Lithium ion battery thermal runaway early warning system and method

Abstract

The invention relates to a lithium ion battery thermal runaway early warning system and method, and belongs to the technical field of lithium ion battery thermal runaway early warning, the system comprises a force signal monitoring unit, a micro processing unit and an early warning unit; the force signal monitoring unit is used for collecting the expansion force of the lithium ion battery; the micro-processing unit is used for analyzing the expansion force through a trained SVM classification model to obtain a potential thermal runaway state judgment result of the lithium ion battery; and the early warning unit is used for generating first early warning information when the potential thermal runaway state is judged to be in the potential thermal runaway state. According to the invention, the timeliness of early warning can be improved, more sufficient time is reserved to handle the abnormality of the lithium ion battery so as to prevent the occurrence of thermal runaway of the lithium ion battery, and the occurrence probability of chain thermal runaway is reduced.

Description

Technical Field

[0001] This invention relates to the field of lithium-ion battery thermal runaway early warning technology, and in particular to a lithium-ion battery thermal runaway early warning system and method. Background Technology

[0002] Driven by the "dual carbon" goal, lithium-ion batteries, as the core carrier of clean energy storage, have been widely used in electric vehicles, energy storage power stations, and other fields. However, frequent safety accidents have hindered their widespread construction and large-scale application. A large number of catastrophic fires and explosions are mainly attributed to thermal runaway of batteries. During the thermal runaway of a single cell, the heat generated is transferred to adjacent cells, ultimately leading to the thermal runaway of the entire battery pack.

[0003] Traditional thermal runaway early warning technologies mainly rely on monitoring parameters such as voltage, temperature, and gas. However, these signals often show significant abrupt changes only during the intense heat generation phase, resulting in a noticeable lag in early warning (usually only a few minutes in advance), making it difficult to effectively prevent chain thermal runaway reactions.

[0004] For example, temperature signals rely on the process of heat transfer from the battery interior to the casing. In the early stages of thermal runaway (such as SEI film decomposition or negative electrode lithium plating), the internal reaction heat generation rate is low, and the external temperature rise is significantly delayed. Voltage drops typically only occur in the internal short-circuit stage triggered by thermal runaway (such as after separator melting). At this point, the battery has already entered an irreversible and violent exothermic reaction, losing its early warning significance. Gases (such as CO, H2, and HF) need to accumulate to a certain concentration before they can be detected by sensors, but the airtightness of the battery packaging structure delays gas escape. Summary of the Invention

[0005] In view of this, it is necessary to provide a lithium-ion battery thermal runaway early warning system and method to solve the problem of significant early warning lag in existing thermal runaway early warning technologies.

[0006] To address the aforementioned problems, in a first aspect, the present invention provides a lithium-ion battery thermal runaway early warning system, comprising: a force signal monitoring unit, a microprocessor unit, and an early warning unit; The force signal monitoring unit is used to collect the expansion force of the lithium-ion battery; The microprocessor unit is used to analyze the expansion force through a trained SVM classification model to obtain the judgment result of the potential thermal runaway state of the lithium-ion battery. The early warning unit is used to generate a first early warning message when the judgment result of the potential thermal runaway state is that it is in a potential thermal runaway state.

[0007] In one possible implementation, it also includes: a control unit; The control unit is used to generate a command to limit battery power and a command to start air cooling when the potential thermal runaway state is determined to be in a potential thermal runaway state.

[0008] In one possible implementation, it also includes: an electrical signal monitoring unit; The electrical signal monitoring unit is used to collect the voltage information of the lithium-ion battery; The microprocessor unit is used to analyze the voltage information using a trained SVM classification model to obtain voltage offset features, and to obtain the early thermal runaway state judgment result of the lithium-ion battery based on the voltage offset features; wherein, the SVM classification model is obtained by training the voltage offset features under normal dynamic operating conditions of the lithium-ion battery and the voltage offset features in the early stage of thermal runaway. The early warning unit is used to generate a second early warning message when the early thermal runaway state judgment result indicates that the thermal runaway state is in its early stage.

[0009] In one possible implementation, the trained SVM classification model is used to output a judgment result indicating that the lithium-ion battery is in the early stage of thermal runaway when the voltage of the lithium-ion battery reaches a local maximum value and then decreases, and the voltage offset value during the decrease reaches an offset value threshold.

[0010] In one possible implementation, it also includes: a control unit; The control unit is used to generate a power cut-off command and a cooling phase change material command when the early thermal runaway state determination result is that the thermal runaway early state is in the early thermal runaway state.

[0011] In one possible implementation, it also includes: a gas signal monitoring unit; The gas signal monitoring unit is used to collect gas information from the lithium-ion battery; The microprocessor unit is used to analyze the gas information of the lithium-ion battery through a trained SVM classification model to obtain the thermal runaway state judgment result of the lithium-ion battery. The early warning unit is used to generate a third early warning message when the thermal runaway state determination result is that the thermal runaway state is in a state of thermal runaway.

[0012] In one possible implementation, the gas information is mixed gas information of multiple gases, and the microprocessor unit is further configured to decompose the mixed gas information into gas information components of each gas through a state Kalman filter, determine the diffusion coefficient of each gas based on the gas information components of each gas, and determine the target gas from each gas based on the diffusion coefficient of each gas. The microprocessor unit is used to analyze the gas information components of the target gas using a trained SVM classification model to obtain the thermal runaway state judgment result of the lithium-ion battery.

[0013] In one possible implementation, the microprocessor unit is further configured to compare the diffusion coefficient of each gas with the diffusion coefficient of a preset interfering gas. If the difference between the diffusion coefficient of the gas and the diffusion coefficient of the preset interfering gas is greater than a preset diffusion coefficient deviation threshold, and the energy ratio of the gas information component of the gas is greater than the energy ratio threshold, then the gas is determined to be the target gas.

[0014] In one possible implementation, it also includes: a control unit; The control unit is used to generate a command to blow the main fuse and a command to activate the fire extinguishing device when the thermal runaway state determination result is that the thermal runaway state is in a thermal runaway state.

[0015] Secondly, the present invention also provides a method for early warning of thermal runaway in lithium-ion batteries, applied to the lithium-ion battery thermal runaway early warning system described in any one of the above claims, the method comprising: The expansion force of the lithium-ion battery is collected by the force signal monitoring unit. The microprocessor unit is used to analyze the expansion force based on the trained SVM classification model to obtain the judgment result of the potential thermal runaway state of the lithium-ion battery. When the potential thermal runaway state is determined to be in a potential thermal runaway state by the early warning unit, a first early warning message is generated.

[0016] The beneficial effects of this invention are: When lithium-ion batteries experience thermal runaway, the expansion force of the lithium-ion battery responds earlier and faster than characteristic information such as temperature, voltage, and gas production. Therefore, this invention uses a force signal monitoring unit to monitor the expansion force of the lithium-ion battery in real time. A microprocessor unit analyzes this expansion force based on a trained SVM classification model to obtain a judgment result on the potential thermal runaway of the lithium-ion battery. This allows for timely detection of early signs of thermal runaway and effective monitoring of the early stages. Finally, an early warning unit generates warning information based on the potential thermal runaway judgment result, improving the timeliness of the warning and giving users or processing units more time to handle the abnormality of the lithium-ion battery to prevent thermal runaway and reduce the probability of cascading thermal runaway. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A schematic diagram of a structure of an embodiment of the lithium-ion battery thermal runaway early warning system provided by the present invention; Figure 2 This is a schematic diagram illustrating the training of an SVM classification model provided by the present invention. Figure 3 A flowchart of a lithium-ion battery thermal runaway early warning system provided by the present invention; Figure 4 This is a schematic diagram of a typical experimental platform for monitoring and early warning of thermal runaway in lithium-ion batteries provided by the present invention; Figure 5 This invention provides a typical data curve for monitoring and early warning of thermal runaway in lithium-ion batteries. Figure 6 This is a flowchart illustrating an embodiment of the lithium-ion battery thermal runaway early warning method provided by the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0020] In the description of the embodiments of this invention, unless otherwise stated, "multiple" means two or more. "And / or" describes the relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following associated objects have an "or" relationship.

[0021] In the embodiments of this invention, the terms "first," "second," etc., are used to distinguish similar objects, and are not used to describe a specific order or sequence, nor to indicate or imply their relative importance or implicitly specify the number of technical features indicated. It should be understood that such data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class, and the number of objects is not limited; for example, the first object can be one or more.

[0022] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0023] Reference Figure 1 The diagram shows a structural schematic of an embodiment of the lithium-ion battery thermal runaway early warning system 10 provided by the present invention. The system 10 includes: a force signal monitoring unit 110, a microprocessor unit 120, and an early warning unit 130. The force signal monitoring unit 110 is used to collect the expansion force of the lithium-ion battery; The microprocessor unit 120 is used to analyze the expansion force through a trained SVM classification model to obtain the judgment result of the potential thermal runaway state of the lithium-ion battery; The early warning unit 130 is used to generate a first early warning message when the judgment result of the potential thermal runaway state is that it is in a potential thermal runaway state.

[0024] Expansion force refers to the internal stress or external force generated by the volume expansion of a lithium-ion battery when it undergoes physicochemical changes.

[0025] In this embodiment, the force signal monitoring unit 110 can be connected to both the lithium-ion battery and the microprocessor unit 120 to collect the expansion force of the lithium-ion battery and transmit the expansion force to the microprocessor unit 120. Specifically, the force signal monitoring unit 110 may include a force sensor, such as an expansion force sensor. The microprocessor unit 120 may include a controller.

[0026] The microprocessor unit 120 can also be directly or indirectly connected to the early warning unit 130. For example, the microprocessor unit 120 can be directly connected to the early warning unit 130, or as... Figure 1The control unit 140 is indirectly connected to the early warning unit 130. The microprocessor unit 120 is used to analyze the expansion force using a trained support vector machine (SVM) classification model to obtain the judgment result of the potential thermal runaway state of the lithium-ion battery. The judgment result of the potential thermal runaway state includes whether it is in a potential thermal runaway state or not.

[0027] For example, when the input expansion force exceeds the expansion force threshold, the SVM classification model outputs the classification result: the system is in a potential thermal runaway state. The expansion force threshold can be learned during the training of the SVM classification model.

[0028] Potential thermal runaway refers to a state in which a lithium-ion battery shows abnormal signs before serious damage. In this state, sensor data detects parameters that deviate abnormally from normal operating conditions, but have not yet reached the critical threshold that directly threatens battery safety or function. Such anomalies may be early warning signs of thermal runaway, requiring limiting operating parameters and activating primary protection mechanisms to prevent escalation of risk. Causes of this state include localized micro-short circuits, slight thickening of the solid electrolyte interphase (SEI) film, loose connections, or minor electrolyte leakage.

[0029] The early warning unit 130 may include an alarm, used to generate a first early warning message when the judgment result of the potential thermal runaway state is that a potential thermal runaway state is in progress. The first early warning message may include an early warning level, the judgment result of the thermal runaway state, and the corresponding handling method for the judgment result of the thermal runaway state. The early warning level in the first early warning message may be a level one early warning.

[0030] When lithium-ion batteries experience thermal runaway, the expansion force of the lithium-ion battery responds earlier and faster than characteristic information such as temperature, voltage, and gas production. Therefore, in this embodiment, the force signal monitoring unit 110 monitors the expansion force of the lithium-ion battery in real time. The microprocessor unit 120 analyzes the expansion force based on a trained SVM classification model to obtain a judgment result on the potential thermal runaway of the lithium-ion battery. This allows for timely detection of early signs of thermal runaway and effective monitoring of the early stages of thermal runaway. Finally, the early warning unit 130 generates early warning information based on the potential thermal runaway judgment result, improving the timeliness of the warning and giving users or processing units more time to handle the abnormality of the lithium-ion battery to prevent thermal runaway and reduce the probability of cascading thermal runaway.

[0031] In some embodiments of the present invention, the lithium-ion battery thermal runaway early warning system 10 further includes: a control unit 140; The control unit 140 is used to generate a command to limit battery power and a command to start air cooling when the potential thermal runaway state is determined to be in a potential thermal runaway state.

[0032] In some embodiments of the present invention, the lithium-ion battery thermal runaway early warning system 10 further includes: an electrical signal monitoring unit 150; The electrical signal monitoring unit 150 is used to collect voltage information of the lithium-ion battery; The microprocessor unit 120 is used to analyze the voltage information through a trained SVM classification model to obtain voltage offset features, and to obtain the early state judgment result of thermal runaway of the lithium-ion battery based on the voltage offset features; wherein, the SVM classification model is obtained by training the voltage offset features under normal dynamic operating conditions of the lithium-ion battery and the voltage offset features in the early stage of thermal runaway. The early warning unit 130 is used to generate a second early warning message when the early stage of thermal runaway is determined to be in the early stage of thermal runaway.

[0033] Voltage information can be time-series voltage data, and voltage offset features can be voltage offset values, which are the differences between the voltages of two adjacent time points.

[0034] In this embodiment, the electrical signal monitoring unit 150 can be connected to both the lithium-ion battery and the microprocessor unit 120 to collect voltage information from the lithium-ion battery and transmit the voltage information to the microprocessor unit 120. Specifically, the electrical signal monitoring unit 150 may include a voltage sensor.

[0035] The microprocessor unit 120 is used to analyze the voltage information using a trained SVM classification model to obtain the early thermal runaway state judgment result of the lithium-ion battery. The early thermal runaway state judgment result of the lithium-ion battery includes: being in the early thermal runaway state and not being in the early thermal runaway state.

[0036] The early warning unit 130 is used to generate a second early warning message when the early stage of thermal runaway is determined to be in the early stage of thermal runaway. The second early warning message may include an early warning level, which may be a level two early warning.

[0037] Early stage of thermal runaway refers to the irreversible exothermic reaction that begins inside the battery, but has not yet triggered a chain reaction.

[0038] In the early stages of thermal runaway, sensor data clearly indicates a serious safety threat that could immediately trigger thermal runaway or irreversible damage, requiring the execution of the highest priority protective actions to prevent catastrophic consequences. Causes of this state include severe overcharging (lithium plating on the cathode, electrolyte decomposition), metal dendrite penetration caused by puncture / extrusion, and heat accumulation due to cooling failure.

[0039] The response speed of expansion force is the fastest, followed by the response speed of voltage information. Therefore, this embodiment uses voltage information to monitor the next stage of thermal runaway in lithium-ion batteries (early stage of thermal runaway).

[0040] In addition, dynamic operating conditions (such as high-rate charging and discharging or low-temperature operation) can cause normal voltage fluctuations in lithium-ion batteries. These fluctuations are difficult to distinguish from the slight voltage deviations in the early stages of thermal runaway, resulting in a false alarm rate of over 20%.

[0041] Therefore, this embodiment uses an SVM classification model to learn and distinguish the voltage offset characteristics of lithium-ion batteries under dynamic operating conditions and the voltage offset characteristics in the early stages of thermal runaway. This allows for rapid differentiation between the two modes when the lithium-ion battery voltage changes, improving the accuracy of identifying the early stages of thermal runaway in lithium-ion batteries.

[0042] Furthermore, the SVM classification model can accurately classify voltage dynamic anomalies based on the optimal hyperplane constructed from support vectors, and balance false alarms and false negatives by adjusting the penalty factor (such as the C parameter).

[0043] In some embodiments of the present invention, the trained SVM classification model is used to output a judgment result indicating that the lithium-ion battery is in the early stage of thermal runaway when the voltage of the lithium-ion battery reaches a local maximum and then decreases, and the voltage offset value during the decrease reaches an offset value threshold. The offset value threshold can be learned during the training of the SVM classification model.

[0044] In some embodiments of the present invention, the control unit 140 is further configured to generate a power cut-off command and a cooling phase change material command when the early thermal runaway state determination result indicates that the early thermal runaway state is in the early thermal runaway state.

[0045] In some embodiments of the present invention, the lithium-ion battery thermal runaway early warning system 10 further includes: a gas signal monitoring unit 160; The gas signal monitoring unit 160 is used to collect gas information from the lithium-ion battery; The microprocessor unit 120 is used to analyze the gas information of the lithium-ion battery through a trained SVM classification model to obtain the judgment result of the thermal runaway state of the lithium-ion battery. The early warning unit 130 is used to generate a third early warning message when the thermal runaway state judgment result is that the thermal runaway state is in a state of thermal runaway.

[0046] In this embodiment, the gas signal monitoring unit 160 can be connected to both the lithium-ion battery and the microprocessor unit 120 to collect gas information from the lithium-ion battery and transmit the gas information to the microprocessor unit 120. The gas information may include the gas concentration.

[0047] The microprocessor unit 120 is used to analyze gas information using a trained SVM classification model to obtain the thermal runaway state judgment result of the lithium-ion battery. The thermal runaway state judgment result of the lithium-ion battery includes: in a thermal runaway state and not in a thermal runaway state.

[0048] For example, an SVM classification model might output the classification result "in a thermal runaway state" when the concentration of the input hydrogen (H2) exceeds a first concentration threshold. Similarly, it might output the classification result "in a thermal runaway state" when the concentration of the input carbon monoxide (CO) exceeds a second concentration threshold. The first and second concentration thresholds can be learned during training.

[0049] Thermal runaway refers to a situation in which a battery has undergone a series of chain reactions that release energy at a rate that exceeds the controllable range.

[0050] The early warning unit 130 is used to generate a third early warning message when the thermal runaway state determination result indicates that the thermal runaway state is in progress. The third early warning message may include an early warning level, which can be a level three early warning.

[0051] Gas information responds more slowly than voltage information; therefore, this embodiment uses gas information to monitor the critical stage of thermal runaway in lithium-ion batteries.

[0052] In some embodiments of the present invention, the gas information is mixed gas information of multiple gases. The microprocessor unit 120 is further configured to decompose the mixed gas information into gas information components of each gas through a state Kalman filter, determine the diffusion coefficient of each gas according to the gas information components of each gas, and determine the target gas from each gas according to the diffusion coefficient of each gas. The microprocessor unit 120 is used to analyze the gas information components of the target gas through a trained SVM classification model to obtain the judgment result of the thermal runaway state of the lithium-ion battery.

[0053] In some embodiments of the present invention, the microprocessor unit 120 is further configured to compare the diffusion coefficient of each gas with the diffusion coefficient of a preset interfering gas. If the difference between the diffusion coefficient of the gas and the diffusion coefficient of the preset interfering gas is greater than a preset diffusion coefficient deviation threshold, and the energy ratio of the gas information component of the gas is greater than the energy ratio threshold, the gas is determined to be the target gas. The target gas may be CO, and / or H2.

[0054] This embodiment utilizes the difference in gas molecule diffusion coefficients to identify true positive signals. A true positive gas signal refers to a genuine hazardous gas leak event that is correctly identified and triggers an alarm. In complex gas detection scenarios, the target gas X may sometimes overlap with interfering gas signals, leading to a high false positive rate with traditional single-sensor methods. To address this issue, this embodiment employs a target gas X identification strategy combining dual Kalman filtering (DKF) with differences in gas diffusion coefficients. This strategy specifically includes: firstly, establishing a gas diffusion dynamics model with a time constant... τ Strictly related to the molecular diffusion coefficient D ( τ ∝1 / D). The early warning system 10 decomposes the mixed signal into gas concentration components in real time using a state Kalman filter, while the parameter Kalman filter dynamically optimizes the diffusion coefficient estimate ( ). ), utilizing the target gas X The residual is calculated based on the theoretical difference between the diffusion coefficient and the diffusion coefficient of the interfering gas. When the residual value is less than the recommended residual threshold and the sensor signal energy ratio exceeds the energy ratio threshold, the target gas X is determined to be a true positive. True positive signal identification significantly improves the reliability and efficiency of the early warning system 10. Through dual verification (diffusion coefficient residual + concentration threshold), the early warning system 10 can accurately distinguish between real leaks and interference signals, greatly reducing the false alarm rate.

[0055] In some embodiments of the present invention, the control unit 140 is also used to generate a fuse blow command and a fire extinguishing device activation command when the thermal runaway state determination result is that the thermal runaway state is in a thermal runaway state.

[0056] When the thermal runaway state is determined to be in a state of thermal runaway, it is highly likely to cause an explosion or deflagration, and the highest level of isolation, suppression and safety protection measures must be activated.

[0057] In some embodiments of the present invention, the lithium-ion battery thermal runaway early warning system 10 further includes a temperature signal monitoring unit 170. The temperature signal monitoring unit 170 can be connected to both the lithium-ion battery and the microprocessor unit 120, and is used to collect temperature information from the lithium-ion battery and transmit the temperature information to the microprocessor unit 120. Specifically, the temperature signal monitoring unit 170 may include at least one temperature sensor. The microprocessor unit 120 can be used to analyze the temperature information using a trained SVM classification model. The temperature information is used to assist in obtaining the thermal runaway state judgment result of the lithium-ion battery by analyzing expansion force, voltage information, and gas information.

[0058] Reference Figure 2 This diagram illustrates the training schematic of an SVM classification model provided by the present invention. The inputs to the SVM classification model include the expansion force of the lithium-ion battery, voltage information (voltage time-series data), and gas information (…). Information such as CO concentration and temperature (surface / internal gradient of the battery module) is used. Before input, a high-dimensional feature vector can be constructed from the above information, and then normalized and reduced in dimensionality by Principal Component Analysis (PCA) before input. During the training process of the SVM classification model, in order to minimize the Hinge Loss function, i.e., maximize the classification margin, the Hinge Loss function and constraints are derived into a dual problem. Then, the Lagrange multipliers in the dual problem are quickly solved using the Sequential Minimal Optimization (SMO) algorithm to minimize the loss function. Then, the accuracy of the obtained SVM classification model is evaluated using the test set. If the classification accuracy of the SVM classification model meets the requirements, the SVM classification model is determined to be feasible, and training ends; if the classification accuracy of the SVM classification model does not meet the requirements, the SVM classification model is determined to be infeasible, and the structure and hyperparameters of the SVM classification model are further optimized. The optimization of structure and hyperparameters includes readjusting at least one of its penalty parameters, kernel function, and loss function.

[0059] Reference Figure 3 This diagram illustrates the workflow of a lithium-ion battery thermal runaway early warning system 10 provided by the present invention. The lithium-ion battery thermal runaway early warning system 10 monitors multi-source information from the lithium-ion battery and analyzes this information using an SVM classification model to obtain classification results. These results include: in a potential thermal runaway state, in an early stage of thermal runaway, and in a thermal runaway state. When in a potential thermal runaway state, the lithium-ion battery thermal runaway early warning system 10 generates a power limiting command and a wind-cooling command to initiate heat dissipation, and issues an alarm. When in an early stage of thermal runaway, it generates a power cut-off command and a phase change material cooling command, and issues an alarm. When in a thermal runaway state, it generates a main fuse blow command and a fire extinguishing device activation command, and issues an alarm.

[0060] This embodiment monitors multiple sources of signals simultaneously. When multiple classification conditions are met simultaneously, an alert is issued based on the most severe classification result. For example, if lithium ions simultaneously meet the classification conditions of being in a potential thermal runaway state and in the early stage of thermal runaway (i.e., the expansion force is greater than the expansion force threshold and the lithium ion battery voltage reaches a local maximum value and then decreases, with the voltage deviation value during the decrease reaching the deviation value threshold), an alert is issued based on the classification result of being in the early stage of thermal runaway.

[0061] Reference Figure 4This diagram illustrates a typical experimental platform for monitoring and early warning of thermal runaway in lithium-ion batteries, provided by the present invention. The experimental platform includes an explosion-proof enclosure, an expansion force sensor, thermocouples, a battery cycle meter, a gas monitoring system, a data acquisition instrument, and a computer. During the experiment, the lithium-ion battery being monitored is placed in the explosion-proof enclosure. The expansion force sensor, thermocouples, data acquisition instrument, and gas monitoring system are used to collect information on the expansion force, voltage, gas content, and temperature of the monitored lithium-ion battery, respectively. The computer is used to analyze the collected information.

[0062] Reference Figure 5 The diagram illustrates a typical data curve for monitoring and early warning of thermal runaway in lithium-ion batteries provided by this invention. The first-level warning is based on expansion force; when the expansion force exceeds a threshold, a first-level warning is initiated. The second-level warning is initiated when the voltage reaches a local maximum value and then begins to decrease, and the voltage deviation ΔU during the decrease exceeds 0.22V. The third-level warning is based on gas information; when the concentration of the target gas exceeds a certain value, a third-level warning is initiated.

[0063] Referring to Table 1, a schematic diagram of key node parameters of a typical data curve for monitoring and early warning of thermal runaway in lithium-ion batteries provided by this invention is shown. Table 1 shows that monitoring thermal runaway of lithium-ion batteries based on expansion force has the fastest response speed, detecting potential thermal runaway states 3103-1564=1539 s in advance. The second fastest method is voltage offset characteristics, which can detect early thermal runaway states 3103-1873=1230 s in advance. When a level 3 warning is triggered by gas information monitoring, thermal runaway is already imminent, and at this point, it can be directly treated as a thermal runaway state.

[0064] Table 1. Schematic diagram of key node parameters at each early warning stage

[0065] Reference Figure 6 The diagram illustrates a flowchart of an embodiment of the lithium-ion battery thermal runaway early warning method provided by the present invention. This method is applied to any of the above-mentioned lithium-ion battery thermal runaway early warning systems and includes: S601 collects the expansion force of the lithium-ion battery through a force signal monitoring unit; S602 uses a microprocessor unit to analyze the expansion force based on a trained SVM classification model, and obtains the judgment result of the potential thermal runaway state of the lithium-ion battery. S603, when the judgment result of the potential thermal runaway state is that the early warning unit is in a potential thermal runaway state, the first early warning information is generated.

[0066] It should be noted that the implementation principle or process of the above method can be referred to the aforementioned embodiment of the lithium-ion battery thermal runaway early warning system, and will not be elaborated here.

[0067] Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by a computer program instructing related hardware, and the program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.

[0068] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A lithium-ion battery thermal runaway early warning system, characterized in that, include: Force signal monitoring unit, microprocessor unit, and early warning unit; The force signal monitoring unit is used to collect the expansion force of the lithium-ion battery; The microprocessor unit is used to analyze the expansion force through a trained SVM classification model to obtain the judgment result of the potential thermal runaway state of the lithium-ion battery. The early warning unit is used to generate a first early warning message when the judgment result of the potential thermal runaway state is that it is in a potential thermal runaway state.

2. The lithium-ion battery thermal runaway early warning system according to claim 1, characterized in that, Also includes: Control unit; The control unit is used to generate a command to limit battery power and a command to start air cooling when the potential thermal runaway state is determined to be in a potential thermal runaway state.

3. The lithium-ion battery thermal runaway early warning system according to claim 1, characterized in that, It also includes: an electrical signal monitoring unit; The electrical signal monitoring unit is used to collect the voltage information of the lithium-ion battery; The microprocessor unit is used to analyze the voltage information using a trained SVM classification model to obtain voltage offset features, and to obtain the early thermal runaway state judgment result of the lithium-ion battery based on the voltage offset features; wherein, the SVM classification model is obtained by training the voltage offset features under normal dynamic operating conditions of the lithium-ion battery and the voltage offset features in the early stage of thermal runaway. The early warning unit is used to generate a second early warning message when the early thermal runaway state judgment result indicates that the thermal runaway state is in its early stage.

4. The lithium-ion battery thermal runaway early warning system according to claim 3, characterized in that, The trained SVM classification model is used to output a judgment result indicating that the lithium-ion battery is in the early stage of thermal runaway when the voltage of the lithium-ion battery reaches a local maximum value and then decreases, and the voltage offset value during the decrease reaches the offset value threshold.

5. The lithium-ion battery thermal runaway early warning system according to claim 3, characterized in that, Also includes: Control unit; The control unit is used to generate a power cut-off command and a cooling phase change material command when the early thermal runaway state determination result is that the thermal runaway early state is in the early thermal runaway state.

6. The lithium-ion battery thermal runaway early warning system according to claim 1, characterized in that, Also includes: Gas signal monitoring unit; The gas signal monitoring unit is used to collect gas information from the lithium-ion battery; The microprocessor unit is used to analyze the gas information of the lithium-ion battery through a trained SVM classification model to obtain the thermal runaway state judgment result of the lithium-ion battery. The early warning unit is used to generate a third early warning message when the thermal runaway state determination result is that the thermal runaway state is in a state of thermal runaway.

7. The lithium-ion battery thermal runaway early warning system according to claim 6, characterized in that, The gas information is a mixture of multiple gases. The microprocessor unit is also used to decompose the mixture of gas information into gas information components of each gas through a state Kalman filter, determine the diffusion coefficient of each gas based on the gas information components of each gas, and determine the target gas from each gas based on the diffusion coefficient of each gas. The microprocessor unit is used to analyze the gas information components of the target gas using a trained SVM classification model to obtain the thermal runaway state judgment result of the lithium-ion battery.

8. The lithium-ion battery thermal runaway early warning system according to claim 7, characterized in that, The microprocessor unit is further configured to compare the diffusion coefficient of each gas with the diffusion coefficient of a preset interfering gas. If the difference between the diffusion coefficient of the gas and the diffusion coefficient of the preset interfering gas is greater than a preset diffusion coefficient deviation threshold, and the energy ratio of the gas information component of the gas is greater than the energy ratio threshold, the gas is determined to be the target gas.

9. The lithium-ion battery thermal runaway early warning system according to claim 6, characterized in that, Also includes: Control unit; The control unit is used to generate a command to blow the main fuse and a command to activate the fire extinguishing device when the thermal runaway state determination result is that the thermal runaway state is in a thermal runaway state.

10. A method for early warning of thermal runaway in lithium-ion batteries, characterized in that, The method, applied to the lithium-ion battery thermal runaway early warning system according to any one of claims 1 to 9, comprises: The expansion force of the lithium-ion battery is collected by the force signal monitoring unit. The microprocessor unit is used to analyze the expansion force based on the trained SVM classification model to obtain the judgment result of the potential thermal runaway state of the lithium-ion battery. When the potential thermal runaway state is determined to be in a potential thermal runaway state by the early warning unit, a first early warning message is generated.

Images