Research on Self-Decision Control Strategy in Distributed Photovoltaic Intelligent Monitoring Device

This study proposes an intelligent monitoring self-decision control strategy based on multi-agent reinforcement learning algorithm for distributed photovoltaic systems to improve the efficiency and stability of the system under variable environments. In order to cope with the uncertainty in the dist...

Full description

Saved in:
Bibliographic Details
Published in:IEEE International Conference on Automation, Electronics and Electrical Engineering (Online) pp. 515 - 519
Main Authors: Lei, Tinghao, Cai, Xinya, Wu, Yang, Guo, Yingjie, Li, Yao
Format: Conference Proceeding
Language:English
Published: IEEE 27.12.2024
Subjects:
Collaboration
Distributed photovoltaic
intelligent monitoring
Inverters
Monitoring
multi-agent reinforcement learning algorithm
Optimization
Photovoltaic systems
Power grids
Power system stability
Real-time systems
Reinforcement learning
self-decision-making control
Stability analysis
ISSN:2831-4549
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This study proposes an intelligent monitoring self-decision control strategy based on multi-agent reinforcement learning algorithm for distributed photovoltaic systems to improve the efficiency and stability of the system under variable environments. In order to cope with the uncertainty in the distributed photovoltaic power generation process, this paper designs a self-decision control algorithm to achieve real-time monitoring and optimization control of photovoltaic components, inverters and grid interfaces through multi-agent collaboration. Each agent represents a different control unit, which is independent and cooperative with each other, and continuously trains under the reinforcement learning framework to improve the overall performance. The model verification based on the simulation platform shows that the algorithm can significantly improve the response speed and power generation efficiency of the photovoltaic system under different weather and load conditions. The data analysis results show that after adopting this control strategy, the output fluctuation of photovoltaic power generation is reduced by 15%, the inverter efficiency is improved by 12%, the system stability is enhanced, and the impact on the power grid is effectively reduced. In addition, the algorithm has good generalization performance and can be applied to large-scale distributed photovoltaic networks.
ISSN:2831-4549
DOI:10.1109/AUTEEE62881.2024.10869759