Summary
Under China’s dual-carbon goals and new power system transition, distributed PV is entering large-scale market participation, yet high dispersion and intermittency lead to poor forecastability, deviation risk, and unfair benefit allocation. This paper proposes a multi-agent aggregation model integrating LSTM forecasting with dynamic credit assessment in a
Read more Read lessStackelberg bi-level framework. The aggregator agent performs robust bid correction and credit-weighted revenue allocation, while PV agents use LSTM to learn temporal output patterns and maximize profit under credit constraints. Simulations with real data from a highpenetration Chinese province show reduced day-ahead bidding errors and improved participant revenues. The dynamic credit mechanism enhances incentive compatibility and suppresses speculative behavior, offering a practical pathway and policy insights for scalable
DPV market operation.
Additional informations
| Publication type | Session Materials |
|---|---|
| Reference | C5_11524_2026 |
| Publication year | |
| Publisher | CIGRE |
| Country | China, People's Republic of |
| Study committees | |
| File size | 438 KB |
| Price for non member | 30 € |
| Price for member | 30 € |
Authors
MA Li - State Grid; ZHAO Zheng - State Grid Energy Research Institute; SUN Qingkai - State Grid Energy Research Institute; LIN Sen - State Grid; GUO Kuifeng - State Grid; LIU Runzi - State Grid; TANG Chenghui - State Grid Energy Research Institute
Keywords
