting the property of the subproblem. Lastly, we show that the bounds on the expected costs can be obtained for the proposed model. Computational experiments were conducted to show the effectiveness and efficiency of the proposed model.-
dc.description.abstract발전계획 문제는 전력 수요와 운영상의 제약을 만족하면서 전체 운영 비용을 최소로 하는 발전기별 운전상태를 찾는 것을 목표로 한다. 수요의 불확실성에 효율적으로 대처하기 위해, 1단계에서 운전상태를 결정하고 2단계에서 발전량을 결정하는 2단계 추계적 최적화 모형이 문헌에서 널리 사용되어 왔으나, 수요 시나리오의 수가 증가함에 따라 과도한 계산 부담을 겪는 경우가 많다. 본 논문에서는 이러한 단점을 보완하기 위해 수요가 시점별로 독립이라는 가정 하에서 발전구간 기반 2단계 추계적 최적화 모형을 제안한다. 해당 모형에서는 각 시점별로 발전기의 운전 여부뿐만 아니라 발전구간, 즉 발전량의 범위 또한 함께 결정된다. 이는 2단계 문제가 시점별로 분해될 수 있도록 하고, 많은 수의 시나리오가 필요하지 않게끔 해준다. 또한, 부문제의 성질을 이용하여 압축된 벤더스 모형을 제안한다. 마지막으로, 제안한 모형에 대해 기대비용의 한계치를 구할 수 있음을 보인다. 수치적 실험을 수행하여 제안한 모형의 효과성과 효율성을 입증하였다.-
dc.description.tableofcontentsChapter 1 Introduction 1
1.1 Background 1
1.2 Problem Description 3
1.3 Literature Review 6
1.3.1 Unit Commitment Problem 6
1.3.2 Unit Commitment Models Under Uncertainty 6
1.3.3 Solution Approaches for the Stochastic Unit Commitment Model 8
1.4 Motivation and Contributions 10
1.5 Organization of the Thesis 12
Chapter 2 Two-stage Stochastic Unit Commitment Model 13
2.1 Modeling Framework 13
2.2 Model Formulation 14
Chapter 3 Interval-based Two-stage Stochastic Unit Commitment Model 18
3.1 Modeling Framework 18
3.1.1 Finite Interval Design 20
3.1.2 Infinite Interval Design 25
3.2 Model Formulation 26
3.2.1 Finite Interval Design 26
3.2.2 Infinite Interval Design 29
3.2.3 Comparison of the Objective Values 31
Chapter 4 Solution Approaches 34
4.1 Bounding Method 34
4.2 Benders Reformulation 38
Chapter 5 Computational Experiments 45
5.1 Experiment Setting 45
5.2 Evaluation Method 47
5.3 Experiment Results 47
5.3.1 Performance Comparison of the Models 47
5.3.2 Robustness Test on the Different Demand Distributions 50
5.3.3 Efficiency of the Bounding Method 52
5.3.4 Efficiency of the Benders Reformulation 54
5.3.5 Comparative Analysis of Various Interval Design Methods 55
Chapter 6 Conclusion 58
Bibliography 60
국문초록 67
-
dc.format.extentviii, 67-
dc.language.isoeng-
dc.publisher서울대학교 대학원-
dc.subjectUnit Commitment-
dc.subjectStochastic Optimization-
dc.subjectInterval-
dc.subjectBounding Method-
dc.subjectBenders Reformulation-
dc.subject.ddc670.42-
dc.titleAn Interval-Based Two-Stage Stochastic Optimization Model for the Unit Commitment Problem Under Demand Uncertainty-
dc.title.alternative수요의 불확실성 하에서의 발전구간 기반 2단계 추계적 발전계획 모형-
dc.typeThesis-
dc.typeDissertation-
dc.contributor.AlternativeAuthor한국어-
dc.contributor.department공과대학 산업공학과-
dc.description.degree석사-
dc.date.awarded2023-02-
dc.identifier.uciI804:11032-000000174246-
dc.identifier.holdings000000000049▲000000000056▲000000174246▲-
Appears in Collections:
Files in This Item:

Altmetrics

Item View & Download Count

  • mendeley

Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.

Share