.png){kind=icon}
.png){kind=icon}
Return
Multi-Objective Energy Management Strategy for Distribution Network With Distributed Renewable Based on Learning-Driven Model Predictive Control
.png){kind=icon}
PRE
AI
DOI：10.1109/TSG.2025.3600071.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
This paper investigates the energy management of distribution network with distributed renewable. A novel energy management strategy is proposed based on learning-driven model predictive control. To address the uncertainty of renewable, a hybrid TCN framework is proposed and the wavelet packet decomposition approach is adopted to capture temporal-frequency features. This paper considers generation cost and environmental cost as two objective functions respectively. An improved MOPSO is proposed, the initialization process and learning coefficients are optimized. The Pareto frontier is evaluated by TOPSIS based on objective weights. The proposed hybrid TCN framework is validated under sunny and cloudy days. The proposed energy management strategy is validated under 33 bus and 118 bus test system with real-world data. Simulation results verify the effectiveness of proposed methods.
Keywords:
Energy management
Optimization
Renewable energy sources
Distribution networks
Predictive control
Uncertainty
Predictive models
Power systems
Costs
Partial discharges
Energy management strategy
active distribution network
temporal convolutional network
receding horizon optimization
multi-objective optimization
Journal
IF:
9.8
Papers:
5.7K
Citations:
4.3W