Shen Wan Hongyuan: Computing and electricity collaboration is included in the top-level power plan, and the entire industry chain has entered a period of accelerated implementation
The Zhitong Finance App learned that Shen Wan Hongyuan released a research report saying that the policy lists the collaborative development of electricity and computing power in a separate section, and makes it clear in the deployment of major projects that focus on laying out a number of collaborative computing projects in the electricity computing collaborative development area. At the same time, emphasis is placed on comprehensively strengthening green power consumption and grid investment requirements, and it is proposed to determine guidance targets for the utilization rate of new energy sources by region. The bank recommended focusing on the four main lines of direct green power supply and source network load storage integration, new energy and energy storage, AIDC power supply and distribution upgrades, and intelligent scheduling and virtual power plants.
Shen Wan Hongyuan's main views are as follows:
incident
On August 3, the National Development and Reform Commission and the National Energy Administration issued the “Fifteenth Five-Year Plan” (hereinafter referred to as the “Plan”), which clarifies the strategic goal of initially constructing a new power system by 2030. For the first time, computational and electricity collaboration was systematically incorporated into the core tasks of new power system construction, and proposed “coordinating the allocation of energy resources and the construction of computing power facilities, and collaboratively planning and laying out computing power and power projects to achieve power generation through electricity calculation”. The plan includes a separate section on “Promoting the Collaborative Integrated Development of Electricity and Computing Power” in Chapter 10, “Promoting the Collaborative Development of Electricity and Computing Power”, and makes it clear in the deployment of major projects that focus on laying out a number of collaborative electricity computing projects in areas of collaborative computing and development.
Computing and telecommunication collaboration has been incorporated into the top-level design of power development, and policy positioning has been strengthened once again
In December 2023, the “Implementation Opinions on Further Implementing the “East Digital and Western Computing” Project to Accelerate the Construction of a National Integrated Computing Power Network” first proposed the concept of “computing power and power collaboration”; the March 2026 “Report on the Work of the Government” first included “computing and electricity collaboration” in a new national infrastructure project; this plan included separate entries and supporting projects in the “15th Five-Year Plan” for the power industry. The subsequent provincial implementation plan and project list are expected to be implemented at an accelerated pace.
Plan the system to deploy four implementation paths to comprehensively strengthen green power consumption and power grid investment needs
The first is the integration of direct green power connection with source grid load storage. The plan clearly shows that in scenarios such as carbon reduction, zero carbon (low carbon) parks, and computing power facilities increase the proportion of green electricity, etc., to promote new energy sources not directly connected to public grids, supply green electricity to power users through direct lines, and lay out a number of smart microgrid projects with coordinated source grid load storage; the second is to regulate capacity building. The “15th Five-Year Plan” will add about 160 million kilowatts of new energy storage. The distribution grid needs to support 900 million kilowatts of distributed energy access; third, the distribution grid needs to support 900 million kilowatts of distributed new energy access; third, the distribution grid needs to support 900 million kilowatts of distributed new energy access; the third is a new type of grid mainly equipped with micro collaboration. Electric power transmission capacity exceeds 420 million kilowatts, and 15 additional times were put into operation HVDC green power channel; fourth, power-side collaboration and market mechanisms to vigorously promote the development of virtual power plants. At the same time, the “1+N” rules system for the unified national electricity market will be improved, and the share of electricity traded in the market will increase to 70% by 2030.
Partition control of consumption targets to marginally improve the economy of the Western Computing and Telecommunication Collaboration Project
The plan innovatively proposes to determine the guiding target for the utilization rate of new energy by region. In principle, the first tier region should not be less than 95%, the second tier should not be less than 90%, and the third tier should not be less than 85%. The national new energy utilization rate remains around 90%, and a unified national utilization rate red line will no longer be used. Regions rich in new energy sources are the main carriers of “resource-based” electricity computing collaborative projects (Zhongwei in Ningxia, Ulanqab in Inner Mongolia, Xinjiang, etc.). Tiered control has released some flexibility in consumption in these regions, and the economy of matching green electricity and computing power loads locally is expected to improve.
Investment analysis opinion: it is recommended to focus on the four main lines
1) Integration of direct green power supply and source network load storage: extension of new energy assets to the computing power load side, focusing on project contracts, green power resources, etc.; 2) New energy and energy storage: increasing demand for wind and solar storage configurations, moving energy storage from backup power to core conditioning resources for source load matching; 3) AIDC power supply and distribution upgrades: high power, high efficiency, DC drive equipment iteration, and focus on relevant companies' AIDC orders, unit value and HVDC/SST progress; 4) Intelligent scheduling and virtual power plants: computing power to participate in the “+ forecasting demand” of the electricity market “Scheduling+transaction” closed-loop capability, power AI scheduling and virtual The power plant platform ushered in an accelerated deployment window.
Risk warning: The pace of implementation of provincial supporting policies and project lists falls short of expectations; computing power and capital expenditure falls short of expectations; the economy of green power directly connected projects falls short of expectations; and the progress of electricity spot and auxiliary services market construction falls short of expectations.