TY - JOUR
T1 - Multi-party stochastic energy scheduling for industrial integrated energy systems considering thermal delay and thermoelectric coupling
AU - Chen, Liudong
AU - Liu, Nian
AU - Li, Chenchen
AU - Wu, Lei
AU - Chen, Yubing
N1 - Publisher Copyright:
© 2021 Elsevier Ltd
PY - 2021/12/15
Y1 - 2021/12/15
N2 - Multi-dimensional stochastic factors challenge the interactive energy scheduling of the industrial integrated energy system (IIES). Previous research focuses on either deterministic energy scheduling or individual stochastic scheduling while neglecting complicated interactions among uncertain parties, which brings the research gaps about stochastic multi-party's interaction. In this regard, a multi-party stochastic energy scheduling approach in IIES is proposed based on the stochastic game. A decentralized decision support system is considered, and a stochastic utility model is designed for decentralized IUs with multi-dimensional stochastic factors from photovoltaic (PV) production and IIES parameters, enabling them to participate in the multi-energy scheduling with their own strategies. A stochastic game model is developed considering the thermoelectric coupling and the IUs’ interaction. The co-decision mechanism, recognizing different transfer times of electrical and thermal energy, is built based on the state transition within the game. Moreover, a distributed solution algorithm that includes the Markov decision process and iterative method is designed to address the problem of the “curse of dimensionality” arising from multiple stochastic factors. Finally, case studies with realistic data from an industrial park in Guangdong Province, China, are designed to show the effectiveness of the proposed approach, which enhances IUs’ profits by 9.4% and fits flexible load strategies and price strategies. The decentralized system can also reduce the computation time by 70.1% compared to the centralized system. Through analyzing different number of scenarios and intervals for PV generation, electrical and thermal load, the conclusion has obtained that increase the number of scenarios has a negative effect on IUs’ decision, but increase the number of load intervals contributes to more specific results and higher utility.
AB - Multi-dimensional stochastic factors challenge the interactive energy scheduling of the industrial integrated energy system (IIES). Previous research focuses on either deterministic energy scheduling or individual stochastic scheduling while neglecting complicated interactions among uncertain parties, which brings the research gaps about stochastic multi-party's interaction. In this regard, a multi-party stochastic energy scheduling approach in IIES is proposed based on the stochastic game. A decentralized decision support system is considered, and a stochastic utility model is designed for decentralized IUs with multi-dimensional stochastic factors from photovoltaic (PV) production and IIES parameters, enabling them to participate in the multi-energy scheduling with their own strategies. A stochastic game model is developed considering the thermoelectric coupling and the IUs’ interaction. The co-decision mechanism, recognizing different transfer times of electrical and thermal energy, is built based on the state transition within the game. Moreover, a distributed solution algorithm that includes the Markov decision process and iterative method is designed to address the problem of the “curse of dimensionality” arising from multiple stochastic factors. Finally, case studies with realistic data from an industrial park in Guangdong Province, China, are designed to show the effectiveness of the proposed approach, which enhances IUs’ profits by 9.4% and fits flexible load strategies and price strategies. The decentralized system can also reduce the computation time by 70.1% compared to the centralized system. Through analyzing different number of scenarios and intervals for PV generation, electrical and thermal load, the conclusion has obtained that increase the number of scenarios has a negative effect on IUs’ decision, but increase the number of load intervals contributes to more specific results and higher utility.
KW - Energy scheduling
KW - Game theory
KW - Industrial integrated energy system
KW - Multi-dimensional stochastic factors
KW - Multi-party
UR - http://www.scopus.com/inward/record.url?scp=85115924489&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85115924489&partnerID=8YFLogxK
U2 - 10.1016/j.apenergy.2021.117882
DO - 10.1016/j.apenergy.2021.117882
M3 - Article
AN - SCOPUS:85115924489
SN - 0306-2619
VL - 304
JO - Applied Energy
JF - Applied Energy
M1 - 117882
ER -