Stochastic dual dynamic programming for multi-stage stochastic programming of sustainable utility systems

Fossil fuels remain a predominant energy source in traditional industrial utility systems, resulting in significant carbon emissions. This study proposes a sustainable utility system that integrates renewable energy and energy storage. A Multi-stage Stochastic Programming (MSSP) model is developed t...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	Energy (Oxford) Ročník 338; s. 138943
Hlavní autoři:	Liu, Nianxin, Yang, Kangyuan, Zhao, Liang, Ye, Zhencheng
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	Elsevier Ltd 30.11.2025
Témata:	Multi-stage stochastic programming Optimization under uncertainty Renewable energy Stochastic dual dynamic programming Sustainable utility system Multi-stage stochastic programming Stochastic dual dynamic programming Renewable energy Sustainable utility system Optimization under uncertainty
ISSN:	0360-5442
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Popis
Shrnutí:	Fossil fuels remain a predominant energy source in traditional industrial utility systems, resulting in significant carbon emissions. This study proposes a sustainable utility system that integrates renewable energy and energy storage. A Multi-stage Stochastic Programming (MSSP) model is developed to accommodate system flexibility in dynamic environments characterized by multiple uncertainties, including wind speed, solar irradiance, and multi-level steam demand. To address the computational challenges of large-scale stochastic optimization, a Stochastic Dual Dynamic Programming (SDDP) algorithm is employed, enhanced with scenario sampling, reduction techniques, and Benders decomposition. Case studies from real industrial utility systems demonstrate that the proposed method reduces total operational costs and carbon-related costs by 2.3 % and 5.7 %, respectively, while achieving up to 90 % reductions in problem size under the same scenario tree. As a result, the proposed approach for optimizing sustainable utility systems is both effective and scalable. •Multi-stage stochastic programming method is used to handle multi-scale uncertainties.•Stochastic dual dynamic programming is used to for solving the MSSP model.•Emission reduction of sustainable utility systems is quantified.
ISSN:	0360-5442
DOI:	10.1016/j.energy.2025.138943