Hybrid Proximal Policy Optimization—Wasserstein Generative Adversarial Network Framework for Hosting Capacity Optimization in Renewable-Integrated Power Systems

Jun Han (), Chao Cai, Wenjie Pan, Hong Liu and Zhengyang Xu
Additional contact information
Jun Han: State Grid Jiangsu Electric Power Economic and Technological Research Institute, No. 251 Zhongshan Road, Gulou District, Nanjing 210009, China
Chao Cai: State Grid Jiangsu Electric Power Economic and Technological Research Institute, No. 251 Zhongshan Road, Gulou District, Nanjing 210009, China
Wenjie Pan: State Grid Jiangsu Electric Power Economic and Technological Research Institute, No. 251 Zhongshan Road, Gulou District, Nanjing 210009, China
Hong Liu: School of Electrical and Information Engineering, Tianjin University, No. 92 Weijin Road, Nankai District, Tianjin 300072, China
Zhengyang Xu: School of Electrical and Information Engineering, Tianjin University, No. 92 Weijin Road, Nankai District, Tianjin 300072, China

Energies, 2024, vol. 17, issue 24, 1-22

Abstract: The rapid integration of distributed energy resources (DERs) such as photovoltaics (PV), wind turbines, and energy storage systems has transformed modern power systems, with hosting capacity optimization emerging as a critical challenge. This paper presents a novel Hybrid Proximal Policy Optimization-Wasserstein Generative Adversarial Network (PPO-WGAN) framework designed to address the temporal-spatial complexities and uncertainties inherent in renewable-integrated distribution networks. The proposed method combines Proximal Policy Optimization (PPO) for sequential decision-making with Wasserstein Generative Adversarial Networks (WGAN) for high-quality scenario generation, enabling robust hosting capacity enhancement and operational efficiency. Simulation results demonstrate a hosting capacity improvement of up to 128.6% in high-penetration scenarios (90% renewable), with average operational cost reductions of 22%. Voltage deviations are minimized to within ±5% of nominal levels, while energy losses are reduced by 18%. Scenario quality, evaluated using the Wasserstein metric, achieved convergence with an average score of 0.95 after 80 iterations, highlighting the WGAN’s ability to generate realistic and diverse scenarios. This study advances the state of the art in distribution network optimization by integrating machine learning techniques with robust mathematical modeling. The PPO-WGAN framework enhances scalability, ensures grid stability, and promotes efficient renewable integration, providing a robust foundation for future applications in modern power systems.

Keywords: hosting capacity optimization; proximal policy optimization (PPO); wasserstein generative adversarial network (WGAN); renewable integration; temporal-spatial resource coordination; distributed energy resources (DERs); grid stability and robustness (search for similar items in EconPapers)
JEL-codes: Q Q0 Q4 Q40 Q41 Q42 Q43 Q47 Q48 Q49 (search for similar items in EconPapers)
Date: 2024
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/1996-1073/17/24/6234/pdf (application/pdf)
https://www.mdpi.com/1996-1073/17/24/6234/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jeners:v:17:y:2024:i:24:p:6234-:d:1541180

Access Statistics for this article

Energies is currently edited by Ms. Agatha Cao

More articles in Energies from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().