A Control Strategy Based on Deep Reinforcement Learning Under the Combined Wind-Solar Storage System

Huang, Shiying; Yang, Ming; Zhang, Changhang; Yun, Jiangyang; Gao, Yuan; Li, Peng

doi:10.1109/scems48876.2020.9352436

Cited by 6 publications

(1 citation statement)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Optimization modeling and power system applications have also been explored. Soroudi [28] provided a comprehensive guide to power system optimization modeling using GAMS, offering insights into modeling techniques applicable to hybrid microgrid configurations. Shah and Mithulananthan [29] conducted a review of optimal power flow literature, emphasizing the application of traditional and evolutionary methods in solving nonlinear power system problems.In the realm of microgrid integration and energy storage, Basu et al [30] presented a review of microgrid concepts, underlining the importance of proper integration of diverse energy sources for enhanced performance.…”

mentioning

confidence: 99%

Optimal Power Flow in Hybrid AC/DC Microgrid using ANN for Cost Minimization

Yamuna,

2024

JMC

View full text Add to dashboard Cite

Currently, this work lays the ground work for sophisticated control methods and decision support systems in hybrid microgrid operations by providing insightful information about integrating artificial intelligence for improved microgrid control. In this work, a neural network (NN) method is proposed for power flow analysis in an IEEE 12-bus-based Hybrid AC/DC Microgrid. The study optimizes power dispatch, minimizes expenses, and minimizes losses in both AC and DC components. Simulation is carried using MATLAB software and the results are presented and analysed. The accuracy of the NN’s predictions of active power flows is demonstrated by training it on historical data and validating it on real-time observations. Regression plots comparing anticipated and real values demonstrate the effectiveness of NN-based analysis in reaching the ideal power distribution.

show abstract

mentioning

confidence: 99%