Published on Dec 17, 2025

A subsidiary of CSG Energy Storage received China’s first export order for an AI-based data platform for pumped-storage hydropower in early December, the Chinese science newspaper Zhongguo Kexue Bao (in Chinese) reported.

It is the first platform of its kind for pumped-storage hydropower plants independently developed by CSG. Guangzhou Qiming Intelligent Energy Technology Co., Ltd., a subsidiary of CSG Energy Storage, won the tender to supply the system to the Stung Tatay hydropower project in Cambodia.

The big-data analytics system will be used there for AI-based monitoring of the hydropower plant as well as the upstream Upper Stung Tatay Hydropower Station, the newspaper wrote. Under its “AI Plus” initiative, the Chinese government is promoting the deployment of artificial intelligence in the operation of pumped-storage hydropower plants. This is intended to increase the energy efficiency of such facilities.

China regards pumped-storage hydropower as a key pillar of its strategy to stabilise the power grid during the rapid transition to renewable energy. Wind and solar power supply electricity intermittently and often not at times of peak demand.

Pumped-storage hydropower is considered one of the most effective methods of stabilising electricity grids. When there is excess power, electricity is used to pump water from one reservoir to a second reservoir at a higher elevation. When electricity is scarce, the water flows back downhill through turbines, generating power.

The government in Beijing has set ambitious climate targets and decided to almost triple the country’s installed solar and wind capacity by 2035, from 1.4 terawatts at the end of 2024 to 3.6 terawatts.

At the same time, the number of pumped-storage hydropower projects in China is to be expanded significantly. While 151 new pumped-storage plants are currently under construction or in the planning stage, this figure is expected to rise to 340 over the next five years.

Chinese manufacturers are therefore rapidly expanding their capacity to build such facilities and are increasingly exporting them to so-called Belt and Road countries in Africa, South America, and Southeast Asia. Cambodia is one of them.

The first export of AI technology for pumped-storage hydropower is being interpreted by Chinese state media as evidence that China is now able to offer internationally competitive equipment in the field of grid and system control.

“From energy-system control to total power generation capacity, we are now at the global forefront,” said Lin Boqiang, director of the China Center for Energy Economics Research at Xiamen University, in an interview with the Global Times.

“The key question now is how we deploy artificial intelligence on this basis, building on more than a decade of digitalisation that has created a solid foundation,” he added.

Artificial intelligence can support the development of a modern and sustainable energy infrastructure by being integrated across the entire value chain, the renewable-energy expert said.

“Upstream, AI can improve exploration and reserve assessment; midstream, it can better match supply and demand and optimise maintenance; and downstream, the technology can strengthen response and early-warning systems,” Lin said.

With the help of specialised AI models, real-time adjustments and predictive maintenance can be optimised, reducing failure rates, conserving resources, and maximising energy output, the expert added. China’s power producers are also using artificial intelligence in pumped-storage hydropower plants to simulate the interaction between electricity demand, weather conditions, and water-flow rates. This increases the efficiency of pumped-storage facilities and, in turn, optimises the storage of renewable energy.