TY - JOUR
T1 - Risk-Averse Stochastic Midterm Scheduling of Thermal-Hydro-Wind System
T2 - A Network-Constrained Clustered Unit Commitment Approach
AU - Yin, Yue
AU - He, Chuan
AU - Liu, Tianqi
AU - Wu, Lei
N1 - Publisher Copyright:
© 2010-2012 IEEE.
PY - 2022/7/1
Y1 - 2022/7/1
N2 - With the increasing penetration of renewable energy, designing the efficient midterm (i.e., annual) scheduling for multi-source power systems is valuable in providing substantial cost reductions and flexibilities while ensuring system security against uncertainties of renewables. This paper presents a risk-averse two-stage stochastic midterm scheduling model for thermal-hydro-wind systems against short-term uncertainties of renewable generation. A network-constrained clustered unit commitment (NC-CUC) model is proposed to effectively incorporate short-term operation constraints into the midterm scheduling model, for improving the solution accuracy while reducing computational burden. Specifically, as the traditional CUC model cannot ensure consistent unit commitment status of generators across all scenarios, a power flow-based unit clustering method is proposed to improve the accuracy of short-term operation results. In addition, the conditional value-at-risk (CVaR) is used to quantify the system operational risk caused by uncertainties of wind generation and natural water inflows. Case studies show that the risk-averse two-stage stochastic midterm scheduling model can ensure operation dynamics and solution efficiency, and the proposed clustering method can achieve a tradeoff between solution accuracy and computational burden for the stochastic network-constrained CUC model.
AB - With the increasing penetration of renewable energy, designing the efficient midterm (i.e., annual) scheduling for multi-source power systems is valuable in providing substantial cost reductions and flexibilities while ensuring system security against uncertainties of renewables. This paper presents a risk-averse two-stage stochastic midterm scheduling model for thermal-hydro-wind systems against short-term uncertainties of renewable generation. A network-constrained clustered unit commitment (NC-CUC) model is proposed to effectively incorporate short-term operation constraints into the midterm scheduling model, for improving the solution accuracy while reducing computational burden. Specifically, as the traditional CUC model cannot ensure consistent unit commitment status of generators across all scenarios, a power flow-based unit clustering method is proposed to improve the accuracy of short-term operation results. In addition, the conditional value-at-risk (CVaR) is used to quantify the system operational risk caused by uncertainties of wind generation and natural water inflows. Case studies show that the risk-averse two-stage stochastic midterm scheduling model can ensure operation dynamics and solution efficiency, and the proposed clustering method can achieve a tradeoff between solution accuracy and computational burden for the stochastic network-constrained CUC model.
KW - Midterm scheduling
KW - cluster
KW - renewable energy
KW - unit commitment
UR - http://www.scopus.com/inward/record.url?scp=85124845905&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85124845905&partnerID=8YFLogxK
U2 - 10.1109/TSTE.2022.3150918
DO - 10.1109/TSTE.2022.3150918
M3 - Article
AN - SCOPUS:85124845905
SN - 1949-3029
VL - 13
SP - 1293
EP - 1304
JO - IEEE Transactions on Sustainable Energy
JF - IEEE Transactions on Sustainable Energy
IS - 3
ER -