Iran Energy News Oil, Gas, Petrochemical and Energy Field Specialized Channel
Related Articles
China is increasingly integrating artificial intelligence into its energy system to increase the sector’s efficiency and enhance national energy security. From managing energy storage to more accurately forecasting energy supply and demand, large language models can make the whole sector smarter, by streamlining energy systems and minimizing waste.
As early as 2020, the Chinese state-owned natural gas giant Gas Holdings Ltd. was using artificial intelligence to more accurately predict complex and dynamic shifts in natural gas demand. These AI-driven models can use both historical data and real-time inputs to calculate shifts in demand faster than human employees ever could.
As China’s grid becomes increasingly powered by renewable energies, however, the issue of forecasting becomes even trickier and even more important. China leads the world in renewable energy deployment. In May alone, China added so much solar and wind power to its grid that the generation capacity of those additions alone rivals the entire electricity output of Poland. As China continues to break its own renewable energy records, its grid becomes much more dependent on variable energies, and therefore much more weather-dependent. Accurate forecasting is therefore paramount of national energy security.
To this end, a new transformer-based weather forecasting system has been piloted in China to monitor and make calculated predictions around domestic electricity markets. The project, fronted by the firms TerraQuanta and Horizon Power, marks “a major step forward in aligning AI innovation with a weather-driven decarbonizing grid” according to the World Economic Forum (WEF). The large language model employed in the project has been trained on over 40 years of worldwide weather data and is powered by 5 billion parameters to produce between 7 and 15 forecasts a day. The system is reportedly 50,000 times more efficient than traditional modelling methods. “Now used across Horizon’s 2,000 GWh annual trading portfolio, It enables faster smarter decisions in volatile electricity spot markets reducing risk and boosting profitability,” reports the WEF.
All that variable energy also requires massive amounts of energy storage infrastructure. Since wind and solar power respond to weather rather than demand, supply and demand are often mismatched. This requires vast storage capacity to capture excess energy produced when the sun is shining and the wind is blowing, which can then be fed back into the grid as needed. This, too, can be massively improved through the use of artificial intelligence.
AI is boosting both the design and the safety mechanisms of battery storage. The timelines to develop new, more efficient batteries, has been slashed by magnitudes thanks to the integration of large language models. “Engineers simply type a goal — ‘raise energy density’ — and the cloud platform runs physics-grade AI simulation models across 70 variables, returning an optimal cell blueprint with full citations,” reports the WEF in relation to a Chinese battery venture led by software company Electroder and Tsinghua University.
What’s more, AI integration can make battery storage more failure-proof through better management and maintenance strategies. “AI-driven energy storage systems can ensure safety and intelligent operation and maintenance,” explained Shi Zinan, director of the Smart Microgrid Institute at Tsinghua Sichuan Energy Internet Research Institute, to CGTN. “AI also empowers energy storage transactions, enhances economic benefits, and assists in grid efficiency regulation.”
Of course, AI is a double-edged sword in the context of the energy sector. Training large language models carries a gargantuan energy footprint, and running AI queries is far more-energy intensive than traditional searches – though exact figures are hard to come by. But while AI is placing a strain on energy grids, it also has enormous potential to improve grid management. And China is far ahead of the curve in terms of tapping into that potential.
