The increasing penetration of variable renewable energy sources necessitates enhanced flexibility and reliability in modern power grids. Among the various solutions, the Battery Energy Storage System (BESS) has emerged as a critical asset due to its rapid response and versatility. An energy storage battery can provide a spectrum of services, from fast frequency regulation to energy time-shifting. However, these services impose vastly different demands on the battery’s operation, leading to significant differences in potential revenue streams, technical performance requirements, and, most critically, the impact on the battery’s cycle life. A primary challenge lies in the inherent trade-off: while services like frequency regulation can be highly lucrative, they involve frequent, shallow cycling that accelerates battery degradation. Conversely, energy arbitrage involves deeper but less frequent cycles, potentially allowing for a longer operational life but with different revenue dynamics. Therefore, developing a planning methodology that holistically optimizes the deployment of an energy storage battery across its entire lifespan is paramount for maximizing economic viability.
This article proposes a novel whole-life-cycle planning framework for an energy storage battery, explicitly designed to navigate the trade-offs between multiple, sequential applications. The core philosophy is to strategically allocate the battery’s capacity and usage profile across different stages of its life. Specifically, the method proposes that a new, high-capacity energy storage battery is first deployed in a high-revenue, high-stress application like frequency regulation in ancillary service markets. Once the battery’s capacity has degraded to a predetermined threshold, it is then repurposed (or its utilization strategy is shifted) to a less strenuous application, such as energy arbitrage in wholesale energy markets. This sequential multi-scenario approach aims to maximize the cumulative net present value over the battery’s entire operational life, extending its useful service period compared to single-application strategies. To validate the proposed method, a case study is conducted on a modified IEEE 33-bus distribution system, comparing the economic and longevity outcomes against traditional single-service planning approaches.
The planning and operation of an energy storage battery within power systems have been extensively studied. Prior research can be broadly categorized into single-service optimization and multi-service co-optimization. In single-service studies, the focus is often on optimal sizing and siting for specific applications like peak shaving, renewable energy integration, or frequency support. For instance, some models determine BESS capacity to minimize grid upgrade costs or system operating costs, while others evaluate the economic benefits from a singular perspective like arbitrage or government subsidies. A critical shortcoming of these approaches is the frequent omission of battery degradation dynamics. Operating an energy storage battery in a demanding service without considering its lifespan can lead to severely underestimated long-term costs and overstated profitability.
Recent studies have begun to explore the co-optimization of multiple services from a single energy storage battery asset to increase revenue density. Control schemes have been proposed to allocate the battery’s power and energy capacity dynamically to provide several services simultaneously, such as combined peak shaving and frequency regulation. These operational studies have shown the potential for super-linear economic gains. However, they often lack an integrated planning perspective that determines the optimal initial sizing and location of the energy storage battery with the explicit intent of sequential, lifespan-based service provision. This gap motivates the present work, which integrates the long-term degradation model directly into a forward-looking planning optimization model that schedules the battery’s primary application across its usable life.
Multi-Service Application Modeling for Energy Storage Battery
The proposed methodology considers two distinct electricity market applications for the energy storage battery, each with its own operational model and revenue mechanism.
Ancillary Service Market: Frequency Regulation
The primary value of an energy storage battery in ancillary markets lies in its speed and accuracy. For primary frequency regulation (PFR), the BESS responds to system frequency deviations on a timescale of seconds. A standard droop control characteristic is adopted. To prevent unnecessary wear, a deadband is implemented. Following typical grid codes, a deadband of ±0.033 Hz around the nominal frequency ($f_{rated}$ = 50 Hz) is used. The output power of the energy storage battery is determined by a piecewise linear P-f characteristic.
The frequency regulation droop coefficient $\sigma$ and the charging/discharging power $P^{\xi}_{ch/dis,i,t}$ for BESS $i$ at time $t$ in scenario $\xi$ are given by:
$$
\sigma = – \frac{\Delta f / f_{rated}}{\Delta P / P^{bat,max}_{i}}
$$
$$
P^{\xi}_{ch/dis,i,t} = \pm P^{bat,max}_{i} \cdot \frac{\Delta f(t) / f_{rated}}{\sigma}
$$
where $\Delta f(t)$ is the instantaneous frequency deviation, $\Delta P$ is the corresponding power adjustment, and $P^{bat,max}_{i}$ is the rated power capacity of BESS $i$ allocated for ancillary services.
In many regulation markets, compensation consists of two parts: a capacity payment for being available and a performance payment for accurate response. This model focuses on the capacity payment for the first life stage. The annual revenue from providing frequency regulation reserve is:
$$
\text{Profit}_1 = C_{reserve} \times C^{bat,max}_{i}
$$
where $C_{reserve}$ is the market price for regulation capacity (\$/MW/year) and $C^{bat,max}_{i}$ is the energy capacity (MWh) of BESS $i$ in its first life cycle stage.
Energy Arbitrage Market
In the energy arbitrage market, the energy storage battery exploits temporal price differences. It charges during low-price (off-peak) periods and discharges during high-price (peak) periods. This is a cyclical, energy-intensive service typically scheduled on an hourly basis over a 24-hour horizon.
The daily revenue from energy arbitrage for BESS $i$ in scenario $\zeta$ is calculated as:
$$
\text{Profit}_2 = \sum_{h} \left( \lambda_{peak}(h) \cdot P^{\zeta}_{dis,i,h} – \lambda_{offpeak}(h) \cdot P^{\zeta}_{ch,i,h} \right)
$$
where $P^{\zeta}_{dis,i,h}$ and $P^{\zeta}_{ch,i,h}$ are the discharge and charge powers during hour $h$, and $\lambda_{peak}(h)$ and $\lambda_{offpeak}(h)$ are the corresponding market energy prices (\$/MWh).
Whole-Life-Cycle Planning Model for Energy Storage Battery
The overarching framework is illustrated conceptually as a sequential decision process. The planning model determines the optimal sizing ($C^{bat,max}_{i}$), siting, and the capacity threshold for transitioning from the first to the second life cycle stage. The objective is to maximize the total net present value (NPV) over the planning horizon, encompassing revenues, operational costs, degradation costs, and initial investment.
The objective function is formulated as follows:
$$
\begin{aligned}
\max \quad & \Bigg\{ \left[ \sum_{y=1}^{Y^{1st}} \left( \text{Profit}_1 – \frac{365}{S} \sum_{\xi=1}^{S} \sum_{t} (C_{opm} + C_{deg} + C_{penalty}) \right) \right] + \\
& \left[ \sum_{y=Y^{1st}+1}^{Y^{2nd}} \left( \frac{365}{S’} \sum_{\zeta=1}^{S’} (\text{Profit}_2 – \sum_{h} (C’_{opm} + C’_{deg})) \right) \right] – C_{inv} \Bigg\}
\end{aligned}
$$
where $Y^{1st}$ and $Y^{2nd}$ are the durations (in years) of the first and second life stages, $S$ and $S’$ are numbers of representative daily scenarios for each stage, and $t$ and $h$ are time intervals within a day for each respective service.
The detailed cost and penalty components are defined below:
- Operating & Maintenance (O&M) Cost: Assumed proportional to the installed power capacity.
$$ C_{opm} = \sum_i (P^{bat,max}_{i} \cdot C_{O,BESS}), \quad C’_{opm} = \sum_i (P’^{bat,max}_{i} \cdot C_{O,BESS}) $$
where $C_{O,BESS}$ is the O&M cost coefficient (\$/MW/year). - Degradation Cost: Models the wear and tear on the energy storage battery. It is calculated based on the effective “aging” per cycle relative to the battery’s rated cycle life $\Gamma_R$.
$$ C_{deg} = (C_{inv} / \Gamma_R) \times \sum_{\xi} \sum_{t} \Gamma_{eff}(t) $$
$$ C’_{deg} = (C’_{inv} / \Gamma’_R) \times \sum_{\zeta} \sum_{h} \Gamma’_{eff}(h) $$
Here, $\Gamma_{eff}$ represents the effective cycle aging equivalent for a specific operating point (often a function of depth of discharge, charge rate, and temperature). $C’_{inv}$ is the remaining capital value of the battery at the beginning of the second stage. - Penalty Cost (for Regulation): Accounts for the cost of imperfect tracking of the regulation signal. If the actual power $P^{bat}$ deviates from the required signal $P^{reg}$, a penalty is incurred.
$$ C_{penalty} = \frac{365}{S} \sum_{\xi} \sum_{t} \left( \rho \times | \Delta P^{\xi}_{dev}(t) | \times \Delta t \right) $$
where $\Delta P^{\xi}_{dev}(t) = P^{\xi}_{reg}(t) – P^{\xi}_{bat}(t)$ and $\rho$ is the penalty price (\$/MWh). - Investment Cost: The annualized capital cost of the energy storage battery.
$$ C_{inv} = \frac{d(1+d)^{(Y^{1st}+Y^{2nd})}}{(1+d)^{(Y^{1st}+Y^{2nd})} – 1} \times \sum_i (C_E \cdot C^{bat,max}_{i}) $$
where $d$ is the discount rate and $C_E$ is the unit energy cost of the battery (\$/kWh).
The model is subject to a comprehensive set of constraints including: power and energy balance equations for the distribution network; BESS operational constraints (state-of-charge limits, charge/discharge power limits, efficiency); logical constraints ensuring sequential stage operation (e.g., second stage begins only when first-stage capacity fade reaches a set point); and network security constraints (voltage limits, branch flow limits).
Case Study and Analysis
The proposed whole-life-cycle planning method for the energy storage battery is tested on a modified IEEE 33-bus radial distribution system. Two wind farms are integrated at buses 19 and 32 to introduce variability. Key technical and economic parameters for the lithium-ion energy storage battery are summarized in Table 1.
| Parameter | Value |
|---|---|
| BESS Operational Constraints | |
| State-of-Charge (SOC) Limits | 20% (min), 80% (max) |
| Initial SOC | 50% |
| Charge/Discharge Efficiency | 75% (each way) |
| Self-discharge Rate | 0.24% per day |
| BESS Cost Parameters | |
| Capital Cost ($C_E$) | 325 \$/kWh |
| O&M Cost ($C_{O,BESS}$) | 20 \$/MW/day |
| First-Life Stage (Regulation) | |
| Control Time Step / Horizon | 15 seconds / 24 hours |
| Droop Coefficient ($\sigma$) | 5% |
| Regulation Capacity Price ($C_{reserve}$) | 120,000 \$/MW/year |
| Second-Life Stage (Arbitrage) | |
| Control Time Step / Horizon | 1 hour / 24 hours |
| Peak / Off-Peak Price Ratio | Approx. 2:1 |
Table 2 outlines the overarching planning parameters.
| Parameter | Value |
|---|---|
| Total Investment Budget | 1,500,000 \$ |
| Planning Horizon | 15 years |
| Discount Rate ($d$) | 5% |
| Regulation Performance Penalty ($\rho$) | 20 \$/MWh |
Four distinct planning cases are analyzed for comparison:
- Case 1: BESS planned and used only for energy arbitrage over its full life.
- Case 2: BESS planned and used only for frequency regulation over its full life.
- Case 3: BESS planned for and operated in simultaneous provision of both services from day one.
- Case 4: The proposed sequential whole-life-cycle method (first regulation, then arbitrage after capacity fade).
The optimization results for sizing, location, lifetime, and economics are compiled in Table 3.
| Metric | Case 1 (Arbitrage Only) | Case 2 (Regulation Only) | Case 3 (Simultaneous) | Case 4 (Sequential) |
|---|---|---|---|---|
| Application | Arbitrage | Regulation | Both | Regulation → Arbitrage |
| Optimal Locations (Bus) | 31, 18 | 29, 14 | 30, 13 | 30, 14 |
| BESS Capacity (MWh) | 3.15, 1.46 | 3.09, 1.52 | 3.26, 1.32 | 2.77, 1.84 |
| Useful Lifetime (years) | 12.5, 12.5 | 4.35, 4.4 | 2.16, 2.1 | 10.66, 10.8 |
| Investment Cost (\$) | 813,330 | 1,210,299 | 1,339,614 | 889,049 |
| Total O&M Cost (\$) | 28,904 | 10,191 | 4,946 | 19,958 |
| Regulation Revenue (\$) | – | 2,563,531 | 1,246,161 | 589,407 |
| Arbitrage Revenue (\$) | 2,300,088 | – | 202,407 | 2,438,818 |
| Net Present Value (\$) | 1,380,864 | 1,343,040 | 1,784,347 | 2,119,217 |
The results demonstrate the significant impact of application choice on the energy storage battery’s lifespan and economics. Case 1 (sole arbitrage) yields the longest battery life (12.5 years) due to gentle, once-daily cycling but offers the lowest NPV. Case 2 (sole regulation) provides high annual revenue but causes extremely rapid degradation, exhausting the battery in about 4.4 years, which ultimately limits its total financial return. Case 3 (simultaneous services) represents the worst scenario for battery longevity (just over 2 years), as it combines the high cycle frequency of regulation with the deep cycles of arbitrage, leading to accelerated aging that outweighs the combined revenue gains.
In contrast, the proposed sequential method in Case 4 achieves a superior balance. It captures high-value revenue from regulation during the battery’s “prime” high-capacity life. Once capacity degrades to a level unsuitable for reliable regulation service, it is transitioned to the less demanding arbitrage service, extending its useful economic life to over 10 years. This strategic phasing allows the energy storage battery to generate substantial revenue in its first life stage while still securing a long, profitable second life. Consequently, Case 4 achieves the highest overall Net Present Value, outperforming the best single-service case (Case 1) by approximately 34.8% and the simultaneous service case (Case 3) by 15.8%. This clearly validates the economic advantage of a planned, lifecycle-aware approach to energy storage battery deployment.
Conclusion
This article presents a comprehensive whole-life-cycle planning methodology for grid-scale energy storage battery systems. The method addresses a critical gap in traditional planning by explicitly modeling the sequential deployment of a battery asset across different service markets—first in high-stress, high-reward ancillary services, followed by repurposing for energy arbitrage after significant capacity fade. The core optimization model integrates detailed technical models for each service, a battery degradation cost model, and a full economic assessment within a unified framework aimed at maximizing net present value over the system’s lifespan.
Simulation results on a standard test network underscore the limitations of single-application or simultaneous multi-service strategies, which either underutilize the battery’s revenue potential or drastically shorten its life. The proposed sequential strategy, however, demonstrates a pathway to significantly enhance the economic viability of energy storage battery investments. By aligning the battery’s operational duty cycle with its aging characteristics, planners can extract greater total value, improve asset utilization, and contribute to a more sustainable and cost-effective energy storage integration strategy. Future work will explore more granular degradation models, the integration of third-life applications (e.g., behind-the-meter or recycling), and the planning of heterogeneous energy storage battery fleets within complex market environments.