Introduction
The rapid integration of renewable energy sources, particularly photovoltaic (PV) and wind power, into modern power grids has introduced challenges related to intermittency, grid stability, and energy curtailment. Large-scale battery energy storage system (ESS) have emerged as a critical solution to mitigate these challenges by enabling peak shaving, frequency regulation, and enhanced grid flexibility. This study focuses on optimizing the capacity configuration and control strategies of battery energy storage system (ESS) to maximize economic benefits while ensuring system reliability.
Modeling of Photovoltaic-Energy Storage Systems
A typical grid-connected PV-ESS structure comprises PV arrays, bidirectional DC-DC converters, energy storage units, and inverters. The output power of a PV array is modeled using the single-diode equivalent circuit:Ipv=Iph−I0{exp(q(Upv+IpvRs)AkT)−1}−Upv+IpvRsRshIpv=Iph−I0{exp(AkTq(Upv+IpvRs))−1}−RshUpv+IpvRs
where IphIph is the photocurrent, I0I0 is the diode saturation current, RsRs and RshRsh are series and shunt resistances, and AA is the ideality factor. battery energy storage system (ESS) dynamics are governed by the state of charge (SOC):SOC(t)=(1−σ)SOC(t−1)+PcΔtηcE(charging)SOC(t)=(1−σ)SOC(t−1)+EPcΔtηc(charging)SOC(t)=(1−σ)SOC(t−1)−PdΔtEηd(discharging)SOC(t)=(1−σ)SOC(t−1)−EηdPdΔt(discharging)
where Pc/PdPc/Pd are charging/discharging power, ηc/ηdηc/ηd are efficiencies, σσ is self-discharge rate, and EE is rated capacity. Operational constraints include:SOCmin≤SOC(t)≤SOCmax,0≤Pc,Pd≤PratedSOCmin≤SOC(t)≤SOCmax,0≤Pc,Pd≤Prated
Economic Analysis of Energy Storage Systems
The lifecycle cost of battery energy storage system (ESS) encompasses installation, operation, maintenance, and residual value. Key cost components are summarized below:
| Cost Category | Description |
|---|---|
| System Cost | Energy (CeCe) and power (CpCp) costs based on storage technology (Table 1). |
| Power Conversion Cost | Cpc=λpc⋅CpCpc=λpc⋅Cp, where λpcλpc is the power-to-energy ratio. |
| Construction Cost | 3–10% of system cost: Ctc=λt⋅CeCtc=λt⋅Ce, Ctp=λt⋅CpCtp=λt⋅Cp. |
| Operation & Maintenance | 1–10% of system cost: Cyc=λy⋅CeCyc=λy⋅Ce, Cyp=λy⋅CpCyp=λy⋅Cp. |
| Residual Value | 3–40% of system cost: Crc=λr⋅CeCrc=λr⋅Ce, Crp=λr⋅CpCrp=λr⋅Cp. |
Table 1: Energy and Power Costs of Storage Technologies (Unit: 10,000 CNY/MWh)
| Technology | Energy Cost (CeCe) | Power Cost (CpCp) |
|---|---|---|
| Pumped Hydro | 120–170 | 550–700 |
| Lithium Iron Phosphate (LFP) | 150–230 | 180–300 |
| Vanadium Redox Flow | 350–420 | 800–1500 |
| Lead-Carbon | 200–240 | 400–600 |
| Supercapacitors | 60,000–80,000 | 1000–1500 |
The levelized cost of energy (LCOE) for battery energy storage system (ESS) is calculated as:Cd=CsumEsum=Cd=EsumCsum=
where NN is cycle life, KDODKDOD is depth of discharge, ηη is efficiency, and ϵϵ is capacity retention. For LFP batteries, CdCd ranges from 0.62–0.82 CNY/kWh. Sensitivity analysis shows that reducing material costs by 10% or increasing cycle life to 7,000 cycles can lower CdCd to 0.3 CNY/kWh, enabling commercial viability.
Optimal Capacity Configuration
A multi-objective optimization model maximizes annual revenue while minimizing PV curtailment:maxF=ω1M+ω2CmaxF=ω1M+ω2C
where MM is curtailed PV energy, CC is economic benefit, and ω1/ω2ω1/ω2 are weights. Constraints include power balance:PPV(t)+Pess(t)+Pgrid(t)=Pload(t)PPV(t)+Pess(t)+Pgrid(t)=Pload(t)
and SOC limits (SOCmin=20%SOCmin=20%, SOCmax=80%SOCmax=80%). Using an improved particle swarm optimization (PSO) algorithm, the optimal power/capacity ratio for LFP-based ESS in Qinghai’s Haixi region was determined as 1:4 (840 MW power, 3,600 MWh capacity), reducing curtailment from 10.6% to 0.3%. Economic simulations (Table 2) indicate that a purchase price ≤0.4 CNY/kWh ensures an 8% annual return over a 12-year lifecycle.
Table 2: Annual Benefits of ESS Under Different Purchase Prices (Unit: 10,000 CNY)
| Plant | 0 CNY/kWh | 0.3 CNY/kWh | 0.5 CNY/kWh |
|---|---|---|---|
| A | 13,085.75 | 8,048.75 | 4,690.75 |
| B | 19,344.16 | 11,898.16 | 6,934.16 |
| C | 5,348.09 | 3,289.49 | 1,917.09 |
| D | 4,323.99 | 2,659.59 | 1,549.99 |
| E | 7,965.24 | 4,899.24 | 2,855.24 |
Variable-Parameter Power Difference Control Strategy
Traditional constant-parameter strategies often neglect SOC constraints, risking overcharge/discharge. The proposed variable-parameter strategy divides load into three zones and SOC into five states:
- Load Zones:
- Zone I (Charging): Pload≤P2Pload≤P2
- Zone II (Inactive): P2<Pload<P1P2<Pload<P1
- Zone III (Discharging): Pload≥P1Pload≥P1
- SOC States:
- Extreme High (SOC≥SmaxSOC≥Smax), High (Sup≤SOC<SmaxSup≤SOC<Smax), Normal (Sdown≤SOC<SupSdown≤SOC<Sup), Low (Smin≤SOC<SdownSmin≤SOC<Sdown), Extreme Low ().
Control parameters a,b,c,da,b,c,d adjust charging/discharging power in High/Low SOC states:Pc=a(P2−Pload)(High SOC)Pc=a(P2−Pload)(High SOC)Pd=b(Pload−P1)+(1−b)min(Prated,Pload−P2Pd=b(Pload−P1)+(1−b)min(Prated,Pload−P2
An adaptive PSO algorithm optimizes Sup,Sdown,a,b,c,dSup,Sdown,a,b,c,d every 15 minutes using ultra-short-term forecasts. Case studies in Qinghai demonstrated a 15% reduction in load variance and zero SOC violations compared to constant-parameter strategies.
Case Study: Qinghai Haixi Region
A 200 MW PV plant with 50 MW/50 MWh LFP battery energy storage system (ESS) was simulated. Key results include:
- SOC Management: The variable-parameter strategy maintained SOC within 20–80%, whereas constant-parameter strategies caused frequent violations.
- Peak Shaving: Load variance decreased by 22% during peak hours.
- Economic Benefit: At 0.4 CNY/kWh purchase price, the project achieved a 12.7% internal rate of return.
Conclusion
This study establishes a framework for optimizing large-scale energy storage systems, emphasizing economic viability and adaptive control. Key findings include:
- LFP Batteries: Optimal power/capacity ratio of 1:4 reduces PV curtailment to 0.3% with LCOE ≤0.4 CNY/kWh.
- Variable-Parameter Control: Enhances SOC management and peak-shaving efficacy, validated through real-world simulations.
- Policy Implications: Subsidies or market mechanisms are essential to achieve grid parity for battery energy storage system (ESS) in high-renewable grids.
Future work will explore hybrid storage systems and AI-driven predictive controls to further improve battery energy storage system (ESS) performance.