Artificial intelligence for the new power system: Trends and challenges

Driven by the goals of carbon peak and carbon neutrality, the new power system has developed rapidly, exhibiting main characteristics such as strong stochastic volatility, high power electronization, and significant susceptibility to external influences, which pose numerous challenges to the secure, stable, and economic operation of power grids. Simultaneously, the swift advancement of artificial intelligence has enabled it to initially possess the capabilities of autonomous knowledge discovery and inference, thereby driving a transformation of the scientific research paradigm from traditional mechanism- and experiment-driven approaches to data- and computing-driven ones, which provides a new path to address the issues in the new power system. From a system-level perspective, this paper focuses on artificial intelligence technologies tailored for the new power system, summarizing the current development status and emerging trends in core application areas, including simulation analysis, planning, dispatch and operation, stability control, and electricity market transactions. Furthermore, it discusses several technical bottlenecks and key challenges associated with the artificial intelligence-assisted new power system.