Objective This study aims to systematically review the business models, key optimization techniques, and application practices of energy storage arbitrage in electricity markets, and to analyze revenue structures, risk characteristics, and development challenges under different market mechanisms, providing theoretical references and practical guidance for energy storage participation in market trading and value enhancement.

Method A survey of energy storage arbitrage practices in representative domestic and international electricity markets is conducted, and various optimization and artificial intelligence methods are compared in terms of applicability and performance. The techno-economic adaptability of electrochemical storage, pumped hydro storage, and hydrogen storage to different arbitrage pathways is also examined.

Result Findings indicate that AI algorithms can increase arbitrage profits by 10%–20% in highly volatile markets; multi-market arbitrage significantly outperforms single-market models; and the suitability of storage technologies must be aligned with market mechanisms, with lithium-ion batteries fitting high-frequency responses, while pumped hydro and hydrogen storage are better suited for long-cycle peak shaving. Current challenges include immature multi-market coordination mechanisms, insufficient real-time response capability, and degradation issues.

Conclusion Energy storage arbitrage constitutes the core mechanism for commercial operation. Future directions include the development of real-time forecasting and coordinated control, carbon-electricity coupled arbitrage, and digital twin simulation, along with improved market rules and cross-product trading mechanisms, so as to realize the multiple value streams of energy storage in modern power systems.