Issues And Recommendations For China’s Virtual Power Plants – Analysis


By Kung Chan and Xia Ri

With China working on its dual-carbon goals and under the current high temperature, virtual power plants have been consistently dominating the country’s new energy sector since the beginning of this year. In fact, in 2022, the 14th Five-Year Plan for Modern Energy System proposed to promote flexible resources such as energy storage facilities, virtual power plants, and user-interruptible loads to participate in ancillary services.

According to Huaxi Securities’ estimation, by 2025, the investment and construction market size of virtual power plants in China will exceed RMB 30 billion, with an average annual investment and construction scale ranging from RMB 10.5 billion to RMB 20 billion. It is expected that by 2025, the operational market size of virtual power plants in the country will reach RMB 96.8 billion. By 2030, this scale is projected to exceed RMB 450 billion. With policy support and the attraction of potential markets, several listed companies have started to lay out and participate in this area. Researchers at ANBOUND believe that based on the development experience of the United States and Germany, China’s virtual power plants, comparatively latecomers, are mainly government and state-owned enterprises-led and face many practical issues, without having formed a relatively mature industry chain.

First of all, there is a low level of technological modernization and an unstable industrial foundation. Currently, virtual power plant technology has evolved through three generations. The first generation focused on centralized aggregation and coordination of distributed power sources, based on traditional industrial automation SCADA architecture, mainly serving the internal grid. The second generation emphasized the coordination of generation and load, added load-side resource aggregation functionality, still using the traditional SCADA architecture, and participated in market-based transactions (such as spot markets and ancillary service markets) in addition to serving grid operators. The third generation centers around the extensive aggregation of diverse digital resources, utilizing new digital models and new digital architecture (cloud-edge-end coordination), creating a highly complex collaborative ecosystem, and even incorporating new technology elements like blockchain and DAO to facilitate the integration of various resources such as tri-generation, energy storage, vehicle-to-grid (V2G) technology, small-scale cogeneration, and thermal energy storage. However, based on China’s current practice, most virtual power plants are still at the first-generation level, with a few projects exhibiting some characteristics of the second generation. This indicates that China’s virtual power plant technology is still far from mature.

The second issue is the low level of marketization in trading, with evident narrow profit margins. Currently, electricity marketization trading has not been fully implemented nationwide in China and is still operating under the traditional electricity trading model. Electric energy is purchased and sold by grid companies from power plants to users, and the buying and selling prices are determined by relevant national departments. Therefore, at least in the short term, the country’s virtual power plants cannot expect the high returns seen in mature electricity markets abroad in terms of demand response. For instance, the annual duration of a certain load value can be roughly divided into three levels: The first is a 3% peak load with a duration of about 10 hours per year. This is the scarcest electric power resource, mainly controlled by the grid company, included in transmission and distribution prices, to ensure the safety and stability of the grid. The second is the 3% to 8% peak load, with a duration of about several tens of hours per year. This part of the electric power resource is mainly obtained through two ways: one is market-oriented. Although it enters the auxiliary service market, the threshold for trading is relatively high, making it difficult for small and medium-sized load aggregators to participate. The other is non-market-oriented, usually through direct adjustment resource compensation to respond to policy-driven demands. Currently, the frequency of execution is relatively low each year, and the prices are non-market-driven, making it insufficient to provide sufficient incentives for participants. The third level is those below 8%, with a duration of tens to hundreds of hours per year. This part is mainly responded to by retail power companies through a deviation cost control method, and the current level of marketization is also relatively low.

The third issue is the low market awareness of users and the lack of a supply-demand circulation system. Virtual power plants are a load-side management technology, and between the management and the managed, there needs to be a strong mutual trust established in load management, transaction structure, interest structure, and deep-level energy operation and service capabilities. However, currently, there is a lack of mutual trust between domestic load management and the managed (electricity users, mainly businesses), making it difficult to form a positive supply-demand system. On one hand, electricity users have a low level of market awareness. Under the existing electricity pricing system, most companies lack load management capabilities and find it difficult to truly understand the time value of loads and the cost of implementing off-peak operations. Consequently, without a sound incentive compensation mechanism, their willingness to participate is not high. On the other hand, China lacks genuine load operators. Presently, virtual power plants are mainly established by three types of entities: first, power grid companies invest and operate through project construction, which involves large investments and numerous customers but lacks deep-level interaction; second, a small number of entrepreneurial teams undertake construction, but this business itself lacks profitability and needs to be supplemented by other businesses; third, power generation companies establish virtual power plants through pilot projects, technology projects, etc., but most of them do not have the possibility of sustainable commercial operation. All in all, these virtual power plant developers pay little attention to the usage demands of the businesses behind them.

In the face of the issues existing in the development of China’s virtual power plants, researchers at ANBOUND suggest accelerating technology introduction and R&D by increasing policy support and utilizing modern digital technology. This aims to enhance the accessibility of it to modern technology. Additionally, there is the need to explore mature commercial profit models by promoting the reform of the electricity market system, establishing reasonable competition mechanisms, improving electricity market operation, creating a mature market environment, and attracting more private capital. Furthermore, efforts should be made to enhance the market awareness of electricity users. This can be achieved through collaborations with higher education institutions and other organizations to cultivate load operators with market awareness and professional capabilities, thereby improving energy management and load management capabilities for electricity users. In parallel, various training initiatives will need to be employed to strengthen electricity users’ understanding of virtual power plants and foster market-oriented habits. Simultaneously, establishing a relatively comprehensive compensation and incentive mechanism can further encourage active participation from electricity users.

Final analysis conclusion:

Entry into the virtual power plant race has become a hot investment trend for many companies in China. However, there are still issues with low levels of technological modernization, limited marketization of transactions, and low market awareness among electricity users in the country’s virtual power plant sector. As a result, mature service operators have not been nurtured, and a relatively mature industry chain has not been formed. Therefore, it is necessary to promote reforms in the electricity system, establish reasonable competition mechanisms, improve electricity market operation mechanisms, and enhance the market awareness of electricity users.

Kung Chan and Xia Ri are researchers at ANBOUND


Anbound Consulting (Anbound) is an independent Think Tank with the headquarter based in Beijing. Established in 1993, Anbound specializes in public policy research, and enjoys a professional reputation in the areas of strategic forecasting, policy solutions and risk analysis. Anbound's research findings are widely recognized and create a deep interest within public media, academics and experts who are also providing consulting service to the State Council of China.

Leave a Reply

Your email address will not be published. Required fields are marked *