Construction of uncertainty models for renewable energy systems on multiple time scales

Yuxuan Zheng; Shihua Luo; Weihao Hu; Zhenyuan Zhang; Qi Huang; Zhe Chen; Sayed Abulanwar

doi:10.1109/EI256261.2022.10116873

Construction of uncertainty models for renewable energy systems on multiple time scales

Yuxuan Zheng, Shihua Luo, Weihao Hu, Zhenyuan Zhang, Qi Huang, Zhe Chen, Sayed Abulanwar

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution book

Abstract

The output of the renewable energy system has strong randomness and volatility, and the construction of the uncertainty model of the system is of great significance to the power system planning. Scenario generation is a common method to analyze uncertainty problems by constructing deterministic scenarios. Since traditional scenario generation methods are sensitive to abnormal data, they cannot handle large-dimensional data well when the time scale is large. K-Medoids and Dynamic Time Warping (K-Medoids+DTW) methods can reduce the impact of abnormal data and generate scenarios for renewable energy output and load changes at multiple time scales. On this basis, the vector autoregressive model (VAR) is used to construct supply and demand matching patterns in different scenarios. This paper uses the historical data of renewable energy, adopts the improved clustering algorithm to generate typical scenarios, and builds a source-load supply-demand matching model, and finally verifies the validity of the model.

Original language	English
Title of host publication	2022 IEEE 6th Conference on Energy Internet and Energy System Integration (EI2)
Pages	561-567
Number of pages	7
ISBN (Electronic)	979-8-3503-4715-9
DOIs	https://doi.org/10.1109/EI256261.2022.10116873
Publication status	Published - 10 May 2023
Event	2022 IEEE 6th Conference on Energy Internet and Energy System Integration (EI2) - Chengdu, China Duration: 11 Nov 2022 → 13 Nov 2022

Conference

Conference	2022 IEEE 6th Conference on Energy Internet and Energy System Integration (EI2)
Country/Territory	China
City	Chengdu
Period	11/11/22 → 13/11/22

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 7 Affordable and Clean Energy

Keywords

Renewable energy sources
Reactive power
Uncertainty
Supply and demand
Clustering algorithms
System integration
Wind power generation

Access to Document

10.1109/EI256261.2022.10116873

Cite this

@inproceedings{f10a6600fe30411c89dd56ca109ec369,

title = "Construction of uncertainty models for renewable energy systems on multiple time scales",

abstract = "The output of the renewable energy system has strong randomness and volatility, and the construction of the uncertainty model of the system is of great significance to the power system planning. Scenario generation is a common method to analyze uncertainty problems by constructing deterministic scenarios. Since traditional scenario generation methods are sensitive to abnormal data, they cannot handle large-dimensional data well when the time scale is large. K-Medoids and Dynamic Time Warping (K-Medoids+DTW) methods can reduce the impact of abnormal data and generate scenarios for renewable energy output and load changes at multiple time scales. On this basis, the vector autoregressive model (VAR) is used to construct supply and demand matching patterns in different scenarios. This paper uses the historical data of renewable energy, adopts the improved clustering algorithm to generate typical scenarios, and builds a source-load supply-demand matching model, and finally verifies the validity of the model.",

keywords = "Renewable energy sources, Reactive power, Uncertainty, Supply and demand, Clustering algorithms, System integration, Wind power generation",

author = "Yuxuan Zheng and Shihua Luo and Weihao Hu and Zhenyuan Zhang and Qi Huang and Zhe Chen and Sayed Abulanwar",

year = "2023",

month = may,

day = "10",

doi = "10.1109/EI256261.2022.10116873",

language = "English",

isbn = "979-8-3503-4716-6",

pages = "561--567",

booktitle = "2022 IEEE 6th Conference on Energy Internet and Energy System Integration (EI2)",

note = "2022 IEEE 6th Conference on Energy Internet and Energy System Integration (EI2) ; Conference date: 11-11-2022 Through 13-11-2022",

}

Zheng, Y, Luo, S, Hu, W, Zhang, Z, Huang, Q, Chen, Z & Abulanwar, S 2023, Construction of uncertainty models for renewable energy systems on multiple time scales. in 2022 IEEE 6th Conference on Energy Internet and Energy System Integration (EI2)., 10116873, pp. 561-567, 2022 IEEE 6th Conference on Energy Internet and Energy System Integration (EI2), Chengdu, China, 11/11/22. https://doi.org/10.1109/EI256261.2022.10116873

TY - GEN

T1 - Construction of uncertainty models for renewable energy systems on multiple time scales

AU - Zheng, Yuxuan

AU - Luo, Shihua

AU - Hu, Weihao

AU - Zhang, Zhenyuan

AU - Huang, Qi

AU - Chen, Zhe

AU - Abulanwar, Sayed

PY - 2023/5/10

Y1 - 2023/5/10

N2 - The output of the renewable energy system has strong randomness and volatility, and the construction of the uncertainty model of the system is of great significance to the power system planning. Scenario generation is a common method to analyze uncertainty problems by constructing deterministic scenarios. Since traditional scenario generation methods are sensitive to abnormal data, they cannot handle large-dimensional data well when the time scale is large. K-Medoids and Dynamic Time Warping (K-Medoids+DTW) methods can reduce the impact of abnormal data and generate scenarios for renewable energy output and load changes at multiple time scales. On this basis, the vector autoregressive model (VAR) is used to construct supply and demand matching patterns in different scenarios. This paper uses the historical data of renewable energy, adopts the improved clustering algorithm to generate typical scenarios, and builds a source-load supply-demand matching model, and finally verifies the validity of the model.

AB - The output of the renewable energy system has strong randomness and volatility, and the construction of the uncertainty model of the system is of great significance to the power system planning. Scenario generation is a common method to analyze uncertainty problems by constructing deterministic scenarios. Since traditional scenario generation methods are sensitive to abnormal data, they cannot handle large-dimensional data well when the time scale is large. K-Medoids and Dynamic Time Warping (K-Medoids+DTW) methods can reduce the impact of abnormal data and generate scenarios for renewable energy output and load changes at multiple time scales. On this basis, the vector autoregressive model (VAR) is used to construct supply and demand matching patterns in different scenarios. This paper uses the historical data of renewable energy, adopts the improved clustering algorithm to generate typical scenarios, and builds a source-load supply-demand matching model, and finally verifies the validity of the model.

KW - Renewable energy sources

KW - Reactive power

KW - Uncertainty

KW - Supply and demand

KW - Clustering algorithms

KW - System integration

KW - Wind power generation

U2 - 10.1109/EI256261.2022.10116873

DO - 10.1109/EI256261.2022.10116873

M3 - Conference contribution book

SN - 979-8-3503-4716-6

SP - 561

EP - 567

BT - 2022 IEEE 6th Conference on Energy Internet and Energy System Integration (EI2)

T2 - 2022 IEEE 6th Conference on Energy Internet and Energy System Integration (EI2)

Y2 - 11 November 2022 through 13 November 2022

ER -

Construction of uncertainty models for renewable energy systems on multiple time scales

Abstract

Conference

UN SDGs

Keywords

Access to Document

Fingerprint

Cite this