Climate Resilience–Oriented Power Source Expansion Planning Under Supply–Demand Perspective

The power composition of the modern power system will shift from fossil fuels as the dominant source to large‐scale renewable energy generation. The resource attribute of renewable energy that relies on the weather has made the meteorological challenges of the power system increasingly prominent. Taking the climate risk of the power system as the core, this paper summarizes the mechanisms of various extreme weather events affecting the power system. Based on the denoising variational autoencoder algorithm, a method for generating extreme scenarios is proposed. The extreme scenarios are generated by combining historical data of a southeastern province in China and the IEEE 39‐node standard calculation example. By adopting the generated extreme scenarios into the long‐term scheduling model, the proposed method in this paper is verified.