With the rapid proliferation of distributed photovoltaic systems and electric vehicles, the demand for energy storage in low-voltage distribution networks has become increasingly prominent. As a researcher focused on advancing battery energy storage system (BESS) technologies, I have explored the application of cascaded H-bridge topologies to enhance safety and efficiency in residential and commercial areas. Traditional two-level BESS configurations often involve large-scale battery series-parallel connections, which pose significant management challenges and safety risks. In contrast, the cascaded battery energy storage system (CHB-BESS) reduces the battery string scale by an order of magnitude, enabling finer battery management and improved operational safety. This paper presents the design, implementation, and testing of a 380 V CHB-BESS prototype, emphasizing its main circuit and control system architecture. The findings demonstrate that the output waveforms meet expectations, validating the design’s correctness and paving the way for broader adoption in low-voltage grids.
The core of any battery energy storage system lies in its power conversion and control mechanisms. For low-voltage applications, the CHB-BESS topology offers modularity and scalability, which are critical for integrating diverse energy sources and loads. In this work, the main circuit of the CHB-BESS consists of multiple H-bridge submodules connected in series per phase, allowing direct connection to the 380 V grid without bulky transformers. Each submodule incorporates a battery pack, a full-bridge converter, and associated control elements. The system parameters are summarized in Table 1, which outlines key specifications such as rated voltage, power, and submodule count. This design minimizes battery imbalance issues, a common drawback in conventional BESS setups, by localizing energy management at the submodule level.
| Parameter | Value |
|---|---|
| Rated Voltage (UN) | 380 V |
| Rated Power (SN) | 6.6 kW |
| Number of Submodules per Phase (N) | 16 |
| Submodule Battery Voltage (UB) | 25.6 V |
| Grid-Connected Inductance (Ls) | 10 mH |
The submodule circuit, as illustrated in the topological diagram, employs power MOSFETs as switching devices due to their low on-resistance and high-frequency capabilities. For instance, the MOSFETs (e.g., model SFG180N10KF) are rated for 100 V and 150 A at 25°C, providing ample voltage and current margins. The soft-start circuit includes relays and resistors to precharge the DC-link capacitor, ensuring safe initialization. The capacitance C1 is set to 40.8 mF to stabilize the DC voltage and suppress ripples. The power flow in each submodule can be described by the following equations, which govern the operation of the battery energy storage system during charging and discharging modes:
$$P_{\text{sub}} = V_{\text{bat}} \cdot I_{\text{bat}}$$
$$V_{\text{dc}} = \frac{1}{C_1} \int I_{\text{cap}} \, dt$$
where \(P_{\text{sub}}\) is the submodule power, \(V_{\text{bat}}\) is the battery voltage, \(I_{\text{bat}}\) is the battery current, \(V_{\text{dc}}\) is the DC-link voltage, and \(I_{\text{cap}}\) is the capacitor current. These equations highlight the importance of precise control in maintaining stability across the cascaded battery energy storage system.
Control system architecture is pivotal for the reliable performance of the battery energy storage system. In this CHB-BESS, a master-slave configuration is adopted, with the master controller utilizing a heterogeneous structure of DSP and FPGA. The DSP handles high-level power control algorithms, such as phase-locked loop (PLL) synchronization and power reference tracking, while the FPGA manages real-time tasks like modulation and fiber-optic communication. The control cycle is set to 100 μs, with submodule controllers achieving μs-level synchronization via daisy-chained fiber-optic loops. This isolation ensures safety and noise immunity, critical for a battery energy storage system operating in proximity to sensitive loads. The power control algorithm, implemented in the DSP interrupt routine, calculates modulation indices based on grid voltage and current feedback, as expressed below:
$$I_d^* = K_p (P_{\text{ref}} – P_{\text{meas}}) + K_i \int (P_{\text{ref}} – P_{\text{meas}}) \, dt$$
$$I_q^* = K_p (Q_{\text{ref}} – Q_{\text{meas}}) + K_i \int (Q_{\text{ref}} – Q_{\text{meas}}) \, dt$$
where \(I_d^*\) and \(I_q^*\) are the d-axis and q-axis current references, \(P_{\text{ref}}\) and \(Q_{\text{ref}}\) are the active and reactive power references, and \(K_p\) and \(K_i\) are proportional and integral gains. This decoupled control enables independent regulation of active and reactive power, enhancing the flexibility of the battery energy storage system in grid support functions.
Submodule controllers, based on FPGA, execute switching commands received from the master controller and monitor local parameters such as battery voltage and temperature. The fiber-optic communication protocol ensures low-latency data exchange, with each phase forming an independent loop to handle the 16 submodules. This modular approach not only simplifies maintenance but also allows for graceful degradation in case of submodule failure—a significant advantage for a cascaded battery energy storage system in dynamic environments. Key functions of the submodule controller include soft-start/stop sequences, fault detection, and driver signal generation for the MOSFETs. The state machine for submodule operation can be summarized as follows: idle → precharge → active → fault handling. This ensures that the battery energy storage system operates within safe limits, minimizing risks of overcurrent or overvoltage.
To validate the design, extensive testing was conducted on the CHB-BESS prototype. Open-loop voltage output tests were performed first, where the modulation waves were derived from grid synchronization. The d-axis voltage reference \(U_d\) was set to 1 per unit, and the q-axis reference \(U_q\) to 0, resulting in balanced three-phase voltages after inverse Park transformation. The output line voltage closely matched the grid voltage in phase and magnitude, confirming the accuracy of the sampling and modulation processes. For closed-loop power control, tests involved both charging and discharging scenarios at 0.8 times the rated power. During charging, the grid supplied power to the battery energy storage system, with measured current peaks aligning with theoretical values. Similarly, during discharging, the battery energy storage system injected power into the grid, demonstrating stable current waveforms with minimal distortion. The test results, summarized in Table 2, verify the system’s ability to track power references accurately.
| Test Scenario | Power Reference | Measured Current Peak | Phase Shift |
|---|---|---|---|
| Charging | 0.8 × SN | 8 A | 30° |
| Discharging | 0.8 × SN | 8 A | 60° |
The harmonic performance of the battery energy storage system was also evaluated using Fourier analysis. The output voltage total harmonic distortion (THD) was found to be below 5%, meeting IEEE standards for grid-connected inverters. This is attributed to the cascaded multilevel topology, which generates a staircase waveform that approximates a sine wave more closely than two-level converters. The modulation strategy employed carrier phase-shift PWM, where each submodule operates with a phase-shifted carrier signal. The effective switching frequency \(f_{\text{sw,eff}}\) is given by:
$$f_{\text{sw,eff}} = N \cdot f_{\text{sw}}$$
where \(N\) is the number of submodules per phase and \(f_{\text{sw}}\) is the individual submodule switching frequency. For \(N = 16\) and \(f_{\text{sw}} = 10\) kHz, the effective switching frequency reaches 160 kHz, significantly reducing output filter requirements and electromagnetic interference.
Battery management is a critical aspect of any battery energy storage system. In the CHB-BESS, the reduced battery string scale—each submodule handles only 25.6 V—simplifies state-of-charge (SOC) balancing and thermal management. Compared to conventional BESS designs, where hundreds of cells are connected in series, this approach localizes faults and enhances overall system reliability. The SOC balancing algorithm operates at both the submodule and system levels, using measured voltages and currents to compute SOC estimates:
$$\text{SOC} = \text{SOC}_0 – \frac{1}{C_{\text{bat}}} \int I_{\text{bat}} \, dt$$
where \(\text{SOC}_0\) is the initial state of charge and \(C_{\text{bat}}\) is the battery capacity. Active balancing techniques, such as shifting power between submodules, ensure uniform energy distribution across the cascaded battery energy storage system, prolonging battery lifespan.
In conclusion, the cascaded battery energy storage system presented here offers a robust solution for low-voltage distribution networks. By leveraging modular power electronics and advanced control strategies, the CHB-BESS achieves high safety standards and operational flexibility. The design reduces battery series-parallel complexity, enabling more granular management and fault tolerance. Future work will focus on optimizing the control algorithms for larger-scale deployments and integrating renewable energy sources. This research underscores the potential of cascaded topologies in advancing battery energy storage system technologies for smart grid applications.