Research on a high-reliability-guided hierarchical dynamic optimal scheduling method for photovoltaic systems.

Area of Science:

• Electrical Engineering
• Power Systems Engineering
• Renewable Energy Integration

Background:

• Hierarchical scheduling in active distribution networks (ADNs) faces challenges in maintaining nodal power fluctuation consistency and forecast coherence.
• Existing methods struggle to balance system security with scheduling efficiency under dynamic conditions.

Purpose of the Study:

• To propose a high-reliability-guided hierarchical dynamic optimal scheduling method for photovoltaic (PV) power in ADNs.
• To improve the consistency of nodal power fluctuation characteristics and hierarchical forecast coherence.

Main Methods:

• Utilized Crossformer for nodal power forecasting and introduced frequency-domain features for forecast reliability assessment.
• Designed a dynamic hybrid reconciliation strategy that adaptively switches between bottom-up aggregation and top-down decomposition based on reliability.
• Employed a constraint-based neural network (ConsNN) for nodal forecast reconstruction, ensuring aggregation-consistency.

Main Results:

• The proposed method effectively balances system security and scheduling efficiency in simulations.
• Demonstrated improved consistency in nodal power fluctuation characteristics.
• Achieved hierarchical forecast coherence through adaptive reconciliation.

Conclusions:

• The developed method offers a robust solution for optimal scheduling of PV power in ADNs.
• The integration of reliability assessment and adaptive reconciliation enhances the performance of hierarchical scheduling.
• The approach contributes to more secure and efficient operation of power distribution systems.