In the era of climate change and sustainable development, the main task of China’s power sector is not to expand production, but to promote the transition to a low-carbon energy system. On the generation side, the power sector needs to reduce the share of thermal power, especially coal power, and increase that of nonfossil energy sources; on the demand side, the power sector needs to support electrification of residential and transportation sectors to achieve deep decarbonization and air quality goals. This will significantly increase the uncertainties on both the generation and consumption ends, since nonfossil energy sources, such as photovoltaics and wind power, are highly unpredictable, and large-scale electrification of heating and transportation can increase the volatility of power demand. Therefore, the future power system requires a high degree of technological and operational flexibility to cope with uncertain supply and demand. Our analysis suggests that market-oriented reforms in electricity dispatch and pricing are critical to provide the right incentives to electricity generators and consumers, as well as to promote the adoption of new technologies to facilitate China’s transition to low-carbon energy.

The increasing penetration of renewable resources has changed the characteristics of power system and market operations, from one relying primarily on deterministic and static planning to one involving highly stochastic and dynamic operations. In such new operation regimes, the ability of adapting changing environments and managing risks arising from complex scenarios of contingencies is essential. To this end, an operation tool that provides probabilistic forecasting that characterizes the underlying probability distribution of variables of interest can be extremely valuable. A fundamental challenge in probabilistic forecasting for system and market operations is the scalability. As the size of system and the complexity of stochasticity increase, standard techniques based on direct Monte Carlo and machine learning techniques become intractable. This chapter outlines an alternative approach based on an online learning to overcome barriers of computation complexity.