Battery Energy Storage Systems: Energy Market Review, Challenges, and Opportunities in Frequency Control Ancillary Services

Abstract

Battery energy storage systems (BESS) are considered a good energy source to maintain supply and demand, mitigate intermittency, and ensure grid stability. The primary contribution of this paper is to provide a comprehensive overview of global energy markets and a critical analysis of BESS’ participation in frequency control ancillary service (FCAS) markets. This review synthesises the current state of knowledge on the evolution of the energy market and the role of battery energy storage systems in providing grid stability, particularly frequency control services, with a focus on their integration into evolving high-renewable-energy-source (RES) market structures. Specifically, solar PV and wind energy are emerging as the main drivers of RES expansion, accounting for approximately 61% of the global market share. A BESS offers greater flexibility in storage capacity, scalability and rapid response capabilities, making it an effective solution to address emerging security risks of the system. Moreover, a BESS is able to provide active power support through power smoothing when coupled with solar photovoltaic (PV) and wind generation. In this paper, we provide an overview of the current status of energy markets, the contribution of battery storage systems to grid stability and flexibility, as well as the challenges that BESS face in evolving electricity markets.

1. Introduction

Achieving the net-zero emission target by mid-century requires a paradigm shift in the energy sector and radical decarbonisation. The transition towards low-carbon-emission energy systems has accelerated the integration of large-scale renewable energy sources (RESs), such as solar PV and wind, into power systems. The accelerated integration of renewable energy sources (RESs) is displacing traditional power generation by synchronous generators. While RESs are critical to decarbonisation, their inherent intermittency significantly challenges power system reliability and flexibility [1]. In addition, the phasing out of synchronous generation is causing low system strength due to low inertia, further degrading power system stability. Due to the low system inertia, frequency stability is emerging as a more dominant issue in high-RES power grids [2,3]. One solution to mitigate this issue is to ensure that adequate frequency control ancillary services (FCASs) are provided. Battery energy storage systems (BESSs) are considered a good energy source to maintain supply and demand, mitigate intermittency, and ensure grid stability. A BESS can provide fast and precise responses, making it an ideal candidate for FCASs [4]. With the advancement of technologies, a BESS offers a wide range of energy storage solutions, enabling frequency regulation, voltage support, energy arbitrage, peak shaving, and smoothing ancillary services that support the grid with higher penetration of RESs [4,5,6,7].

Despite BESSs’ technical capabilities to provide frequency response in seconds, many electricity markets lack clarity on the role of BESSs in the ancillary markets [7]. The market rules are predominantly designed for thermal generation, with minimum durations, response rates, and ramp rates that are suitable for traditional generating units [8,9,10,11,12]. Furthermore, while BESSs provide FCASs, frequent cycling and high-power discharges accelerate battery degradation [13,14]. Without proper degradation-aware control strategies and compensation, BESS operators may face challenges of accelerated degradation and reach the end of life (EOL) much sooner, resulting in a financial loss. The primary contribution of this paper is to provide a comprehensive overview of global energy markets and a critical analysis of battery energy storage systems (BESS) participation in frequency control ancillary service (FCAS) markets.

Contribution of This Paper

Understanding the rapid changes in the energy market and emerging technical challenges has become imperative. This review synthesises the current state of knowledge on the evolution of the energy market, the role of battery energy storage systems (BESSs) in providing grid stability, particularly frequency control services, with a focus on their integration into evolving high-renewable-energy-source (RES) market structures. While much of the existing literature has focused on the technical aspects of battery storage, this review goes a step further by connecting these technologies to real-world market dynamics and policy developments. It brings together insights from engineering, economics, and regulation to highlight the role of BESSs in grid stability and flexibility, as well as the challenges that BESSs face in evolving electricity markets. The key highlights of this study are as follows:

• Comprehensive Overview of Global Energy Markets: This review presents an in-depth analysis of the evolution and current landscape of global energy markets, with a particular emphasis on the increasing penetration of renewable energy sources (RESs), such as solar photovoltaics (PV) and wind power.
• Focus on Degradation Mechanisms: The study concentrates on degradation from high-cyclic and high-C-rate conditions typical in FCAS operations.
• Critical Analysis of BESSs in frequency control ancillary service (FCAS) Markets: This review critically examines the evolving role of battery energy storage systems (BESSs) in frequency control ancillary service (FCAS) markets. The review further discusses emerging cost-recovery mechanisms, such as pay-for-availability combined with energy throughput limits.
• Identification of Revenue Stacking Opportunities and Long-Term Value Pathways for BESSs: To complement its in-depth discussion of technical and market challenges, this paper also offers a critical and constructive analysis of value-creation opportunities for battery energy storage systems (BESSs) in ancillary service markets.

2. Global Energy Market Overview: Renewables Growth and Energy Market Evolution

Globally, the energy market is undergoing a drastic transformation. The projections show a significant increase in global electricity demand, estimated at 16,885 GW by 2030, up 39% from 2020, and increasing further to 30,227 GW by 2050, equating to a 166% increase from 2020. The higher penetration of RESs, primarily the generation of wind and solar photovoltaics (PV), has also accelerated the modernisation of power systems. In the next few decades, the integration of RESs is expected to grow significantly. A recent study suggests that RESs achieved a major milestone in 2023, when RESs exceeded 30% of global electricity generation for the first time [15]. This showed a steady increase in RESs of 2% from 27% in 2019 to 29% in 2020 [16,17]. The years 2020 to 2024 showed a notable increase in RESs, with installed capacity growing by about 117%, rising from 270.1 GW in 2020 to 585.2 GW in 2024 (

Table 1. Global annual renewable capacity additions by technology (2020–2024) in GW [19].

Year	Total RES (GW)	Solar PV (GW)	Wind (GW)
2020	270.1	131.0	110.8
2021	263.0	143.2	90.6
2022	302.9	193.3	78.8
2023	484.1	353.0	116.4
2024	585.2	451.9	113.2

). This growth accounted for 46% of the global power capacity by the end of 2024. Notably, supportive policies and the decline in the costs of renewable technologies are the main driver of this growth. By 2030, RES capacity is projected to be 10,300 GW, representing 61% of the installed capacity worldwide, and projected to reach 26,600 GW by 2050 (

Table 2. Projected global renewable capacity outlook for 2030 and 2050 [18].

Category	2030	2050
Total Installed Capacity (GW)	16,885	30,227
RES Installed Capacity (GW)	10,300	26,600
RES Shares (%)	61%	88%
Solar PV and Wind Generation (GW)	6754	20,555
Solar PV and Wind Shares (%)	40%	68%

), accounting for 88% of total installed capacity [18].

2.1. Solar PV and Wind

Solar PV technology demonstrated the most rapid expansion among all technologies, achieving a 245% increase in installed capacity between 2020 and 2024, while wind energy remained stable throughout that time frame (Table 1). Studies suggest that the rapid development of RESs is driven by lowering technological costs, commitments to net-zero emissions, increased investment in major global energy markets, including China, the European Union, and the United States, and geopolitical uncertainty in fuel and energy supply [18,20,21]. A recent report from IRENA [19] claimed that current growth was geographically concentrated in China, the European Union, and the United States, which accounted for 489 GW (83.6%) of renewable capacity installed in 2024.

Recently, at the 28th United Nations Climate Change Conference, COP28, 130 countries pledged to accelerate the energy transition and committed to triple RESs with an installed capacity of at least 11,000 GW by 2030 [22]. The tripling-RES goal aligns with the Paris Agreement target of limiting global warming to 1.5 °C [23]. Despite the challenges, recent forecasts (Table 2) predict that by 2050, approximately 90% of electricity generation could come from RESs, with around 61% supplied by solar PV and wind [18,24].

In Australia, a CSIRO forecast (Figure 1) indicates that the transformation of the National Electricity Market (NEM) follows global trends, and it is expected that total generation will increase from 225 TWh in 2025 to 439 TWh in 2050, an increase of 95%. Study have shown rapid growth in RESs, particularly solar PV with an increase of 200% and wind with an increase of 113% by 2050, projecting to reach 64% of variable energy sources by 2030 and 95% by 2050, predominantly from solar PV and wind. The CSIRO forecast projects an exponential growth in dispatchable storage from 10 TWh in 2025 to 75 TWh by 2050, with an increase in capacity of 650%. The BESS growth also follows the RES trends, and the expected dispatchable electricity storage capacity is expected to grow by 6.3 GW in 2023 and 44–96 GW/550–950 GWh by 2050 [7,25,26].

The synthesis of the studies and recent data presented above supports the conclusion that RESs are growing rapidly. Specifically, solar PV and wind energy are emerging as the main drivers of RES expansion, accounting for approximately 61% of the global market share. This substantial integration of variable RES technologies is progressively displacing conventional synchronous generation, leading to a reduction in overall system inertia. As a result, the decline in system inertia poses a significant challenge to the stability and security of power systems.

2.2. Battery Energy Storage System (BESS)

In recent years, technological advancements have positioned the battery energy storage system (BESS) as a viable solution to address emerging security risks of the system. Research on electricity storage technologies has underscored the critical role of BESSs in grid applications [27]. Compared to alternative technologies, a BESS offers greater flexibility in storage capacity, scalability, and rapid response capabilities, making it an effective solution to enhance grid stability. A BESS provides voltage and frequency control, load following, black start capability, deferral of power line upgrades, and other vital services (Figure 2).

In recent years, the deployment of BESSs has experienced exponential growth (

Table 3. Global and key regional BESS installed capacity additions (GW/GWh), 2023–2024.

Region	Year	Added Capacity (GW)	Added Capacity (GWh)	Cumulative Capacity End-2024 (GW/GWh)	Sources
Global	2023	42–45	90–97	90–97/190	[7,32,33]
Global	2024	150	363	150/363	[31]
China	2024	36–42	101–107	74/168	[34,35]
USA	2024	10.4	28	26/72	[35,36]
Europe	2024	11.9	22.4	35/58.3	[37,38]
Australia	2024	2	4	5/11	[34,39]

and Table 4). Global BESS additions surged in 2023 and 2024, reaching a cumulative capacity of approximately 150 GW/363 GWh by the end of 2024 [31]. Recent research suggests that China, the United States, Europe, and Australia are emerging as key growth drivers in the global battery energy storage system (BESS) market [7]. In 2023, the global BESS capacity saw significant growth, with China adding 23 GW, the United States 8 GW, Europe 6 GW, and Australia 1.3 GW [7].

The IEA Net-Zero Emissions by 2050 (NZEs) scenario reported strong growth in BESSs and suggested that 1200–1500 GW of storage may be required by 2030 (Table 4). This growth is expected to be 8-fold by 2050 from the estimated installed capacity of 150 GW of BESSs in 2024. The BESS capacity projected under various scenarios is shown in Table 4. Technological advancement and a steep decline in manufacturing costs, i.e., from 2511 USD/kW in 2010 to 274 USD/kW in 2023, have contributed to the rapid expansion of battery energy storage systems (BESSs) [24,40].

Table 4. Global BESS capacity projections (GW/GWh) for 2030 and 2050 under key scenarios [7,18,33,38,39,41,42,43,44,45].

Region	Scenario	Year	Projected BESS Capacity (GW)	Projected BESS Capacity (GWh)	Key Assumptions
Global	IEA NZE	2030	1200–1500	N/A	1.5 °C aligned pathway.
Global	IEA STEPS	2030	790	N/A	Policies, BESS addition rate.
Global	BNEF (April 2024)	2030	137 (Annual Adds)	442 (Annual Adds)	Updated forecast, 21% CAGR.
Global	IEA NZE	2050	3100	N/A	1.5 °C aligned pathway.
Europe	EASE	2050	200–600	>600	Based on the system needs assessment for net-zero emissions.
USA	NREL	2050	22.4	35/58.3	Range across scenarios based on costs, PV penetration.

Table 4. Global BESS capacity projections (GW/GWh) for 2030 and 2050 under key scenarios [7,18,33,38,39,41,42,43,44,45].

Region	Scenario	Year	Projected BESS Capacity (GW)	Projected BESS Capacity (GWh)	Key Assumptions
Global	IEA NZE	2030	1200–1500	N/A	1.5 °C aligned pathway.
Global	IEA STEPS	2030	790	N/A	Policies, BESS addition rate.
Global	BNEF (April 2024)	2030	137 (Annual Adds)	442 (Annual Adds)	Updated forecast, 21% CAGR.
Global	IEA NZE	2050	3100	N/A	1.5 °C aligned pathway.
Europe	EASE	2050	200–600	>600	Based on the system needs assessment for net-zero emissions.
USA	NREL	2050	22.4	35/58.3	Range across scenarios based on costs, PV penetration.

The outlook for the growth of battery energy storage systems (BESSs) is increasingly promising, driven by a combination of strong policy support, growing market interest, commitments to net-zero emissions, and rapid technological advancements. Ambitious scenarios, such as the IEA Net-Zero Emissions (NZEs) pathway, highlight the crucial role storage could play in building a cleaner and more resilient energy system. Even conservative projections like the IEA’s STEPS forecast indicate consistent growth aligned with the current trend.

3. High-RES Impact on Power System Frequency

The primary concern is the reduction in system inertia, which is critical to maintaining grid stability during disturbance events [26]. RESs are inverter-based resources (IBRs), which are not directly connected to the power grid but are behind converters and decouple the RESs from the power systems (Figure 3). This decoupling reduces the natural support provided by the kinetic energy stored in the form of inertia and weakens the power system [46,47,48]. In addition, solar photovoltaic systems do not have a rotating mass and do not provide inertia support. The weak grid with low system inertia introduces a higher rate of change of frequency (RoCoF), lower frequency nadir (the minimum level the frequency reached after a disturbance event), and weakens the damping performance, making grids more susceptible to frequency fluctuations [49]. The RoCoF (Hz/s) reflects the speed with which the system frequency changes following a disturbance event and indicates the strength of the system. An aggressive RoCoF may not provide sufficient time for under-frequency load shed (UFLS) schemes to operate and poses a risk of system blackouts [50].

In low-inertia systems, following a disconnection of loads/generators, the RoCoF increases rapidly [48,49,50,51,52]. The RoCoF is the time derivative of the power system frequency (df/dt). The average RoCoF for N synchronous generators and loads is computed using the equation below:

$${\left.{\displaystyle \frac{d\Delta f}{dt}}\right|}_{t=0^{+}}={\displaystyle \frac{f^0{P}_k}{2{\sum }_{i=1, i\ne k}^N{H}_i{S}_i}}$$

(1)

where $\Delta f$ is a frequency deviation from the initial frequency $f^0$ at time $t=0^{+}$ just a moment after the disconnection of loads/generation $P_k$. $H_i$ and $S_i$ are the inertia constant and the apparent power of the synchronous generator unit i, with i ranging from 1 to N [51]. Haque et al. [50] investigated the frequency stability at 10%, 20%, 30%, and 40% of RES integration into the power system and concluded that the frequency nadir $f_{\mathrm{nadir}}$ (Hz) decreased and RoCoF increased with an increase in RESs. A similar investigation was conducted by Saleem and Saha [48] at 0%, 10%, 30%, 50% and 70% of RESs and concluded that with an increase in RESs, the $f_{\mathrm{nadir}}$ decreased and RoCoF increased significantly. A key limitation of that study was the lack of empirical validation of the RoCoF and frequency nadir equations under high-renewable penetration conditions (i.e., 100% of RESs), due to limited availability of high-resolution post-contingency data from such grids; as a result, the broader applicability of these models remains to be fully established and is a key area for future research. Wind and solar energy inherently depend on environmental conditions and exhibit significant intermittency, variability, and uncertainty. This intermittent and variable nature of RESs introduces operational risk to the power system security and causes frequency, voltage, and ramp rate stability issues [40]. As an example, wind gusts cause rapid fluctuations in power generation and ramp rate. Similarly, cloud cover and irradiation intensity are the primary causes of power generation and ramp rate fluctuations in solar PV generation [53]. Power generation variability and ramp rate fluctuations cause frequency deviations in power systems. The system frequency deviation is determined by taking into account the dynamic effects of loads and generations, including inertia, primary, and secondary control [54]. The system frequency deviation is as follows:

$$\Delta f={\displaystyle \frac{1}{(2H_S+D)}}\left(\Delta P_m+\Delta P_W+\Delta P_{PV}+\Delta P_{VI}-\Delta P_L\right)$$

(2)

where

$$\Delta P_m={\displaystyle \frac{1}{1+s{T}_t}}\left(\Delta P_g\right)$$

(3)

$$\Delta P_g={\displaystyle \frac{1}{1+s{T}_g}}\left(ACE-{\displaystyle \frac{1}{R}}(\Delta f)\right)$$

(4)

$$ACE={\displaystyle \frac{K}{s}}\left(\beta ·\Delta f\right)$$

(5)

$$\Delta P_W={\displaystyle \frac{1}{1+s{T}_{WT}}}\left(\Delta P_{\mathrm{wind}}\right)$$

(6)

$$\Delta P_{PV}={\displaystyle \frac{1}{1+s{T}_{PV}}}\left(\Delta P_{\mathrm{solar}}\right)$$

(7)

$$\Delta P_L=P_I+P_R$$

(8)

$$\Delta P_{VI}={\displaystyle \frac{J_{VI}s+D_{VI}}{1+s{T}_{ESS}}}\left(\Delta f\right)$$

(9)

In the above equations, $\mathrm{ACE}$ means the Area Control Error, $\Delta P_{\mathrm{Wind}}$ means the initial wind power change, $\Delta P_W$ means generated power from wind-generating systems, $\Delta P_{\mathrm{solar}}$ is the initial solar power change, $\Delta P_{\mathrm{PV}}$ means generated power from solar PV systems, $\Delta P_g$ means power generated by turbines, $J_{\mathrm{VI}}$ is the virtual inertia constant, $D_{\mathrm{VI}}$ is the virtual damping constant, $T_{\mathrm{ESS}}$ is the time constant of the inverter-based ESS, and $\Delta P_{\mathrm{VI}}$ means virtual inertia power. $P_I$ and $P_R$ are industrial and residential loads, respectively.

Equations (2), (6), and (7) support the notion that active power variation ($\Delta P_{\mathrm{solar}}$ and/or $\Delta P_{\mathrm{wind}}$) due to environmental factors such as irradiation and wind influences grid frequency. A recent investigation by Buch [55] demonstrated that solar photovoltaic (PV) systems exhibited significant variations in power generation due to changes in irradiation under cloudy conditions. Similarly, a study by Yan et al. [56] concluded that fluctuations in irradiation led to rapid changes in ramp rate, which consequently affected power output and system frequency. Liu et al. [57] investigated the effects of wind fluctuation and its impact on power system security, concluding that wind fluctuations greatly influenced power output.

The increasing integration of renewable energy sources (RESs) represents a critical advancement toward achieving net-zero-emission targets. However, with increasing RESs, the system inertia is expected to reduce significantly. This will increase the rate of change of frequency (RoCoF) in power systems, thereby elevating the risk to power system stability and reliability. In this context, battery energy storage systems (BESSs) are becoming a viable and effective solution for addressing RES intermittency, thereby promoting grid stability, enhancing system reliability, and enabling the increased deployment of low-carbon-energy technologies.

4. BESS Role in Modern Power Systems

BESSs are playing a critical role in modern power systems with a high integration of RESs. A BESS provides active power support through power smoothing when coupled with solar PV and wind generation. It also offers power levelling during low- and high-demand periods and peak shaving during high-demand periods. In addition, a BESS contributes to voltage and frequency regulation. A BESS also provides reactive power support and helps regulate power system voltage, as well as fault ride-through capabilities [4,58]. Grid-forming inverter technology positions BESS as an ideal solution for delivering emergency services such as black start capability, islanding grid support, and congestion relief (Figure 4). Due to the intermittent nature of RESs, voltage and frequency control have become a challenge for utilities. This problem is worsening with increasing RESs and decreasing system inertia [28]. This is a well-recognised issue that necessitates a better approach to managing safe and reliable grid operation.

Globally, grid frequency is managed through various mechanisms, including frequency control services and mandatory frequency support. The frequency regulation market design varies with markets and network conditions. The section below summarises the frequency control services in different global markets and their characteristics.

4.1. Frequency Control Ancillary Services (FCASs)

Historically, grid stability, particularly frequency stability, has been supported by the inherent physical property of inertia provided by large, rotating synchronous generators (e.g., thermal, hydro). The kinetic energy stored in these rotating masses naturally resists changes in system frequency, slowing down the rate at which frequency deviates (rate of change of frequency, RoCoF) following a disturbance, such as the sudden loss of a large generator or load (Figure 5). As RESs provide asynchronous generation and are typically behind the inverters, the inertial support from RESs is limited [59,60]. Transmission System Operators (TSOs) or Independent System Operators (ISOs) procure grid support services commonly known as ancillary services to regulate grid frequency within the acceptable range. Among the most critical services are the frequency control ancillary services (FCASs) in the National Electricity Market (NEM), the Frequency Co-Optimised Essential System Services (FCESSs) in the Wholesale Electricity Market (WEM), and frequency response services, Balancing Services and Frequency Containment Reserve (ENTSOE), particularly within European electricity markets [10,61,62,63,64].

Several studies have explored the various frequency control services across major electricity markets and suggest that frequency control is typically achieved through a layered approach, involving different types of responses acting over various timescales following a frequency disturbance [66]. These layers work collaboratively to arrest the initial frequency deviation, stabilise the frequency of the system, restore it to its nominal value, and restore the reserves utilised [65]. Typically, a response is required when grid frequency operates outside the defined deadbands (Figure 6).

The commonly known frequency responses are the following.

4.1.1. Inertial Response

This is the near-instantaneous (sub-second) response inherent in the physics of synchronous machines (Figure 5). When frequency changes, the kinetic energy stored in their rotating mass is naturally released (during low-frequency events) or absorbed (during high-frequency events), counteracting the change [51]. However, a BESS does not have any rotating mass and cannot provide inertia support. With technological advancements, BESSs with grid-forming inverters (GFM) can provide synthetic inertia by mimicking the active power response proportional to the RoCoF [67].

4.1.2. Fast Frequency Response (FFR)

Fast-frequency-response services require an injection or rejection of active power (MW) in a couple of seconds to maintain grid stability [8,68,69]. FFR responds to frequency excursions much faster than conventional generators and is the first responder after the inertia response. AEMC presents a comprehensive study on this subject, highlighting the close interaction between FFR services and the response to inertia [68]. The study identifies the challenge of differentiating FFR from the inertia response during the initial stage of the frequency response and the difficulty in explicitly distinguishing their contributions. Batteries, because of their fast response rate, are considered a good candidate for FFR service. In the Australian electricity market, the fast-frequency-response (FFR) provision must be activated within one second of a frequency disturbance event [10,68]. The United Kingdom market also has a similar response requirement, with the addition of sustained response for an additional 15 min [63]. In contrast, in the US and Canadian markets, the service is called Inertial Control and requires a full response in 0–12 s, offering flexibility. Due to its rapid response capabilities, BESSs can deliver a FFR in 1–2 s. However, this rapid power alteration at a very high C-rate might induce a high current in the BESS, leading to a quick temperature increase and accelerated battery degradation. Typically, the rapid response required within a few seconds leads to steep current transients ${\displaystyle \frac{dI}{dt}}$ during fast charging or discharging events, resulting in elevated I²R losses. Furthermore, during FFR events, batteries are subjected to aggressive ramp rates, for example, up to 1 C/s in response to a 1 Hz frequency deviation with a 2% droop setting, compared to typical ramp rates of approximately 0.0033 C/s (equivalent to a 20%/minute rated active power ramp rate) under normal operating conditions in the Australian market, representing an increase of nearly 300 times the standard rate. Although the total energy delivered/consumed during such events is relatively low, the high current over a short duration can result in significant degradation of battery components. In addition, high-C-rate operations can, in certain cases, induce irreversible changes in the cathode lattice structure, thereby accelerating battery degradation. Chemistries such as nickel cobalt manganese (NCM), lithium manganese oxide (LMO), and lithium cobalt oxide (LCO) exhibit relatively poor thermal stability; participation in FFR markets under such conditions may lead to premature battery failure and an increased risk of thermal runaway. Furthermore, these batteries are susceptible to structural changes in the cathode material at elevated temperatures, which further accelerate the likelihood of failure. In contrast, lithium iron phosphate (LiFePO₄) batteries, when supported by effective thermal management strategies, are better suited for providing FFR services. However, they still experience an accelerated cycle-life degradation under high-C-rate operation, necessitating careful commercial evaluation and operational decision-making.

4.1.3. Primary Frequency Response (PFR)

This is the first active control response, typically activated within seconds of a frequency deviation, typically within 30 s, as shown in Figure 5 [70]. PFR or Frequency Containment Reserve (FCR) resources independently adjust their power output in proportion to the magnitude of the frequency deviation based on the droop settings. The primary goal of the PFR is to arrest the frequency fall or rise outside of the predefined frequency deadbands. This response is generally sustained for several minutes until slower acting reserves, typically the secondary frequency response (SFR), can take over [10].

The PFR requirements in the Australian and UK markets are similar but differ slightly in their support for the response. There are no sustained response requirements in the National Electricity Market (NEM), but the Wholesale Market (WEM) on the West Coast of Australia requires sustained response for 15 min. In contrast, the UK market requires a sustained response of 20 s. Similar to FFR, the PFR is required within a short timeframe, resulting in higher-C-rate charge or discharge cycles for BESSs. In BESSs, during PFR, a high C-rate with a deep depth of discharge (DoD) leads to rapid temperature rises, resulting in capacity and power degradation. The rate of response (typically 4% frequency droop) and degradation when providing a PFR is lower than that of an FFR. At a 4% frequency droop, the battery will ramp up or down at 0.167 C/s over 6 s, which is 50 times higher than the standard, safe ramp rate of 0.0033 C/s. Furthermore, following the initial primary frequency response (PFR), both the Australian and UK markets require a sustained response lasting from several seconds (10 s) to a few minutes (30 m). Battery energy storage systems (BESSs) may face limitations in meeting these sustained response requirements, as their capability is constrained by the availability of remaining stored energy and state-of-charge dynamics. This constraint can be mitigated through the implementation of a comprehensive energy management strategy, which may include backup support from slower-responding generation units, the use of battery technologies that are not optimal for PFR but are capable of sustaining longer-duration responses, and the integration of energy-balancing mechanisms to maintain system stability and ensure sustained frequency support. A LiFePO₄ battery with appropriate heat management system is suitable for providing a PFR. LiFePO₄ batteries exhibit lower degradation rates compared to other lithium-ion chemistries and offer superior thermal stability. However, similar to the FFR, participation in PFR services imposes significant cycling stress, which can impact battery life and requires careful commercial evaluation and informed operational decision making.

4.1.4. Secondary Frequency Response (SFR)

Following the initial containment by PFR/FCR, the SFR/aFRR is activated to restore the system frequency to its nominal value. The SFR is typically slow compared to the FFR and PFR, typically 30 s to a few minutes. The SFR/aFRR is typically remotely controlled by the TSO/ISO, which sends automated signals (usually through automatic generation control (AGC) systems) to participating generating units, instructing them to adjust their power output [71]. The activation time for the SFR/aFRR is slower than that of the PFR/FCR, ranging from tens of seconds to several minutes. In Europe, this is known as aFRR. Markets like PJM differentiate within this category, offering signals such as RegD (dynamic regulation) tailored for faster and more precise adjustments, alongside traditional regulation signals (RegA). Similarly, CAISO procures the Regulation Up and Regulation Down services. BESS performance within the SFR offers better safety compared to the FFR and PFR, primarily due to its regulated charge–discharge rates and moderate temperature rise, which result in less battery wear and an extended BESS lifespan.

4.1.5. Tertiary Frequency Response (TFR)

The TFR services are executed to manage ongoing current and future contingencies. This is the slowest response frequency regulation service, typically activated manually by grid operators over minutes to hours. The TFR/mFRR involves adjusting the dispatch of generation units (or large loads) using SCADA to fully restore the balance between generation and load in the longer term, replenish the faster-acting reserves (PFR/FCR and SFR/aFRR), and manage anticipated changes or persistent imbalances. In the TFR market, a BESS undergoes frequent minor charge and discharge cycles. The long-term impact of repeated charging and discharging in micro- and mini-cycles has not been sufficiently studied and warrants further research.

4.1.6. Battery Operation in Other Markets

Grid-scale battery energy storage systems (BESSs) offer a wide range of services beyond frequency control ancillary services (FCASs), including energy arbitrage, peak shaving, renewable energy forming and shaping, and grid support. In profit-driven markets, energy arbitrage, when charging during low-price periods and discharging during peak demand, remains a key revenue stream. For example, the Hornsdale Power Reserve (HPR) in South Australia participates in both FCASs and energy arbitrage, with 30 MW of energy arbitrage and 70 MW of power reliability purposes [72]. The Hornsdale Power Reserve (HPR) is incorporated into the System Integrity Protection Scheme (SIPS), designed to reduce the risk of South Australia’s separation from the NEM due to sudden increases in flow on the Heywood Interconnector. The SIPS detects high interconnector flows and rapidly dispatches the HPR at 100 MW, with optional controlled load shedding to further mitigate the risk of separation. These customised schemes facilitate greater battery integration and contribute significantly to grid stability. Further, peak shaving helps reduce peak demand and mitigate price spikes, while BESSs also smooth the variability of solar and wind generation to better align supply with demand. Moreover, strategically sited storage can defer costly transmission and distribution upgrades. Studies suggest that optimising BESS operations through strategic charging and discharging minimises trading costs, maximises profits, and enables participation in markets such as FCAS, energy arbitrage, and peak shaving [73,74]. Ling et al. [75] introduced a MILP model aimed at maximising operational revenue by incorporating battery dynamics. They used a simplified equivalent circuit model to linearise non-linear constraints. The model focused on minimising operational costs by optimising charge and discharge power in response to electricity prices and conversion efficiencies at each time step. Revenue was achieved by charging during low-price periods (off-peak period) and discharging during high-price periods (peak period). The study also highlighted that higher C-rates, typically observed during FCASs and other ancillary services, enhance battery power capabilities, allowing more frequent cycling and greater exploitation of price differentials, thereby increasing profit.

4.2. Frequency Control Markets Comparison

The frequency control services across major electricity markets globally vary in terms of response time and sustained response. The Australian markets require a fast frequency response, typically within a few seconds, <2 s for FFR services, and within 6 s for PFR services

Table 5. Comparison of key FCAS markets.

Market	Country	Region	Ancillary Services	Normalised Service	Response Time (s/m/h)	Source/ References
AEMO	Australia	East Coast	Very Fast Fast Slow Delayed Regulation	FFR PFR SFR TFR TFR	1 s 6 s 60 s 300 s (5 min) 300 s (5 min)	[10,68]
AEMO	Australia	West Coast 1	Very fast Fast Fast Regulation Regulation	FFR PFR PFR SFR SFR	<2 s 6 s 6 s 300 s (5 min) 300 s (5 min)	[61]
National Grid ESO	UK	Great Britain 2	Fast Fast Fast	MFR-PFR MFR-SFR MFR-PFR	10 s 30 s 10 s	[63]
National Grid ESO	UK	Great Britain 3	Very fast Very fast Fast	DC-PFR DM-SFR DR-PFR	1 s 1 s 30 s	[63]
EirGrid	Ireland	Ireland	Fast Fast Slow Slow Very slow	PFR SFR TFR TFR TFR	5–15 s 15–90 s 90–300 s 5–20 m 20 m to 4 h	[64]
NORDIC	Norway, Sweden, Finland, Eastern Denmark	All (except Western Denmark)	Fast Fast Fast Slow	FFR FCR-D FCR-N FRR	5–30 s 5–30 s 180 s 2–5 m	[77]
NERC	US/ Canada	US/ Canada	Very fast Fast Slow Very slow	FFR PFR SFR TFR	0-12 s 10-60 s 1–10 m 10 m to h.	[78]

¹ PFR = Contingency UP and Down; SFR = Regulation UP and Down; Regulation UP and Down response lasts for 15 min; ² MFR = Mandatory Frequency Response; MFR-PFR (primary response) = sustained for 20 s; MFR-SFR (secondary response) = sustained for 30 s; MFR-PFR (high-frequency response) = sustained indefinitely; ³ DC = Dynamic Containment; DC-FFR (Dynamic Containment) = 1 s (sustained 15 min); DM-FFR (Dynamic Moderation) = 1 s (sustained 30 min); DR-PFR (Dynamic Regulation) = 10 s (sustained 60 min).

. The Wholesale Electricity Market (WEM) on the West Coast of Australia has established Frequency Co-optimised Essential System Services (FCESSs) equivalent to FCASs in the National Electricity Market (NEM). The WEM has a cost-recovery mechanism and incentivises market participants for frequency control services such as FFR, PFR, SFR, and TFR. One of the cost-recovery mechanisms in the Wholesale Electricity Market (WEM) is the inclusion of variable operating and maintenance (VOM) costs. Battery energy storage system (BESS) operators, as market participants, are permitted to incorporate VOM costs into their market offers, reflecting wear and tear associated with charging and discharging cycles. Additionally, market participants are allowed to recover costs related to round-trip efficiency losses, which are typically amortised and expressed on a AUD/MWh basis. Moreover, the market framework permits the inclusion of opportunity costs in the offers submitted by battery energy storage system (BESS) operators. These opportunity costs may reflect the foregone revenue associated with discharging energy during future Dispatch Intervals with higher prices or from providing alternative Market Services. The Wholesale Electricity Market (WEM) also provides cost-recovery mechanisms to market participants for the provision of Frequency Co-Optimised Essential System Services (FCESSs). FCESS Uplift Payments are issued to market participants when a generator is dispatched to provide FCESSs instead of energy and the market-price-based compensation is insufficient to cover the associated costs [76].

In the National Electricity Market (NEM), the current cost-recovery mechanism is based on net energy import/export. Regulation Services’ payments are recovered through the Frequency Performance Payments (FPPs) framework, which operates as a double-sided incentive mechanism. Under this system, market participants are rewarded or penalised based on their contribution to system frequency control. Participants with positive contribution factors receive incentive payments, while those identified as primary contributors to frequency deviations, referred to as “causes”, are subject to penalties. Payments for Contingency Raise Services are recovered from Cost Recovery Market Participants (CRMPs) based on their Adjusted Sent-Out Energy (ASOE), while payments for Contingency Lower Services are recovered from CRMPs according to their Adjusted Consumed Energy (ACE) values [79].

$$\mathrm{Payment}={\displaystyle \frac{MWE\times CP}{12}}$$

(10)

where MWE is the amount of MW enabled by NEMDE for the service being settled, and CP is the regional clearing price for the service in that trading interval. (note that as the bids and clearing price are defined as dollars per MW per hour, dividing the result by 12 brings the payment back in line with the five-minute trading interval) [79].

The UK electricity market appears to operate primarily through a balancing mechanism, with no dedicated cost-recovery framework currently in place for battery energy storage system (BESS) operators. Similar findings have been observed in other European and North American markets, where the cost recovery of BESSs is typically achieved through participation in balancing markets and exposure to market-based pricing mechanisms. Cost-recovery mechanisms that rely on balancing markets may place battery energy storage system (BESS) operators at a disadvantage. For instance, when a battery provides frequency control ancillary services (FCASs) and sustains a response over an extended period, it may deplete its capacity at the prevailing balancing market price. This can result in missed opportunities to capitalise on potentially higher prices in subsequent market intervals. In addition, variable operating and maintenance (VOM) cost recovery for battery energy storage systems (BESSs) in PJM and other U.S. electricity markets is typically assumed to be zero. However, in practice, when batteries provide frequency control ancillary services (FCASs), particularly fast frequency response (FFR) and primary frequency response (PFR), intensive cooling is required, the demand for which increases with both the rate of response and the duration of service. Moreover, such operations accelerate battery capacity degradation beyond the rated levels. These additional costs and performance impacts are not currently compensated in most major electricity markets, which serves as a disincentive for BESS developers and operators to participate in FCAS markets.

4.3. Regulatory Requirements: Grid Code, Generator Performance Standards

Grid codes, also known as Generator Performance Standards (GPSs) in some energy markets, are comprehensive technical documents that outline the rules, regulations, operational mandates, and performance benchmarks for power-generating units. In Western Australia, for instance, GPSs are formally defined within Chapter 3A and Appendix 12 of the Wholesale Electricity Market (WEM) Rules [9]. These standards are designed to ensure that each connected generation asset adheres to its operational obligations and specified parameters, thus mitigating risks such as cascading failures and maintaining voltage and frequency within acceptable limits [64]. The current market rules are generic in nature and do not distinguish between traditional generators and battery energy storage systems (BESSs).

To regulate frequency, generating units such as BESSs must supply and sustain active power output according to the specified droop setting. The droop sets the rate of the active power response with respect to a frequency change, that is, in WEM, a 5% active power change is required with every 0.1 Hz frequency change, making it a 4% droop response. In some markets, a BESS is expected to provide a rapid response, i.e., FFR, PFR, LVRT, HVRT, FDRT, FRT, which causes a rapid charge or discharge of energy into the grid within a short timeframe (

Table 6. Grid code (GC)/Generator Performance Standards’ (GPSs) requirements.

Grid Code/GPS Requirements	Response Time	References
Low-Voltage Ride Through (LVRT)	0.1–3 s	[80,81]
Multiple-Disturbance Ride Through (MDRT)	0.1–3 s	[9,80,81,82]
Frequency Support	1–30 s	[9,61,62,63,68,71,78]
Voltage Support	0.45–5 s	[9]
Partial/Full-Load Rejection Support	A few seconds to minutes	[9]
Fault Ride Through (FRT)	Milliseconds to a few seconds	[9]

). This rapid charge and discharge cycle at elevated C-rates increases battery temperature, lithium depletion, and accelerates capacity degradation. Furthermore, a BESS has limited stored energy to support the grid during such an event and may not sustain a response beyond a certain level. This causes a lack of visibility and uncertainty in grid stability.

5. FCAS Impact on BESS’s Operational Life

When providing FCASs, many factors significantly influence the performance of BESSs, including the nature of charge/discharge and the depth of discharge, the temperature, and the chemistry of the battery cell. Several mature battery chemistries are currently in use, including lithium iron phosphate (LiFePO₄), nickel cobalt manganese (NCM), sodium ion (Na-ion), lithium cobalt oxide (LCO), lithium manganese oxide (LMO), and nickel cobalt aluminium (NCA). Each of these chemistries possesses distinct electrochemical characteristics that critically influence their performance under high-cycle conditions, particularly at higher C-Rates, when used in FCASs (Table 7).

Lithium iron phosphate (LiFePO₄) is considered the safest among other battery technologies. This is largely attributed to its stable olivine crystal structure, which minimises risks associated with thermal runaway [83]. The thermal runaway threshold for lithium iron phosphate (LiFePO₄) is typically reported in the range of 250–300 °C, significantly higher than that of many other lithium-ion chemistries [84]. Additionally, the minimal ion displacement within the LiFePO₄ structure contributes to its enhanced thermal stability. In addition, LiFePO₄ batteries exhibit a flat discharge voltage profile and an exceptionally long cycle life, typically ranging from 2400 to over 8000 cycles, which makes them suitable for applications such as frequency control services. LiFePO₄ batteries exhibit a low capacity degradation rate under high-C-rate operation, with reported degradation rates of approximately 0.005–0.02% per cycle at 1 C and 0.02–0.05% per cycle at 10 C, highlighting their robustness in high-power, long-duration applications [77,85,86].

Lou et al. [87] reported significantly enhanced cycling performance of LiFePO₄/C cathodes with carbon coating, achieving a specific energy of 255 Wh/kg at a 30 C-rate, corresponding to a specific power of 7650 W/kg. While still in the development phase, these batteries demonstrate considerable promise for delivering high-performance energy storage in applications that require both high power and stability in FCASs. Although LiFePO₄ batteries are safe and thermally stable, their lower energy density compared to NMC, LCO, and NCA batteries limits their applicability in compact energy storage systems. Additionally, when providing frequency control services, i.e., FFR and PFR, with a response time of 1–6 s, the C-rate-based battery degradation accelerates by 2.5 to 4 times.

Lithium nickel cobalt manganese oxide (NCM) batteries provide better power and energy density compared to LiFePO₄ batteries. Their layered crystal structure facilitates rapid ion transport, enabling fast-charging and discharging capabilities. However, NCM-based lithium-ion batteries exhibit thermal sensitivity, with capacity degradation accelerating significantly under elevated temperatures and high-C-rate conditions, such as those encountered in frequency control services. For instance, the degradation rate at a 1 C-rate increases from 0.02% per cycle at 25 °C to 0.03% at 40 °C. In contrast, at a 5 C-rate, it rises from 0.05% per cycle at 25 °C to 0.06% at 40 °C, highlighting the detrimental impact of combined thermal and C-rate-induced stress on long-term performance. These conditions exacerbate cathode degradation, reduce coulombic efficiency, and increase the risk of thermal runaway, which limits their large-scale deployment in energy storage systems and FCASs.

Lithium cobalt oxide (LCO) batteries are rich in energy and power density, but they are sensitive to overvoltage. Under high-voltage operating conditions, lithium cobalt oxide (LCO) cathode materials undergo irreversible phase transformations, which compromise their structural integrity and thermal stability [88]. This degradation behaviour, coupled with the material’s sensitivity to rapid capacity loss at elevated temperatures, exhibiting a degradation rate of approximately 0.16% to 0.20% per cycle at a 1.5 C-rate, 10 times that of LiFePO₄ batteries, raises significant concerns regarding the thermal and structural stability of the battery under high-C-rate operational conditions. Also, the thermal runaway threshold for LCO-based batteries is substantially lower (180 °C) compared to that of lithium iron phosphate (LiFePO₄) batteries, which ranges from approximately 180 °C to 350 °C. Similar to LCO, lithium manganese oxide (LMO) and lithium nickel cobalt aluminium oxide (NCA) batteries offer a high energy density but limited thermal stability, lower thermal runaway thresholds, faster degradation at higher C-rate operation, and reduced cycle life. These limitations render them less suitable for demanding applications such as FCASs, where thermal resilience and long-term durability are critical.

Sodium-based cathode materials, such as NaFePO₄, exhibit excellent cycling stability, retaining approximately 95% of their initial capacity after 1000 cycles at a 1 C-rate. Additionally, Na₃V₂(PO₄)₃/C composite cathodes demonstrate outstanding high-rate performance, maintaining around 90% capacity retention over 2000 cycles at a 5 C-rate, highlighting their strong potential for frequency control services [89].

Table 7. Comparative performance and degradation characteristics of battery chemistries for grid services.

Parameter	LFP (LiFePO4)	NCM	LCO	LMO	NCA	Na-Ion	References
Cycle Life (@80% DoD)	2400–8000+ cycles	1200–5600+ cycles	1500–2600 cycles	500–2200+ cycles	250–1500 cycles	2000–5000+ cycles (emerging)	[90,91,92,93,94,95,96,97,98]
Thermal Stability	High	Moderate	Poor	Moderate	Poor	High	[99,100]
Thermal Runaway Threshold (@ > 50% SoC)	High (180–350 °C)	Medium (150–200 °C)	Medium (180 °C)	Medium (150–220 °C)	Medium (150–200 °C)	High (190–200 °C)	[84,101,102,103,104,105]
Degradation Rate (High C-rate) Per Cycle	Low (at 1 C-rate: 0.005–0.02%; at 10 C-rate: 0.02–0.05%)	High (at 1 C-rate: 0.02% @25 °C; 0.03% @40 °C; at 5 C-rate: 0.05% @25 °C; 0.06% @40 °C)	Very high (at 1.5 C-rate: 0.16–0.2%)	High (at 2 C-rate: 0.01–0.04% @25 °C; 0.04–0.08% @55 °C)	High (at 1 C-rate: 0.05–0.07%)	Very low (0.0012%) (emerging)	[77,85,106,107,108,109,110,111]
C-rate Capability	1–3 C	1–2 C	1–3 C	1–3 C	1–3 C	1–7 C (emerging)	[84,95,112]
Power Density (W/kg)	1500–3000	1044–2760	1500–3900	3500–6880	2200–3000	670–4000	[89,113,114,115,116,117,118,119,120]
Energy Density (Wh/kg)	90–150	150–200	150–392	100–350	200–260	120–150	[89,118,119,121,122,123,124]
Cost (USD/kWh)	USD 101–400	USD 94–230	USD 357	USD 133–251	USD 98–418	USD 262	[125,126,127,128,129,130,131]
Safety	Very high	Moderate (risk of thermal runaway)	Low (risk of thermal runaway)	Moderate (risk of thermal runaway)	Low (risk of thermal runaway)	High	[132]

Table 7. Comparative performance and degradation characteristics of battery chemistries for grid services.

Parameter	LFP (LiFePO4)	NCM	LCO	LMO	NCA	Na-Ion	References
Cycle Life (@80% DoD)	2400–8000+ cycles	1200–5600+ cycles	1500–2600 cycles	500–2200+ cycles	250–1500 cycles	2000–5000+ cycles (emerging)	[90,91,92,93,94,95,96,97,98]
Thermal Stability	High	Moderate	Poor	Moderate	Poor	High	[99,100]
Thermal Runaway Threshold (@ > 50% SoC)	High (180–350 °C)	Medium (150–200 °C)	Medium (180 °C)	Medium (150–220 °C)	Medium (150–200 °C)	High (190–200 °C)	[84,101,102,103,104,105]
Degradation Rate (High C-rate) Per Cycle	Low (at 1 C-rate: 0.005–0.02%; at 10 C-rate: 0.02–0.05%)	High (at 1 C-rate: 0.02% @25 °C; 0.03% @40 °C; at 5 C-rate: 0.05% @25 °C; 0.06% @40 °C)	Very high (at 1.5 C-rate: 0.16–0.2%)	High (at 2 C-rate: 0.01–0.04% @25 °C; 0.04–0.08% @55 °C)	High (at 1 C-rate: 0.05–0.07%)	Very low (0.0012%) (emerging)	[77,85,106,107,108,109,110,111]
C-rate Capability	1–3 C	1–2 C	1–3 C	1–3 C	1–3 C	1–7 C (emerging)	[84,95,112]
Power Density (W/kg)	1500–3000	1044–2760	1500–3900	3500–6880	2200–3000	670–4000	[89,113,114,115,116,117,118,119,120]
Energy Density (Wh/kg)	90–150	150–200	150–392	100–350	200–260	120–150	[89,118,119,121,122,123,124]
Cost (USD/kWh)	USD 101–400	USD 94–230	USD 357	USD 133–251	USD 98–418	USD 262	[125,126,127,128,129,130,131]
Safety	Very high	Moderate (risk of thermal runaway)	Low (risk of thermal runaway)	Moderate (risk of thermal runaway)	Low (risk of thermal runaway)	High	[132]

A BESS, when providing FCASs, undergoes many micro, mini, partial, and full charge and discharge cycles at different C-rates throughout its lifecycle [133,134,135]. Under these dynamic conditions, varying C-rates make degradation non-linear, predicting the remaining useful life (RUL), typically between 70 and 80% of the full capacity of the BESS [136,137], is a significant challenge that substantially impacts its reliability and process safety.

There are three stages in battery capacity degradation, as shown in Figure 7. The first stage, the acceleration stage, is believed to be caused by the formation of the initial solid electrolyte interphase layer (SEI), which causes a rapid increase in the internal resistance of the battery and, consequently, a rapid decrease in capacity [138]. The second stage is the stabilisation stage, also known as Useful Life or Linear Aging, where the battery operates under normal conditions, providing energy and FCAS support. This stage is more useful, stable, and predictable in terms of the BESS response [138,139]. The third and final stage is the saturation stage, also known as non-linear aging. This stage occurs predominantly due to the loss of lithium, acid dissolution, or the higher internal resistance of the batteries. Batteries are considered unsafe to operate once they reach that stage. The point where the saturation stage starts, also known as the knee point, is considered the End of Life (EOL) of a battery [138,139,140].

Battery performance deteriorates sharply beyond the knee point, and the failure rate increases significantly. Therefore, manufacturers recommend stopping the operation when the knee point is reached or when the saturation stage is reached. To manage BESS effectively in FCAS and energy markets, several control strategies are deployed in grid-connected BESS. The next section discusses the commonly used control strategies.

5.1. Calendar Aging Impact

The aging degradation in BESSs is mainly due to variations in ambient temperature. Several studies have investigated the impact of temperature on battery aging. A recent review study by Zhang et al. [40] investigated factors affecting the life expectancy of a lithium-ion battery and highlighted the importance of temperature on battery life. Zhang et al. [40] pointed out that battery performance was ideal at a room temperature of 25 °C, and performance varied if used at higher or lower temperatures. In lithium-ion batteries, low temperature is one of the major contributors to irreversible depletion of lithium ions and dendrite formation [141]. Similarly, recent studies on the high-temperature impact on batteries discovered that battery aging degradation accelerated at high temperatures, and the major contributing factor to degradation was the growth of the solid electrolyte interphase (SEI) (Deshpande et al., 2017) [142]. Guan et al.’s [143] experimental data suggest that at elevated temperatures (45 °C), a thick and unstable layer of SEI forms on the cathode, which contributes to the battery’s higher internal resistance and leads to a battery performance decrease. Their study also emphasises that 75% of capacity degradation on the anode is caused by SEI formation. Recently, Yang et al. [144] revealed that the continuous growth of the SEI layer led to a reduction in anode porosity and the presence of lithium ions accelerated the reduction in anode porosity and promoted lithium plating. A study by Baghdadi et al. [145] highlighted that during prolonged battery operation at elevated temperatures, calendar aging became the dominant contributor to total aging. With advancements in thermal management systems, this issue is being mitigated through effective cooling techniques, including forced-air and liquid-cooled battery operations [146]. That study investigates degradation mechanisms occurring under high-cycle and high-C-rate conditions, which more accurately reflect the operational profile of battery energy storage systems (BESS) during the provision of frequency control ancillary services (FCAS). Under such conditions, calendar aging is considered less significant and is therefore assumed to remain constant.

5.2. Cyclic Aging

Unlike aging degradation, cyclic degradation is more nonlinear and primarily caused by various internal and external factors such as SEI layer growth, lithium plating, mechanical stress on electrodes, electrolyte–electrodes reaction, loss of lithium, thermal effects, electrolyte decomposition particles cracking, dendrites formation, excessive temperature due to rapid charging and discharging, and cycling loading (Figure 8) [144,147].

5.2.1. Solid Electrolyte Interphase (SEI) Formation

The solid electrolyte interphase (SEI) Formation is one of the major causes of battery performance degradation. The SEI formation process results in lithium-ion loss and, consequently, performance degradation [148]. This relationship is further supported in a recent study [138], which suggests that SEI formation interacts with other degradation mechanisms such as lithium plating. The SEI formation continues in both resting and operating modes, accelerating with high temperatures and high current throughputs, which leads to irreversible capacity loss and an increase in internal battery impedance [149]. When the battery is used to provide FCASs such as FFRs (response in 1–2 s) and PFRs (response in 6 s), it undergoes frequent and rapid charge–discharge cycles at higher C-rates (higher I²R losses), which elevate the cell temperature. Under such conditions, the SEI layer becomes mechanically unstable as a result of repeated cracking and reformation, further exacerbating degradation and affecting long-term battery performance (Figure 8).

During FCAS support, a battery undergoes various magnitudes of depth of discharge (DoD) and energy throughput. This results in elevated temperatures in battery cells, which are directly related to the formation of the SEI and cracking. The primary concern with LiFePO₄ is I²R losses and active material loss due to SEI layer formation. In contrast, lithium cobalt oxide (LCO) cathodes are prone to rapid capacity degradation primarily due to lattice disorder, irreversible phase change, and gas evolution resulting from oxygen release during high-voltage operation.

5.2.2. Lithium Plating and Loss of Active Material

Lithium plating occurs on the surface of the Negative Electrode (NE) during charging [150]. The low-temperature operation leads to poor solid diffusivity, which restricts the movement of lithium ions and results in their accumulation on the electrode surface. The factors affecting lithium plating include fast charging. These findings are closely related to FCASs, where the response from a BESS is expected within a short period. This lithium plating increases the risk of short-circuiting within the cell and is a contributing factor in the loss of cycle life. Numerous studies have confirmed that lithium plating is an irreversible process and one of the primary reasons for dendrite formation in battery cells [138,144,150].

5.3. Temperature Impact

A BESS’s temperature is another critical parameter that determines the state of health (SoH) and readiness of the BESS for FCASs. If the temperature exceeds the permissible level, it can lead to accelerated battery performance degradation and thermal runaway [151,152]. The performance degradation of BESSs is categorised into two categories, capacity and power degradation [153]. During both charging and discharging processes, battery energy storage systems (BESSs) generate heat primarily due to internal resistance (I²R losses), lattice structure disorder, and electrochemical activity. Furthermore, the temperature and reactions within the cell cause the depletion of active materials like lithium ions, leading to a reduction in BESS capacity. The capacity loss in LiFePO₄ is primarily attributed to I²R losses and the loss of active material. In contrast, cathode materials such as NMC, LMO, and NCA exhibit more pronounced degradation at elevated temperatures, where irreversible changes in the lattice structure and significant loss of active material occur, which exacerbates degradation and leads to permanent capacity loss. Ali et al. [154] examined the behaviour of lithium-ion batteries at temperatures exceeding 40 °C and determined that elevated temperatures accelerated chemical degradation processes, such as the dissolution of LiFePO₄ particles and thickening of the SEI layer, resulting in increased internal resistance and higher I²R losses, which further increased temperature and subsequently reduced capacity. The study also highlighted that at higher temperatures, the battery caused the evaporation of electrolyte solvents, which increased the internal pressure and posed safety risks. At higher C-rates during BESS charging and discharging, the internal and surface temperatures of the battery increase, which accelerates battery degradation. Bandhauer et al. [153] reviewed thermal challenges in lithium-ion batteries, noting a reduction in power output with rising internal impedance.

Main Causes of Heat Generation in BESSs

There are four fundamental sources of heat generation in BESSs: (1) ohmic losses (Joule heat losses) resulting from the movement of electrons and the internal impedance of the battery, (2) electrochemical reactions, (3) phase changes, and (4) mixing effects. Bernardi et al. [155] proposed a comprehensive and general formulation of the energy balance for an electrochemical system, under the assumptions of uniform temperature and negligible pressure effects. Bernardi et al.’s [155] equation can be simplified to:

$$\dot{\mathbf{Q}}=\left[IV-\sum _i{I}_i{U}_i\right]+\sum _i{I}_{i}T{\displaystyle \frac{d{U}_{i,}}{dT}}+M{C}_p^{\mathrm{m}}{\displaystyle \frac{dT}{dt}}$$

(11)

where $\dot{\mathbf{Q}}$ is the rate of heat generated or consumed, M is the mass per unit area of one cell, $C_p^{\mathrm{m}}$ means specific heat capacity at constant pressure in J/g-K, and T is absolute temperature in K.

Equation (11) can be reformulated as follows:

$$\dot{\mathbf{Q}}=I\left(V-U+T{\displaystyle \frac{\partial U}{\partial T}}\right)+C_p{\displaystyle \frac{dT}{dt}}$$

(12)

For simplification, Equation (12) can be rearranged:

$$\dot{\mathbf{Q}}=I\left(V-U\right)+\left(IT{\displaystyle \frac{\partial U}{\partial T}}\right)+C_p{\displaystyle \frac{dT}{dt}}$$

(13)

Here, V is the cell potential and U is the thermodynamic (open-circuit) potential of a reaction, evaluated at a reference electrode of a given kind. The term (V–U) is a potential drop equivalent to the term IR. Therefore, Equation (13) can be simplified as follows:

$$\dot{\mathbf{Q}}=I^{2}R+\left(T{\displaystyle \frac{\partial U}{\partial T}}\right)+C_p{\displaystyle \frac{dT}{dt}}$$

(14)

In Equation (14), the first term $I^{2}R$ represents ohmic loss, which is irreversible heat loss and is directly related to C-rates, that is, at a higher C-rate, when a battery discharges or charges at a higher current, the losses $I^{2}R$ result in increased battery temperature. Heat generation varies under different operating conditions, such as state of charge (SoC). In Equation (14), the second term, $\left(T{\displaystyle \frac{\partial U}{\partial T}}\right)$, denotes the entropic heat coefficient. This term reflects the variation in open-circuit voltage, U, as a function of temperature T, related to reversible entropic heat. Liu et al. [156] examined thermal generation in lithium-ion batteries during charging and discharging, considering several influencing variables. The investigation concluded that heat generation was largely determined by operational factors such as ambient temperature, state of charge, and C-rates, which impacted both irreversible and reversible heat losses. The term $\left(C_p{\displaystyle \frac{dT}{dt}}\right)$ characterises the rate of change in the internal temperature of the battery over time and is due to the electrochemical reactions and phase changes.

6. Conclusions

The rapid integration of renewable energy sources (RESs) into the evolving electricity market presents a significant risk to power system stability. Previous studies indicated that battery energy storage systems (BESSs), with their fast response capability, were suitable for enhancing grid stability. This review analysed the evolving energy market and the role of battery energy storage systems (BESSs) in frequency regulation. It also examined the technical, regulatory, and economic factors that shaped the operation of BESSs in these markets, as well as the associated challenges and opportunities.

The analysis revealed that the RES market share is projected to reach 61% by 2030, with solar photovoltaic (PV) and wind power collectively contributing 40%. By 2050, it is projected to increase to 88% of RES, with solar photovoltaic (PV) and wind energy collectively accounting for 68%. BESS integration is also forecasted to grow significantly, reaching 1200 GW by 2030, reflecting an increase of 1050 GW since 2024. This upward trajectory is expected to persist, reaching 3100 GW by 2050.

Rapidly changing market conditions, the high integration of RESs and BESSs into frequency control ancillary service (FCAS) operations without fully understanding the BESS limitations in the FCAS market present several challenges. The main challenges include degradation from cycling, thermal management challenges at high C-rates, and system-level risks arising from poor visibility of a BESS’s state of health (SoH) and uncertainties in its state of charge (SoC) and the effects of partial cycling. These factors can significantly impact the economic viability and reliability of BESS deployments, particularly in long-duration or high-frequency-response applications. The effects of mini-, micro- and partial cycles have not yet been thoroughly investigated, potentially disadvantaging BESS operators in capacity payment markets such as the Wholesale Electricity Market (WEM), where BESS operators receive compensation for variable cost elements.

Despite recent advancements, significant hurdles remain in optimising BESS performance and extending lifespan. Future research should focus on the development of advanced degradation models, adaptive control algorithms, and novel hybrid system configurations, such as battery energy storage systems (BESSs) integrated with tailored algorithms and a deeper understanding of cyclic aging mechanisms to address these gaps and unlock further performance improvements.

In conclusion, while BESSs have certain drawbacks, their role in frequency control markets is projected to increase. Ongoing research and regulatory advancements are crucial to overcome current barriers and fully exploit BESSs and low-emission energy systems.