From September 17 to 18, the 2026 Green Energy Innovation and Development Conference was held in Tianjin. Niu Shubin, vice president of GCL Energy Technology, said that GCL's comprehensive AI energy trading service solution has achieved 200 billion kilowatt-hours of trial management electricity, which has verified its stability and reliability to support large-scale electricity transactions and operations on a large scale. Currently, it covers 23 major provinces, more than 100 prefectures and cities, and thousands of market players, and has built AI capabilities for power trading to serve the whole country. Niu Shubin explained that in terms of predictive capabilities, GCL's AI energy trading service integrates algorithm capabilities such as timing models, machine learning, and deep learning, combines multi-source meteorological integration with meteorological power mapping, and deeply embeds electricity market rules, pricing mechanisms, and settlement logic. Through a unified prediction engine, the platform's prediction accuracy is significantly improved on key indicators such as recent electricity prices, real-time electricity prices, optical power, wind power, and proxy user electricity, providing high-precision data support for transaction decisions.

From September 17 to 18, the 2026 Green Energy Innovation and Development Conference was held in Tianjin. Niu Shubin, vice president of GCL Energy Technology, said that GCL's comprehensive AI energy trading service solution has achieved 200 billion kilowatt-hours of trial management electricity, which has verified its stability and reliability to support large-scale electricity transactions and operations on a large scale. Currently, it covers 23 major provinces, more than 100 prefectures and cities, and thousands of market players, and has built AI capabilities for power trading to serve the whole country. Niu Shubin explained that in terms of predictive capabilities, GCL's AI energy trading service integrates algorithm capabilities such as timing models, machine learning, and deep learning, combines multi-source meteorological integration with meteorological power mapping, and deeply embeds electricity market rules, pricing mechanisms, and settlement logic. Through a unified prediction engine, the platform's prediction accuracy is significantly improved on key indicators such as recent electricity prices, real-time electricity prices, optical power, wind power, and proxy user electricity, providing high-precision data support for transaction decisions.