T/JES 010-2025 Active Group standards

T/JES 010-2025 Technical Specifications for Dynamic Grid Division and Output Prediction of Distributed Photovoltaic Clusters

T/JES 010-2025 Distributed Photovoltaic Cluster Dynamic Mesh Partitioning and Output Prediction Technical Specification

Publish Date: 2025-05-29 Implement Date: 2025-05-29 For services related to genuine standard inquiry, procurement, translation, and other related services in China, please Contact Us

Price: $ 20.00

Out of stock

Basic Information

Standard Code: T/JES 010-2025

Standard Type: Group standards

Standard Status: Active

is_force_gb: no

CCS Name: Electric systems and equipment

ICS Name: Generator sets

Publish Date: 2025-05-29

Implement Date: 2025-05-29

Scope

Main technical content: 1. **Dynamic grid division technology specification**: - Division principle: Trigger grid adaptive reorganization based on meteorological abruptness sensitivity (cloud movement speed > 10 m/s); - Grid area dynamic range: Urban dense areas ≤ 5 km2, rural areas ≤ 15 km2. - Algorithm requirements: Use an improved DBSCAN clustering algorithm to define meteorological similarity thresholds (temperature difference ≤ 2 °C, irradiance deviation ≤ 50 W/m2); - Grid merging/splitting conditions: When the photovoltaic output fluctuation rate within a grid exceeds 20% and lasts for 30 minutes, initiate grid reorganization. 2. **Spatio-temporal coupling modeling technology specification**: - Time series modeling: Use the Transformer architecture with ≥ 4 attention heads and ≥ 24-hour historical sequence length; - Time resolution: Ultra-short-term forecast ≤ 15 minutes, short-term forecast ≤ 1 hour. - Spatial modeling: Adopt graph convolutional networks (GCN) and define adjacency matrix weight calculation rules; - Cross-grid correlation threshold: Grids with a spatial correlation coefficient ≥ 0.6 must be forcibly connected. 3. **Multi-source data fusion technology requirements**: - Data categories: Meteorological data: Satellite cloud maps, ground weather stations; - Equipment data: Inverter operating parameters (active power, temperature, upload frequency); - Geographical data: Latitude/longitude coordinate error ≤ 10 meters, shading map accuracy ≥ 90%. - Modal alignment methods: - Time alignment: Use Beidou timing signals as the reference; - Spatial alignment: Use bilinear interpolation to match meteorological data to the center of the dynamic grid. 4. **Prediction accuracy verification standards**: - Normal scenarios: Sunny/cloudy weather: RMSE ≤ 8%, MAE ≤ 5%; - Rainy weather: RMSE ≤ 12%, MAE ≤ 8%; - Extreme weather: Typhoons/snowstorms: RMSE ≤ 15%, and th