Abstract
The integration of renewable energy sources into power grids has significantly reduced system inertia, challenging frequency stability. This paper proposes a novel adaptive primary frequency control strategy for battery energy storage system (BESS), combining trend control, inertia-droop control, and MPPT-inspired optimization. By dynamically adjusting system coefficients based on grid frequency acceleration, rate of change, and deviation, the strategy enhances frequency stabilization while preventing battery overcharge/discharge. A hybrid SOC estimation method, merging ampere-hour integration and unscented Kalman filtering, further improves accuracy. Simulation results demonstrate superior performance in reducing frequency deviations (20.7–50.0% improvement over benchmarks) and SOC estimation errors (0.01% average error). The integration of MPPT principles ensures optimal power extraction and utilization, aligning frequency regulation with energy efficiency.
1. Introduction
Modern power grids face escalating instability due to high renewable penetration and reduced inertia. Traditional battery energy storage system (BESS) control strategies often exhibit poor adaptability, fixed coefficients, and inadequate SOC estimation, limiting their effectiveness in primary frequency regulation. Inspired by Maximum Power Point Tracking (MPPT) techniques—widely used in solar systems to optimize energy harvest—this work introduces adaptive mechanisms to maximize battery energy storage system (BESS) efficiency during frequency events. By dynamically adjusting trend, inertia, and droop coefficients, the proposed strategy mimics MPPT’s real-time optimization, ensuring rapid response and minimal frequency deviation.
Key Contributions:
- MPPT-inspired adaptive control: Coefficients adapt to grid dynamics, akin to MPPT’s power optimization.
- Hybrid SOC estimation: Combines ampere-hour integration and unscented Kalman filtering for high accuracy.
- Frequency-acceleration-driven trend control: Mitigates frequency恶化 and accelerates recovery.
2. System Model and MPPT Integration
2.1 BESS Primary Frequency Regulation Model
The battery energy storage system (BESS) interacts with thermal units and loads to stabilize grid frequency. The transfer functions for thermal units and battery energy storage system (BESS) IS defined as:
Thermal Unit Governor:Gg(s)=11+sTgGg(s)=1+sTg1
Thermal Turbine:Gth(s)=1+sTrh(1+sTr)(1+sTt)Gth(s)=(1+sTr)(1+sTt)1+sTrh
BESS Dynamics:Gc(s)=11+sTcGc(s)=1+sTc1
Where Tg,Trh,Tr,Tt,TcTg,Trh,Tr,Tt,Tc are time constants.
2.2 MPPT-Inspired Adaptive Control
MPPT algorithms adjust operating points to extract maximum power under varying conditions. Similarly, this strategy adapts battery energy storage system (BESS) coefficients to maximize frequency regulation efficiency:ΔPc=MbΔfΔt+KbΔf+Tbd2fdt2ΔPc=MbΔtΔf+KbΔf+Tbdt2d2f
- Trend coefficient TbTb: Adjusted via frequency acceleration d2fdt2dt2d2f, analogous to MPPT’s gradient ascent.
- Inertia coefficient MbMb: Tuned using frequency rate dfdtdtdf.
- Droop coefficient KbKb: Optimized based on frequency deviation ΔfΔf.
Adaptive Rules:Tb=(T0+Ti∣d2fdt2∣)⋅Pc/dTb=(T0+Ti∣dt2d2f∣)⋅Pc/dMb=(M0+Mi∣dfdt∣)⋅Pc/dMb=(M0+Mi∣dtdf∣)⋅Pc/dKb=(K0+Ki∣Δf∣)⋅Pc/dKb=(K0+Ki∣Δf∣)⋅Pc/d
Here, Pc/dPc/d represents charge/discharge constraints derived from logistic curves (Table 1).
Table 1: SOC-Dependent Charge/Discharge Constraints
| SOC Range | Discharge (PdPd) | Charge (PcPc) |
|---|---|---|
| SOC≤0.2SOC≤0.2 | 0 | Logistic curve |
| 0.2<SOC<0.30.2<SOC<0.3 | S-curve transition | – |
| SOC≥0.3SOC≥0.3 | Full discharge | – |
| SOC≤0.7SOC≤0.7 | – | Full charge |
| 0.7<SOC<0.80.7<SOC<0.8 | – | S-curve transition |
| SOC≥0.8SOC≥0.8 | – | 0 |
3. Hybrid SOC Estimation with MPPT Parallels
Accurate SOC estimation prevents overcharge/discharge and optimizes battery energy storage system (BESS). Traditional methods (e.g., ampere-hour integration) accumulate errors, while Kalman filters face initialization inaccuracies. Our hybrid approach blends both, akin to MPPT’s multi-objective optimization:
State Equations:[St+1U1,t+1U2,t+1]=A[StU1,tU2,t]+BIt+wtSt+1U1,t+1U2,t+1=AStU1,tU2,t+BIt+wtA=[1000e−Δt/R1C1000e−Δt/R2C2],B=[−η/SmaxR1(1−e−Δt/R1C1)R2(1−e−Δt/R2C2)]A=1000e−Δt/R1C1000e−Δt/R2C2,B=−η/SmaxR1(1−e−Δt/R1C1)R2(1−e−Δt/R2C2)
Weighted SOC Estimation:S={bSah+cSukf,ΔUah⋅ΔUukf<0bSah−cSukf,ΔUah⋅ΔUukf>0S={bSah+cSukf,bSah−cSukf,ΔUah⋅ΔUukf<0ΔUah⋅ΔUukf>0b+c=1,c=Δt60b+c=1,c=60Δt
This method transitions from ampere-hour dominance (high initial accuracy) to Kalman filtering (low long-term drift), mirroring MPPT’s adaptive weighting of environmental variables.
4. Simulation Results and MPPT Performance Enhancement
4.1 Step Load Disturbance
A 60 MW load surge at 10 s tests battery energy storage system (BESS) response. The proposed strategy outperforms benchmarks (Table 2), reducing frequency deviations by 20.7–40.0% through MPPT-like coefficient adaptation.
Table 2: Frequency Deviation Comparison
| Method | Max Deviation (Hz) | Steady-State Deviation (Hz) |
|---|---|---|
| No BESS | -0.096 | -0.063 |
| Variable-K | -0.082 | -0.058 |
| Adaptive | -0.067 | -0.053 |
| Proposed | -0.061 | -0.049 |
SOC Estimation Accuracy:
| Method | Max Error (%) | Avg Error (%) |
|---|---|---|
| Ampere-hour | 0.42 | 0.21 |
| Kalman Filter | 0.88 | 0.13 |
| Proposed | 0.09 | 0.01 |
4.2 Continuous Load Disturbance
For 10-minute and 30-minute perturbations, the strategy maintains frequency stability while balancing SOC consumption (Table 3). MPPT principles ensure minimal energy waste during regulation.
Table 3: Continuous Load Performance
| Scenario | Avg Deviation (Hz) | Peak-Peak (Hz) | SOC Fluctuation |
|---|---|---|---|
| 10-min (Proposed) | 0.038 | 0.126 | 0.052 |
| 30-min (Proposed) | 0.035 | 0.126 | 0.048 |
5. Conclusion
This work presents a comprehensive adaptive strategy for battery energy storage system (BESS) primary frequency regulation, integrating MPPT-inspired optimization to enhance adaptability and efficiency. Key outcomes include:
- MPPT-like Coefficient Adaptation: Trend, inertia, and droop coefficients dynamically adjust to grid conditions, reducing frequency deviations by 20.7–50.0%.
- Hybrid SOC Estimation: Merging ampere-hour and Kalman filtering achieves 0.01% average error, critical for preventing battery degradation.
- MPPT-Enhanced Energy Utilization: By optimizing出力 during frequency events, the strategy aligns with MPPT’s goal of maximizing system efficiency.
Future work will explore deeper integration of MPPT algorithms with multi-objective frequency regulation, further bridging renewable energy optimization and grid stability.