Power battery thermal runaway fault detection method and system based on multi-source parameters

Abstract

The invention discloses a power battery thermal runaway fault detection method and system based on multi-source parameters, and belongs to the technical field of battery management, and the method comprises the steps: inputting a high-frequency signal generated by a signal generator to a positive electrode feed point of a battery cell, connecting a coaxial line to a positive electrode, and grounding a negative electrode to form a grounding loop; the collected dendritic crystal reflection signals are analyzed, a quantitative correlation model is established, the growth condition of dendritic crystals is analyzed according to the quantitative correlation model, and preliminary early warning is carried out; electrochemical parameters and auxiliary parameters of the battery cell are collected, and cross validation is carried out on the electrochemical parameters and the auxiliary parameters and preliminary early warning of dendritic crystals; and establishing a dynamic risk and grading response mechanism, dividing the risk into different grades, and setting corresponding response measures. In the implementation process of the technical scheme provided by the invention, by combining dendritic crystal growth prediction and electrochemical parameter analysis, accurate monitoring of the internal state of the power battery is realized, the timeliness and accuracy of thermal runaway early warning are effectively improved, and safe and stable operation of a battery system is ensured.

Description

Technical Field

[0001] The present application relates to the field of battery management technology, and specifically to a method and system for detecting thermal runaway faults of power batteries based on multi-source parameters. Background Art

[0002] With the development of new energy vehicles, the safety of power batteries has received more and more attention from car companies and OEMs. Currently, power batteries are usually lithium-ion batteries. Lithium batteries are prone to thermal runaway when overcharged, short-circuited or physically damaged, leading to the risk of fire or even explosion.

[0003] In existing technologies, a variety of solutions have been adopted to address the thermal runaway problem of lithium-ion batteries, such as using a high-nickel positive electrode plus a silicon-carbon negative electrode in combination with a low-impedance electrolyte to improve the stability of the lithium-ion battery, adopting a tab-free design to reduce internal resistance, and integrating various parameters of the lithium-ion battery through the BMS (battery management system) to achieve thermal runaway warning, etc., to reduce the risk of thermal runaway of lithium-ion batteries in various environments. At present, BMS has become the mainstream response solution for automobile companies, which can realize full life cycle data management and thermal runaway detection services.

[0004] However, in practical applications, BMS technology still has some problems. For example, when lithium-ion batteries are overcharged, lithium dendrites will form at the negative electrode. These lithium dendrites will pierce the battery separator and cause internal short circuits. The existing BMS technology can only analyze whether it is in a normal state through the electrical parameters of the battery. However, it is difficult to accurately judge whether chronic, irreversible damage such as dendrites has occurred inside the battery, resulting in delayed thermal runaway warning and inability to capture early hidden dangers in time, which in turn affects the overall safety performance.

[0005] Therefore, it is necessary to provide a power battery thermal runaway fault detection method and system based on multi-source parameters to solve the above problems.

[0006] It should be noted that the above information disclosed in this Background section is only for understanding the background technology of the present application concept, and therefore, it may contain information that does not constitute prior art. Summary of the Invention

[0007] Based on the above-mentioned problems existing in the prior art, the problem to be solved by this application is: to provide a power battery thermal runaway fault detection method and system based on multi-source parameters, so as to combine the growth prediction of dendrites inside the battery cell with the traditional electrochemical parameter prediction to realize thermal runaway risk detection of the power battery.

[0008] The technical solution adopted by this application to solve the technical problem is: a power battery thermal runaway fault detection method based on multi-source parameters, including: A high-frequency signal generated by a signal generator is input to the positive electrode feed point of the battery cell, and a coaxial line is connected to the positive electrode. The negative electrode is grounded to form a ground loop. After the high-frequency signal propagates directionally in the electrolyte, the dendrite reflection signal is collected through the coaxial line. Analyze the collected dendrite reflection signals and establish a quantitative correlation model between specific frequency bands and dendrite growth. Use this quantitative correlation model to analyze dendrite growth and provide preliminary warnings. Collect the electrochemical parameters and auxiliary parameters of the battery cell and cross-verify them with the initial warning of dendrites. The electrochemical parameters include voltage, current, temperature and internal resistance, and the auxiliary parameter is gas content. Establish a dynamic risk and graded response mechanism, combine historical data and real-time monitoring results, divide risks into different levels, and set corresponding response measures for different levels.

[0009] During the implementation of the technical solution of this application, by combining dendrite growth prediction with electrochemical parameter analysis, accurate monitoring of the internal state of the power battery is achieved, which effectively improves the timeliness and accuracy of thermal runaway warning and ensures the safe and stable operation of the battery system.

[0010] Furthermore, a matching resistor is connected in series to the end of the coaxial line on the positive pole, and the resistance of the matching resistor is ohm. The matching resistor is used to achieve optimal matching of the signal during transmission, thereby improving the signal collection quality.

[0011] Furthermore, when performing high-frequency signal scanning, it is also necessary to establish a self-triggered scanning mechanism, which has a fixed starting point and dynamic scanning power. The fixed starting point is the frequency point when the internal impedance of the battery cell reaches the minimum value, and the dynamic scanning power is adjusted according to the actual state of the battery.

[0012] Furthermore, the quantitative correlation model is established through a specific frequency band and dendrite growth, wherein the specific frequency is the center frequency of the high-frequency signal generated by the signal generator, 2.4 GHz.

[0013] Furthermore, analyzing the growth of dendrites based on the quantitative correlation model and issuing preliminary warnings further include: judging the surface dendrite position, dendrite coverage area and growth rate based on the amplitude change, phase jump and time domain change of the reflected signal; quantifying the risk value of the battery cell based on the growth rate and coverage area of the dendrite, and setting a risk threshold. When the quantified risk value exceeds the risk threshold, a dendrite growth warning is issued.

[0014] Furthermore, the coverage area of the dendrite is determined by establishing a mapping model between the reflection coefficient and the coverage area. The mapping model is a direct proportional model, in which the independent variable is the reflection coefficient and the dependent variable is the coverage area. The mapping model parameters are fitted by experimental data or historical data to calculate the coefficients of the direct proportional model.

[0015] Furthermore, the position information of the dendrite is calculated using the time domain reflection method. The connection position between the signal generator and the battery cell is used as the starting point and initial moment. The time difference of the high-frequency signal from the emission point to the dendrite reflection point and the propagation speed of the signal are obtained, thereby calculating the distance between the dendrite and the starting point.

[0016] Furthermore, quantifying the risk value of the battery cell specifically includes: establishing a risk value quantification relationship, which takes into account the dendrite growth rate, coverage area and internal resistance change of the battery cell. The risk value quantification relationship is a linear weighted model, and first normalizes the different parameters, maps each parameter to the range of 0 to 1, and then obtains the comprehensive risk value through weighted summation. Among them, the normalization of the dendrite growth rate, coverage area and internal resistance change of the battery cell adopts the maximum value standardization method. Each parameter is proportional to the risk value, and weights are assigned to them respectively. The weight coefficient is determined according to the actual application scenario and experimental data.

[0017] Furthermore, the graded response mechanism includes at least three risk levels. The first risk level is determined by the dendrite growth rate exceeding the threshold and the temperature rising suddenly. The second risk level is the frequent voltage fluctuations and abnormal increase in internal resistance. The third risk level is the abnormal increase in gas content and the continuous increase in gas concentration near the battery cell pressure relief valve.

[0018] A power battery thermal runaway fault detection system based on multi-source parameters, the system comprising: The high-frequency signal input module is used to input the high-frequency signal generated by the signal generator to the positive electrode feed point of the battery cell. The coaxial line is connected to the positive electrode and the negative electrode is grounded to form a ground loop. After the high-frequency signal propagates directionally in the electrolyte, the dendrite reflection signal is collected through the coaxial line. The reflection signal analysis module is used to analyze the collected dendrite reflection signals and establish a quantitative correlation model between specific frequency bands and dendrite growth. Based on the quantitative correlation model, the dendrite growth situation is analyzed and a preliminary warning is issued. The cross-validation module is used to collect the electrochemical parameters and auxiliary parameters of the battery cell and cross-validate them with the initial warning of dendrites. The electrochemical parameters include voltage, current, temperature and internal resistance, and the auxiliary parameter is gas content. The dynamic risk and graded response module is used to establish a dynamic risk and graded response mechanism. It combines historical data and real-time monitoring results to divide risks into different levels and set corresponding response measures for different levels.

[0019] The beneficial effects of the present application are: the present application provides a power battery thermal runaway fault detection method and system based on multi-source parameters, which realizes accurate monitoring of the internal state of the power battery by combining dendrite growth prediction and electrochemical parameter analysis, effectively improves the timeliness and accuracy of thermal runaway warning, ensures the safe and stable operation of the battery system, and can accurately calculate the actual position of the dendrite, facilitating the detection and maintenance of the power battery.

[0020] In addition to the above-described purposes, features and advantages, the present application also has other purposes, features and advantages. The present application will be further described in detail below with reference to the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The drawings that constitute part of this application are used to provide a further understanding of this application. The illustrative embodiments of this application and their descriptions are used to explain this application and do not constitute an improper limitation on this application. In the drawings: Figure 1 This is an overall schematic diagram of a power battery thermal runaway fault detection method based on multi-source parameters in this application; Figure 2 This is a connection diagram of the matching resistor; Figure 3 This is a schematic diagram of the module structure of a power battery thermal runaway fault detection system based on multi-source parameters in this application. DETAILED DESCRIPTION

[0022] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in this application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0023] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.

[0024] Example 1: Figure 1 As shown, the present application provides a power battery thermal runaway fault detection method based on multi-source parameters. The method is applicable to lithium-ion battery application scenarios such as electric vehicles and energy storage power stations. The method includes: S110, input a high-frequency signal generated by a signal generator to the positive electrode feed point of the battery cell, connect a coaxial line to the positive electrode, and ground the negative electrode to form a ground loop. After the high-frequency signal propagates directionally in the electrolyte, the dendrite reflection signal is collected through the coaxial line; Dendrites are crystals that form on the negative electrode surface during the charging process of lithium-ion batteries. These crystals may pierce the diaphragm and cause a short circuit, posing a safety hazard. They are a common cause of thermal runaway in lithium-ion batteries. During the cyclic charge and discharge process of lithium-ion batteries, the early stages of dendrite growth will generate sound waves or impedance change signals of a specific frequency. By inputting a high-frequency signal into the battery cell, the specific frequency sound waves or impedance change signals extracted can be used to monitor dendrite growth and provide timely warnings. Among them, the positive pole of the battery cell is provided with a feeding point, and the output end of the signal generator is connected at the feeding point. The signal generator generates a high-frequency signal with an output frequency of 2.2 GHz to 2.6 GHz, and the center frequency of the high-frequency signal is 2.4 GHz. The high-frequency signal is input into the battery cell. Since the negative pole of the battery cell is grounded to form a ground loop, the high-frequency signal will form a direction propagation in the electrolyte of the battery. In this embodiment, the chip model of the signal generator is ADF4350, or other chips that can output high-frequency signals that meet the requirements, and a coaxial line is synchronously connected to the positive pole of the battery cell. In the process of directional propagation of the high-frequency signal, part of the reflected signal will return along the coaxial line connected to the positive pole, and part of the transmitted signal will pass through the negative pole ground loop to generate a reference level. The reflected signal returned along the coaxial line is because dendrites appear inside the battery. The impedance change, capacitance, inductance and other characteristics brought by the dendrites will cause signal reflection. When performing dendrite analysis on lithium-ion batteries, it is only necessary to collect the reflected signal of the positive electrode. Therefore, it is also necessary to connect a coaxial line to the positive electrode and connect a signal acquisition device to the coaxial line to collect the reflected signal. In this embodiment, the model of the acquisition device is not limited. Any device with a high sampling rate and high precision can be used, such as a data acquisition card, and the collected reflected signal is transmitted to a processing unit for analysis. The processing unit can be an embedded processor in the vehicle system. Therefore, there is a data exchange channel between the acquisition device and the processing unit. The channel is a high-speed Ethernet interface or a high-speed CAN interface. Through this interface, the acquisition device can transmit the reflected signal data to the embedded processor in real time. It should be noted that due to the different manufacturing tolerances of different battery cells, the problem of standing wave ratio deterioration may occur. Therefore, a matching resistor is required to be connected in series at the end of the coaxial line on the positive electrode. The resistance of the matching resistor is generally around 50 ohms. Its purpose is to obtain the best matching during the transmission process of the signal, thereby improving the signal collection quality to ensure that the collected signal can accurately reflect the growth state of the dendrite. The specific wiring method can be referred to Figure 2, wherein other pins of the signal generator (such as the enable pin, the ground pin, etc.) are not shown, only the output pin is shown. For the wiring of other pins, please refer to the chip manual of this model, and no detailed description is given in this embodiment; When performing high-frequency signal sweeping, a self-triggered sweeping mechanism also needs to be established. The self-triggered sweeping mechanism has a fixed starting point and a dynamic sweeping power. The fixed starting point is the frequency point when the internal impedance of the battery cell reaches the minimum value. When the internal impedance of the battery cell is the minimum, the heat generation of the battery itself is the minimum, and the actual capacity is closest to the theoretical maximum value. The internal impedance of the battery cell can be measured using an existing BMS. The dynamic sweeping power is adjusted according to the actual state of the battery. Specifically, the current sweeping power is equal to the actual capacity when the internal impedance of the battery cell is the minimum multiplied by a specific adjustment coefficient to ensure that the sweeping power is adapted to the state of the battery cell. For example, when the internal impedance of the battery cell is the minimum, the actual capacity of the battery cell at this time is 90% multiplied by the rated battery capacity in kilowatt-hours, and the adjustment coefficient is 0.1. The corresponding dynamic sweeping power should be 9% multiplied by the rated battery capacity in milliwatts. Through the self-triggered sweeping mechanism, the high-frequency signal can be accurately swept when the battery cell is closest to the actual capacity. In addition, compared with the traditional fixed-cycle detection mode, the determination of the dynamic sweeping power can select the sweeping power according to the operating state of the battery cell to avoid polarization interference. S120, analyzing the collected dendrite reflection signal, and establishing a quantitative correlation model between a specific frequency band and dendrite growth, analyzing the dendrite growth according to the quantitative correlation model, and issuing a preliminary warning; After collecting the dendrite reflection signal, the growth of the dendrite can be analyzed based on the signal. The growth of the dendrite is fed back to the reflection signal as amplitude change, phase jump and time domain reflection. Based on these changes, the dendrite can be quantitatively analyzed. In order to quantitatively analyze the dendrite according to the changes in different parameters, in this embodiment, a quantitative correlation model between a specific frequency band and dendrite growth is established, wherein the specific frequency is the center frequency of the high-frequency signal generated by the signal generator, 2.4 GHz. The purpose of selecting the center frequency is to ensure that the frequency band covers the characteristic frequency of dendrite growth, enhance the signal recognition ability, and reduce interference from other frequencies. Specifically, analyzing the growth of the dendrite according to the quantitative correlation model and making a preliminary warning further includes: S220, determining the surface dendrite position, dendrite coverage area, and growth rate based on the amplitude change, phase jump, and time domain change of the reflected signal; When dendrites grow, the internal impedance of the battery cell will suddenly change, which will increase the reflection coefficient. At the same time, the conductivity of the dendrites produced will be higher than that of the electrolyte (such as metallic lithium dendrites), which will cause the reflection coefficient to approach a negative value. In either case, the amplitude of the reflected signal will change. Therefore, the change in the amplitude of the reflected signal can be used to determine whether dendrites have occurred. In addition, the amplitude of the reflected signal is approximately linearly related to the dendrite coverage rate. Therefore, when the change in the amplitude of the reflected signal is obtained, the coverage area of the dendrite can be determined by establishing a mapping model between the reflection coefficient and the coverage area. Specifically, the mapping model can be a direct proportional model, in which the independent variable is the reflection coefficient and the dependent variable is the coverage area. The mapping model parameters are fitted with experimental data or historical data to calculate the coefficients of the direct proportional model, thereby achieving an accurate estimation of the dendrite coverage area. The specific calculation method is not described in detail in this embodiment. After obtaining the dendrite coverage area, the dendrite growth rate can be calculated based on the change of the coverage area over time, thereby evaluating the health status of the battery cell. At the same time, the capacitive or inductive properties of the dendrite will cause the phase shift of the reflected signal. This is because when the dendrite is in a capacitive state, it is not fully conductive, which will cause the phase of the reflected signal to advance. When it is in an inductive state, the opposite is true, resulting in a phase lag. Therefore, by real-time monitoring of the phase shift of the reflected signal, the capacitive and inductive states of the dendrite can be effectively identified. When this change process is subjected to a time-varying characteristic analysis, its growth rate and position information can be obtained. Specifically, the position information of the dendrite is calculated using the time domain reflection method. The connection position of the signal generator and the battery cell is used as the starting point and the initial moment. The time difference t of the high-frequency signal from the emission point to the dendrite reflection point and the signal propagation speed v are obtained, so that the distance d from the dendrite to the starting point can be calculated. The propagation speed v is calculated by the relative dielectric constant of the electrolyte and the speed of light. For details, please refer to the calculation method combining the speed of light and the relative dielectric constant in the existing formula. Therefore, considering the emission and reflection process of the signal, the distance d from the dendrite to the starting point is d=v*t / 2; For example, if the signal is injected from the surface of the positive terminal and the relative dielectric constant of the electrolyte is 30, the calculated propagation velocity v is 5.5*10 7 Meters per second, if the time difference of the reflected signal arrival is 10 picoseconds, then the dendrite depth d=5.5*10 7 *10*10 -12 / 2=0.275 microns, that is, the dendrite is located about 0.275 microns below the surface of the positive terminal, thus accurately determining the location of the dendrite; It should be noted that the above process is an analysis of the situation where the dendrites are located inside the battery cell. Therefore, only the relative dielectric constant of the electrolyte needs to be considered. For the dendrites on the electrode surface, the propagation of the signal in the electrode needs to be considered. Therefore, it is not described in this embodiment. The image analysis method can be used to determine the formation of dendrites on the electrode surface. For details, reference can be made to the analysis method of the growth characteristics of dendrites on the electrode surface in the prior art. In addition, since the specific frequency band selected is 2.4 GHz, the resonance effect of the dendrite will produce a characteristic frequency deviation or secondary reflection peak near this frequency band. The dendrite length can be reversed based on the visual simulation calculation in the existing technology. For details, please refer to the Chinese invention patent publication number CN118553342A. S230: quantify the risk value of the battery cell based on the growth rate and coverage area of the dendrite, and set a risk threshold. When the quantified risk value exceeds the risk threshold, issue a dendrite growth warning. When dendrites appear in a battery cell, they don't immediately cause a short circuit or other risks in the early stages. However, as the dendrites grow and expand, their coverage area and growth rate gradually increase, causing changes in the cell's internal resistance, which in turn affects battery performance and may eventually lead to a short circuit or even thermal runaway. Existing methods don't quantify the dendrite growth process and cell risks, making it difficult to effectively warn of risks. Quantifying the risk value of a battery cell specifically includes: Establish a risk value quantification relationship, which takes into account the dendrite growth rate, coverage area, and internal resistance change of the battery cell. The risk value quantification relationship is a linear weighted model. Since the dimensions of different parameters are not uniform, normalization processing is required first. Each parameter is mapped to the range of 0 to 1, and then a weighted sum is performed to obtain a comprehensive risk value. Among them, the normalization processing of the dendrite growth rate, coverage area, and internal resistance change of the battery cell adopts the minimum-maximum normalization method. Each parameter is proportional to the risk value and is weighted. The weight coefficient is determined according to the actual application scenario and experimental data. For example, the weight of the dendrite growth rate is set to 0.5, the weight of the coverage area is 0.3, and the weight of the resistance change is 0.2. The comprehensive risk value is then calculated by weighted summation. When the comprehensive risk value exceeds the preset threshold, the early warning mechanism is automatically triggered and the early warning information is sent to the monitoring system in real time through the data transmission module, prompting maintenance personnel to intervene in time. Intervention measures include but are not limited to replacing battery cells, adjusting the charge and discharge strategy (for example, dissolving some dendrites through low-current charge and discharge), or performing in-depth maintenance to ensure the safe and stable operation of the battery system. S130: Collect electrochemical parameters and auxiliary parameters of the battery cell and cross-verify them with the initial warning of dendrites. The electrochemical parameters include voltage, current, temperature and internal resistance, and the auxiliary parameter is gas content. In the above steps, the dendrite growth characteristics of the battery cell are monitored in real time, which can issue a warning in time at the early stage of dendrite growth and take corresponding measures. However, the thermal runaway problem of the battery cell is not only caused by dendrite growth, but may also be caused by external short circuit, overcharge and over-discharge and other factors. Therefore, it is necessary to conduct a comprehensive analysis based on electrochemical parameters and auxiliary parameters. Among them, electrochemical parameters include the fluctuation range of voltage, abnormal changes in current, abnormal increase in temperature and significant increase in internal resistance. Auxiliary parameters such as abnormal increase in gas content, that is, increased concentration of carbon monoxide or hydrogen near the battery cell pressure relief valve, abnormal changes in these parameters can serve as precursors to thermal runaway; Specifically, the electrochemical parameters are jointly analyzed with the preliminary warning of dendrites, because the growth of dendrites is coupled with abnormal changes in electrochemical parameters. For example, when the voltage fluctuation intensifies, the dendrite growth rate tends to accelerate, resulting in increased internal resistance, increased temperature, and abnormal increase in gas content. Therefore, the linkage effect of each parameter is comprehensively considered during the analysis. The existing BMS (battery management system) can be used to add a multi-parameter linkage analysis module on the basis of the original monitoring function. This module applies a machine learning algorithm to analyze the correlation between each parameter and adjust the risk value of the aforementioned process based on the analysis results. For example, when voltage fluctuations and temperature increases occur at the same time, the weight of the dendrite growth rate is automatically adjusted dynamically. Since voltage fluctuations and temperature increases are highly correlated with the dendrite growth rate, the weight of the dendrite growth rate is adjusted to 0.6, the coverage area weight is adjusted to 0.2, and the resistance change weight is adjusted to 0.2 to ensure that the comprehensive risk value more accurately reflects the actual situation and improves the accuracy of the warning.

[0025] S140: Establish a dynamic risk and graded response mechanism, combine historical data and real-time monitoring results, divide risks into different levels, and set corresponding response measures for different levels.

[0026] Thermal runaway risk in battery cells is not an isolated event. It requires comprehensive consideration of multiple factors, real-time assessment through a dynamic risk model, and triggering appropriate intervention measures based on the risk level. For example, a Level 1 risk requires immediate shutdown and inspection, a Level 2 risk requires focused monitoring and the preparation of emergency plans, and a Level 3 risk requires enhanced daily inspections to ensure the system can respond effectively to different risk levels and guarantee the long-term stable operation of the battery system. The dynamic risk mechanism uses temporal correlation analysis to identify risk points by determining whether there are temporal correlations between different parameters. For example, when dendrite growth and temperature rise change synchronously and exhibit temporal correlation, it indicates a significant increase in the risk of thermal runaway and requires an immediate increase in the risk level. The graded response mechanism includes at least three risk levels. The first risk level is determined by the dendrite growth rate exceeding the threshold and the temperature rising suddenly. The second risk level is frequent voltage fluctuations and abnormal increase in internal resistance. The third risk level is abnormal increase in gas content and continuous increase in gas concentration near the battery cell pressure relief valve. The judgment conditions corresponding to each level can be combined according to actual conditions. For example, the dendrite growth rate and increased gas content are used as the first risk level, voltage fluctuations and temperature rise are used as the second risk level, and abnormal internal resistance and continuous increase in gas concentration are used as the third risk level. This ensures that the judgment conditions of each level can be flexibly adjusted, reflects the system status in real time, and provides accurate early warning and effective control of thermal runaway risks, avoids misjudgment of a single indicator, improves overall prevention and control efficiency, and thus achieves a comprehensive multi-dimensional risk assessment. The intervention measures of the graded response mechanism include but are not limited to power-off protection, optimized charging strategies, active thermal management (directional cooling, temperature limiting, etc.), power limiting, pressure relief, etc., which are linked to the dynamic risk level to ensure that effective measures are taken when risks first appear.

[0027] Example 2: Figure 3 As shown, the present application also provides a power battery thermal runaway fault detection system based on multi-source parameters, which runs the detection method in Example 1 and includes: The high-frequency signal input module is used to input the high-frequency signal generated by the signal generator to the positive electrode feed point of the battery cell. The coaxial line is connected to the positive electrode and the negative electrode is grounded to form a ground loop. After the high-frequency signal propagates directionally in the electrolyte, the dendrite reflection signal is collected through the coaxial line. The reflection signal analysis module is used to analyze the collected dendrite reflection signals and establish a quantitative correlation model between specific frequency bands and dendrite growth. Based on the quantitative correlation model, the dendrite growth situation is analyzed and a preliminary warning is issued. The cross-validation module is used to collect the electrochemical parameters and auxiliary parameters of the battery cell and cross-validate them with the initial warning of dendrites. The electrochemical parameters include voltage, current, temperature and internal resistance, and the auxiliary parameter is gas content. The dynamic risk and graded response module is used to establish a dynamic risk and graded response mechanism. It combines historical data and real-time monitoring results to divide risks into different levels and set corresponding response measures for different levels.

[0028] The above description is merely a preferred embodiment of the present application and is not intended to limit the present application. Various modifications and variations are possible for those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present application shall be included within the scope of protection of the present application.

Claims

1. A method for detecting thermal runaway faults in power batteries based on multi-source parameters, characterized by: include: A high-frequency signal generated by a signal generator is input to the positive electrode feed point of the battery cell, and a coaxial line is connected to the positive electrode. The negative electrode is grounded to form a ground loop. After the high-frequency signal propagates directionally in the electrolyte, the dendrite reflection signal is collected through the coaxial line. Analyze the collected dendrite reflection signals and establish a quantitative correlation model between specific frequency bands and dendrite growth. Use this quantitative correlation model to analyze dendrite growth and provide preliminary warnings. Collect the electrochemical parameters and auxiliary parameters of the battery cell and cross-verify them with the initial warning of dendrites. The electrochemical parameters include voltage, current, temperature and internal resistance, and the auxiliary parameter is gas content. Establish a dynamic risk and graded response mechanism, combine historical data and real-time monitoring results, divide risks into different levels, and set corresponding response measures for different levels.

2. The method for detecting thermal runaway faults of power batteries based on multi-source parameters according to claim 1, characterized in that: A matching resistor is connected in series to the end of the coaxial line on the positive pole. The resistance of the matching resistor is 50 ohms. The matching resistor is used to achieve the best matching of the signal during transmission, thereby improving the signal collection quality.

3. The method for detecting thermal runaway faults of power batteries based on multi-source parameters according to claim 1, characterized in that: When performing high-frequency signal sweeping, a self-triggered sweeping mechanism also needs to be established. This self-triggered sweeping mechanism has a fixed starting point and dynamic sweeping power. The fixed starting point is the frequency point when the internal impedance of the battery cell reaches the minimum value, and the dynamic sweeping power is adjusted according to the actual state of the battery.

4. The method for detecting thermal runaway faults of power batteries based on multi-source parameters according to claim 1, characterized in that: The quantitative correlation model is established through a specific frequency band and dendrite growth, wherein the specific frequency is the center frequency of the high-frequency signal generated by the signal generator, 2.4 GHz.

5. The method for detecting thermal runaway faults of power batteries based on multi-source parameters according to claim 1, characterized in that: Analyzing dendrite growth based on a quantitative correlation model and issuing a preliminary warning further includes: determining the surface dendrite position, dendrite coverage area, and growth rate based on the amplitude change, phase jump, and time domain change of the reflected signal; quantifying the risk value of the battery cell based on the dendrite growth rate and coverage area, and setting a risk threshold. When the quantified risk value exceeds the risk threshold, a dendrite growth warning is issued.

6. The method for detecting thermal runaway faults of power batteries based on multi-source parameters according to claim 5, characterized in that: The coverage area of the dendrite is determined by establishing a mapping model between the reflection coefficient and the coverage area. The mapping model is a direct proportional model, in which the independent variable is the reflection coefficient and the dependent variable is the coverage area. The mapping model parameters are fitted by experimental data or historical data to calculate the coefficients of the direct proportional model.

7. The method for detecting thermal runaway faults of power batteries based on multi-source parameters according to claim 5, characterized in that: The time domain reflection method is used to calculate the position information of the dendrite. The connection position between the signal generator and the battery cell is used as the starting point and the initial moment. The time difference of the high-frequency signal from the emission point to the dendrite reflection point and the propagation speed of the signal are obtained, thereby calculating the distance between the dendrite and the starting point.

8. The method for detecting thermal runaway faults of power batteries based on multi-source parameters according to claim 5, characterized in that: The risk value quantification of battery cells specifically includes: establishing a risk value quantification relationship, which takes into account the dendrite growth rate, coverage area and internal resistance changes of the battery cell. The risk value quantification relationship is a linear weighted model, and first normalizes the different parameters, maps each parameter to the range of 0 to 1, and then obtains the comprehensive risk value through weighted summation. Among them, the normalization of the dendrite growth rate, coverage area and internal resistance changes of the battery cell adopts the maximum value standardization method. Each parameter is proportional to the risk value, and weights are assigned to them respectively. The weight coefficient is determined according to the actual application scenario and experimental data.

9. The method for detecting thermal runaway faults of power batteries based on multi-source parameters according to claim 1, characterized in that: The graded response mechanism includes at least three risk levels. The first risk level is determined by the dendrite growth rate exceeding the threshold and the temperature rising suddenly. The second risk level is the frequent voltage fluctuations and abnormal increase in internal resistance. The third risk level is the abnormal increase in gas content and the continuous increase in gas concentration near the battery cell pressure relief valve.

10. A power battery thermal runaway fault detection system based on multi-source parameters, used to implement the detection method according to any one of claims 1 to 9, characterized in that: The system includes: The high-frequency signal input module is used to input the high-frequency signal generated by the signal generator to the positive electrode feed point of the battery cell. The coaxial line is connected to the positive electrode and the negative electrode is grounded to form a ground loop. After the high-frequency signal propagates directionally in the electrolyte, the dendrite reflection signal is collected through the coaxial line. The reflection signal analysis module is used to analyze the collected dendrite reflection signals and establish a quantitative correlation model between specific frequency bands and dendrite growth. Based on the quantitative correlation model, the dendrite growth situation is analyzed and a preliminary warning is issued. The cross-validation module is used to collect the electrochemical parameters and auxiliary parameters of the battery cell and cross-validate them with the initial warning of dendrites. The electrochemical parameters include voltage, current, temperature and internal resistance, and the auxiliary parameter is gas content. The dynamic risk and graded response module is used to establish a dynamic risk and graded response mechanism. It combines historical data and real-time monitoring results to divide risks into different levels and set corresponding response measures for different levels.

Images