Robust optimization-based dynamic power generation mix evolution under the carbon-neutral target

A transition towards long-term sustainability in energy systems based on a low-carbon generation mix could mitigate growing global warming threats to human society. However, the optimal structure of future systems and potential transition paths are still open questions, especially for China's power generation sector dominated by fossil fuels. In this research, robust optimization-based dynamic generation expansion planning is proposed to describe the carbon-neutral transition path for China's power generation sector. The steps required to enable a realistic transition that prevents societal disruption, and the impact of pricing policies (i.e., carbon trading and tax) on neutrality are also discussed. Simulation results show that there exist multiple potential evolution paths for China's power generation system to reach carbon neutral. For the next decades-long journey, this radical transition will require steady but evolutionary changes. The low-share (under 10%) coal scheme is more likely a better option for the carbon-neutral transition of China's power generation sector. Under the low-coal scenario, the emissions peak would be seen by 2025 with around 4543 Mt (20% above the 2015 level) of CO2, and the milestone of neutrality would be reached in 2057. By 2060, wind and solar production could provide 63% of the electricity demand, and the share of non-fossil energy generation would approach 84%. The total cost of the low-coal plan is 14% lower than that under the 100% clean energy supply scenario.

Automated summary

This research proposes a robust optimization-based dynamic generation expansion planning to describe the carbon-neutral transition path for China's power generation sector. The study shows that the low-share coal scheme is the better option for China's power generation sector to achieve carbon neutrality, with a milestone reached in 2057. The power generation system in China needs steady but evolutionary changes in the next few decades.