Abstract: This paper addresses the issue of large voltage errors when using the state of charge (SOC) as the equalization variable for multi-energy storage battery packs at the end of charging and discharging in DC microgrids. A segmented voltage-SOC equalization control strategy for multi-energy storage battery packs is proposed. Based on the variation curve of open-circuit voltage and SOC, this strategy fully exploits the advantages of low computational complexity in voltage equalization and effective SOC equalization. At the beginning and end of charging and discharging, voltage is used as the equalization variable, while SOC is used during the voltage plateau phase. Droop control is employed to achieve equalization control among multiple battery packs. To verify the effectiveness of the equalization strategy, a simulation model is built in MATLAB/Simulink. Simulation results demonstrate that the battery packs can maintain good voltage and SOC balance after charging and discharging, effectively improving consistency among multiple battery packs.
1. Introduction
Locally configured energy storage systems (ESS) ensure the stability of microgrid operations. However, to maximize the allocation efficiency of load power resources among various energy storage units, the energy supplied by each unit depends on its output range [6]. Various control strategies have been proposed in the literature to balance SOC among energy storage units, but they have limitations. For instance, some methods only consider discharging factors and neglect charging operations, leading to potential overcharging or over-discharging issues due to the introduction of virtual impedance in droop control. Others rely on real-time SOC measurements for adjusting droop control parameters, but SOC estimation accuracy can affect battery equalization control, especially at the beginning and end of charging and discharging when voltage balance among batteries cannot be well maintained solely based on SOC.
2. Literature Review
Table 1 summarizes the key features and limitations of existing equalization control strategies for energy storage in microgrids.
| Reference | Strategy | Advantages | Limitations |
|---|---|---|---|
| [7] | SOC-based droop control with virtual impedance | Achieves SOC balance | Limited to discharging scenarios, risks of overcharging/discharging |
| [8] | Adaptive droop control with SOC adjustment | Fast SOC balance, improved safety | Complexity in adjusting virtual impedance |
| [9-10] | Multi-layer communication-based droop control | Ensures SOC consistency among storage elements | Increased complexity and reduced reliability due to communication elements |
| [11] | Communication-less droop control | Simplifies control, reduces complexity | Limited to specific load power and initial SOC errors |
3. Proposed Voltage-SOC Segmented Equalization Control Strategy
3.1 Traditional I-U Droop Control Principle
The equivalent circuit for n paralleled energy storage units (DESUi, i=1,2,…,n). The droop control expression for DC microgrids is given by:
Uref_i = Uref – Ii Ri (1)
Where Uref_i is the output voltage reference, Uref is the rated bus voltage, and Ri is the droop coefficient.
3.2 Voltage-SOC Segmented Equalization Control Strategy
Due to the inherent errors in SOC estimation using the coulomb counting method, this paper proposes a voltage-SOC segmented equalization control strategy. Table 2 outlines the control stages and primary equalization variables.
| Stage | SOC Range | Primary Equalization Variable |
|---|---|---|
| Charging/Discharging Initial Phase | 0% < SOCave ≤ 20% | Voltage (CCV) |
| Voltage Plateau Phase | 20% < SOCave < 80% | ΔSOC |
| Charging/Discharging Final Phase | 80% ≤ SOCave < 100% | Voltage (CCV) |
For the voltage-based equalization at the beginning and end of charging/discharging, the droop coefficient Ri adjustment strategy is:
Ri = R0i (CCVave / CCVi)r1 (Ii > 0) or R0i (CCVi / CCVave)r2 (Ii < 0) (6)
Where R0i is the initial droop coefficient, CCVave is the average open-circuit voltage of the energy storage system, r1 and r2 are balancing factors, Ii > 0 indicates discharging, and Ii < 0 indicates charging.
4. Simulation and Results
4.1 Simulation Setup
Parameters for the simulation model are listed in Table 3.
| Parameter | Value |
|---|---|
| DC bus voltage | 380 V |
| Load power | 2500 W |
| Battery pack rated capacity | 0.1 Ah |
| Fully charged battery voltage | 221.1576 V |
| Battery cut-off voltage | 142.5 V |
| Battery pack rated discharge current | 4.3478 A |
4.2 Simulation Results
Three operating conditions were designed to validate the proposed strategy:
- Condition 1: Average SOC in the voltage balance range (80%-100%), batteries discharging.
- Condition 2: Average SOC between 20% and 80%, batteries discharging.
- Condition 3: Average SOC in the initial range (0%-20%), batteries charging.
Table 4: SOC Comparison for Condition 2
| Group | Initial SOC (%) | SOC after 10s (%) | ΔSOC (%) |
|---|---|---|---|
| DESU1 | 67 | 65.035 | -1.965 |
| DESU2 | 70 | 66.968 | -3.032 |
| DESU3 | 75 | 69.898 | -5.102 |
Results show that the proposed strategy effectively balances SOC among battery packs, especially during the critical charging and discharging phases, maintaining good voltage and SOC balance.
5. Conclusion
This paper presents a segmented voltage-SOC equalization control strategy tailored for lithium-ion battery packs in DC photovoltaic energy storage microgrids. By leveraging the relationship between open-circuit voltage and SOC, the strategy minimizes errors from SOC estimation and maintains better voltage balance at the beginning and end of charging/discharging, enabling more precise equalization control among DESUs. Simulations in MATLAB/Simulink validate the effectiveness of the proposed strategy.