Active safety of lithium-ion power stations refers to the timely detection and elimination of faults through real-time monitoring and early warning before a grid energy storage system with potential safety hazards fails or causes serious consequences, thereby avoiding subsequent losses.
Most existing active safety protection methods at home and abroad are based on the following: according to the battery thermal runaway mechanism, the main factors of the battery thermal runaway state are characterized and analyzed, relevant features are extracted, and then the feature quantities are collected and processed to realize signal warning or linkage protection control.
Currently, there are three main methods for identifying battery thermal runaway in China: First, the battery management system (BMS) monitors data such as temperature, voltage, and current during battery operation in real time; second, the internal operating status and temperature of the battery can reflect the thermal runaway state of the battery during charging and discharging; and third, the types and concentrations of gases released at different stages of thermal runaway in the battery module are different, and real-time monitoring of the types and concentrations of gases in the battery module can help determine battery thermal runaway.
Thermal runaway detection based on battery internal temperature
A Battery Management System (BMS) can monitor battery surface current, voltage, temperature, and other data in real time to determine whether the battery module is in a state of thermal runaway. The drawback is that the battery is a completely sealed unit, and the surface state of charge cannot fully reflect the internal working state of the battery, especially during high-power charging and discharging, when the temperature difference between the inside and outside of the battery module can be significant (up to 20°C).
The internal temperature of a battery is the most direct and effective signal reflecting the battery's safety status. There are two main methods for thermal runaway early warning based on the internal temperature of a battery: the embedded sensor measurement method and the method of measuring the internal impedance-temperature correspondence.
Embedding sensors within batteries (such as embedded Bragg fiber optic sensors) alters the battery structure, making it difficult to integrate with existing battery manufacturing processes and hindering its widespread application.
The impedance phase shift method for monitoring the internal temperature of a battery effectively compensates for the shortcomings of embedded sensors. During battery operation, there is a strong correlation between the internal impedance phase shift and the internal temperature. In the early stages of thermal runaway, the surface temperature shows no significant change, but the internal impedance phase shift is noticeably abnormal. However, this method has the disadvantage of requiring precise measuring instruments, resulting in high costs.
The method of measuring the internal impedance-temperature relationship is based on the fact that when the battery's operating state changes, the refractive index and wavelength of the light received by the fiber optic sensor will change accordingly. Combined with the BMS, it can accurately monitor the battery's internal temperature and other performance indicators in real time, and provide effective early warning of battery thermal runaway.
Specifically, at the beginning of battery overcharging, the slope of the dynamic impedance changes from negative to positive in the 30-90Hz frequency band. Cutting off charging at this point can successfully prevent the battery from emitting warning signals, with the warning time being 580 seconds earlier than thermal runaway. Moreover, this method does not require complex mathematical models and parameters, which is conducive to large-scale application and has a low cost.
Based on thermal runaway detection of battery gases
In the early stages of thermal runaway in lithium-ion batteries, the internal electrochemical reaction releases a large amount of gas. Placing gas sensors around the energy storage battery module to detect the types and concentrations of these gases is an effective way to provide early warning of battery thermal runaway.
The types and concentrations of gases produced at different stages of battery thermal runaway are different.
In the early stages of thermal runaway, the battery module casing remains intact with no significant temperature rise, but it will produce large amounts of gases such as carbon dioxide, carbon monoxide, ethyl methyl carbonate, dimethyl carbonate, and methane.
As the thermal runaway worsened, the battery module casing cracked, the temperature rose sharply, and the battery's electrochemical reaction produced a large amount of gas, increasing the gas production rate. Harmful gases such as dimethyl ether, methyl formate, and ethylene began to be produced.
Research on lithium iron phosphate battery modules has found that hydrogen concentration changes are the most sensitive during battery thermal runaway, and it can be used as an early warning gas for battery thermal runaway.
Thermal runaway detection based on internal battery pressure
The internal pressure method is based on the following: when the battery is working normally, the internal pressure is the same as the atmospheric pressure; when the battery experiences thermal runaway, the internal electrochemical reaction releases a large amount of gas in a short period of time, and the internal pressure of the battery rises sharply.
The revised method has a high false alarm rate because it is limited by the individual capacity and volume of a single battery. When a single battery experiences thermal runaway, the gas generated cannot reach the preset gas pressure threshold, making it difficult to trigger a thermal runaway warning.
Furthermore, the peak time of pressure change caused by thermal runaway in the battery is relatively short (generally around 100ms), followed by a rapid increase in pressure that causes the pressure relief valve to open, and the internal pressure of the battery to drop rapidly. Due to factors such as the sampling frequency of the pressure sensor, it may be unable to detect the rapid change in internal pressure of the battery in time, thus failing to trigger an early warning.
Thermal runaway discrimination based on battery expansion force
The expansion force detection method is based on the following: During the charging and discharging of a lithium-ion battery cell, the insertion/deposition of lithium ions causes a change in the cell thickness, resulting in cell expansion. During charging, lithium ions are extracted from the positive electrode and inserted into the negative electrode, increasing the distance between the negative electrodes and causing the cell to expand. In the event of thermal runaway in a lithium-ion battery, the cell expansion force will change significantly.
However, this method is still in the experimental research stage, and the problems mainly manifest in the following aspects: First, the internal expansion force changes differently depending on the type and size of the battery (the expansion force of lithium iron phosphate batteries is significantly less than that of ternary lithium batteries); second, for batteries with the same or similar body material system, the expansion force changes differently when charging and discharging for two batteries with different negative electrode systems (when 811 ternary cathode material batteries are used with silicon-carbon anode material, the expansion force is significantly greater than when used with graphite anode material); third, the expansion force changes differently when the battery is in different states of charge (the expansion force of ternary 523 cells changes most significantly when the state of charge is 80%), etc.
Thermal runaway discrimination based on acoustic signals
Battery exhaust acoustic signals exhibit specificity under thermal runaway and normal operating conditions. After feature extraction, an effective identification feature set can be formed. Thermal runaway early warning based on battery exhaust acoustic signals can achieve an accuracy rate of 92.31%.
The advantages of this method are fast implementation speed, high sensitivity, low cost, easy detection of acoustic signals, and wide application range; the disadvantage is that it can only identify the battery exhaust sound signal to make a judgment on its presence or absence, and cannot accurately locate the thermal runaway fault unit.
