TY - JOUR
T1 - Designing a standalone wind-diesel-CAES hybrid energy system by using a scenario-based bi-level programming method
AU - Xu, Xiao
AU - Hu, Weihao
AU - Cao, Di
AU - Huang, Qi
AU - Liu, Wen
AU - Liu, Zhou
AU - Chen, Zhe
AU - Lund, Henrik
N1 - Funding Information:
This work was supported by the National Key Research and Development Program of China ( 2018YFB0905200 ).
Publisher Copyright:
© 2020 Elsevier Ltd
Copyright:
Copyright 2020 Elsevier B.V., All rights reserved.
PY - 2020/5
Y1 - 2020/5
N2 - Compressed air energy storage (CAES) systems are promising for the application of a standalone hybrid system. This study adopts a scenario-based bi-level programming method to design a standalone hybrid system that mainly contains wind turbines, diesel generators (DGs), and a CAES system. The demand response is considered as a deferrable load. The uncertainties on the wind power outputs and load demand are modeled using scenario generation and reduction techniques. The generated scenarios are used in a bi-level programming model for designing the hybrid energy system (HES). The model is composed of an outer planning layer and an inner operation layer. The outer layer optimizes the size of each component in the HES using a quantum particle swarm optimization (QPSO) method with the objective of minimizing daily total costs including daily investment costs and daily operating costs. On the other hand, the inner layer optimizes the operational strategies of the HES using a sequential quadratic programming method with the objective of minimizing the total operating costs, including the generation and emission costs of the DGs and the degradation cost of the CAES system. The well-established HES tool HOMER is used to validate the results obtained by the developed and adopted models. The results indicate that 1) the QPSO method performs better than the particle swarm optimization and genetic algorithm methods. 2) The results obtained by the scenario-based bi-level programming method have an average similarity of approximately 97%, which is very high compared to that of the results obtained by HOMER.
AB - Compressed air energy storage (CAES) systems are promising for the application of a standalone hybrid system. This study adopts a scenario-based bi-level programming method to design a standalone hybrid system that mainly contains wind turbines, diesel generators (DGs), and a CAES system. The demand response is considered as a deferrable load. The uncertainties on the wind power outputs and load demand are modeled using scenario generation and reduction techniques. The generated scenarios are used in a bi-level programming model for designing the hybrid energy system (HES). The model is composed of an outer planning layer and an inner operation layer. The outer layer optimizes the size of each component in the HES using a quantum particle swarm optimization (QPSO) method with the objective of minimizing daily total costs including daily investment costs and daily operating costs. On the other hand, the inner layer optimizes the operational strategies of the HES using a sequential quadratic programming method with the objective of minimizing the total operating costs, including the generation and emission costs of the DGs and the degradation cost of the CAES system. The well-established HES tool HOMER is used to validate the results obtained by the developed and adopted models. The results indicate that 1) the QPSO method performs better than the particle swarm optimization and genetic algorithm methods. 2) The results obtained by the scenario-based bi-level programming method have an average similarity of approximately 97%, which is very high compared to that of the results obtained by HOMER.
KW - Compressed air energy storage
KW - Demand response
KW - Quantum particle swarm optimization
KW - Scenario-based bi-level programming
KW - Wind-diesel-CAES hybrid energy system
UR - http://www.scopus.com/inward/record.url?scp=85082427167&partnerID=8YFLogxK
U2 - 10.1016/j.enconman.2020.112759
DO - 10.1016/j.enconman.2020.112759
M3 - Journal article
AN - SCOPUS:85082427167
SN - 0196-8904
VL - 211
JO - Energy Conversion and Management
JF - Energy Conversion and Management
M1 - 112759
ER -