Multi-modal multi-task artificial intelligence model for active distribution network scheduling with multi-agent reinforcement learning

doi:10.1016/j.epsr.2025.112091

Multi-modal multi-task artificial intelligence model for active distribution network scheduling with multi-agent reinforcement learning

2025-08-29

0

PRE

AI

DOI：10.1016/j.epsr.2025.112091

Original

Original request for help

Save

Abstract

En 中文

• A multi-modal multi-task AI framework is proposed for ADN scheduling. • Grid state and solar panel image data are considered as real-time inputs. • Hybrid CNN-ViT encoder enables real-time image regression for soiling estimation. • Proposed framework is validated on the large-scale 2289-bus Nizwa network.

Keywords:

Deep learning

Multi-modal framework

Renewable energy

Reinforcement learning

Artificial intelligence

Multi-modal multi-task artificial intelligence model for active distribution network scheduling with multi-agent reinforcement learning

Abstract

Journal

Organization