Battery stacks based on lithium ion (Li-ion) battery cells are widely used in various applications, such as hybrid electric vehicles (HEV), electric vehicles (EV), renewable energy storage for future use, and power grid energy storage for various purposes (power grid stability, peak shaving and time shift of renewable energy, etc). This article will show you the technical developments of measuring the state of charge (SOC) and the state of health (SOH) of battery cells, as well as the relevant solutions introduced by ADI.
Accurate estimation of battery SOC can prevent batteries from overcharge and discharge
In the application of electric vehicles and energy storage systems, it is very important to measure the state of charge (SOC) of battery cells. SOC is defined as available capacity (in Ah), expressed as a percentage of rated capacity. SOC parameters can be regarded as a thermodynamic quantity, which can be used to evaluate the potential energy of the battery. It is also important to estimate the state of health (SOH) of the battery. SOH measures the battery's ability to store and deliver electrical energy compared with a new battery.
However, it is a very complicated task to determine the SOC of a battery, which is related to the type of battery and its application. Therefore, in recent years, a lot of R&D efforts have been made to improve the SOC estimation accuracy. Accurately estimating SOC is one of the main tasks of battery management systems, which is helpful to improve system performance and reliability and prolong battery life.
In fact, accurate estimation of the SOC of a battery can avoid unpredicted system interruption and prevent the battery from overcharge and discharge (which may lead to permanent damage of the battery, depending on the internal structure of the battery). However, battery charging and discharging involve complex chemical and physical processes, and it is not easy to accurately estimate SOC under different operation conditions.
The general method to measure SOC is to measure the amount of electricity (coulombs) and current flowing into and out of the cell stack under all operating conditions, as well as the voltage of each battery cell in the stack, and then use this data and the previously loaded cell pack data that are exactly the same as the monitored cells to obtain an accurate estimate of SOC. Other data required for this calculation includes battery temperature, battery mode (whether the battery is charged or discharged at the time of measurement), cell age, and other relevant cell data obtained from the cell manufacturers.
Sometimes, data about the performance characteristics of Li-ion batteries under different operating conditions can be obtained from manufacturers. After determining the SOC, the system is responsible for updating the SOC during subsequent operation, which is basically counting the amount of electricity (coulombs) flowing into and out of the cell. If the accuracy of the initial SOC is not high enough, or is influenced by other factors, such as cell self-discharge and leakage effect, the accuracy of this method may not be satisfactory.'
Equivalent circuit model for a lithium-ion battery pack
Evaluation platform measures SOC and SOH of typical energy storage modules
In order to measure the SOC and SOH of typical energy storage modules, it involves the design and development of a coulomb counting evaluation platform. The evaluation platform is mainly composed of the following parts: hardware system (including MCU and required interfaces and peripherals), embedded software (which can be used to realize SOC and SOH algorithms), application software based on PCs (which is used as the user interface for system configuration), data display, and analysis.
The evaluation platform periodically measures the voltage value of each battery cell and the current and voltage of the battery pack through appropriate ADC and sensors, and runs the SOC estimation algorithm in real time. This algorithm will use the measured voltage and current values, some other data collected by temperature sensors and/or provided by PC-based software programs (for example, the manufacturer's specifications from the database). The output of an SOC estimation algorithm will be sent to a PC graphical user interface for dynamic display and database updates. SOC and SOH estimation mainly use three methods, including the Coulomb counting method, voltage method, and Kalman filter method. These methods are suitable for all battery systems, especially hybrid electric vehicles (HEV), electric vehicles (EV) and photovoltaic (PV) applications.
The coulomb counting method, also known as ampere-hour counting and current integration method, is the most commonly used technology to calculate SOC. This method calculates the SOC value by integrating the battery current reading with the service time. The coulomb counting method calculates the remaining capacity by accumulating the charges transferred into or out of the battery. The accuracy of this method mainly depends on the precise measurement of battery current and the accurate estimation of initial SOC. With a pre-known capacity (which can be memorized by memory or initially estimated by operating conditions), the SOC of the battery can be calculated by integrating the charging and discharging current with the operating periods.
The voltage method is that the SOC (that is, its remaining capacity) of the battery can be determined by the discharge test under controlled conditions. The voltage method uses the known discharge curve of the battery (the relationship between voltage and SOC) to convert the battery voltage reading into the equivalent SOC value. However, due to the electrochemical kinetics and temperature of the battery, the influence of battery current on voltage is more serious. Using a correction term proportional to the battery current to compensate the voltage reading, and using the look-up table of the battery open circuit voltage (OCV) and temperature can make this method more accurate.
The Kalman filter method is an algorithm that can estimate the inner state of any dynamic system, and can also be used to estimate the SOC of the battery. Compared with other estimation methods, the Kalman filter can automatically provide dynamic error bounds for its own state estimation. By modeling the battery system to include the required unknown quantity (such as SOC) in its state description, the Kalman filter estimates its value and gives the estimated error bound. Then, it becomes a model-based state estimation technology, which uses an error correction mechanism to provide real-time predictions of SOC.
Kalman filter principle
Choose the appropriate SOC and SOH estimation methods
When choosing a suitable SOC estimation method, many criteria should be considered. First of all, SOC and SOH estimation techniques should be applicable to HEV and EV applications, renewable energy storage for future use, and Li-ion batteries for power grid energy storage. In addition, the key point is that the selected method should be an online real-time technology with low computational complexity and high accuracy (low estimation error). In addition, the estimation method is required to use measured values of voltage and current, and other data collected by temperature sensors and/or provided by PC-based software applications.
In order to overcome the shortcomings of the Coulomb counting method and improve its estimation accuracy, an enhanced Coulomb counting algorithm has been proposed to estimate the SOC and SOH parameters of Li-ion batteries. The initial SOC is obtained from the loaded voltage (charging and discharging) or the open circuit voltage. The loss is compensated for by considering the charging and discharging efficiency. By dynamically recalibrating the maximum releasable capacity of the operating battery, the SOH of the battery can also be estimated at the same time, which will further improve the precise of the SOC estimation.
The battery has three working modes: charging, discharging, and open circuit. In the charging stage, when the battery is charged in constant current and constant voltage (CC-CV) mode, the manufacturer usually explains the changes of the battery voltage and current. When the charging current is constant, the battery voltage gradually increases until it reaches the threshold. Once the battery is charged in constant voltage mode, the charging current will decrease rapidly at first, and then decrease slowly. Finally, when the battery is fully charged, the charging current tends to zero. This charging curve can be converted into a relationship between the SOC and charging voltage in the constant current stage and the SOC and charging current in the constant voltage stage, and the initial SOC during charging can be calculated from these relationships.
In the discharge stage, the typical voltage curves of the battery when discharged at different currents are given by the manufacturer. As the operating time goes by, the terminal voltage will decrease. The greater the current, the faster the terminal voltage drops, so the shorter the operating time. In this way, the relationship between SOC and discharge voltage under different currents can be obtained, and then the initial SOC in discharge stage can be deduced.
The open circuit phase needs the relationship between OCV and SOC. Before disconnecting the load, the battery discharges at different currents. If the rest time is long, the OCV can be used to estimate SOC. The operating efficiency of the battery can be evaluated by Coulomb efficiency, which is defined as the ratio of the charge that can be obtained from the battery during discharge to the charge that enters the battery during charging.
Wired Battery Management System (BMS)
A variety of solutions to meet the needs of battery monitoring
In order to solve various problems in battery monitoring, ADI has also introduced a number of product solutions, including ADBMS6815, a multi-cell battery stack monitor, which can measure up to 12 series-connected battery cells with a total measurement error (TME) of less than 1.5 mV. The ADBMS6815 has a battery measurement range of 0 V to 5 V, which is suitable for most battery chemistry applications. All 12 battery cells can be measured within 304 μs, and a lower data acquisition rate is selected for noise reduction.
In addition, several ADBMS6815 devices can be connected in series to monitor a long high-voltage battery string at the same time. Each ADBMS6815 has an isoSPI™ interface for long-distance high-speed communication without interference from radio frequencies. Multiple devices are connected in daisy chain, and are connected to the host processor through the top or bottom devices. The daisy chain can be operated in bidirectionally, ensuring communication integrity even if the communication path is fault.
The battery stack can directly supply power to the ADBMS6815, or it can be powered by an isolated power supply. The ADBMS6815 includes a passive balancing for each battery, and each cell can be individually controlled by the pulse width modulation (PWM) duty cycle. Other features include an onboard 5 V regulator, seven general-purpose input/output (GPIO) lines, and a sleep state that can reduce current consumption to 5.5 µA. The ADBMS6815WFS model is designed for ISO 26262 applications with an automobile safety integrity level capability D (ASIL D).
On the other hand, ADI also introduced the LTC2949, a high-precision current, voltage, temperature, charge and energy meter suitable for EVs, HEVs, and other isolated current detection applications. By simultaneously monitoring the voltage drop across up to two sense resistors and the battery pack voltage, it can infer the amount of electricity and energy flowing into and out of the battery pack.
In addition, the LTC6820 of isoSPI isolated communication interfaces introduced by ADI can provide bidirectional SPI communication between two isolated devices through a single twisted pair connection. Each LTC6820 encodes a logic state into a signal and transmits the signal to another LTC6820 across an isolation barrier. The receiving LTC6820 decodes the transmission signal and drives the slave bus to appropriate logic states. The isolation barrier can be bridged by a simple pulse transformer to achieve isolation of several hundred volts.
Conclusion
Whether it is an electric vehicle or an energy storage system application, the operating efficiency of the battery is an important key to improve the efficiency of related products. By monitoring the SoC and SoH of the battery, the battery can be guaranteed to operate in an efficient and stable manner. ADI's related solutions for battery monitoring applications will improve the efficiency and safety of battery operation. For more technical and product details, please contact ADI or Arrow Electronics for more detailed information.