A distributed photovoltaic scheduling operation method and system based on load clustering
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- STATE GRID BEIJING ELECTRIC POWER CO
- Filing Date
- 2023-02-07
- Publication Date
- 2026-06-19
AI Technical Summary
After distributed photovoltaic power is connected to the grid in industrial parks, the types and characteristics of loads change frequently. Existing dispatching methods fail to effectively consider the socio-economic impact of the loads, resulting in large fluctuations in grid voltage and power, making it difficult to optimize dispatching.
A load clustering-based scheduling method is adopted, which classifies and clusters the load using K-means and hierarchical agglomerative clustering algorithms, establishes a load prediction model, sets voltage and power variance thresholds, and combines energy storage to optimize and adjust load and photovoltaic operation data to ensure that voltage and power are within the threshold range.
It enables refined regulation of the load side, reduces grid voltage and power fluctuations, optimizes the operation of distributed photovoltaic and energy storage, and improves grid stability and efficiency.
Smart Images

Figure CN115954958B_ABST