Multidimensional Evaluation and Research of Energy Storage Technologies for Nuclear Power Frequency Regulation Scenarios

Abstract

Under the drive of the “dual carbon” goals, the insufficient frequency regulation capability of nuclear power as a baseload source and the dynamic demand of integrating a high proportion of renewable energy into the grid have increasingly highlighted conflicts. The inherent minute-level regulation inertia of nuclear power units struggles to cope with second-level frequency fluctuations in the grid, leading to an increased risk of system instability. There is an urgent need for energy storage technologies to fill the millisecond-level power support gap for nuclear power frequency regulation. This paper, focusing on nuclear power frequency regulation scenarios, constructs a “Technology–Economy–Policy” multidimensional energy storage evaluation system for the first time. Through a systematic analysis of 11 key indicators, such as response time and safety, the paper selects energy storage technologies suitable for nuclear power frequency regulation scenarios and proposes a hybrid energy storage optimization strategy. The research provides a systematic evaluation framework and empirical support for the selection of energy storage for nuclear power frequency regulation, with significant practical value in improving grid dynamic stability and promoting the construction of new power systems under the “dual carbon” goals.

1. Introduction

With the global energy structure accelerating towards a low-carbon transition, nuclear energy, as an efficient and stable baseload power source, occupies a core position in ensuring electricity supply security and achieving the “dual carbon” goals. According to the International Atomic Energy Agency’s forecast, with the continuous growth of electricity demand in Asia and countries fulfilling their commitments under the 2015 Paris Agreement, global nuclear power installed capacity is expected to increase by 42% by 2030, and the world is working towards large-scale deep decarbonization [1,2]. However, due to the technical characteristics of nuclear power plants, they typically operate at a constant power output and find it difficult to quickly respond to grid frequency fluctuations. The large-scale integration of renewable energy has further exacerbated the randomness and complexity of grid frequency [3]. The installed capacity of nuclear power units in China is steadily increasing, and to alleviate the burden of frequency regulation on the grid, there is an increasing demand for large-capacity nuclear power plants to participate in primary frequency regulation [4]. As a baseload power source, nuclear power has strong inertia in its regulation and slow response speed (minute-level), which makes it difficult to meet the second-level or even millisecond-level frequency regulation requirements of the grid [5]. For example, when the grid frequency fluctuates instantaneously, nuclear power cannot quickly adjust its output, which can easily lead to frequency deviations or system instability. Energy storage technology, with its fast response (such as flywheel energy storage with response time as low as 10 ms) and high regulation accuracy (such as lithium-ion batteries with efficiency greater than 95%), has become an ideal supplement for nuclear power frequency regulation [6].

Policy support has provided important impetus for integrating energy storage with nuclear power. In 2025, the National Development and Reform Commission issued the “Notice on Deepening the Market Reform of the New Energy Grid-Connected Electricity Price,” which explicitly requires new energy projects to enhance their regulation capabilities through energy storage to mitigate the risk of revenue fluctuations due to expanded spot market price differences [7]. Additionally, due to the mandatory requirement in the new “Power System Security and Stability Guidelines” for new energy plants to equip flexible regulation power sources to ensure grid stability. The flow battery, with its rapid response capability, long service life, and relatively safer characteristics for nuclear power plants, has become the preferred solution for frequency regulation in power systems that include nuclear power [8]. In terms of market mechanisms, the shared energy storage model (such as the Qinghai demonstration project) has significantly reduced marginal costs and shortened investment recovery periods by enabling coordinated scheduling among multiple entities, providing a replicable business model for large-scale nuclear power frequency regulation.

Due to the characteristics of nuclear power units, they pose significant challenges in terms of safety, technology, and economics [9]. From a safety perspective, the physical characteristics of nuclear reactors limit the amplitude and rate of their frequency regulation, especially during base-load operation, where frequency regulation actions may cause the unit to exceed its power capacity, violating operational technical standards and increasing safety risks [10]. Technically, nuclear power units face slow response times, strict safety constraints, complex fuel cycles, thermal stress, and regulatory requirements, all of which weaken their flexibility in regulation. From an economic perspective, frequency regulation actions reduce fuel efficiency, increase the generation of radioactive waste, and raise operational costs, thereby affecting the economic benefits and environmental friendliness of nuclear power plants. Therefore, these operational characteristics make nuclear power units subject to multiple constraints when participating in grid frequency regulation, requiring external flexible regulation power sources such as flow batteries to maintain the overall stability and balance of the power system.

Given the characteristics of nuclear power units, relying solely on their own capabilities to quickly respond to frequency regulation demands is difficult. Therefore, adding energy storage is the primary choice to enhance the frequency regulation capabilities of nuclear power units. This also places higher demands on the safety and reliability of energy storage technologies. The operating environment of nuclear power units is harsh, so the introduction of energy storage systems must not only meet the frequency regulation needs of the power system but also ensure that they do not affect the normal operation of the nuclear power units. Compared to conventional power systems, nuclear power units have low tolerance for system instability or equipment failure. If an energy storage system malfunctions or its performance is unstable, it may cause power system fluctuations, which could affect the safe operation of nuclear power units and even lead to serious accidents. Therefore, nuclear power energy storage systems not only need to focus on the power output capacity of the energy storage system but also need to give special consideration to its reliability under extreme conditions. The energy storage system should have rapid response capabilities to quickly provide power support when frequency fluctuations occur in the grid, ensuring the stable operation of the nuclear power unit. Additionally, the design of the energy storage system should include a high level of fault tolerance to reduce the risk of system failure under extreme conditions and ensure the safe and stable operation of the power system. Furthermore, the operation and maintenance of the energy storage system must align with the lifecycle management of nuclear power units to ensure that its safety and economic viability are not compromised over the long term [11].

In recent years, with the increasingly prominent role of nuclear power as a low-carbon base load power source in frequency stability and regulation of power systems, research on related technologies and theoretical foundations has gradually deepened. There have been literature systematically exploring various technical paths to improve the frequency regulation performance of nuclear power units. In reference [1], Bindra and Revankar proposed a hybrid integration mode of nuclear energy and energy storage systems, pointing out that energy storage technology can effectively compensate for the inherent limitations of slow response speed in nuclear power and enhance the frequency regulation capability of the power grid. Reference [2] conducted research on hydrogen energy storage optimization based on carbon nanomaterials from the perspective of materials science, providing interdisciplinary theoretical support for the design and application of distributed energy storage systems. The above research has laid a solid theoretical and technical foundation for the selection and integration of nuclear power frequency regulation energy storage technology. In addition, reference [12] systematically reviewed the performance characteristics and application potential of various energy storage technologies such as electrochemical energy storage, mechanical energy storage, and thermal energy storage, emphasizing that electrochemical energy storage, with its efficient and fast response characteristics, is suitable for fast frequency response and peak shaving needs. Reference [13] studied the integration scheme of nuclear power units and flywheel energy storage. The simulation results showed that flywheel energy storage can effectively compensate for the frequency response delay of nuclear power units, achieve second level frequency regulation, and improve the stability of the power grid. Reference [4] systematically summarizes the application advantages of flow batteries in high safety scenarios of nuclear power, emphasizing the outstanding performance of water-based flow batteries in safety and cycle life, making them the preferred energy storage technology for nuclear power frequency regulation. Reference [14] proposes a multi index comprehensive evaluation model that combines Analytic Hierarchy Process (AHP) and Grey Relational Analysis (GRA) to effectively solve the problems of weight assignment and uncertainty, providing reference for the construction of a multidimensional evaluation system in this study. Reference [15] discusses the dynamic frequency regulation strategy of energy storage based on model predictive control (MPC), highlighting the necessity of real-time response and fault-tolerant mechanism for the safe and stable operation of nuclear power frequency regulation system.

In terms of economy and returns, reference [16] constructed a profit model that includes peak valley arbitrage, government subsidies, multi-channel transaction deductions, and demand response, verifying the good investment return rate of energy storage projects; Reference [17] uses Analytic Hierarchy Process to establish a comprehensive evaluation index system for energy storage, conducts technical and economic evaluations of multiple types of energy storage technologies, and verifies the effectiveness of the method through simulation. Although significant progress has been made in these studies, they mostly focus on single technologies or user side applications, lacking a multidimensional comprehensive evaluation system for nuclear power frequency regulation scenarios, which cannot fully reflect the adaptability and overall value of energy storage technology. In addition, there is insufficient analysis of the safety and reliability of energy storage and nuclear power unit collaboration, and the extreme working conditions and fault tolerance capabilities in complex frequency regulation environments have not been fully considered. In addition, economic analysis generally ignores the full lifecycle cost and market dynamics impact, and the evaluation model parameter assignment relies heavily on expert experience, lacking systematic sensitivity and uncertainty analysis, which affects the scientificity and robustness of the results.

In addition, reference [18] provides a detailed analysis of the collaborative frequency regulation effect of energy storage in high proportion new energy grids, pointing out that energy storage can achieve short-term power output while optimizing energy configuration, supporting stable regulation of nuclear power units. Reference [19] compared the performance of multiple energy storage technologies in nuclear power frequency regulation, proposed a hybrid energy storage scheduling model, and optimized the energy storage combination and nuclear power unit frequency regulation capability. Reference [20] constructs a safety design and risk assessment framework for nuclear energy storage, meeting strict safety regulations. Reference [21] explores the contribution of energy storage to enhancing the dynamic inertia of nuclear power units based on the theory of power system frequency control. Reference [22] summarizes the advantages of hybrid energy storage systems in frequency regulation and power quality improvement, with a focus on the combination optimization scheme of flywheel and flow battery. Reference [23] analyzed the economic challenges and cost per kilowatt hour of nuclear power frequency regulation energy storage, providing guidance for investment and operational strategy formulation. Reference [24] discusses the driving forces behind the promotion of energy storage technology in nuclear power systems from the perspectives of policy and market mechanisms. Reference [25] verifies the complementary performance of flywheel and lithium battery hybrid energy storage in second level frequency response and proposes an optimized control scheme. Reference [26] emphasizes the lifecycle cost and environmental impact of energy storage technology, advocating that nuclear energy storage applications should take into account sustainable development. The comparison of relevant literature is shown in

Table 1. Comparison of Relevant Literature.

Reference Number	Research Technology Path	Analysis of Energy Storage Technology Characteristics	Integration of Energy Storage and Nuclear Power	Safety Analysis	Economic and Benefit Analysis	Multi-Dimensional Comprehensive Evaluation System	Extreme Working Conditions	Multi Energy Storage Hybrid Scheduling	Life Cycle Cost and Environmental Impact	Policy And Market Mechanism Analysis
[1]	√	×	√	×	×	×	×	×	×	×
[2]	×	√	×	×	×	×	×	×	×	×
[4]	×	√	×	√	×	×	×	×	×	×
[12]	√	√	×	×	×	×	×	×	×	×
[13]	√	√	√	×	×	√	×	×	×	×
[14]	×	×	×	×	×	√	×	×	×	×
[15]	×	×	√	√	×	×	√	×	×	×
[16]	×	×	×	×	√	×	×	×	×	×
[17]	×	√	×	×	√	√	×	×	×	×
[18]	×	×	√	×	×	×	×	√	×	×
[19]	×	√	×	×	×	×	×	√	×	×
[20]	×	×	×	√	×	×	×	×	×	×
[21]	√	×	×	×	×	×	×	×	×	×
[22]	×	√	×	×	×	×	×	√	×	×
[23]	×	×	×	×	√	×	×	×	×	×
[24]	×	×	×	×	×	×	×	×	×	√
[25]	×	√	√	×	×	×	×	√	×	×
[26]	×	×	×	×	×	×	×	×	√	×

Note: √ indicates that the issue has been studied in this literature, while × indicates that it has not been studied.

.

In conclusion, the existing research covers multiple aspects of nuclear power frequency regulation energy storage, including technology integration, safety control, economic evaluation, and dispatching optimization, highlighting the necessity and advancement of hybrid energy storage and the construction of a multi-dimensional evaluation system. The focus of this study is on the primary and secondary frequency regulation of nuclear power systems, excluding inertial response, which complies with the technical requirements of the “Guidelines for Safety and Stability of Power Systems”. Specifically, the first frequency adjustment targets short-term frequency deviations (0–10 s after interference) to prevent frequency collapse, while the second frequency adjustment targets medium-term deviations (10–300 s) to restore the frequency to the rated value. Inertial response is not within the scope of research because nuclear power units have already provided sufficient moment of inertia (≥5 s) through their steam turbine generator sets. The energy storage technologies under study (such as flywheels and flow batteries) are more suitable for compensating for the insufficient primary/secondary regulation capacity of nuclear power. In response to the above deficiencies, the innovation points of this paper are reflected in the following aspects: First, a comprehensive evaluation system covering multiple dimensions such as technicality (response time, safety, cycle life) and economy (cost per kilowatt-hour, investment payback period) has been constructed to systematically assess the adaptability of energy storage technology in the nuclear power frequency regulation scenario and achieve a scientific balance of multiple factors. Second, the focus is on the safety of energy storage technology and its stability under extreme working conditions. In combination with the operational characteristics of nuclear power units, design requirements for the fault-tolerant capacity of energy storage systems are proposed to provide technical guidance for the safe and coordinated operation of nuclear power and energy storage systems. Thirdly, sensitivity analysis is introduced to deeply explore the impact of core evaluation indicators and weight configuration on the comprehensive evaluation results, thereby enhancing the robustness of the model and its reference value for decision-making. Through the above innovations, this paper provides theoretical support and practical guidance for nuclear power units to safely and efficiently participate in grid frequency regulation, promoting the deep integration and coordinated development of nuclear power and energy storage.

2. Introduction to Energy Storage Technologies

2.1. Electrochemical Energy Storage

Electrochemical energy storage, as a flexible regulation resource, can provide auxiliary services such as peak shaving and frequency regulation for the power grid. It enhances the regulation capacity of the grid under normal operating conditions and serves as an emergency power source to ensure the safe operation of the power system during faults and abnormal conditions. Additionally, it can operate in conjunction with fossil fuels and non-fossil energy sources to improve the utilization rate of thermal and nuclear power units, promote the integration and efficient utilization of renewable energy, or be used for demand-side response on the user side, helping to flatten demand curves and achieve supply–demand balance [27].

Battery energy storage is an electrochemical energy storage device that operates through the redox reactions occurring at the positive and negative electrodes during charging and discharging. Depending on the internal materials and the electrochemical reaction mechanisms, battery energy storage can be divided into several types, such as lead-acid batteries, lithium-ion batteries, sodium-sulfur batteries, and flow batteries. The core structure inside different types of batteries is fundamentally the same, consisting of a positive electrode, a negative electrode, a separator, and an electrolyte. During charging, oxidation reactions occur at the active material on the positive electrode, causing it to lose electrons. At the same time, cations move towards the negative electrode through the electrolyte under the influence of the electric field. The lost electrons flow through the external circuit towards the negative electrode, where they combine with the active material on the negative electrode to undergo a reduction reaction. The discharging process is the reverse of the charging process [28]. Different types of battery energy storage have distinct characteristics, development levels, and applications, and the following will specifically describe their characteristics in frequency regulation scenarios.

Lithium-ion battery storage technology primarily utilizes the high energy density and high power density characteristics of lithium-ion batteries for energy storage and release. The fast response speed and excellent frequency regulation performance of lithium-ion batteries allow them to play a vital role in high-frequency, short-duration frequency regulation scenarios, significantly improving the efficiency and reliability of the power system [22]. Lithium iron phosphate (LiFePO4) batteries have significant advantages in solving nuclear power frequency regulation issues. Firstly, they have extremely high safety and stability, with a low risk of thermal runaway, making them ideal for high-safety environments such as nuclear power plants. Secondly, lithium iron phosphate batteries have a long cycle life, with more than 5000 charge–discharge cycles, and relatively low overall lifecycle costs, offering strong economic benefits [29]. Additionally, they have a fast response speed, capable of switching between charge and discharge in under 200 ms, which helps to effectively smooth out grid frequency fluctuations and meet the timeliness requirements of frequency regulation.

Flow batteries have an extremely long cycle life, with more than 10,000 charge–discharge cycles, and low overall lifecycle costs, making them suitable for long-duration frequency regulation needs. The safety of flow batteries is high, as the electrolyte is a water-based solution, which is non-flammable and non-explosive, making them highly suitable for high-safety environments such as nuclear power plants [30]. Additionally, the power and capacity of flow batteries can be independently designed, offering high flexibility to adjust the system size according to actual needs.

Sodium-sulfur (NaS) batteries have high energy density and large storage capacity, making them suitable for long-duration frequency regulation needs. Additionally, NaS batteries have high cycle efficiency and low energy loss, and they can operate stably in high-temperature environments without requiring additional temperature control systems. The response speed of NaS batteries is relatively fast, capable of switching between charge and discharge in 500 ms, making them suitable for specific frequency regulation needs [31]. By deploying NaS battery storage systems, nuclear power can quickly respond to grid frequency fluctuations in high-temperature environments, providing continuous and stable power support while improving the efficiency and reliability of the power system.

Lead-carbon battery storage technology offers low cost, high safety, and a mature technological system, providing economic advantages for large-scale applications. On one hand, the low cost of lead-carbon batteries makes them widely applicable in scenarios with limited budgets, reducing the initial investment and operational costs of energy storage systems. On the other hand, their high safety and mature recycling system allow for resource reuse and reduce environmental impact.

2.2. Flywheel Energy Storage

Flywheel Energy Storage System (FESS) is an energy storage system based on electromechanical energy conversion. It stores energy in the form of kinetic energy in a high-speed rotating flywheel, using physical methods to achieve energy storage. It has the advantages of high power density, wide application range, strong adaptability, high efficiency, long lifespan, and no pollution. However, its main disadvantages are low energy density and relatively high self-discharge rate.

The flywheel energy storage system consists of a bearing support system, high-speed flywheel, motor/generator, power converter, electronic control system, vacuum pump, emergency backup bearings, and other additional equipment. During periods of low load, the flywheel energy storage system absorbs electrical energy from the grid to accelerate the flywheel to high speeds, storing energy in the form of kinetic energy. During peak load periods, the high-speed rotating flywheel drives a generator to generate electricity, which is then converted into appropriate voltage and current through a power converter. Flywheel energy storage systems typically operate in high-vacuum environments to reduce losses caused by wind resistance. Flywheel storage systems require minimal maintenance and have a long lifespan (they can complete 20 years of deep charge–discharge cycles), and they have no adverse impact on the environment [6].

In the energy storage technologies suitable for nuclear power frequency regulation, the flywheel energy storage system, with its rapid response, precise regulation, and high power density, is an ideal choice for short-duration high-frequency frequency regulation. As a base-load power source, nuclear power typically operates at a constant power level and finds it difficult to quickly respond to grid frequency fluctuations. Flywheel energy storage can provide instantaneous power support through rapid charge–discharge cycles the moment the grid frequency changes, effectively smoothing out frequency fluctuations and compensating for the lack of frequency regulation capacity of nuclear power. Moreover, the cycle life of flywheel energy storage can exceed 100,000 cycles, making it suitable for high-frequency regulation scenarios, and it requires minimal maintenance, which results in lower overall lifecycle costs. By working in synergy with nuclear power, flywheel energy storage not only enhances the frequency stability of the power system but also extends the operational life of nuclear power units, reducing the impact of frequency regulation on the safe operation of nuclear power.

2.3. Supercapacitors

Supercapacitors have the characteristics of high power density, high energy density, long cycle life, and a large number of charge–discharge cycles. They can output short bursts of high power, making them suitable for applications that require frequent charge and discharge [32]. Supercapacitors offer the following advantages in nuclear power unit frequency regulation: Firstly, their power density is extremely high, allowing them to respond to frequency regulation commands within milliseconds, enabling rapid charge and discharge and quickly smoothing out grid frequency fluctuations. Secondly, supercapacitors have a long cycle life, capable of more than 500,000 deep charge–discharge cycles, and do not require replacement throughout their lifecycle, which reduces maintenance costs and offers strong economic benefits. Additionally, supercapacitors are highly safe; they do not catch fire or explode, are capable of operating in a wide temperature range from −40 °C to 70 °C, and maintain stable performance, making them very suitable for high-safety, high-reliability applications like frequency regulation in nuclear power plants.

3. Technical and Economic Evaluation of Various Energy Storage Technologies

In the application of nuclear power frequency regulation, the selection of energy storage technology is fundamental to ensuring the stability, reliability, and efficiency of the frequency regulation process. Nuclear power frequency regulation is aimed at addressing the impact of power load fluctuations on the grid and maintaining the stability of grid frequency. The choice of the right energy storage technology not only impacts frequency regulation efficiency but also directly affects the system’s safety, economic viability, and long-term sustainability. Therefore, conducting an in-depth technical evaluation of various energy storage technologies is a crucial step in ensuring that the energy storage solutions can effectively meet the frequency regulation needs of nuclear power plants [21,22].

3.1. Technical Analysis of Energy Storage Technologies

The necessity of technical evaluation is reflected in several aspects. On one hand, nuclear power frequency regulation imposes strict requirements on the response speed and regulation accuracy of energy storage technologies. Grid frequency changes are often instantaneous and complex, requiring the energy storage system to respond rapidly within milliseconds and precisely regulate the output. Therefore, the response speed and regulation accuracy of the energy storage system will directly affect the effectiveness of nuclear power frequency regulation and the stability of the grid. On the other hand, the lifecycle of the energy storage system is also an important aspect of technical evaluation. The operational cycle of nuclear power plants is typically long, often spanning several decades, and the lifespan of the energy storage system directly impacts its long-term stability and economic viability. If the energy storage system has a short lifespan and requires frequent replacement or maintenance, it will increase the operational costs of the nuclear power frequency regulation system. Additionally, the maturity of the storage technology is another important dimension for evaluation. Different energy storage technologies on the market have varying levels of development; some technologies may be theoretically feasible but may present numerous uncertainties and technical challenges in practical applications. Therefore, assessing the maturity of energy storage technologies, particularly their successful cases and practical experiences in similar application scenarios, is crucial for determining whether they are suitable for nuclear power frequency regulation. Safety is also the most critical issue in nuclear power frequency regulation. Therefore, the energy storage technology integrated with nuclear power must have high safety standards to effectively prevent accidents such as fires, explosions, or leaks. By evaluating the safety performance of various energy storage technologies, it is essential to ensure that the chosen energy storage solution will not pose any safety hazards to the nuclear power plant or the grid during long-term operation.

In conclusion, a comprehensive technical evaluation of energy storage technologies can ensure that the selected technology meets the frequency regulation needs of nuclear power, enhances the frequency regulation capability and stability of the grid, reduces risks, and improves economic efficiency. This evaluation process not only helps in scientifically selecting the appropriate energy storage technology but also provides important decision-making support for subsequent implementation plan design and technical route planning. A summary of the characteristics of various energy storage technologies is shown in

Table 2. Characteristics of Various Energy Storage Technologies.

Energy Storage Classification	Response/Duration	Power Range	Usage Duration	Cycle Life	Efficiency/%	Safety	Maturity
Flywheel [33,34,35,36]	10 ms/15 s–15 min	0–1 MW	15–20 years	>10,000 cycles	80–95	High	Relatively Mature
Lithium Iron Phosphate Battery [37,38]	<1 s/1–4 h	1 kW–10 MW	8–15 years	3000–6000 cycles	95–98	Relatively High	Mature
Lithium-ion Battery [33,34,35,39,40]	min/6–20 h	1–10 MW	5–12 years	1000–10,000 cycles	90–98	Moderate	Mature
Lead-Acid Battery [33,34,35,36,40]	s/1 min–8 h	0–20 MW	3–8 years	500–1000 cycles	63–80	Relatively High	Mature
Sodium-Sulfur Battery [33,34,35,36]	s/1 min–1 h	1 kW–10 MW	10–15 years	4500 cycles	75–90	Moderate	Relatively Mature
Lead-Carbon Battery [41,42,43,44]	s/1–5 h	1 kW–10 MW	5–10 years	2000–3000 cycles	85–90	Low	Relatively Mature
Flow Battery [34,40,42,43,44,45]	s/4–24 h	1–10 MW	15–25 years	12,000 cycles	65–85	High	Relatively Mature
Supercapacitor [33,34,35,36]	s/1 s–1 min	0–1 MW	5–15 years	100,000 cycles	70–80	High	Mature

[33,34,35,36,37].

3.2. Economic Evaluation

It is not sufficient to consider only the technical feasibility of energy storage for frequency regulation; economic evaluation is an essential step to ensure that the chosen energy storage technology not only meets technical requirements but is also economically feasible and sustainable. Nuclear power frequency regulation typically involves high-frequency charge and discharge operations and long-term stable operation, which places higher demands on the cost-effectiveness of the energy storage system. Therefore, when selecting an energy storage solution, it is necessary to conduct a comprehensive evaluation from an economic perspective, in addition to the technical evaluation. Economic evaluation helps assess the economic feasibility of different energy storage technologies in nuclear power frequency regulation and provides scientific support for investment decisions, operational strategies, and economic returns.

The necessity of economic evaluation is reflected in several aspects. First, unit investment cost is the basis for evaluating the economic viability of energy storage technology. The construction of energy storage systems often requires significant initial investment, which is crucial to the overall economic benefits of the nuclear power plant. By comparing the unit investment costs of different energy storage technologies, clear guidance can be provided for financial planning and resource allocation in nuclear power frequency regulation projects, ensuring the project can proceed smoothly within the budget. Second, operation and maintenance (O&M) costs are an important factor affecting the long-term economic viability of the energy storage system. The O&M costs directly influence the long-term sustainability of nuclear power frequency regulation projects. If the O&M costs are too high, especially maintenance, management, and replacement of spare parts, it could significantly increase overall operational costs and reduce the economic return of the project. The cost per kWh (electricity generation cost) is an important indicator of the economic benefits of energy storage technology. Since nuclear power frequency regulation systems require frequent power adjustments, the cost per kWh of the energy storage system directly impacts the price of frequency regulation services and its market competitiveness. Analyzing the cost per kWh of energy storage technology can help decision-makers assess the long-term economic viability of different technologies and choose the most cost-effective solution.

Furthermore, power price is also a key factor in evaluating the economic viability of energy storage technology. Power price reflects the ability of the energy storage system to provide high power output over a short period, directly influencing the market value and profitability potential of the energy storage system. Therefore, a thorough analysis of this factor is necessary during economic evaluation to ensure that investments in energy storage systems will generate reasonable returns.

In conclusion, economic evaluation is crucial for the feasibility of nuclear power frequency regulation projects. By analyzing factors such as unit investment cost, O&M costs, cost per kWh, and power price, it is possible to provide a clear economic basis for the selection of energy storage technologies, ensuring that the project can achieve high economic benefits while meeting technical requirements. This approach reduces long-term operational risks and enhances overall economic sustainability. A summary of the economic characteristics of various energy storage technologies is shown in

Table 3. Economic Characteristics of Various Energy Storage Technologies.

Energy Storage Classification	Unit Investment Cost ($/MW)	Operation and Maintenance Cost ($/year/MW)	Cost per KW·h ($/KW·h)	Power Cost ($/KW)
Flywheel [36]	171,429–285,714	28,571–71,429	0.11–0.21	36–50
Lithium Iron Phosphate Battery [46]	400,000–628,571	21,429–42,857	0.04–0.09	43–93
Lithium-ion Battery [36,47,48,49]	42,857–60,000	28,571–57,143	0.06–0.11	171–571
Lead-Acid Battery [47]	42,857–71,429	42,857–85,714	0.09–0.17	43–86
Sodium–Sulfur Battery [36,47]	42,857–71,429	114,286–214,286	0.10–0.14	143–429
Lead–Carbon Battery [41,48,49,50,51]	100,000–142,857	35,714–71,429	0.06–0.10	800–1000
Flow Battery [40,45,52]	42,857–142,857	71,429–142,857	0.07–0.13	114–143
Supercapacitor [36,53]	714,286–1,142,857	285,714–571,429	0.71–1.43	86–214

[35,36,37,38,39].

The scoring thresholds for technical and economic indicators in Table 2 and Table 3 are quantified using the method of discrete interval division, which integrates relevant industry standards, authoritative literature, and the practical experience of consulting experts in this research institute at home and abroad. The determination of the threshold is first based on systematic research on historical data related to energy storage technology and typical project market application cases, combined with multiple rounds of evaluation and consensus by invited expert teams, to ensure that the scoring criteria not only meet the actual technical level, but also have strong universality and scientificity. The threshold range of economic cost indicators refers to the open bidding prices and operation and maintenance data of the energy storage industry in recent years; The technology maturity and security indicators draw on the internationally recognized technology maturity level system and security evaluation standards. This scoring system provides a reliable and easy to operate foundation for subsequent multi factor comprehensive evaluation by structuring multidimensional complex indicators into standardized scores.

In addition, to enhance the robustness of the evaluation results, this article explored the reasonable range of the scoring threshold in sensitivity analysis, verified the influence of threshold setting on comprehensive scoring and technical ranking, and further improved the scientific and practical value of the model.

4. Energy Storage Rating Criteria

In terms of the frequency regulation requirements needed for nuclear power, the dependency on energy storage power capacity has reached a level of tens of megawatts. For short-term output, the regulation time scale must be in milliseconds. Considering economic efficiency, the optimal system should have an efficiency of over 80%. Additionally, based on the different needs of system frequency regulation, there are specific requirements for energy storage in terms of lifespan, power rating, and other aspects. A summary of the rating criteria for different types of energy storage technologies is shown in

Table 4. Scoring Criteria for Indicators.

Indicator Score	Poor (1 Point)	Medium (2 Points)	Good (3 Points)	Excellent (4 Points)
Maturity	Concept Stage	In Development	Relatively Mature	Mature
Safety	Low	Medium	Relatively High	High
Response Time	h	min	10 s	s
Maximum Duration	≤15 min	15 min~1 h	1~2 h	≥2 h
Cycle Life (per 1000 cycles)	≤3	3–5	5–10	≥10
Conversion Efficiency (%)	≤80	80–84	84–90	90–100
Power Level (MW)	≤0.5	0.5–1	1–10	≥10
Usage Duration (years)	≤10	10–15	15–20	≥20
Investment Cost ($/MW)	>500,000	285,714–500,000	142,857–285,714	<142,857
Operation and Maintenance Cost/($/year)	>28,571	14,286–28,571	7143–14,286	<7143
Cost per KW·h ($/KW·h)	>0.21	0.14–0.21	0.07–0.14	<0.07
Power Cost ($/kW)	≥429	143–429	71–143	≤71

Note: To avoid ambiguity, two key economic indicators are defined in this table: ① Cost per kilowatt-hour: It refers to the average cost of generating 1 kilowatt-hour of electricity throughout the entire life cycle of the energy storage system. The calculation method is the total life cycle cost divided by the total power generation; unit: $/kWh. For instance, the “0.07–0.13 US dollars per kilowatt-hour” of the flow battery in the table reflects its long-term average energy supply cost. ② Electricity cost: It refers to the cost corresponding to the power capacity of the energy storage system, that is, the cost required to provide 1 kilowatt of power output (unit: CNY/kW), mainly including the investment cost of power-related components. The “$36–$50 per megawatt” of the flywheel in the table represents the electricity cost converted to megawatt-level capacity. The distinction between these two indicators ensures the accurate assessment of “energy supply cost” (cost per kilowatt-hour) and “power supply cost” (electricity cost) under different frequency regulation scenarios (primary regulation mainly based on electricity demand and secondary regulation mainly based on energy demand).

.

At the same time, due to the varying focus of different indicators, their impact on the energy storage system and the specific system requirements differs in terms of key importance. Therefore, when conducting a comprehensive evaluation, it is necessary to assign weights to each indicator based on its significance, meaning that the weighted average of each indicator should be calculated. According to existing research, since there is no specific external correlation between the different indicators, the active weighting method can be applied for analysis based on expert survey results, with the corresponding weight coefficients for each indicator shown in

Table 5. Weight Coefficients of Each Indicator.

Evaluation Indicators	Technical Indicators (0.85)							Economic Indicators (0.15)
	Maturity	Power Level	Response Time	Duration	Conversion Efficiency	Cycle Life	Safety	Investment Cost	Operation and Maintenance Cost	Cost per kW·h	Power Cost
Weight	15%	15%	20%	3%	2%	10%	20%	5%	3%	5%	2%

.

1.

High-weight Indicators (Weight ≥ 15%)—Nuclear Power Frequency Regulation Requires Absolute Reliability and Instant Control; Technology Maturity and Safety Are Prerequisites.

(1) Response Time (20%): As a base-load power source, nuclear power has significant inherent inertia (minute-level response) and cannot meet the grid’s second-level or even millisecond-level frequency regulation demands. Energy storage needs to quickly fill the delay in nuclear power regulation (within <100 ms); otherwise, it may lead to frequency violations or grid instability.

(2) Safety (20%): Nuclear power plants have strict safety requirements for equipment. If energy storage experiences an explosion or failure, it may threaten the safety of the nuclear island and even trigger the risk of nuclear leakage.

(3) Maturity (15%): The maturity of the technology directly affects the feasibility and risks of project implementation. Nuclear power, being a highly sensitive environment, tends to choose technologies that have already been widely commercialized.

(4) Power Requirements (15%): Nuclear power frequency regulation needs to address instantaneous power fluctuations caused by renewable energy integration (such as sudden changes in wind power or photovoltaic output). Energy storage must have the capability to provide short-term high power output (2 C~4 C rate).

2.

Medium-weight Indicators (5%~10%)—Power and Lifespan Directly Affect Frequency Regulation Performance and Economic Viability.

(1) Cycle Life (10%): Nuclear power frequency regulation requires frequent charge and discharge of energy storage (e.g., dozens of times a day). If the lifespan is insufficient (e.g., lead-acid batteries have only 500–1000 cycles), the replacement costs will significantly erode the economic viability.

(2) Investment Cost (5%) & Cost per kWh (5%): Technologies with high initial costs but long lifespans (e.g., flow batteries) may be more advantageous over the full lifecycle. However, nuclear power frequency regulation places more focus on short-term power support (not long-duration storage), so high-power density technologies (e.g., lithium batteries) are more economically suitable for the short term.

3.

Low-weight Indicators (≤ 5%)—Can Be Supplemented Through Technological Iteration or Policy Support; Non-core Constraints.

(1) Duration (5%): Nuclear power frequency regulation mainly involves seconds to 10 min fluctuations, so energy storage does not need to discharge for extended periods (e.g., hourly peak shaving). Therefore, there is lower sensitivity to duration.

(2) High Efficiency (e.g., lithium batteries at 93%): High efficiency reduces frequency regulation energy losses, but the benefits of frequency regulation (e.g., 0.7 yuan/kWh) in nuclear power frequency regulation far outweigh the electricity cost loss, so efficiency has a limited impact on economic viability.

(3) Operation and Maintenance (O&M) Costs (3%) & Power Price (2%): O&M costs can be reduced through intelligent management (e.g., AI predictive maintenance). Power price has a lower weight because frequency regulation revenue is already dominated by “mileage compensation,” and after market mechanisms mature, the revenue stability is higher.

5. Comprehensive Energy Storage Scoring Results

Based on the scoring criteria and weight coefficients, the comprehensive scores for each type of energy storage are calculated, as shown in

Table 6. Comprehensive Evaluation of Each Type of Energy Storage.

Energy Storage Classification	Technical Indicators (0.85)							Economic Indicators (0.15)				Comprehensive Score
	Maturity	Power Level	Response Time	Duration	Conversion Efficiency	Cycle Life	Safety	Investment Cost	Operation and Maintenance Cost	Cost per kW·h	Power Cost
Flywheel	3	2	4	1	3	4	4	3	4	3	3	3.32
Lithium Iron Phosphate Battery	4	3	4	3	4	2	3	2	4	3	4	3.27
Lithium-ion Battery	4	3	2	4	4	1	2	4	4	3	3	2.68
Lead-Acid Battery	4	4	4	2	1	1	3	4	3	3	4	3.3
Sodium-Sulfur Battery	3	3	4	1	1	2	2	4	2	3	3	2.82
Lead-Carbon Battery	3	3	4	3	3	1	1	4	4	3	3	2.68
Flow Battery	3	3	4	4	1	4	4	4	3	3	3	3.54
Supercapacitor	4	1	4	1	1	4	4	4	4	1	4	3.25

. The radar charts for each type of energy storage are shown in Figure 1, Figure 2, Figure 3, Figure 4, Figure 5, Figure 6, Figure 7, Figure 8 and Figure 9.

The comprehensive scores for each type of energy storage are shown in Figure 9. Based on the analysis of the comprehensive scores, the following types of energy storage are suitable for nuclear power frequency regulation:

Flow Batteries: Flow batteries have the highest comprehensive score (3.54) and perform excellently in various technical indicators. They have significant advantages in terms of duration, conversion efficiency, and cycle life, making them well-suited to large-scale frequency regulation demands. Due to their ability to provide stable, long-duration output and high conversion efficiency, flow batteries are ideal for nuclear power frequency regulation systems that require long-term stable operation, particularly in scenarios where there are significant fluctuations in power demand.

Flywheel Energy Storage: Flywheel energy storage ranks second in the comprehensive score (3.32) and stands out in power requirements, response time, and safety. The advantage of flywheel energy storage is its ability to provide large amounts of power in a short time and its quick response time, allowing rapid adjustment of power output to respond to fast load changes. Although its duration is shorter, its efficient energy storage and release characteristics make it highly suitable for short-term, high-power frequency regulation scenarios, such as those required in nuclear power frequency regulation.

Lead-Acid Batteries: Lead-acid batteries have a comprehensive score of 3.30, showing high maturity and response time. However, their safety is not as good as that of flow batteries and flywheels, and their conversion efficiency is low, making them less recommended for selection.

The flow battery obtained the highest comprehensive score (3.54 points) in this evaluation system. Although its economic indicators such as cost and efficiency are relatively average, this result reflects the comprehensive consideration of multiple factors such as technical performance, safety, and policy support in the multidimensional evaluation framework. Specifically, flow batteries have received high ratings in technical indicators due to their long cycle life, fast response capability, and high safety level. These technological advantages are particularly important in extreme operating conditions such as nuclear power frequency regulation. In addition, compared to other energy storage technologies, flow batteries have good environmental adaptability and lower lifecycle costs, which is compensated to a certain extent in the comprehensive economic evaluation. Although the initial investment cost of flow batteries is higher and the energy conversion efficiency is slightly lower than that of lithium-ion batteries, their longer service life and lower maintenance requirements help reduce long-term operating costs. Technical and safety indicators account for a significant share in weight allocation, enabling flow batteries to demonstrate their core competitiveness in the overall score. Overall, the high comprehensive score of flow batteries accurately reflects their comprehensive value for nuclear power frequency regulation applications, indicating that flow batteries have significant application advantages in the pursuit of system safety, stability, and long-term economy.

Overall, flow batteries are ideal for large-scale frequency regulation scenarios due to their excellent duration and conversion efficiency. Flywheel energy storage is better suited for high-power, short-term responses, providing rapid regulation, while lead-acid batteries offer economic advantages but come with safety risks.

Individually, none of these technologies can meet the large-scale frequency regulation requirements of the system under the integration of renewable energy. On one hand, the rapid development of renewable energy generation increases the pressure on grid frequency regulation, and the dynamic stability and operational characteristics of the grid become more complex, making frequency control more difficult. On the other hand, in the electricity spot market, the grid is increasingly strengthening its assessment of the flexibility of frequency regulation capabilities for renewable energy stations connected to the grid, setting higher guidance standards. Additionally, the proportion of short-term high-power support is increasing. Therefore, a hybrid energy storage approach is commonly used to improve the power quality of the system during the grid integration process. In the case of nuclear power frequency regulation, the hybrid energy storage frequency regulation technology combining flywheels and flow batteries can effectively coordinate the frequency regulation between the flywheel and flow battery energy storage devices. This approach leverages the strengths of flywheel energy storage, such as its large instantaneous power, millisecond-level response, and frequent charge–discharge cycles, and the advantages of flow batteries, such as their large storage capacity and high frequency regulation amplitude. This also helps avoid the loss caused by frequent charge–discharge cycles in flow batteries.

6. Sensitivity Analysis

6.1. Sensitivity Analysis of Comprehensive Evaluation of Energy Storage Technology

In the multi index comprehensive evaluation system, key technical parameters and indicator weights are used as input data, and their accuracy and rationality directly affect the ranking and selection decisions of the final energy storage technology. In fact, parameters such as battery life, investment cost, operation and maintenance expenses, response time, etc., have certain variability due to various factors such as on-site environment, manufacturing process, and market fluctuations; In addition, the subjective weighting process of experts may also have biases. Therefore, conducting sensitivity analysis is of great significance to verify the robustness of the evaluation model and the credibility of the results. Its goal is to quantify the impact of input parameters and weight changes on the comprehensive score, identify key influencing factors, evaluate model robustness, assist decision-makers in identifying parameters with the greatest uncertainty, and focus on optimization.

6.2. Sensitivity Analysis Methods

This study adopts the following two types of sensitivity analysis methods:

1.

Single factor sensitivity analysis

Set a fluctuation range of ± 20% for the four key parameters of cycle life, investment cost, operation and maintenance cost, and response time, adjust them one by one, and keep other variables unchanged. Calculate the comprehensive score of each energy storage technology again. Evaluate the sensitivity of each parameter to the results by comparing the magnitude of changes in ratings.

2.

Weight sensitivity analysis

Based on the previous expert weighting results, adjust the weights of technical indicators to 75%, 85%, and 95%, and correspondingly adjust the economic weights to 25%, 15%, and 5%. Observe the stability of the technical comprehensive score and ranking under each weight combination and evaluate the impact of weight configuration on the model output.

This method aims to quickly and simply reveal the impact path and strength of input uncertainty on output results, providing uncertainty management ideas for the scientific selection of energy storage technologies.

6.2.1. Single Factor Sensitivity Analysis of Key Technical Parameters

This article conducted sensitivity analysis on the following core indicators:

Cycle life (number of charge and discharge cycles): As a representative of technical life, it is directly related to equipment replacement frequency and operating costs;

Investment cost (10,000 yuan/MW): represents the pressure of capital investment and the initial investment scale of the project;

Operation and maintenance cost (10,000 yuan/year/MW): reflects the long-term operational economy and affects the sustained revenue of the project;

Response time (in milliseconds to seconds): determines the energy storage system’s ability to respond to fluctuations in grid frequency.

Each parameter fluctuates up and down by 20% based on the benchmark value, and the comprehensive score corresponding to each energy storage technology is calculated. The data is organized as shown in the following Figure 10.

By perturbing the four key parameters of cycle life, investment cost, response time, and safety by ± 20%, the changes in the comprehensive scores of different technologies were quantified. Taking the flow battery as an example, its benchmark score is 3.54. When the cycle life increases by 20%, the score rises to about 3.65, an increase of about 3.1%; When the lifespan decreases by 20%, it drops to 3.42, a decrease of about 3.4%, indicating that cycle life has a significant impact on its evaluation. Flywheel energy storage is particularly sensitive to response time parameters, with a benchmark score of 3.32. After a 20% increase in response time, the score rises to 3.45, an increase of approximately 3.9%; If it decreases by 20%, it will decrease to 3.18, a decrease of about 4.2%, reflecting the core role of response speed in its overall performance. Lead acid batteries are sensitive to changes in investment costs. When the cost decreases by 20%, the rating rises from the benchmark 3.30 to about 3.34, an increase of about 1.2%; When the cost increases by 20%, it drops to 3.27, a decrease of about 0.9%, reflecting the significant impact of economic parameters on it. Overall, the score is within ±4% affected by parameter perturbations, indicating that the multidimensional evaluation model has good robustness to parameter uncertainty. At the same time, the differences in sensitivity also correspond reasonably to the performance characteristics of various technologies, verifying the scientific and practical nature of the model.

6.2.2. Weight Sensitivity Analysis

This study assigns benchmark weights of 85% and 15% to technical and economic indicators, as shown in

Table 7. Weight Change Coefficient Table.

Scheme Number	Technical Weight	Economic Weight
1	75%	25%
2	85%	15%
3	95%	5%

:

As shown in Figure 11, in the scheme where the weights of technical indicators are set at 75%, 85%, and 95%, and the economic weights are adjusted accordingly, the comprehensive score of energy storage technology shows a significant change. Taking the flow battery as an example, its comprehensive scores are 3.50, 3.54, and 3.58, respectively, and the scores increase by about 2.3% with the increase of technical weight; The corresponding rating of lead-acid batteries decreased from 3.33 to 3.28, a decrease of about 1.5%, highlighting the adverse impact of the decrease in economic weight on them. Flywheel energy storage exhibits the highest weight sensitivity, with a score increase from 3.29 to 3.38, an increase of approximately 2.7%, indicating that the performance advantages of this technology are more reflected when the technical weight increases. More importantly, the rating fluctuations caused by weight changes are all within ±3%, and the ranking distribution is basically stable: flow batteries and flywheel energy storage remain in the top two, while lead-acid batteries, lithium iron phosphate batteries, and supercapacitors rank relatively low, indicating that the evaluation model still has strong result stability and credibility under weight configuration changes.

The above data analysis shows that the comprehensive evaluation system for energy storage technology can not only effectively capture the impact of key parameter changes on overall performance but also reflect the sensitive response of technical and economic weight adjustments to the scoring results. Technical parameters such as response time and cycle life have a decisive impact on the rating of specific technologies, while investment cost is an important influencing factor for economically sensitive technologies. Weight sensitivity analysis emphasizes the necessity of reasonable weight allocation, which directly relates to the full manifestation of technological advantages and the support of economic weights for cost advantage technologies. The overall performance of the model shows good robustness and stable ranking, which meets the requirements of parameter and weight uncertainty in actual engineering selection, and has strong application and promotion value.

7. Conclusions

With the increasing importance of nuclear power as a low-carbon base load power source, its inherent minute level regulation inertia limits its response capability in power system second level and millisecond level frequency regulation, and high-performance energy storage technology is urgently needed to supplement it. This article systematically analyzes the applicability of various energy storage technologies in nuclear power frequency regulation scenarios based on a multidimensional comprehensive evaluation framework of “technology economy policy”. The evaluation results indicate that flow batteries have significant advantages in meeting long-term frequency regulation requirements due to their ultra long cycle life (over 12,000 cycles), good safety performance (using water-based electrolytes), and stable conversion efficiency, making them the preferred solution for nuclear power frequency regulation. Flywheel energy storage, with its extremely low response time (about 10 milliseconds) and high power density, has important value in dealing with instantaneous fluctuations in grid frequency and rapid power support. In contrast, although lead-acid batteries have certain advantages in investment costs, their shortcomings in safety and conversion efficiency limit their widespread application.

A single factor sensitivity analysis of ±20% was conducted on key parameters such as cycle life, investment cost, operation and maintenance cost, and response time. The results showed that the comprehensive scores of various energy storage technologies were controlled within ±4% under the influence of parameter disturbances, further verifying the robustness of the constructed evaluation model. The weight sensitivity analysis reveals that the adjustment of technical indicator weights has a significant impact on the overall rating and ranking, especially for flywheel energy storage, where the rating increase reaches about 2.7% when the technical weight is increased. This reflects the reasonable allocation of technical performance weights in the model. The stability of the overall ranking structure indicates that the model can still maintain the reliability of the evaluation results when facing uncertainty in parameters and weights.

The engineering practice case further confirms the feasibility and effectiveness of the hybrid scheme. Sumitomo Electric and Hokkaido Electric jointly deploy a 60 MWh liquid flow battery system to provide frequency regulation support for the power grid and promote the integration of renewable energy; Infinity, a British company, has deployed 300 kWh flow batteries in England and entered the GB frequency response market; the Shanxi 30 MW flywheel energy storage project has achieved a second level frequency modulation response, with an annual carbon reduction of 60,000 tons. These examples fully demonstrate the practical application value of energy storage technology in nuclear power frequency regulation and safe and stable operation of power systems.

Future work directions include further improving the energy density of flow batteries and the self-discharge performance of flywheel systems, combining shared energy storage modes and electricity spot market mechanisms, optimizing the operation and scheduling strategies of energy storage systems, and promoting the development of nuclear power frequency regulation and energy storage technology towards higher efficiency, safety, and economic sustainability.

The multidimensional comprehensive evaluation system and sensitivity analysis constructed in this article systematically reveal the performance and economic characteristics of nuclear power frequency regulation energy storage technology, verify the robustness of the model to key parameters and weight configuration, and provide scientific theoretical basis and practical guidance for the selection of nuclear power frequency regulation energy storage. It is of great significance for achieving the “dual carbon” goal and ensuring the safe and stable operation of the power grid.