China Securities Co., Ltd.: AI reshapes the new blue ocean of power trading, enhancing price volatility to quickly open up demand space.

China Securities Co., Ltd. released a research report stating that China's electricity reform is gradually entering a deep water area. With the comprehensive expansion of the electricity spot market and the increasing complexity of trading rules, harnessing AI to empower electricity trading and "reduce costs, increase revenue, and control risks" has become the core proposition for the development of various market entities. The excavation and application of massive high-value data is the key to building this core competitive barrier. With the smooth progress of national policy implementation and the steady advancement of the spot market, the company believes that the increased volatility of electricity prices will rapidly open up the demand space for AI trading products and services in the electricity market. In a neutral valuation scenario, assuming an AI penetration rate of 75%, the incremental space of the AI-empowered electricity trading market will reach 29.52 billion yuan by 2030. Key points from China Securities Co., Ltd. are as follows: Riding the wave of electricity reform, AI reshapes a new blue ocean in electricity trading. China's electricity system reform is gradually entering a deep water area. At the national level, efforts to accelerate the construction of a unified national electricity market system are ongoing. The core function of the spot market in discovering real-time prices and accurately reflecting supply and demand is being strengthened, with the goal of achieving full operation of the spot market by 2027. By 2025, total electricity consumption in the society will be close to 10 trillion kilowatt-hours, with market-based trading accounting for over two-thirds of the total. With the comprehensive expansion of the electricity spot market and the increasing complexity of trading rules, harnessing AI to empower electricity trading and "reduce costs, increase revenue, and control risks" has become the core proposition for the development of various market entities. The excavation and application of massive high-value data is the key to building this core competitive barrier. Borrowing insights from the global market, exploring and practicing in overseas leading markets. The United States, relying on real-time clearing mechanisms and node pricing systems, drives AI to focus on high-frequency forecasting and refined node arbitrage, making it the "ultimate battlefield" for algorithmic warfare; Europe, with its transnational market coupling and diversified product systems, leads AI in virtual power plant aggregation and large model system-level applications, solving the "complex constraints" of cross-border coordination; Australia, with its extreme pressure field of a high proportion of wind and solar energy storage, where extreme price fluctuations force AI to shift from an efficiency tool to a "survival necessity," leading the arms race in high-frequency trading. Observing industry benchmarks, the models and progress of global electricity trading leaders. Global pioneers in electricity trading such as Octopus Energy, Fluence Energy, and Second Foundation rely on deep industry expertise and cutting-edge AI algorithms, combined with intelligent scheduling of distributed resource aggregation and energy storage systems through multiple approaches, efficiently capturing real-time price fluctuations in the electricity market and successfully converting them into strong, scalable excess returns. Market space estimation With the smooth progress of national policy implementation and the steady advancement of the spot market, the company believes that the increased volatility of electricity prices will rapidly open up the demand space for AI trading products and services in the electricity market. In a neutral valuation scenario, assuming an AI penetration rate of 75%, the incremental space of the AI-empowered electricity trading market will reach 295.2 billion yuan by 2030. Risk analysis (1) Model prediction deviation and market adaptation risk; (2) Data quality and security compliance risk; (3) Algorithm black box and regulatory penetration risk; (4) Technical reliance and system resilience risk; (5) Business model and investment return risk.