The comparison analysis of ancillary service market in new power system: policies, practices and inspirations for China

1 Introduction

By the end of 2024, the cumulative installed capacity of renewable energy in China’s power system reached 1889 GW, accounting for about 56.4% of total capacity (Xinhua 2025). Wind and solar power contributed 521 GW and 887 GW, respectively, representing 74.54% of renewable energy, indicating that China’s current renewable energy has been dominated by new energy. Since the ‘14th Five-Year Plan’, demand from emerging loads, such as data centers, new energy vehicle charging, and renewable energy hydrogen production, have developed rapidly (Wei et al., 2025). As a result, both the supply and demand in the power system, along with the regulation resources, have become highly uncertain. The balance mechanism of the power system is therefore shifting from deterministic generation tracking uncertain load toward two-way matching of uncertain generation and uncertain load (Bi et al., 2024). This shift makes it necessary to design reasonable allocation and optimal combination schemes for regulatory resources, improve dispatch capability, and support the construction of new power systems.

To maintain safe and stable power system operation and ensure the power quality, ancillary services have been introduced and refer to services such as peak shaving, frequency modulation, reserve, and ramping provided by the adjustable resources in addition to the normal power production, transmission, and consumption (National Development and Reform Commission, Energy Bureau 2024). To ensure these services are normally, properly, and officially operated, the ancillary service market has been proposed and refers to the market operation mechanism in which the business entities provide ancillary services through the market-oriented mechanism and obtain corresponding benefits based on market rules (Hirst and Kirby 1996). Reasonable formulation of ancillary service market trading varieties, a perfect market environment, business model, and reasonable call mechanism can effectively encourage diversified flexible resource investment, allocation, and cost recovery, fully release its potential, actively participate in and flexibly support system operation and scheduling, and ensure the adequacy of system regulation ability (National Development and Reform Commission, Energy Bureau 2024; Xu et al., 2022).

The research on the ancillary service market has been closely related to the system challenges brought by the high proportion of renewable energy integration. The literature review of this paper is as follows:

Most studies have demonstrated the necessity of introducing market-oriented mechanisms to stimulate flexible resources (Yang et al., 2023; Guo et al., 2020; Feng 2023; Yongsheng, 2023). However, the traditional planning compensation mode based on the generation side is difficult to fully stimulate the adjustment potential of each link of source, network, load, and storage, especially unable to adapt to the new situation of large-scale access of distributed resources (Eid et al., 2016), which becomes more prominent under the background of frequent extreme weather and deepening green transition (Xu et al., 2024). Therefore, it has been accepted to establish a market-oriented cost-sharing and return mechanism of ‘who provides, who profits; who benefits, who bears’ (Shu and Mays 2024).

After integrating this mechanism and practices, China’s market construction has entered the fast lane of ‘market-oriented exploration’ from ‘planned compensation’, and the pilot effects of ancillary services have begun to show (Asia and Dept 2023; Guo et al., 2020). However, the existing research on China’s ancillary service market focuses more on the development status, local pilot, and existing problems (Wu Shan et al., 2023; WANG Jiahao 2023; LI et al., 2019; Nan et al., 2025), and also points out that the market still faces challenges such as a relatively single variety structure, lagging price signal transmission (IEA 2023), incomplete inter-provincial barriers (Lin and Wetzel 2019; L. Huang 2023), and an imperfect cost recovery mechanism for emerging entities such as energy storage (Wang et al., 2025; Changxiang et al., 2022). For the frontier ancillary service varieties oriented to system stability (Oren 2001), such as inertial support (Poolla et al., 2020), fast frequency response (Fernández-Muñoz et al., 2020) and black startup (Pan et al., 2025), most of the domestic research is still in the stage of technical feasibility discussion and mechanism conception, lacking mature market design and in-depth analysis of practical cases (IEA 2024).

However, research on the international market mechanisms has been more in-depth. Scholars have conducted detailed analyses of the three-level balancing market based on the fine division of response time in Germany, the joint optimization and clearing of electric energy and ancillary services in the PJM market in the United States (Ott 2003), and the frequency control ancillary service (FCAS) mechanism independent of the energy market in Australia (Australian Energy Market Operator Ltd 2015; PW Consulting 2025). These studies focus on product design, pricing model, such as the ‘capacity + energy’ two-part system, and on friendly access rules for distributed resources, providing an important reference for China.

While existing studies have provided valuable case descriptions of individual markets, it is still insufficient to systematically compare the similarities and differences between Chinese and foreign markets, and closely combine the phased needs of China’s new power system construction to propose an operable hierarchical and phased development path (IEA 2024). This paper aims to address these gaps by extracting core experience and enlightenment for China’s market construction through multi-dimensional international comparison.

The main contributions of this paper are as follows: First, it constructs a theoretically grounded, multidimensional comparison framework: rather than offering purely descriptive comparisons, it analyzes market designs through the lens of incentive compatibility and mechanism design theory, explaining why certain mechanisms succeed or fail. Second, it provides a data-informed characterization of the performance gap between China and mature markets across key metrics, including capacity pricing, clearing mechanisms, and the participation depth of distributed resources. Third, it proposes a hierarchical, actionable development path tailored specifically to China’s staged conditions, distinct from generic policy recommendations, by incorporating phased targets for emerging ancillary services such as inertia, black start, and voltage support. The conclusions and prospects presented herein may serve as a reference for refining China’s ancillary service market. The framework of this paper is organized as Figure 1 shows.

2 A comparison analysis of policies and practices of the ancillary service market

In this paper, Germany, the U.S. PJM, and Australia’s National Electricity Markets (NEM) are selected as international comparison samples, primarily based on the following criteria: 1) Market maturity: All are benchmark markets operating for over 15 years with relatively well-established mechanisms; 2) Energy structure representativeness: Germany (with high-proportion renewable energy), PJM (a mix of conventional power sources and market mechanisms), and Australia (an independent grid with high-proportion distributed resources) cover different transition pathways; (3) Mechanism innovation: these markets pioneer areas such as balancing market design, joint clearing, and FCAS markets.

This study focuses on wholesale market mechanism comparisons rather than retail markets or technical implementation details. As markets such as Nord Pool and the UK also hold significant reference value, future research may expand the scope for a more comprehensive summary.

2.1 Policies and practices in China

At present, 16 provinces across the country have established a peaking market, 15 provinces have established a frequency modulation market, two provinces have established a ramping market, and six regions have established frequency modulation, reserve, peaking, and other markets (National Development and Reform Commission 2025), as shown in Table 1.

TABLE 1. Service type, function, dispatch mechanism, and provincial construction status of China’s ancillary service market, based on National Development and Reform Commission (2025) data.

2.1.1 Policies

The evolution of China’s ancillary service market is illustrated in Figure 2. Building on the market-oriented shift discussed above, competitive procurement now helps ensure the safe and stable operation of the power system (Feng et al., 2025) while promoting renewable energy consumption (Asia and Dept 2023). The key points of its mechanism design include (IEA 2023):

• Clarify market players: the scope of business entities has expanded from traditional power generation enterprises to power sales enterprises, power users, and new business entities, and specifically clarified the legal status of energy storage enterprises, virtual power plants, smart microgrids, and V2G operating enterprises, aiming to guide diversified flexible resources to participate in system regulation (China Advances Power Market Reform to Support Renewable Integration and System Flexibility - Sino-German Cooperation on Climate Change 2025).
• Improve the cost transmission mechanism: the core principle of ‘someone who provides profits, and someone who benefits bears’ is established. In the area where the power spot market operates continuously, the cost of ancillary services such as frequency modulation and reserve is shared by the user’s electricity consumption and part of the on-grid electricity, which ensures the reasonable transmission of costs.
• Regulate market varieties and convergence: The Two Rules divide ancillary services into active power control, reactive power control, and accident disposal according to their functions, and standardize the definition of peak shaving, frequency modulation, backup, ramping, and other varieties. It is emphasized that the ancillary service market and the spot market of electric energy should be constructed in coordination. At the initial stage, they can be cleared independently, and when the conditions are mature, the joint clearing should be promoted to achieve overall optimization.
• Establish a unified construction path: Aiming at the problem of large differences in rules across the country in the past, the rules have been designed to follow a scientific process from demand demonstration, simulation trial operation, to formal operation, aiming to promote the integration of decentralized provincial markets into the national ‘one game of chess’.

2.1.2 Practices

Under the framework of unified rules, diversified market practices have been carried out in various places. The main varieties and representative projects are concluded in Table 2. According to it, China’s ancillary service market practice has formed a diversified pattern based on frequency modulation, peak shaving, backup, and ramping services. Some other services are still under exploration, like the inertia service.

TABLE 2. Representative provincial practices of ancillary service markets in China, summarizing frequency modulation, peak shaving, and ramping service pilots by region.

In terms of the frequency regulation market, Hunan (Chao et al., 2021), Chongqing (National Energy Administration Central China Regulatory Bureau 2022), and other places promote the participation of new entities, such as independent energy storage, and pay according to performance, which effectively improves the regulation ability and grid frequency stability.

The peak shaving market (such as Hunan) stimulates the deep peak shaving of thermal power and the response of load-side resources through the market mechanism, which significantly promotes the consumption of new energy. To optimize the allocation of resources, innovative mechanisms such as inter-provincial mutual aid in Northeast China (National Energy Administration Northeast Regional Office 2025) and ‘semi-coupling’ clearing in Shanxi have emerged in reserve services (Zou et al., 2024). According to the latest policy effective October 2025, the price cap for peak-shaving ancillary services within the province shall not exceed the grid-connected electricity price of grid-parity renewable energy projects, i.e., 450 CNY/MWh.

Auxiliary ramping services, such as Guangdong Province (Jiantao Shi and Wen Zhao 2025), have entered a simulated trial operation to cope with short-term and violent fluctuations in new energy output. For new services such as inertia, the current compensation is mainly through the Two Rules in some regions. (Two Rules is a short abbreviation for official document issued by the National Energy Administration, i.e., South China Energy Regulatory Bureau, and launched a revised version of the Two Rules of Southern Region on Feb. 14th, 2025.).

In general, the construction of some relevant regional markets is in the research and exploration stage, all kinds of market practice projects are moving from pilot to deepening, and an ancillary service system covering multiple types of resources and multiple time scales is gradually constructed.

2.2 Policies and practices in other countries

2.2.1 Policies

In some mature electricity markets, represented by Germany (SMARD 2020), PJM in the United States (PJM Interconnection, 2025), Australia, and etc., the auxiliary service mechanism design, i.e., sometimes referred to as ‘balanced smarket’ in Europe, is more refined and highly coupled with the electricity energy market, as Figure 3 shows. Taking the German balancing market as an example, the outstanding features of its mechanism design include:

• Fine product classification: German ancillary service products are strictly divided into three levels according to response time and function: primary frequency modulation reserve (Frequency Containment Reserve, FCR, second-level response), automatic frequency restoration reserve (aFRR, 5 min response), and manual frequency restoration reserve (mFRR, 15 min response). This classification enables accurate procurement of different performance resources.
• ‘Capacity + Energy’ compound pricing model: In the German aFRR and mFRR market (Industry News 2025; Jomaux 2025), participants can not only obtain the fixed capacity income that provides reserve capacity through the capacity market bidding, but also obtain energy compensation when it is called. This model provides a stable and predictable revenue stream for flexible resources (especially resources with high investment costs) and reduces investment risks.
• Friendly design of distributed resources: Germany’s balancing market rules do not require aFRR and mFRR providers to offer both upward and downward adjustment capabilities in equal measure. This rule greatly lowers the threshold for Virtual Power Plants (VPPs) to aggregate resources with only one-way regulation capability (such as load reduction and photovoltaic curtailment), promoting broader participation of distributed resources (Zdunkewicz, 2025).
• Regional integration and cross-border synergy: Europe is promoting the establishment of a unified and balanced cross-border market (Huang et al., 2022). Germany has participated in the balanced resource cooperation project in Europe, realized the transnational procurement and unified call of balanced resources with neighboring countries (such as the Netherlands), optimized the allocation of flexible resources in a larger geographical scope, and improved the overall economy and security (Chen et al., 2019).

2.2.2 Practices

Foreign practice projects profoundly reflect their mechanism design concept. Several typical markets as examples to be illustrated are concluded in Table 3. It can be seen from the table that the foreign mature ancillary service markets present a high degree of differentiation and refinement:

• PJM: Features product segmentation (frequency regulation divided into RegA/RegD signals) and multi-market joint optimization of energy, ancillary services, and capacity markets (Monitoring Analytics 2025). This co-optimization prevents strategic arbitrage between segmented markets, internalizes opportunity costs, and generates a unified price signal.
• Australia (NEM): Operates a frequency control ancillary service (FCAS) market independent of the energy market, with performance-based payments ensuring reliability. In September 2025, battery storage revenue declined to AUD 85,000/MW, while over 6 GW of grid-forming storage capacity has been announced for delivery by 2027 to provide inertia and system strength services.
• Germany: Employs a classic three-tier balancing market (FCR, aFRR, mFRR) with a “capacity + energy” compound pricing model to address underinvestment in high-capital-cost, low-marginal-cost flexible resources. Capacity payments cover fixed investment risk, while energy payments reflect real-time value. Its rules are VPP-friendly (Liu et al., 2025), and in 2025 Germany achieved a legislative breakthrough in inertia service procurement.
• Nord Pool: Leverages a highly integrated regional grid to offer finely differentiated frequency products (milliseconds to minutes), providing high-yield opportunities for fast-responding resources like storage.

TABLE 3. Representative practices of ancillary service markets in Germany, PJM, and Australia’s NEM, summarizing product design, pricing structure, and participation rules.

These markets differ in their approaches to peak shaving, reserve (Hogan and Pope 2019), and ramping (Ott 2003), yet share common trends: increasingly refined product design, performance-based pricing, active integration of distributed and storage resources, and forward-looking exploration of emerging challenges such as system inertia decline (Nemes 2021). The following section synthesizes these findings into a structured comparative analysis.

2.2.3 Limitations and challenges

Despite their maturity, like features shown in Figure 4, these foreign markets also exhibit notable shortcomings that warrant critical attention rather than uncritical emulation. Germany’s fine-grained three-level balancing market has driven up transaction and compliance costs for smaller and distributed participants, and its aFRR/mFRR capacity prices have shown pronounced volatility as renewable penetration rises, complicating long-term investment planning for flexible resources (Jomaux 2025).

PJM’s co-optimized clearing of energy, ancillary service, and capacity markets, while efficient in theory, has faced criticism for insufficient market power mitigation and has produced sharp capacity price spikes. Its 2026/27 capacity auction cleared at $329/MW-day, well above historical averages—raising concerns about affordability and the adequacy of current market power safeguards (Monitoring Analytics 2025).

Australia’s FCAS market, despite its institutional separation from the energy market, has experienced significant revenue volatility for battery storage participants; NEM battery storage revenue fell to AUD 85,000/MW in September 2025, illustrating the risk that performance-based payment models can expose flexible resource investors to unpredictable returns. In addition, the high administrative complexity of maintaining separate FCAS products for multiple response timescales has raised entry barriers for smaller aggregators.

These experiences yield valuable lessons for China: fine-grained product differentiation should be pursued gradually to avoid excessive complexity in early-stage markets; robust market power mitigation mechanisms should be established before deep co-optimization is introduced; and price volatility risk-sharing tools should accompany performance-based payment designs to protect early-stage investors.

2.3 Comparison analysis

By combining the market policies and practices at home and abroad, the comparison of the following dimensions can be summarized in Table 4. China’s ancillary service market has taken a key step in the top-level design. The mechanism framework is systematic, inclusive, and forward-looking, and is committed to solving the systemic challenges brought by the high proportion of new energy access. Compared with the mature, fine, and international market system formed in foreign countries for decades, there are still gaps in the depth of rule implementation, the breadth of subject participation, the sensitivity of price signals, and the maturity of cross-border or regional resource allocation in China’s market.

TABLE 4. Comparative summary of ancillary service market design dimensions between China and mature international markets, covering services, pricing, clearing mechanisms, and participants.

2.3.1 Development path

China’s power ancillary service market is evolving from basic coverage to refined design, currently in a stage of unification and standardization, transitioning from fragmented provincial pilots to national rules due to prominent regional discrepancies (China Energy Storage Alliance 2021). It has expanded from frequency modulation and peak shaving to inertia exploration (IEA 2024; Jenkins et al., 2022).

Mature international markets are now optimizing regional integration through two distinct models: PJM’s centralized, single-price co-optimization across a large multi-state footprint, and Europe’s federated model linking national balancing markets via common platforms while preserving national operators (Carrie 2024). China’s provincial structure more closely resembles the European arrangement. We therefore suggest that cross-provincial pilots draw primarily on the EU Balancing Guideline’s platform-based approach rather than PJM’s centralized clearing. North/East China is proposed as a pilot site due to its existing grid interconnection, comparable renewable penetration, and established inter-provincial coordination mechanisms, which lower institutional barriers. This is an initial siting rationale, not an optimality claim; further quantitative analysis is left for future research.

2.3.2 Features of market policies

• Services: China’s market structure is clear, with established frameworks for frequency modulation (in 15 provinces), peak shaving (16 provinces), and ramping services (2 provinces). However, international markets demonstrate finer product granularity. For instance, Germany strictly segments frequency regulation into three tiers: FCR (30-s response), aFRR (5-min), and mFRR (15-min), enabling precise resource procurement.
• Pricing and compensation: China employs a hybrid model. While some services retain compensatory traits, its market-oriented segments (e.g., frequency modulation) use a “mileage + performance” compound pricing. A common clearing price ceiling is 15 RMB/MW. Internationally, mature markets like PJM widely adopt a transparent “capacity price + energy price” model. For example, PJM’s 2026/27 capacity auction price reached $329/MW-day, strongly signalling system reliability value.
• Clearing Mechanism: In China, independent clearing remains dominant, though pilots like Shanxi’s “semi-coupled” clearing for backup services point toward future integration. In contrast, international markets like PJM extensively use joint optimization (co-optimization) clearing across energy, ancillary service, and capacity markets. This mechanism is key to minimizing total social costs and efficiently allocating flexibility resources.

China’s current ancillary service product set insufficiently matches the multidimensional flexibility needs of renewable variability. We recommend adopting Germany’s response-time-based classification (aFRR/mFRR) as the primary near-term reference, because China’s province-level clearing structure resembles Germany’s zonal system rather than PJM’s centrally co-optimized market; this taxonomy can be implemented incrementally within each province without the deeper redesign that PJM’s signal-based differentiation would require. PJM’s finer product segmentation is better suited as a longer-term reference once provincial markets become more tightly integrated. We emphasize that this recommendation is limited to product taxonomy, which is settlement-structure-agnostic, and does not import pricing or settlement rules from either market, thereby avoiding potential inconsistencies from combining German zonal and PJM nodal designs.

2.3.3 Practices and participants

China’s ancillary service market encourages new entities like storage and VPPs through policy, yet their participation remains limited due to access thresholds, unclear business models, and complex settlement rules. Internationally, barriers are lower, with clear legal status, flexible aggregation rules (e.g., Germany’s one-way resource participation), and performance-based payments (e.g., Australia’s FCAS). We recommend pilot programs to clarify VPPs’ role as “third-party aggregators.” PJM’s settlement approach for aggregated storage offers a readily adaptable template, whereas designing an original domestic methodology would require several years of pilot data China currently lacks, making adaptation the lower-risk near-term option. Additionally, a UK-style public registration platform could reduce search and verification costs for small aggregators, and it operates upstream of settlement rules, so adopting both creates no inconsistency.

2.3.4 Other potential applications

In addition to mature and widely traded ancillary services above, such as frequency regulation, backup, and ramping, some leading foreign electricity markets have begun to explore and establish more forward-looking and targeted ancillary service markets in response to the new challenges brought by the high proportion of renewable energy integration. These markets mainly focus on the basic physical characteristics of system stability (such as inertia) and the recovery ability in extreme cases, which represent the frontier direction of the development of the ancillary service market.

• Inertia and fast frequency response market

As renewable energy and inverter-connected storage replace traditional synchronous generators, system inertia continues to decline, posing challenges to frequency stability (Ulbig et al., 2014). Some markets have initiated commercial procurement of inertia or equivalent “virtual inertia” and “rapid frequency response” services. For instance, Australia’s National Electricity Market (NEM) has established regional minimum inertia requirements, permitting synchronous condensers and grid-forming energy storage to compete as resources providing inertia or millisecond-level rapid frequency response (“Inertia Requirements Methodology” 2024). Meanwhile, the UK’s National Grid has introduced new products like dynamic curtailment and rapid frequency response, mandating response times under 1 s, with battery storage emerging as the primary supplier (“National Energy System Operator” 2020; National Grid ESO 2023).

• 2. Black start service market

Black start refers to the process by which the power system gradually restores power supply through internal self-starting units, without relying on the external power grid, after a major system-wide blackout (Friedlingstein et al., 2026). Mature international markets typically procure this as a specialized ancillary service. Grid operators regularly procure services from hydroelectric and gas-fired units with black start capability through tenders or bilateral contracts. Successful bidders receive fixed capacity fees and additional energy compensation when their services are activated. This mechanism is well-established in markets like PJM, the UK, Germany, ERCOT, and New Zealand (PJM Performance Compliance Department 2024; Black Start Strategy and Procurement Methodology 2021/22, 2021; ERCOT Grid Insights: Addressing Issues Important to Maintaining a Reliable and Resilient Grid 2025). It ensures reliable system restoration while avoiding massive investment in dedicated grid-owned resources.

• 3. Voltage control and reactive power support market

While many markets still mandate reactive power services through technical standards or bilateral agreements, some regions have advanced market-based reactive power transactions. For example, ERCOT operates a reactive power services market, procuring reactive capacity at critical nodes based on system voltage support requirements (Potomac Economics 2024). Generators, synchronous condensers, and inverters with dynamic reactive power capabilities can all participate in bidding to earn revenue. With the large-scale integration of distributed PV, voltage over-limit issues in distribution grids have become increasingly prominent (Sarah and Paslawski 2023). Some regions are exploring aggregating the reactive power regulation capabilities of distributed resources (such as smart inverters) as an effective supplement to transmission grid voltage support (ERCOT reactive capability testing, 2025).

• 4. Distributed resource aggregation and virtual power plant market

This innovative participation model uses digital technology and market rules to aggregate vast, dispersed flexible resources into controllable units, enabling them to participate in existing markets at scale as VPPs (Thomas Feenstra 2025; Peter 2019). Germany/EU balancing markets permit aggregators to submit asymmetric regulation capabilities, significantly lowering barriers for distributed resource participation (Loßner et al., 2017); CAISO and PJM have established dedicated demand response aggregator categories with detailed aggregation, communication, and settlement protocols (final et al., 2025); Australia’s EDGE project validated the complete transaction workflow for massive distributed PV and residential storage systems to simultaneously participate in wholesale electricity markets and provide ancillary services through aggregators while meeting distribution network security constraints, providing critical practical evidence for future market rule reforms (Ama van der Sluys, 2022).

3 Inspirations for China’s ancillary service market

Our recommendations are not wholesale transplants. Each is selected to address a specific Chinese institutional feature (e.g., provincial clearing, hybrid compensation) and filtered against foreign-market limitations in Section 2.3 to avoid known problems or add explicit caveats. When multiple alternatives exist, we prefer the one requiring less deviation from China’s current structure, and we state this rationale explicitly for each choice.

China’s electricity market is currently in the early stage of construction, where government regulation and market mechanism coexist. Under this mechanism, in the early stage of market development, the fixed costs and variable costs of power plants and other power companies are compensated by capacity electricity prices and electricity prices, respectively, to ensure basic profitability and recover part of the fixed costs. However, the two-part electricity price mechanism cannot directly participate in market competition, with relatively weak profitability and insufficient flexibility.

3.1 Future development path level inspirations

3.1.1 The general ideas

To support the construction of a new power system with and solve the challenges of randomness and volatility brought by high-proportion new energy, the general idea of developing the ancillary service market in China in the future is to follow the basic principles of ‘system design, step-by-step implementation, easy first and difficult later’ (Chen et al., 2025) and to take market-oriented reform as the core driving force, and to systematically and step-by-step tap and mobilize all available flexible resources by building a modern power market system with wide access, complete variety, fine pricing and sound mechanism, to provide a solid guarantee for the construction of a new power system with clean, low-carbon, safe and abundant, economical and efficient, and supply and demand coordination (IEA 2023).

Future power market reform should start with coherent top-level design to align energy, auxiliary, and capacity markets, then expand pilots’ step by step into a multi-level, flexibility-oriented system. Reforms must remove regional barriers, broaden access, and ensure equal competition among all generation types while fully integrating new flexible resources into ancillary services. Finally, building on spot and auxiliary markets, a capacity market or similar mechanism should be introduced to secure stable returns for reliable, flexible resources and safeguard long-term supply security.

To address China’s lack of new ancillary services like inertia and the reliance on administrative compensation, we propose two measures. First, pilot an inertia or rapid frequency response market in high-renewable regions (e.g., Northwest/North China), adopting Germany’s fixed-price procurement model, designed for early-stage markets with few bidders, to ensure predictable revenue while avoiding illiquid auctions; a transition to competitive procurement should follow once participation grows. Second, incorporate “inertia” and “voltage support” into national grid operation rules, and adopt PJM’s approach of embedding ramping needs into spot market clearing, which internalizes opportunity costs without requiring a separate product and settlement system, a lower-cost first step for China’s consolidating market, though dedicated ramping products remain viable later.

3.1.2 The suggested period objectives

• Recent market design

In regions without spot-market pilots, reform should prioritize refining large-user direct procurement while preserving system reliability and adequacy. Operating hours of units under direct contracts must be capped to safeguard reserve, frequency, and peak-shaving capabilities, while deviation penalties should be imposed to incentivize accurate forecasting by both generators and consumers. Simultaneously, ancillary-service remuneration frameworks should expand eligibility to include storage, demand-side resources, and third-party providers.

In spot-pilot regions, reform should pursue an integrated market-oriented framework that coordinates energy, ancillary, and retail segments. Medium- and long-term physical contracts must specify daily generation and consumption curves, complemented by day-ahead and real-time spot markets with 15–30-min dispatch intervals. Ancillary services should be competitively tendered day-ahead to procure cost-effective flexible resources across differentiated product lines. In areas with high renewable penetration, piloting inertia or fast-frequency response markets is advisable. Initially, Germany’s regional unified fixed-price procurement model offers a useful benchmark to stabilize investor expectations. On the retail side, time-of-use and dynamic pricing schemes should steer consumer behavior, shifting loads to off-peak periods to better align demand with system conditions.

• 2. Medium-term market design

Further expand the degree of market competition, gradually reduce and cancel priority power generation and priority purchase of electricity, and promote and improve the spot market. The minimum unit trading period of the day-ahead market becomes half an hour, and the intra-day market is established to provide a flexible resource trading platform for all parties. Based on the spot ancillary service market, the medium and long-term ancillary service market is introduced, and regional black start service capacity markets may be explored for establishment, so that the power system can also guarantee the adequacy of flexible resources on a long-term scale. In the retail market, market-oriented users and electricity sales packages are being increased to improve the optimal allocation level of demand-side flexibility resources.

• 3. Long-term market design

In the long term, a fully competitive electricity market should be established, and the market mechanism should be fully utilized in the design of all aspects of the market to improve system flexibility. All kinds of planned power and priority power generation are cancelled. In the day-ahead market and real-time market, the full power optimization clearing method is adopted. The trading period is further subdivided into 5–15 min to realize 15-min or even 5-min real-time economic dispatching and improve the optimal allocation level of short-term power resources. The transmission and distribution price mechanism should adopt the peak load responsibility method to improve the utilization of power grid resources and system flexibility. The retail market will be widely promoted in major regions of the country, and the demand side will be fully encouraged to release flexible adjustment resources. And a comprehensive ancillary services market system may be established encompassing multiple dimensions such as frequency, voltage, inertia, and restoration capabilities.

3.2 Technical level inspirations

3.2.1 In-depth evaluation of flexibility resource aggregation features

In new power systems, VPPs aggregate diverse flexible resources but exhibit multi-dimensional coupling. Temporally, storage and EV response delays vary; spatially, grid congestion constrains cross-regional dispatch; cost-wise, marginal costs fluctuate with equipment degradation and fuel prices. Existing clearing models fail to adequately capture these interdependencies, limiting allocative efficiency, necessitating multi-dimensional joint modeling and AI-assisted real-time dynamic evaluation.

Operational uncertainties further intensify these challenges. Boundedly rational user behavior, stochastic state-of-charge (SOC) swings in storage, and intermittent renewable generation often trigger “adverse selection,” whereby declared resource capability diverges from actual deliverability, impairing market-clearing precision and system reliability. Addressing this issue requires integrating uncertainty quantification with adaptive robust optimization, coupled with high-precision short-term prediction to systematically minimize forecasting errors.

3.2.2 Collaborative control of diversified flexible resources

The regulation objectives and response characteristics of diversified heterogeneous flexible resources, such as energy storage, electric vehicles, and controllable loads, are different. For example, the energy storage system is suitable for fast frequency regulation, and the charging of electric vehicles needs to give priority to meeting the travel needs of users. The two may have a competitive relationship in power distribution. If there is no effective coordination mechanism, it will lead to ‘mutually exclusive operation’ of internal resources and reduce the overall service efficiency. Therefore, it is necessary to establish a resource collaborative allocation algorithm based on multi-agent consensus and design a dynamic priority strategy to realize the organic cooperation between resources.

At the level of communication and perception, with many distributed resources connected to the power of the Internet of Things, the problem of insufficient edge computing power is becoming increasingly prominent. At present, due to factors such as the limited computing power of edge nodes and inconsistent communication protocols, the collection of resource status information is delayed, which directly affects the response quality of real-time ancillary services. Therefore, it is necessary to comprehensively improve the edge intelligence level of the power Internet of Things and deploy an end-edge-cloud collaborative computing framework to provide technical support for precise control.

3.3 Market level inspirations

3.3.1 The improvement of the multi-market coordination mechanism

In the current ancillary service markets, frequency regulation and reserve often operate independently from the electricity market, and the rule system lacks an effective connection. This market segmentation phenomenon makes multi-functional resources such as energy storage face ‘role conflict’: participating in frequency modulation services may not be able to carry out peak-valley arbitrage at the same time, resulting in low resource utilization, and even triggering market chaos such as long arbitrage. To solve this dilemma, it is necessary to promote the joint clearing of the electric energy-ancillary service market, design cross-market coupling transaction contracts, and guide the optimal allocation of resources.

For new market entities such as shared energy storage, when they participate in the wholesale, retail, and ancillary service markets at the same time, they face the adaptation problem of different market rule systems. The differences in clearing cycle, pricing mechanism, and settlement rules among different markets have significantly increased operational complexity and compliance costs. This requires the establishment of a unified interoperability framework for market rules, the development of resource optimization decision support systems for multi-market transactions, and the reduction of market participation thresholds.

3.3.2 Improve the reasonable pricing and contribution quantification mechanism

In the pricing of ancillary services, special services such as inertia are faced with the contradiction of value identification. This kind of service not only has the safety value of ‘full-time guarantee’, but also only reflects the economic value when the system inertia is short. The existing pricing mechanism is difficult to balance these two value dimensions, resulting in unstable supplier returns and affecting investment enthusiasm. A dynamic pricing mechanism based on the scarcity of system inertia can be considered, and a two-tier pricing model combining long-term contracts and spot markets can be introduced to stabilize market expectations.

The quantification of the contribution of aggregated resources in virtual power plants also faces technical challenges. Since it is difficult to directly observe the individual contribution when multiple distributed resources jointly provide ancillary services, the existing allocation methods are likely to cause fairness doubts. In this regard, we can study the income distribution model based on cooperative game theory and trace the contribution degree with big data analysis technology to ensure the rationality and transparency of income distribution.

3.3.3 Improve the resource cost recovery mechanism

China’s ancillary service market, characterized by limited product varieties and an immature institutional framework, faces two critical deficiencies in cost recovery for flexible resources.

First, the underdeveloped spot market renders existing products inadequate to meet diversified flexibility demands. This mismatch undermines investment viability for demand-side resources and new energy storage, thereby eroding their regulatory potential.

Second, capacity compensation and pricing mechanisms remain suboptimal. The narrow eligibility scope and fixed compensation under the current two-part tariff prevent flexible resources from capturing scarcity rents during supply shortages, while also failing to convey timely price signals. Moreover, capacity market pilots are restricted to a few provinces and focus narrowly on peak-shaving, lacking multi-temporal variants for frequency regulation, ramping, and inertia. To address these shortcomings, devising appropriate cost-compensation models and establishing a robust long-term capacity market, including progressively diversifying eligible resource types and trading products, is imperative. Such institutional refinement would ensure long-term resource adequacy and substantially enhance overall power system flexibility.

3.3.4 Mitigating market risks and unbalanced funds

The large-scale access to new energy has led to an increase in the price fluctuation of ancillary services and a significant increase in market risk. For example, the Australian market has experienced an extreme situation in which the inertia price fluctuates 8 times within 8 h, bringing huge risk exposure to participants. To this end, it is necessary to improve the price cap and floor mechanism, explore financial hedging tools, and help market players manage possible market risks.

Resource response bias can also lead to the accumulation of unbalanced funds. In some domestic spot pilot projects, the proportion of unbalanced funds has exceeded 5%, which will affect the financial sustainability of the market in the long run. This requires the establishment of a sound deviation assessment mechanism and a balanced account system, and the establishment of a deviation responsibility traceability system based on post-verification to effectively control the size of the funding gap.

3.4 Policy and standard level inspirations

3.4.1 Improve the technology and measuring standard system

At present, technical requirements such as energy storage, security operation and maintenance, virtual power plant aggregation interface, and inertia service have not yet formed a unified standard. There are obvious differences in the response time requirements of energy storage frequency modulation in different regions, which restrict the large-scale application of equipment and the provision of services across regions. It is urgent to promote the unification of technical standards at the national level and establish a standard conformance testing and certification system.

In terms of measurement standards, ancillary services provided by flexible resources need to be accurately quantified. Taking the inertia service as an example, there are many measurement indices, such as kinetic energy (MW·s), moment of inertia (kg·m2), and so on, and there is no unified standard. The consistency of definition and measurement standards should be promoted. Specifically, high-precision phasor measurement units (PMUs)—devices that record voltage and current phasors with GPS-synchronized time stamps at sub-second resolution—should be deployed at key grid nodes and inertia-providing resources (e.g., synchronous condensers, grid-forming storage) to directly observe frequency response speed and the rate of change of frequency (RoCoF) following a disturbance. These measurements would then serve as the technical basis for verifying and settling inertia and fast-frequency-response transactions, replacing indirect, model-based estimates with directly observed performance data, consistent with the frequency-measurement approaches already used in Australia’s and the UK’s inertia and fast frequency response markets discussed above.

3.4.2 Put forward and improve the regional mechanism and policy difference elimination method

The differences in market access and compensation standards among regions hinder the cross-regional optimal allocation of resources. This regional barrier limits the healthy development of the ancillary service market. It is necessary to strengthen the top-level design at the national level and establish a regional coordination mechanism. At the same time, the problem of convergence after the expiration of local subsidy policies is also worthy of attention. We should promote the smooth transition of subsidy policy to a market-oriented mechanism, stabilize the expectation of project investment return, and promote the sustainable development of the industry by establishing a long-term mechanism, such as a capacity electricity price. Having outlined development-path, technical, market, and policy-level inspirations, the final section synthesizes these findings into overarching conclusions and future research directions.

4 Conclusion and prospects

This comparative analysis yields two synthesized insights rather than a simple restatement of case findings. First, the gap between China’s market and mature international systems lies not in the availability of basic services but in the granularity of product design, the precision of pricing incentives, and the efficiency of cross-regional coordination. Second, China’s localized innovations, such as ‘semi-coupled’ clearing, have transitional value, but a longer-term evolution toward joint optimization with energy markets is necessary to minimize total societal costs.

These findings underscore a distinct policy implication: China’s market construction must transition from a phase focused on “coverage” to one dedicated to “refinement and synergy.” A feasible pathway involves leveraging fine-grained product design to unlock the potential of existing flexible resources while establishing cross-regional coordination mechanisms to expand the boundaries of resource optimization. This provides a reform roadmap that integrates international benchmarks with local adaptability.

Looking ahead, future development should explicitly incorporate emerging service markets into a phased implementation plan. In the near term, pilot regions should initiate technical standard research and simulated operation for inertia and fast frequency response markets. By the medium term, exploration should begin for establishing regional black-start service capacity markets. Ultimately, the long-term vision should be the formation of a comprehensive ancillary service market system encompassing frequency, voltage, inertia, and restoration capabilities. Future research should focus on quantitative cost-benefit analyses for emerging services like inertia and ramping to inform precise market design.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fenrg.2026.1773305/full#supplementary-material

Supplementary material Supplementary material at the source

Footnotes

Edited by: Alireza Dehghanisanij, University of Waterloo, Canada

Reviewed by: Hossein Afshar, Islamic Azad University, East Tehran, Iran

Jingrui Liu, Chongqing University, China