AI system boosts distribution network efficiency in Jiaxing

An AI-driven system developed by the State Grid Jiaxing Power Supply Company has recently been put into operation to enhance the efficiency of power distribution network operation and maintenance, marking another step forward for the company in its push toward intelligent, data-driven grid management.

Powered by the Guangming large language model, the intelligent auxiliary system for distribution transformer areas automatically generates optimized operation and maintenance plans, increasing planning efficiency by nearly 80 percent and accelerating the shift from manual, experience-based control to intelligent decision-making.

Traditionally, operation and maintenance planning required personnel to manually consolidate multiple data sources — including outage records, load curves, equipment inventories and weather information — a process that typically took around 60 minutes and relied heavily on the individual experience. To address these challenges, the system integrates area-level load data, outage records and equipment archives from the company's data middle platform, enabling unified data access and analysis.

By applying time-series forecasting and pattern recognition technologies, the system can automatically identify transformer areas with frequent faults, accurately locate potential risks and forecast load variations for the coming week. This supports the scientific prioritization of inspections and more precise allocation of emergency resources.

Following its deployment, operation and maintenance workers are able to generate an optimized weekly plan with a single click, including inspection priorities, resource allocation recommendations, and emergency team deployment, while allowing for manual adjustments. The system significantly reduces the time required to compile data and shifts power supply assurance from experience-driven arrangements to data-based decision-making, effectively avoiding both excessive and insufficient protection.

The application has enabled more precise deployment of inspection resources, shortened response times, and improved power supply reliability, contributing to safer and more efficient grid operations.

Looking ahead, the State Grid Jiaxing plans to further expand the system's capabilities by integrating functions such as distributed energy resources, electric vehicle charging demand forecasting, and energy storage dispatch optimization. The move is expected to provide stronger intelligent decision support for the construction of a new-type power system and promote safer, more economical and low-carbon grid operation.