As we advance toward carbon neutrality, the construction of a new power system dominated by renewable energy has become a pivotal strategy. With the rapid expansion of wind and solar photovoltaic capacity, the operational characteristics of power systems are undergoing profound transformations. The integration of high proportions of variable renewable energy introduces challenges such as peak regulation difficulties, frequency stability pressures, and increased voltage control complexity. In 2023, China added 370 GW of renewable energy capacity, accounting for 82.7% of total new power generation installations, with wind and solar contributing 76 GW and 220 GW respectively. The total installed capacity of wind, solar, and photovoltaic power exceeded 1,000 GW. Against this backdrop, battery energy storage systems have emerged as indispensable flexible regulation resources, offering fast response, flexible configuration, and high regulation accuracy. They play an increasingly critical role in smoothing renewable output fluctuations, providing ancillary services such as frequency regulation, peak shaving, and spinning reserves, enhancing grid resilience, and improving power supply reliability and quality. By the end of 2023, China’s new energy storage installed capacity exceeded 30 GW, dominated by lithium-ion batteries.
However, the large-scale deployment of battery energy storage systems in new power systems still faces numerous hurdles. High initial investment costs, incomplete lifecycle cost-benefit assessment frameworks, and a lack of sustainable business models hinder profitability. Moreover, the market positioning of storage remains ambiguous, with limited participation channels and underdeveloped trading rules. The absence of comprehensive planning standards, grid integration technical requirements, and operation maintenance specifications further complicates the landscape. Safety concerns, cycle life limitations, and recycling challenges also demand urgent attention. To address these issues, we examine the planning and configuration of battery energy storage systems from two perspectives: market mechanism design and government regulation.
Current Status and Key Issues of Battery Energy Storage Systems
Multifunctional Value and Strategic Role
Battery energy storage systems provide multiple functional values essential for the new power system. First, they offer flexible power regulation, effectively smoothing renewable output fluctuations through rapid charge-discharge switching. Properly sized storage can significantly reduce variability in wind and solar generation, improving power quality. Second, they deliver various ancillary services including primary and secondary frequency regulation, automatic generation control, and spinning reserves. Compared to conventional thermal units, storage achieves millisecond-level response times with higher precision, better meeting grid frequency control needs. Third, storage optimizes grid operation and improves economic efficiency. By implementing peak-valley arbitrage, storage charges during low-demand periods and discharges during peak hours, reducing system peak-valley differences and enhancing equipment utilization. Additionally, storage can alleviate transmission congestion and defer grid upgrade investments, improving asset utilization efficiency. Finally, storage plays a vital role in enhancing supply reliability. As emergency power sources, storage can rapidly transition to islanded operation during grid faults, providing continuous power to critical loads. Under extreme weather or natural disasters, the black-start capability of storage supports rapid grid restoration.
Obstacles in Planning and Configuration
Despite technological advancements and expanding application scenarios, the large-scale development of battery energy storage systems faces prominent challenges:
- Inadequate cost recovery mechanisms and poor investment economics. High upfront costs, coupled with heavy reliance on policy subsidies or mandatory storage mandates, result in a lack of sustainable market-based profitability. In current electricity markets, storage value cannot be fully captured through energy-only markets, especially in regions with immature ancillary service markets and missing capacity compensation mechanisms. Investment payback periods are long and risk is high.
- Imperfect market access and trading mechanisms. The status of storage as an independent market entity is not universally recognized. Barriers exist in access conditions, metering rules, and settlement methods. Trading products are limited, lacking short-term, high-frequency, and cross-market arbitrage mechanisms suited to storage characteristics.
- Uncoordinated dispatch and operation mechanisms. Grid dispatch tends to treat storage as a local resource rather than a system-level asset, with insufficient cross-regional coordination. Storage operation modes are too rigid, failing to fully exploit flexible regulation potential, and lacking synergistic interaction with renewable generation and demand-side resources.
- Missing technical standards and safety regulation. Technical bottlenecks persist in performance degradation, thermal runaway risks, cycle life, and end-of-life recycling. Unified standards for planning, grid integration testing, operation maintenance, and second-life utilization are absent. A full lifecycle regulatory framework has yet to be established, posing safety and environmental risks.
Market Mechanism and Regulatory Policy Innovation Pathways
Multi-Layered Market Mechanisms Tailored to Storage Characteristics
First, improve the energy market mechanism. Allow storage to participate as an independent market entity. We recommend introducing shorter trading intervals, such as 15-minute or 5-minute settlement periods, to accommodate storage’s fast response. Implement locational marginal pricing to reflect grid congestion, guiding storage deployment to the most needed locations.
Second, enhance the ancillary service market system. Expand service categories to include frequency regulation, reserves, reactive power support, and black start. Adopt pay-for-performance mechanisms that compensate based on response speed, regulation precision, and other performance metrics. This approach can significantly boost storage revenue from ancillary services.
Third, establish a capacity market mechanism. Provide stable capacity revenues through capacity compensation or auctions, ensuring fixed cost recovery. Use reliability requirement models to scientifically determine capacity needs and implement scarcity pricing to fully reflect the capacity value of storage.
Fourth, explore green value realization mechanisms. Link green certificate trading with carbon emission reduction benefits, converting the environmental value of storage-facilitated renewable integration into economic returns. Such mechanisms can improve storage project profitability.
Comprehensive Regulatory Framework
Strengthen top-level design and strategic planning. Integrate storage development into national energy strategy and power development plans, setting medium- and long-term targets and technology roadmaps. We recommend a dedicated new energy storage development plan identifying priority regions and applications to avoid disorderly development and resource waste.
Improve technical standard systems. Accelerate the development of standards covering all stages: planning, design, manufacturing, grid connection, operation, maintenance, and decommissioning. Prioritize safety standards including battery safety, electrical safety, and fire protection. Establish a product certification and quality supervision system to ensure performance and reliability.
Innovate business models and operation mechanisms. Promote shared storage models, through capacity leasing or service purchase, improving utilization rates and reducing user costs. Explore storage aggregation models, such as virtual power plants, to integrate distributed storage resources for market participation. Shared storage can boost utilization and shorten payback periods.
Increase policy support. Provide investment subsidies, tax incentives, and low-interest loans to reduce project costs. Establish pricing mechanisms defining how storage is compensated in various market transactions. Enhance financial innovation, promoting storage asset securitization to broaden financing channels.
Strengthen safety assurance and risk management. Build a storage safety regulatory system covering access, operational monitoring, and emergency management. Implement full lifecycle management, from production, transport, installation, operation to decommissioning. Establish storage insurance mechanisms to disperse and transfer safety risks.
Quantitative Analysis and Modeling
To illustrate the economic and technical optimization of battery energy storage systems, we present several key models and data summaries.
Cost-Benefit Analysis Model
The net present value (NPV) of a storage system can be expressed as:
$$
NPV = \sum_{t=1}^{T} \frac{R_t – C_t}{(1+r)^t} – I_0
$$
where \(R_t\) represents annual revenue from energy arbitrage, ancillary services, capacity payments, and green value; \(C_t\) includes operation, maintenance, and degradation costs; \(I_0\) is the initial investment; \(r\) is the discount rate; and \(T\) is the project lifetime.
Revenue from peak-valley arbitrage can be modeled as:
$$
R_{arb} = \sum_{h} (P_{discharge,h} \cdot \pi_{peak,h} – P_{charge,h} \cdot \pi_{off-peak,h} – \eta_{loss})
$$
where \(P\) denotes power, \(\pi\) denotes electricity price, and \(\eta_{loss}\) accounts for round-trip efficiency losses.
Capacity Optimization Formulation
For a storage system co-located with a renewable plant, the optimal capacity can be derived by minimizing the combined cost of storage and curtailment penalties:
$$
\min_{E_{cap}, P_{cap}} \left[ C_{inv}(E_{cap}, P_{cap}) + C_{O\&M} + \sum_{t} C_{curt}(t) \right]
$$
subject to power balance and state-of-charge constraints:
$$
SOC_{t+1} = SOC_t + \eta_{ch} P_{ch,t} \Delta t – \frac{P_{dis,t} \Delta t}{\eta_{dis}}
$$
$$
0 \leq SOC_t \leq E_{cap}, \quad 0 \leq P_{ch,t} \leq P_{cap}, \quad 0 \leq P_{dis,t} \leq P_{cap}
$$
Performance Comparison of Storage Technologies
The following table summarizes key parameters of common battery energy storage systems used in new power systems:
| Technology | Energy Density (Wh/kg) | Cycle Life (cycles) | Round-trip Efficiency (%) | Response Time | Capital Cost ($/kWh) | Maturity |
|---|---|---|---|---|---|---|
| Lithium-ion (LFP) | 130-160 | 4000-8000 | 85-95 | milliseconds | 150-300 | Commercial |
| Lithium-ion (NMC) | 150-250 | 2000-6000 | 85-95 | milliseconds | 200-400 | Commercial |
| Sodium-ion | 100-150 | 3000-6000 | 80-90 | milliseconds | 100-200 | Emerging |
| Flow batteries (Vanadium) | 15-30 | 10000+ | 65-80 | seconds | 300-600 | Early commercial |
| Lead-carbon | 30-50 | 2000-4000 | 70-85 | seconds | 100-200 | Mature |
Projected Storage Market Growth (China)
Based on current policies and technology trends, we project the following installed capacity for battery energy storage systems:
| Year | New Energy Storage Installed Capacity (GW) | Market Size (Billion USD) | Key Driving Factors |
|---|---|---|---|
| 2023 | 30 | 15 | Mandatory storage policies, renewable expansion |
| 2025 | 50 | 30 | Cost decline, ancillary market maturation |
| 2027 | 80 | 55 | Capacity market, green certificates |
| 2030 | 150 | 100 | Full market liberalization, AI integration |
Revenue Composition of a Typical 100 MW/400 MWh Storage System
Assuming participation in energy, ancillary, and capacity markets, the estimated annual revenue breakdown is:
| Revenue Source | Annual Revenue (Million USD) | Percentage (%) |
|---|---|---|
| Energy arbitrage | 6.5 | 35 |
| Frequency regulation | 5.0 | 27 |
| Capacity payment | 4.0 | 22 |
| Black start & other | 1.5 | 8 |
| Green value (carbon/certificate) | 1.5 | 8 |
| Total | 18.5 | 100 |
Integration of Artificial Intelligence with Battery Energy Storage Systems
The theme of the 2025 Energy Green Development Conference, “Energy + AI Leading Green Development,” underscores the transformative potential of artificial intelligence in optimizing the planning, operation, and configuration of battery energy storage systems. AI techniques such as machine learning, deep reinforcement learning, and digital twins can significantly enhance the performance of storage assets.
For instance, AI-driven forecasting models can improve the prediction of renewable generation and load patterns, enabling more accurate sizing and scheduling of storage. Reinforcement learning algorithms can optimize real-time charge-discharge decisions in multiple markets, maximizing revenue while minimizing degradation. Digital twin platforms allow for virtual testing of different control strategies, reducing risks and accelerating deployment.
One concrete example is the use of neural networks to predict battery state-of-health and remaining useful life, enabling predictive maintenance and optimal replacement timing. Another is the application of multi-agent reinforcement learning for coordinated dispatch of distributed storage fleets, improving system-wide efficiency. These AI-driven approaches can reduce levelized cost of storage by 10–20% and increase revenue by 15–30%.
Policy Recommendations for Accelerating Deployment
Based on our analysis, we propose the following actionable policy measures to unlock the full potential of battery energy storage systems in the new power system:
- Establish a comprehensive market framework that recognizes storage as a distinct asset class with multiple value streams. Implement real-time pricing, pay-for-performance ancillary services, and capacity auctions.
- Standardize technical requirements across the full lifecycle, including safety, performance testing, and recycling. Develop mandatory certification protocols for grid connection.
- Promote shared and aggregated storage models to lower entry barriers for small-scale users and improve utilization rates. Provide regulatory sandboxes for innovative business models.
- Launch national demonstration projects focusing on AI-optimized integrated storage solutions, with streamlined permitting and financial incentives such as accelerated depreciation.
- Foster international collaboration on technology standards, data sharing for AI training, and best practices in market design. Joint research efforts can accelerate cost reduction and performance improvement.
Conclusion
Battery energy storage systems are the cornerstone of a decarbonized, resilient, and efficient new power system. Their successful large-scale deployment hinges on the dual drivers of innovative market mechanisms and robust regulatory frameworks. As electricity market reforms deepen, storage technology advances, and costs continue to decline, battery energy storage systems will play an increasingly dominant role. By 2030, China’s new energy storage installed capacity is projected to reach 150 GW, forming a trillion-yuan market. Through the synergy of energy and artificial intelligence, we can create a truly sustainable energy future, achieving the “dual carbon” goals and leading global green development.