In recent years, the depletion of traditional energy sources coupled with stringent environmental protection regulations has triggered a profound transformation in the energy sector. The demand for battery energy storage systems has surged dramatically. Compared with conventional energy technologies, a well-designed battery energy storage system offers exceptional peak-shaving and valley-filling capabilities, enabling balanced power loads, efficient and stable power supply, and ultimately delivering the required electrical energy to users. To fully leverage these advantages, designers must develop comprehensive battery energy storage systems by integrating appropriate hardware and software components, and by delineating functional modules to enhance electrical safety and improve battery utilization. As technology evolves, practitioners must continuously monitor industry trends and adopt novel techniques to optimize the design and development of battery energy storage systems.
Overview of Lithium-Ion Battery Energy Storage Products
Based on the form of energy storage, technologies can be categorized into mechanical, electromagnetic, and electrochemical storage. Mechanical storage is highly susceptible to external environmental factors; if the ambient conditions do not meet requirements, storage efficiency can be significantly compromised. Electromagnetic storage still exhibits a relatively low level of commercialization. Electrochemical storage, however, holds immense potential. In recent years, energy storage technologies have made notable progress and found prominent applications in various fields. Among these, lithium-ion batteries stand out due to their high energy density, high power output, durability, and rapid response. Because of these characteristics, lithium-ion battery energy storage systems are particularly suitable for applications such as power quality regulation and uninterruptible power supply. To ensure the proper deployment of lithium-ion batteries in diverse scenarios, it is essential to scientifically design the battery energy storage system, equipping it with specialized devices and constructing a robust management framework.
Key Design Points of the Battery Energy Storage System
Overall Architecture
To enhance the operational safety of the energy storage system, the design incorporates a series connection of twelve 280 Ah single cells. The battery management system (BMS) is structured in two hierarchical levels. The master BMS is responsible for data acquisition and status monitoring, continuously collecting parameters such as voltage and current of the battery pack 24/7. These data are transmitted in real time to a terminal module for storage and processing, and the results are displayed on a human-machine interface (HMI) to facilitate state assessment and process supervision. The specifications of the single cell are summarized in Table 1.
| No. | Parameter | Value |
|---|---|---|
| 1 | Rated Capacity | 280 Ah |
| 2 | Rated Voltage | 3.2 V |
| 3 | Operating Voltage Range | 2.8 – 3.65 V |
| 4 | Cell Chemistry | LFP |
| 5 | Weight | 5.4 kg |
| 6 | Standard Charge/Discharge Rate | 0.5 C |
| 7 | Cycle Life | 6000 (0.5 C, EOL 70%) |
| 8 | Cell Consistency Screening | ≥ 4 E |
| 9 | Test Standard | GB/T36276‑2018 |
The master BMS is built upon the PC104 industrial motherboard architecture, which comprises multiple functional module boards stacked together to achieve interconnections. This management system features a compact structure with minimal space occupancy and generally operates at low power. Because the PC104 platform supports a wide variety of hardware peripherals, there is no need to design separate AD conversion modules, CAN communication circuits, Ethernet interfaces, or RS232 communication circuits, thereby simplifying the overall system design. Moreover, the battery energy storage system generates vast amounts of data during operation; the PC104’s superior data processing capability enables rapid acquisition, transmission, and processing of data, with results displayed on the HMI in a timely manner.
During operation, the subordinate BMS automatically collects the voltage, current, and temperature of each individual cell. These measurements are compared against standard values to estimate the state of charge (SOC). The data are then uploaded to the higher-level management system via the CAN bus, where they are uniformly processed and displayed. To efficiently store and manage the data, a Freescale 16‑bit microcontroller, MC9S12XS128, is employed as the system’s central controller.
Within the BMS, the LTC6802‑2 voltage acquisition chip is utilized to monitor the voltage of every single cell in the battery pack in real time. The acquired data are transmitted to the MCU through the SPI interface for subsequent processing and retrieval when needed. Additionally, a Hall-effect current sensor is incorporated; however, careful attention must be paid to its installation location and interconnection with other devices to accurately measure the pack’s current. The sensor’s output is also sent along a designated path to the relevant modules for later use. Furthermore, the 18B20 temperature sensor employs a single‑bus structure to link multiple sensors, enabling the collection and transmission of temperature data. After integrating all these data, the MCU performs specialized processing to assess the battery’s SOC. The results are then transmitted to the host computer via the RS232 serial port, where they are compared with pre‑set alarm thresholds. If any measured value reaches an alarm level, the host computer automatically issues an alert, and the MCU, upon receiving the alarm signal, can control the circuit to initiate a series of protective actions.
Hardware Circuit Design
Voltage Acquisition
For lithium iron phosphate (LFP) batteries, the voltage parameter is particularly critical. The open‑circuit voltage of a cell is closely related to its state of charge and safety condition. By accurately measuring the voltage, the system or operator can determine the battery’s SOC and identify anomalies such as over‑charging or over‑discharging. In this battery energy storage system, the LTC6802‑2 chip is employed to measure the voltage of each individual cell. A single chip can measure and convert the voltages of up to twelve cells within 13 ms. Thanks to the effective collaboration of the chip and other system components, the measurement accuracy is high, and the deviation between the acquired voltage and the actual value is minimal.
Each LTC6802‑2 can simultaneously measure twelve lithium‑ion cell voltages and features a 12‑bit analog‑to‑digital conversion capability, supporting high‑voltage multiplexed inputs. This makes it well‑suited for data acquisition and status monitoring. The chip has a built‑in 4‑bit address field, enabling up to 16 chips to be connected in parallel if necessary. This capability is especially advantageous when monitoring a large number of cells in a battery energy storage system. When multiple chips are used together, they can all enter the operational state simultaneously, minimizing the time required to obtain the voltage readings.
The relationship between the measured voltage and the analog input can be expressed as:
$$ V_{\text{cell}} = \frac{\text{ADC\_code}}{2^{12}} \times V_{\text{ref}} $$
where ADC_code is the digital output from the LTC6802‑2 and V_ref is the reference voltage (typically 5 V). The high resolution ensures precise voltage monitoring.
Current Acquisition
In addition to voltage, the battery energy storage system must also collect current values. The characteristics of current monitoring include a simplified sampling channel — since the battery pack is typically connected in series, the current flowing through each cell is identical, so only the total pack current needs to be measured. Moreover, the current sampling frequency is high, as it significantly affects the SOC estimation and system safety. To maintain optimal battery and system performance, the sampling frequency should be increased to detect and respond to abnormal conditions promptly. Although a shunt resistor can be used to measure current, it introduces thermal losses. To overcome this issue, a Hall-effect sensor is employed to capture the battery current. The Hall sensor integrates multiple modern information technologies and only requires power to the amplification circuit to measure current, offering high accuracy.
The relationship between the Hall sensor output voltage and the measured current is given by:
$$ I = k \cdot (V_{\text{out}} – V_{\text{offset}}) $$
where k is the sensitivity coefficient (in A/V) and V_offset is the zero‑current offset voltage.
Temperature Acquisition
To accurately estimate the battery’s SOC, temperature must be carefully considered. Under different temperature conditions, the amount of charge that can be released from a cell varies significantly. Generally, as temperature rises, the internal electrochemical reactions become more vigorous and rapid, accompanied by energy release. However, if the temperature exceeds the normal range, adverse consequences such as cell swelling, electrolyte leakage, or even combustion may occur. Conversely, when the temperature is too low, the electrolyte may freeze, hindering the electrochemical reaction. Given the critical influence of temperature on battery state, the BMS must effectively monitor temperature parameters — including those of individual cells, the ambient environment, and the battery enclosure. If the monitored temperature exceeds the set threshold, a parallel forced‑air cooling system is activated to reduce the temperature. If the temperature reaches an alarm limit, the system immediately disconnects the charging/discharging circuit and sounds an alarm. In the case of excessively low temperatures, a heater or an ambient temperature increase is triggered to slowly raise the battery temperature to an acceptable level.
Several methods are available for temperature detection, such as thermistors, the 18B20 digital sensor, and dedicated ICs. Each method has its own merits, and the optimal choice depends on factors like ease of integration, cost, and accuracy. Considering both measurement precision and cost, the 18B20 sensor is adopted. Based on extensive field experience, the 18B20 can measure temperatures from -55 °C to 125 °C with a maximum resolution of 12 bits. Once temperature data are collected, they are analyzed and compared to promptly identify abnormal conditions and take corrective actions before excessive temperatures compromise battery safety.
The temperature measurement resolution can be expressed as:
$$ T = \frac{\text{Digital\_value}}{2^{12}} \times 125^{\circ}\text{C} $$
where Digital_value represents the 12‑bit temperature reading from the sensor.
CAN Communication
In the design and development of the battery energy storage system, the CAN (Controller Area Network) communication technology is also utilized. Compared to traditional communication protocols, CAN offers advanced technology, low cost, high reliability, and robust security, providing an excellent channel for data transmission and sharing among different modules. In this system, the microcontroller’s built‑in CAN controller is used together with a Philips TJA1050 CAN transceiver to enhance functionality. To ensure the system’s functional advantages, the CAN communication design strictly adheres to industry standards, such as the Chinese mechanical industry standard JB/T 11138‑2011 concerning the interface and communication of lithium‑ion battery packs.
The CAN module design encompasses initialization, data transmission/reception, and error checking. After initialization, the CAN module enters the operational state and automatically stores key information — such as voltage and temperature — in the form of messages in the transmit buffer. These messages are marked as “pending transmission.” Once a transmission command is received, the module automatically executes the corresponding operation, sending the stored data in a specific format to the designated module. On the data reception side of the CAN module, initialization is equally important. To meet operational requirements, the CAN bus frequency must be set appropriately, and the identifier and check codes must be verified to determine whether the received data are necessary.
The CAN bus bit timing parameters can be calculated using the following formula:
$$ \text{Baud Rate} = \frac{f_{\text{CLK}}}{\text{(BRP+1)} \times (T_{\text{seg1}}+T_{\text{seg2}}+3)} $$
where f_CLK is the system clock frequency, BRP is the baud rate prescaler, and T_seg1 and T_seg2 are the time segment values.
Software Design
The hardware and software must work together seamlessly to enable data acquisition, storage, and processing in the battery energy storage system. In this design, the microcontroller program is developed using the Freescale CodeWarrior compilation environment. The MCU firmware includes various routines such as hardware initialization, data acquisition, communication configuration, fault warning, and data storage.
The host computer monitoring interface is built using VC6.0 combined with Microsoft Foundation Classes (MFC). This interface not only displays real‑time monitoring data and graphical curves in a visual manner but also includes multiple buttons for operators to interact with the system according to actual needs. In addition, a Kinco EV5000 series industrial embedded touch screen is employed as the human‑machine interface (HMI), providing convenient operation and control for personnel. In the event of an abnormality, the HMI triggers an alarm, prompting the operator to take rapid corrective action.
Other Design Considerations
For the design and development of the battery energy storage system, attention must also be paid to the battery cells themselves, as this directly affects system safety. During the design phase, it is essential to compare and select high‑performance lithium‑ion cells. National and industrial standards for battery production have been progressively established, including detailed technical specifications. Manufacturers must strictly follow these standards — for example, during cell production, proper protective measures should be implemented to prevent contamination by foreign metals, thereby creating safe electrical conditions. Furthermore, to ensure that the manufactured cells meet the required performance criteria, designers must be familiar with the cell design requirements, optimize the design approach, and pay attention to details. In particular, for electrical safety, the insulation between the positive and negative terminals is critical. Preferred separator materials such as polypropylene (PP) high‑molecular plastics should be chosen to leverage their insulating properties. For instance, in the insulation design of pouch cells, materials with high mechanical strength and high temperature resistance must be used between the positive/negative tabs and the cell casing. Additionally, lithium‑ion cells should be inspected by qualified personnel to verify that the production process is advanced and that the cell’s insulation performance meets the required standards, so that any issues can be addressed early.
To further optimize the functionality of the battery energy storage system, modern technologies such as artificial intelligence and cloud computing should be integrated into the battery control and monitoring processes, thereby establishing an intelligent operation mode in which tasks are automatically performed by the system. However, when employing intelligent technologies, it is necessary to equip the battery or system with various sensors and establish interconnections among these devices.
The complete battery energy storage system is illustrated below.
Conclusion
The design and development of a battery energy storage system present considerable challenges. To enhance the system’s safety, stability, and functionality — particularly in data acquisition and status monitoring — it is essential to adopt new concepts and methods, optimize both hardware and software designs, and equip the system with advanced devices. In future work, researchers and engineers should continue to explore innovative design approaches for battery energy storage systems, incorporating cutting‑edge technologies such as AI‑based predictive maintenance and cloud‑connected analytics. Only by doing so can the full potential of battery energy storage systems be realized, contributing to a more sustainable and resilient energy infrastructure.