Method and device for evaluating safety of lithium ion battery and storage medium

Abstract

The invention discloses a lithium ion battery safety assessment method and device and a storage medium, relates to the technical field of battery detection, and mainly aims to improve the lithium ion battery safety assessment accuracy. Comprising the following steps: carrying out step-by-step enhanced energy input on a to-be-evaluated lithium ion battery, and detecting real-time state data of the to-be-evaluated lithium ion battery after energy input; based on the real-time state data, judging whether the to-be-evaluated lithium ion battery meets a triggering condition for recording abnormal event evolution data, and if yes, collecting the abnormal event evolution data of the to-be-evaluated lithium ion battery in a preset time window in real time, the preset time window refers to a moment when the to-be-evaluated lithium ion battery meets the triggering condition to a moment when the abnormal event of the to-be-evaluated lithium ion battery is evolved completely; and based on the abnormal event evolution data, constructing an abnormal evolution mathematical model of the lithium ion battery to be evaluated, and based on the abnormal evolution mathematical model, evaluating the safety of the lithium ion battery.

Description

Technical Field

[0001] The present invention relates to the technical field of battery detection, and in particular, to a method, device and storage medium for evaluating the safety of lithium-ion batteries. Background Art

[0002] As an efficient energy storage device, lithium-ion batteries have been widely used in many fields due to their significant advantages such as high working voltage, high energy density, high specific capacity, long cycle life and environmental friendliness. From common portable electronic devices such as smart phones and laptop computers to emerging electric vehicles and large-scale energy storage systems, all rely on lithium-ion batteries. Based on this, in order to optimize the manufacturing process of lithium-ion batteries, it is particularly important to evaluate the safety of lithium-ion batteries.

[0003] Currently, the safety of lithium-ion batteries is usually evaluated only by measuring the battery capacity. However, this method of safety evaluation has single data and cannot accurately evaluate the safety of lithium-ion batteries. Summary of the Invention

[0004] The present invention provides a method, device and storage medium for evaluating the safety of lithium-ion batteries, mainly aiming at improving the accuracy of evaluating the safety of lithium-ion batteries.

[0005] According to the first aspect of the present invention, there is provided a method for evaluating the safety of a lithium-ion battery, which is applied to a safety evaluation device. The safety evaluation device includes a detection device and an energy input device, and includes:

[0006] Controlling the energy input device to perform step-by-step enhanced energy input on the lithium-ion battery to be evaluated in the safety evaluation device, and using the detection device to detect the real-time state data of the lithium-ion battery to be evaluated after the energy input;

[0007] Based on the real-time state data, determining whether the lithium-ion battery to be evaluated meets the trigger condition for recording the evolution data of abnormal events. If it meets, triggering the detection device to collect the evolution data of abnormal events of the lithium-ion battery to be evaluated within a preset time window in real time, where the preset time window refers to the time from the moment when the lithium-ion battery to be evaluated meets the trigger condition to the moment when the evolution of abnormal events of the lithium-ion battery to be evaluated is completed;

[0008] Based on the evolution data of abnormal events, constructing an abnormal evolution mathematical model of the lithium-ion battery to be evaluated during the evolution of abnormal events, and based on the abnormal evolution mathematical model, evaluating the safety of the lithium-ion battery to be evaluated.

[0009] Optionally, determining whether the lithium-ion battery to be evaluated meets the trigger condition for recording abnormal event evolution data based on the real-time status data includes:

[0010] Construct time series data from the real-time status data at the current moment and each historical moment before the current moment, and input the time series data into a preset battery status recognition model for running status recognition to obtain the running status of the lithium-ion battery to be evaluated at the current moment;

[0011] If the running status is a healthy running status, it is determined that the lithium-ion battery to be evaluated does not meet the trigger condition for recording abnormal event evolution data. If the running status is an accident running status, it is determined that the lithium-ion battery to be evaluated meets the trigger condition for recording abnormal event evolution data.

[0012] Optionally, the abnormal event evolution data includes image data and various numerical data of the lithium-ion battery to be evaluated during the abnormal event evolution process;

[0013] Constructing an abnormal evolution mathematical model of the lithium-ion battery to be evaluated during the abnormal event evolution process based on the abnormal event evolution data includes:

[0014] Input the image data into a preset image analysis model for image feature extraction and classification prediction to obtain the physical structure change information and flame form change information of the lithium-ion battery to be evaluated during the abnormal event evolution process;

[0015] Input various numerical data into a preset numerical analysis model for regression analysis and classification prediction to obtain the energy release process information and chemical reaction path information of the lithium-ion battery to be evaluated during the abnormal event evolution process;

[0016] Based on the physical structure change information, the flame form change information, the energy release process information, and the chemical reaction path information, construct an abnormal evolution mathematical model of the lithium-ion battery to be evaluated during the abnormal event evolution process.

[0017] Optionally, the energy input device adopts a combination of a high-frequency pulse power supply and a microwave generator;

[0018] Controlling the energy input device to perform step-by-step enhanced energy input on the lithium-ion battery to be evaluated in the safety evaluation device includes:

[0019] Control the high-frequency pulse power supply to deliver a gradually increasing pulsed current with adjustable amplitude and a frequency within a first preset frequency range to the lithium-ion battery to be evaluated in the safety evaluation device, and control the microwave generator to emit a gradually increasing microwave signal with controllable power and a frequency within a second preset frequency range to the lithium-ion battery to be evaluated in the safety evaluation device;

[0020] The detection device includes a plurality of detection units, and each detection unit is composed of a variety of sensors;

[0021] Before using the detection device to detect the real-time state data of the lithium-ion battery to be evaluated after energy input, the method further includes:

[0022] Determine the position information of the lithium-ion battery to be evaluated in the safety evaluation device, and arrange each detection unit in a preset arrangement form based on the position information.

[0023] Optionally, the inner wall of the cabin of the safety evaluation device is provided with multiple layers of heat insulation and buffer materials, a plurality of ventilation openings with different apertures and positions are provided on the cabin, and intelligent regulating valves are equipped at preset positions of the cabin.

[0024] Optionally, after triggering the detection device to collect the abnormal event evolution data of the lithium-ion battery to be evaluated within a preset time window in real time, the method further includes:

[0025] Perform at least one of data cleaning and normalization on the abnormal event evolution data to obtain the preprocessed abnormal event evolution data;

[0026] Classify the preprocessed abnormal event evolution data to obtain the abnormal event evolution data under different classification categories, and encrypt the abnormal event evolution data under each classification category to obtain the encrypted abnormal event evolution data under each classification category;

[0027] Distribute and store the encrypted abnormal event evolution data under each classification category according to the classification category to obtain the abnormal event evolution data under different storage nodes.

[0028] Optionally, the evaluation of the safety of the lithium-ion battery to be evaluated based on the abnormal evolution mathematical model includes:

[0029] Extract at least one safety evaluation parameter of the lithium-ion battery to be evaluated from the abnormal evolution mathematical model, and evaluate the safety of the lithium-ion battery to be evaluated based on each safety evaluation parameter and its corresponding preset parameter threshold.

[0030] According to a second aspect of the present invention, there is provided an evaluation device for the safety of a lithium-ion battery, which is applied to a safety evaluation device. The safety evaluation device includes a detection device and an energy input device, and comprises:

[0031] A control unit, configured to control the energy input device to perform step-by-step enhanced energy input on the lithium-ion battery to be evaluated in the safety evaluation device, and use the detection device to detect the real-time state data of the lithium-ion battery to be evaluated after the energy input;

[0032] An acquisition unit, configured to determine whether the lithium-ion battery to be evaluated meets the trigger condition for recording the evolution data of abnormal events based on the real-time state data. If it meets the condition, trigger the detection device to collect the evolution data of abnormal events of the lithium-ion battery to be evaluated within a preset time window in real time, where the preset time window refers to the moment when the lithium-ion battery to be evaluated meets the trigger condition to the moment when the evolution of abnormal events of the lithium-ion battery to be evaluated is completed;

[0033] An evaluation unit, configured to construct an abnormal evolution mathematical model of the lithium-ion battery to be evaluated during the evolution of abnormal events based on the evolution data of abnormal events, and evaluate the safety of the lithium-ion battery to be evaluated based on the abnormal evolution mathematical model.

[0034] According to a third aspect of the present invention, there is provided a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the above-mentioned evaluation device for the safety of a lithium-ion battery is implemented.

[0035] According to a fourth aspect of the present invention, there is provided a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the above-mentioned evaluation device for the safety of a lithium-ion battery is implemented.

[0036] An evaluation method, device, and storage medium for the safety of a lithium-ion battery provided by the present invention. Compared with the current method of evaluating the safety of a lithium-ion battery only by measuring the battery capacity, in the present invention, an energy input device in a safety evaluation device is used to input energy to the lithium-ion battery to be evaluated. During the energy input process, when it is detected that the lithium-ion battery to be evaluated meets the trigger condition for recording abnormal event evolution data, the abnormal event evolution data of the lithium-ion battery to be evaluated is collected, and an abnormal evolution mathematical model is constructed based on the abnormal event evolution data. Finally, the safety of the lithium-ion battery to be evaluated is evaluated according to the abnormal evolution mathematical model. Thus, by starting to collect abnormal event evolution data only when the lithium-ion battery to be evaluated meets the trigger condition for recording abnormal event evolution data, the time and computing resources wasted in processing data when no abnormality occurs can be avoided, thereby improving the safety evaluation efficiency of the lithium-ion battery and reducing the safety evaluation cost. At the same time, since the abnormal evolution mathematical model contains all the evolution process information of the lithium-ion battery in abnormal events, evaluating the safety of the lithium-ion battery through the abnormal evolution mathematical model can improve the accuracy of the safety evaluation of the lithium-ion battery. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0038] Figure 1 Shows a flowchart of an evaluation method for the safety of a lithium-ion battery provided by an embodiment of the present invention;

[0039] Figure 2 Shows a flowchart of another evaluation method for the safety of a lithium-ion battery provided by an embodiment of the present invention;

[0040] Figure 3 Shows a schematic structural diagram of an evaluation device for the safety of a lithium-ion battery provided by an embodiment of the present invention;

[0041] Figure 4 Shows a schematic structural diagram of another evaluation device for the safety of a lithium-ion battery provided by an embodiment of the present invention;

[0042] Figure 5 Shows a schematic physical structure diagram of a computer device provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0043] The present invention will be described in detail below with reference to the drawings and in conjunction with the embodiments. It should be noted that, without conflict, the embodiments in this application and the features in the embodiments can be combined with each other.

[0044] Currently, the safety of lithium-ion batteries is only evaluated by measuring the battery capacity. The data is single and cannot accurately evaluate the safety of lithium-ion batteries.

[0045] To solve the above problems, an embodiment of the present invention provides a method for evaluating the safety of a lithium-ion battery, as Figure 1 shown, the method includes:

[0046] 101. Control the energy input device to perform step-by-step enhanced energy input on the lithium-ion battery to be evaluated in the safety evaluation device, and use the detection device to detect the real-time state data of the lithium-ion battery to be evaluated after the energy input.

[0047] Based on the safety evaluation device, the embodiment of the present invention evaluates the safety of the lithium-ion battery. The safety evaluation device is composed of a special environment simulation chamber, a detection device, an energy input device, an intelligent trigger mechanism, etc. Special environment simulation chamber: A sealed chamber body is constructed using high-strength, high-temperature resistant, and insulating composite materials. Multiple layers of heat insulation and buffer materials are provided on the inner wall of the chamber. Multiple ventilation openings with different apertures and positions are provided on the chamber body, and intelligent regulating valves are equipped to automatically control the ventilation volume and air flow direction according to the detection requirements. Further, the composite material of the special environment simulation chamber can be a carbon fiber reinforced ceramic matrix composite material, which has excellent high-temperature resistance, high strength, and impact resistance, and can effectively withstand the impact force and high temperature when the battery malfunctions, avoiding the rupture of the chamber body and the excessive heat dissipation affecting the detection results. Its working principle is to use the high strength of carbon fiber to enhance the brittleness of the ceramic, so that it has sufficient mechanical strength while maintaining good heat insulation performance. In the implementation manner, after mixing carbon fiber and ceramic powder in a specific ratio (the specific ratio is set according to actual needs), the chamber body components are prepared by a hot pressing forming process, and then precision assembly and sealing treatment are carried out. The special environment simulation chamber in the embodiment of the present invention can more realistically simulate the environment of the battery under various actual working conditions, effectively reduce the interference of external factors on the detection results, and provide stable and reliable conditions for accurately detecting the abnormal state of the battery.

[0048] The detection device consists of a multi-functional detection array: around the placement position of the lithium-ion battery in the simulation cabin, multiple detection units are distributed according to a specific geometric pattern (the pattern of the figure is set according to actual needs). Each detection unit integrates various data acquisition devices such as multiple sensors, including but not limited to high-precision infrared temperature sensors, highly sensitive electrochemical gas sensors, micro pressure sensors, and high-definition optical imaging sensors, etc. Multiple detection units can be connected to the central data processing unit through low-latency and high-bandwidth data transmission lines for data processing. Further, the infrared temperature sensor in the multi-functional detection array can use a mercury cadmium telluride detector, and its precise measurement value range can be set to -40°C - 1000°C, with a resolution of up to 0.1°C. It can quickly capture the minute temperature changes on the battery surface and the surrounding environment, and timely detect signs of thermal runaway. Its working principle is based on the infrared absorption characteristics of the mercury cadmium telluride material, which converts the received infrared radiation into an electrical signal, and after being processed by the built-in amplification and filtering circuits, it is transmitted to the central data processing unit. In the implementation, the sensor is installed on a specially made high-temperature and corrosion-resistant bracket to ensure that it can be stably aligned with the battery detection area, and is connected to the data processing unit through a shielded cable to reduce electromagnetic interference. Further, the electrochemical gas sensor uses a nanostructured metal oxide sensitive material, which has high sensitivity and selectivity to the gases generated under abnormal battery states such as hydrogen, carbon monoxide, and methane, and can quickly respond and measure the gas concentration at low concentrations. Its working principle is to utilize the adsorption and redox reaction of gas molecules on the surface of the nano metal oxide, resulting in a change in the sensor resistance or potential, and the gas concentration is determined by measuring this change. In the implementation, the sensor chip is integrated on a micro circuit board, and a miniaturized and low-power sensor module is manufactured through micro-electro-mechanical system (MEMS) technology and installed at the corresponding position of the detection array and connected to the data transmission line. The multi-functional detection array in the embodiment of the present invention adopts advanced sensors and a reasonable layout, and can comprehensively and accurately monitor various physical and chemical changes during the abnormal operation of the battery, and obtain richer and more accurate data information.

[0049] Energy input device: It can precisely control the type of energy input to a lithium-ion battery, including but not limited to pulsed current and high-frequency electromagnetic radiation. Further, a specially designed energy input device uses a combination of a high-frequency pulsed power supply and a microwave generator. The high-frequency pulsed power supply generates a pulsed current with a frequency in the range of 1 kHz - 100 kHz and an adjustable amplitude, and the microwave generator emits a microwave signal with a frequency in the range of 1 GHz - 10 GHz and a controllable power. By precisely controlling the output parameters of both, it realizes precise stimulation of the internal chemical reactions of the lithium-ion battery and induces abnormal events (where abnormal events such as thermal runaway events of the battery, events where the temperature exceeds a preset threshold, etc.). Its working principle is that the pulsed current can cause the intensification of the internal electrochemical reactions of the battery, and the microwave radiation can change the internal electromagnetic field distribution and ion migration rate of the battery through interaction with the battery materials, jointly promoting the battery to reach the critical state of abnormal events. In the implementation manner, the high-frequency pulsed power supply and the microwave generator are connected to the battery electrodes through a power amplifier and a matching network, and the output parameters are adjusted through a controller and work in coordination with the detection system to ensure the induction of abnormal events under safe and controllable conditions. The energy input device in the embodiment of the present invention can precisely control the type and intensity of energy input, not only can simulate more diverse battery fault situations, but also can better study the abnormal operation mechanism of the battery under different conditions, providing more valuable references for battery safety design.

[0050] Intelligent trigger mechanism: It is used to automatically trigger the detection device to start comprehensively recording the evolution data of abnormal events. At the same time, an emergency trigger button can be set to immediately start the detection of the evolution data of abnormal events in case of an emergency. The intelligent trigger mechanism in the embodiment of the present invention uses machine learning algorithms to achieve early warning and automatic trigger detection of the abnormal operation state of the battery, with higher accuracy and flexibility, and can timely capture potential safety hazards of the battery.

[0051] Among them, the real-time status data includes real-time temperature, real-time voltage, real-time current, real-time internal resistance, etc. of the lithium-ion battery to be evaluated after energy input.

[0052] Specifically, when receiving a safety evaluation signal for the lithium-ion battery to be evaluated, the energy input device is controlled to perform energy input operations such as step-by-step enhanced pulsed current input and high-frequency electromagnetic radiation on the lithium-ion battery to be evaluated in the safety evaluation device. At the same time, various detection arrays on the detection devices around the lithium-ion battery are used to collect real-time temperature, real-time gas release type, real-time pressure, real-time appearance morphology data, etc. of the lithium-ion battery to be evaluated after energy input.

[0053] 102. Based on the real-time status data, determine whether the lithium-ion battery to be evaluated meets the trigger condition for recording the evolution data of abnormal events. If it meets the condition, trigger the detection device to collect the evolution data of abnormal events of the lithium-ion battery to be evaluated within a preset time window. Here, the preset time window refers to the time from the moment when the lithium-ion battery to be evaluated meets the trigger condition to the moment when the evolution of the abnormal events of the lithium-ion battery to be evaluated is completed.

[0054] Among them, as the energy of the lithium-ion battery to be evaluated gradually increases, an abnormal operating state will occur. The evolution data of abnormal events refers to the evolution process data of the lithium-ion battery to be evaluated in the abnormal operating state, including image data, temperature data, pressure data, gas concentration data, current data, voltage data, internal resistance data, flame attribute data, appearance form data, etc. of the lithium-ion battery to be evaluated in the abnormal state. The abnormal operating state includes a thermal runaway state (such as a sharp rise in battery temperature), a state of out-of-control chemical reactions inside the battery, a state of out-of-control internal pressure of the battery (such as the internal pressure of the battery reaching the pressure critical value), and a state of out-of-control appearance form of the battery (such as battery swelling, splitting, etc.). The length of the preset time window can be set according to actual needs.

[0055] For the embodiments of the present invention, after the detection device detects the real-time status data of the lithium-ion battery to be evaluated after energy input, based on the real-time status data, determine whether the lithium-ion battery to be evaluated reaches an abnormal operating state, that is, use an algorithm based on machine learning to analyze the real-time status data of the battery. When the data characteristics of the real-time status data conform to the preset dangerous trend model, it is determined that the lithium-ion battery to be evaluated reaches an abnormal operating state. At this time, the intelligent trigger mechanism starts to trigger the detection device to collect the evolution data of abnormal events of the lithium-ion battery to be evaluated from the start time of the abnormal operating state to the end time of the evolution of abnormal events. Finally, evaluate the safety of the lithium-ion battery according to the evolution data of abnormal events. Thus, by starting to collect the evolution data of abnormal events only when the lithium-ion battery to be evaluated meets the trigger condition for recording the evolution data of abnormal events, it is possible to avoid wasting time and computing resources for processing data when no abnormality occurs, thereby improving the safety evaluation efficiency of the lithium-ion battery and reducing the safety evaluation cost.

[0056] 103. Based on the evolution data of abnormal events, construct an abnormal evolution mathematical model for the lithium-ion battery to be evaluated during the evolution process of abnormal events, and based on the abnormal evolution mathematical model, evaluate the safety of the lithium-ion battery to be evaluated.

[0057] For the embodiments of the present invention, based on the abnormal event evolution data, information such as the energy accumulation and release process, chemical reaction path, physical structure change, flame form, etc. of the lithium-ion battery to be evaluated during the abnormal event evolution process is determined. Finally, an abnormal evolution mathematical model is constructed according to the information such as the energy accumulation and release process, chemical reaction path, physical structure change, flame form, etc., and finally the safety of the lithium-ion battery to be evaluated is determined according to the abnormal evolution mathematical model. Since the abnormal evolution mathematical model contains all the evolution process information of the lithium-ion battery during the abnormal event, the embodiments of the present invention evaluate the safety of the lithium-ion battery through the abnormal evolution mathematical model, which can improve the accuracy of the safety evaluation of the lithium-ion battery.

[0058] According to an evaluation method for the safety of a lithium-ion battery provided by the present invention, compared with the current method of only evaluating the safety of a lithium-ion battery by measuring the battery capacity, the present invention inputs energy to the lithium-ion battery to be evaluated through the energy input device in the safety evaluation device. During the energy input process, when it is detected that the lithium-ion battery to be evaluated meets the trigger condition for recording abnormal event evolution data, the abnormal event evolution data of the lithium-ion battery to be evaluated is collected, and an abnormal evolution mathematical model is constructed based on the abnormal event evolution data. Finally, the safety of the lithium-ion battery to be evaluated is evaluated according to the abnormal evolution mathematical model. Thus, by starting to collect the abnormal event evolution data only when the lithium-ion battery to be evaluated meets the trigger condition for recording the abnormal event evolution data, the time and computing resources wasted in processing the data when there is no abnormality can be avoided, thereby improving the efficiency of the safety evaluation of the lithium-ion battery and reducing the safety evaluation cost. At the same time, since the abnormal evolution mathematical model contains all the evolution process information of the lithium-ion battery during the abnormal event, evaluating the safety of the lithium-ion battery through the abnormal evolution mathematical model can improve the accuracy of the safety evaluation of the lithium-ion battery.

[0059] Furthermore, in order to better illustrate the above process of evaluating the safety of the lithium-ion battery, as a refinement and extension of the above embodiments, the embodiments of the present invention provide another evaluation method for the safety of the lithium-ion battery, as Figure 2 shown, the method includes:

[0060] 201. Control the energy input device to perform step-by-step enhanced energy input on the lithium-ion battery to be evaluated in the safety evaluation device, and use the detection device to detect the real-time state data of the lithium-ion battery to be evaluated after the energy input.

[0061] For the embodiments of the present invention, multi-layer heat insulation and buffering materials are provided on the inner wall of the cabin of the safety assessment device. A plurality of ventilation openings with different apertures and positions are provided on the cabin, and intelligent regulating valves are equipped at preset positions of the cabin. The preset positions can be set according to actual requirements.

[0062] Specifically, the material selection and preparation process of the special environment simulation cabin in the safety assessment device includes: selecting a carbon fiber reinforced ceramic matrix composite material as the main material of the simulation cabin. This material is obtained by mixing carbon fiber and ceramic powder in an accurate proportion. For example, the proportion of carbon fiber is 30%, and the proportion of ceramic powder is 70%. After mixing, it is placed in a special mold and hot-pressed at high temperature and high pressure (for example, the temperature is about 1500 °C and the pressure is about 20 MPa) to obtain the required cabin components, such as cabin walls, cabin tops, cabin bottoms, etc.; for the heat insulation and buffering materials on the inner wall of the cabin, a multi-layer structure can be adopted. First, a ceramic fiber heat insulation felt with a preset thickness (such as 5 mm) is laid on the inner wall of the cabin to reduce heat transfer; then a 3-mm-thick elastic silicone buffer layer is covered on the surface of the heat insulation felt to prevent the cabin from being damaged by the impact of debris generated by abnormal events of the battery; for the ventilation openings, according to the design requirements, a plurality of ventilation openings are opened at different positions of the cabin, such as circular ventilation openings with different apertures are opened at the top, bottom, and side respectively. For example, the apertures are 10 mm, 20 mm, and 30 mm respectively. The ventilation openings adopt electric intelligent regulating valves. The control system of the valves can automatically adjust the ventilation volume and air flow direction according to the detected different stages and requirements according to the preset program. For example, in the initial stage, the valve opening degree of the ventilation opening is set to 20% to ensure that the gas composition and air pressure in the simulation cabin are in the initial set state; in the stage of inducing abnormal events, the ventilation volume is adjusted in a timely manner according to the data fed back by the sensors to avoid the influence of excessive pressure on the detection results.

[0063] Furthermore, the detection device includes a plurality of detection units, and each detection unit is composed of a variety of sensors; the arrangement method of each detection unit includes: determining the position information of the lithium-ion battery to be evaluated in the safety assessment device, and arranging each detection unit in a preset arrangement form based on the position information. The preset arrangement form can be set as needed according to the position of the battery.

[0064] Specifically, the installation and arrangement process of the multi-functional detection array in the detection device includes: Integration of detection units: For each detection unit, detection devices such as high-precision infrared temperature sensors, highly sensitive electrochemical gas sensors, micro pressure sensors, and high-definition optical imaging sensors are integrated together. For example, the infrared temperature sensor uses a mercury cadmium telluride detector, which is installed on a special metal bracket with good high-temperature resistance and corrosion resistance, capable of withstanding high temperatures above 1000°C. The bracket is fixed in the simulation chamber by screws or welding to ensure that the sensor can accurately align with the key parts of the lithium-ion battery, such as the positive and negative electrodes, center, and edge regions of the battery. The detector and the data transmission line are connected by a shielded cable to reduce electromagnetic interference. The transmission line uses low-latency, high-bandwidth optical fiber or coaxial cable to transmit the collected data to the central data processing unit; The electrochemical gas sensor uses a nanostructured metal oxide sensitive material, which is fabricated on a microcircuit board through microelectromechanical system (MEMS) technology to form a miniaturized and low-power sensor module. The module is installed at a specific position in the detection array to ensure that it can fully contact the gas environment around the battery, and the distance from the battery can be maintained within a preset range, such as between 5-10 cm, so as to accurately measure the concentrations of various gases generated during the abnormal state of the lithium-ion battery; The micro pressure sensors are distributed at different positions in the simulation chamber, especially near the battery. Its range is selected according to the maximum pressure expected in the simulation chamber, such as 0-10 MPa. It is connected to the environment around the battery through a dedicated pressure transmission pipeline, and converts the sensed pressure signal into an electrical signal and transmits it to the data processing unit; The high-definition optical imaging sensor uses a high-pixel camera, which is installed on the top and side of the simulation chamber. Through a special high-temperature-resistant and impact-resistant transparent window, the appearance changes of the battery are observed. The shooting frequency of the camera can be set to 30 frames per second or higher as needed to capture the subtle changes of the battery during abnormal events; For example, a geometric pattern of a regular hexagon or square grid can be used for distribution. Taking the regular hexagon grid as an example, multiple detection units are arranged at the vertices and the center of a regular hexagon centered on the lithium-ion battery. This distribution method can ensure the formation of a relatively uniform and dense monitoring network around the battery. Due to the characteristics of the regular hexagon, the distance between each detection unit and its adjacent unit is relatively balanced, so that there will be no monitoring dead spots or overly sparse local monitoring when detecting various physical and chemical changes during the abnormal operation state of the battery. For example, when the battery undergoes thermal runaway and causes a sharp rise in temperature, no matter from which direction the heat spreads from the battery, the detection units at the vertices and the center of the regular hexagon grid can promptly sense the temperature change and quickly transmit the data to the central data processing unit;For a square grid distribution, the principle is similar. The detection units are located at the vertices and the center of the square, which can cover the area around the battery on the plane. In practical applications, factors such as the shape and size of the battery and the accuracy requirements of detection need to be considered when choosing which geometric distribution to use. If the battery has a relatively regular shape and extremely high requirements for all-round monitoring accuracy, a regular hexagon grid may be more advantageous; if the battery is in a relatively regular installation environment and there are certain requirements for the simplicity of the monitoring layout, a square grid can also meet the basic detection needs. This method of distributing detection units according to a specific geometric pattern can greatly improve the reliability and accuracy of detection compared to random or irregular arrangements, providing a solid foundation for comprehensively understanding various changes in lithium-ion batteries during abnormal operation.

[0065] Further, after the energy input device, detection device, simulation chamber, etc. in the safety assessment device are arranged, when receiving the safety assessment signal of the lithium-ion battery to be evaluated, it is first necessary to control the energy input device to input energy to the lithium-ion battery to be evaluated. Based on this, the method includes: the energy input device uses a combination of a high-frequency pulse power supply and a microwave generator; controlling the high-frequency pulse power supply to deliver a gradually increasing pulsed current with a frequency in the first preset frequency range and adjustable amplitude to the lithium-ion battery to be evaluated in the safety assessment device, and controlling the microwave generator to emit a gradually increasing microwave signal with a frequency in the second preset frequency range and controllable power to the lithium-ion battery to be evaluated in the safety assessment device.

[0066] Among them, both the first preset frequency range and the second preset frequency range can be set according to actual needs. Specifically, for example, the energy input device adopts a combination of a high-frequency pulse power supply and a microwave generator. For example, the high-frequency pulse power supply generates a pulse current with an adjustable frequency between 1 kHz and 100 kHz and an amplitude between 0 and 10 A for the lithium-ion battery to be evaluated, and the microwave generator emits a microwave signal with a controllable frequency between 1 GHz and 10 GHz and a power between 0 and 100 W to the lithium-ion battery to be evaluated. In this way, the operation simulation of the corresponding working conditions of the lithium-ion battery to be evaluated is carried out, and the real-time state data of the lithium-ion battery to be evaluated after energy input is recorded in real time. In the embodiment of the present invention, the high-frequency pulse power supply and the microwave generator can be connected to the electrodes of the lithium-ion battery to be evaluated through a power amplifier and a matching network. The power amplifier amplifies the power output by the power supply to the required level according to the control signal, and the matching network ensures that the input energy matches the load characteristics of the lithium-ion battery to be evaluated, avoiding energy loss caused by energy reflection and damage to the equipment. By starting the input of a low-frequency and low-amplitude pulse current through the controller, such as a pulse current with a frequency of 1 kHz and an amplitude of 1 A, observing the state of the lithium-ion battery to be evaluated, and then gradually increasing the frequency and amplitude of the pulse current, and at the same time adjusting the frequency and power of the microwave generator according to the state of the lithium-ion battery to be evaluated, gradually increasing the energy input until the state of the battery meets the safety detection requirements (such as the battery thermal runaway state, the battery explosion state, the battery rupture state, etc.). During this process, the detection array in the detection device continuously monitors the state change of the battery.

[0067] 202. Based on the real-time state data, determine whether the lithium-ion battery to be evaluated meets the trigger condition for recording the evolution data of abnormal events. If it meets, trigger the detection device to collect the evolution data of abnormal events of the lithium-ion battery to be evaluated within a preset time window in real time, where the preset time window refers to the moment when the lithium-ion battery to be evaluated meets the trigger condition to the moment when the evolution of the abnormal events of the lithium-ion battery to be evaluated is completed. The evolution data of abnormal events includes image data and various numerical data during the evolution process of the abnormal events of the lithium-ion battery to be evaluated.

[0068] For the embodiments of the present invention, after receiving the real-time status data of the lithium-ion battery to be evaluated, it is first necessary to determine whether the lithium-ion battery to be evaluated meets the trigger condition for recording the evolution data of abnormal events based on the real-time status data. Based on this, step 202 specifically includes: constructing time series data from the real-time status data at the current moment and each historical moment before the current moment, and inputting the time series data into a preset battery status recognition model for operating status recognition to obtain the operating status of the lithium-ion battery to be evaluated at the current moment; if the operating status is a healthy operating status, it is determined that the lithium-ion battery to be evaluated does not meet the trigger condition for recording the evolution data of abnormal events, and if the operating status is an accident operating status, it is determined that the lithium-ion battery to be evaluated meets the trigger condition for recording the evolution data of abnormal events.

[0069] For the embodiments of the present invention, in order to improve the status recognition accuracy of the preset battery status recognition model, it is first necessary to train and construct the preset battery status recognition model. Based on this, the method includes: constructing a preset initial battery status recognition model and obtaining a sample data set, where the sample data set includes battery status data such as temperature, voltage, current, and internal resistance with battery status labels; dividing the sample data set into training data and test data, training the preset initial battery status recognition model with the training data, and testing the trained preset initial battery status recognition model with the test data, and finally determining the trained preset initial battery status recognition model that meets the test conditions as the preset battery status recognition model, where meeting the test conditions means that the number of training times meets the requirements or the test accuracy meets the requirements, etc. Specifically, collect a large number of multi-dimensional time series data of sample lithium-ion batteries in normal operation, abnormal operation but not damaged, and damaged states at different stages, including data such as voltage, current, temperature, and internal resistance, divide these data into training data and test data to train and test the preset initial battery status recognition model. During the training process of the preset initial battery status recognition model, set appropriate hyperparameters, such as the number of neurons in the hidden layer is 128, the learning rate is 0.001, the number of iterations is 1000 times, etc., and continuously adjust the parameters to optimize the model performance, so that the model can accurately learn the time series characteristics of battery performance changes. Finally, through continuous training and testing, a preset battery status recognition model with the required accuracy is constructed. For example, the preset battery status recognition model can be a long short-term memory network model, etc.

[0070] Furthermore, the trained preset battery state recognition model is deployed into an embedded system, which is connected to the safety evaluation device of the lithium-ion battery to be evaluated, and the real-time state data of the lithium-ion battery to be evaluated is obtained in real time. For example, data is obtained once per second, and the real-time state data is preprocessed, including operations such as data cleaning and normalization. Then, it is input into the preset battery state recognition model for operating state analysis. When the preset battery state recognition model predicts that the operating state of the lithium-ion battery to be evaluated conforms to the preset dangerous trend model (i.e., the accident operating state), the detection device is automatically triggered to start comprehensively recording the abnormal event evolution data of the lithium-ion battery to be evaluated in the accident operating state. The preset battery state recognition model can effectively process the long-term dependence relationship in time series data through the memory unit and the gating mechanism, and capture the trend of battery performance changes. Each sensor in the multi-functional detection array transmits the collected real-time state data, such as temperature, gas concentration, pressure, image and other information, to a computer system equipped with a high-performance processor and professional analysis software through a data transmission line with low latency and high bandwidth. At the same time, an emergency trigger button can be set in the entire detection device, and this button is connected to the control circuit of the detection device. When an abnormal situation is found (i.e., when the lithium-ion battery to be evaluated is in the accident operating state), pressing the button can immediately start the detection program for the accident operating state to ensure timely response in case of an emergency.

[0071] In another embodiment of the present invention, in order to avoid data loss, after obtaining the real-time state data of the lithium-ion battery to be evaluated after energy input and the abnormal event evolution data (image data and numerical data) of the lithium-ion battery to be evaluated within a preset time window, the real-time state data, abnormal event evolution data and other data can also be stored. Based on this, the method includes: performing at least one of data cleaning and normalization on the abnormal event evolution data to obtain the preprocessed abnormal event evolution data; classifying the preprocessed abnormal event evolution data to obtain the abnormal event evolution data under different classification categories, and encrypting the abnormal event evolution data under each classification category to obtain the encrypted abnormal event evolution data under each classification category; storing the encrypted abnormal event evolution data under each classification category according to the classification category in a distributed manner to obtain the abnormal event evolution data under different storage nodes.

[0072] Specifically, taking the abnormal event evolution data as an example, first, data cleaning processing such as removing outliers is performed on the abnormal event evolution data, and the normalized processing is performed on the cleaned abnormal event evolution data to obtain the processed abnormal event evolution data. Then, the classification processing is performed on the abnormal event evolution data. The specific classification process includes: determining the evolution characteristics corresponding to each abnormal event evolution data, and determining the centroid characteristics corresponding to the centroid data under the standard classification category; calculating the cosine similarity between each evolution characteristic and each centroid characteristic respectively, and classifying the abnormal event evolution data into the standard classification category based on the cosine similarity. Then, the encrypted processing and transmission are performed on the classified abnormal event evolution data to ensure data security. The abnormal event evolution data is transmitted to the remote server in real time through the encrypted wireless network for backup and further analysis, which facilitates the collaborative research and data sharing of expert teams in multiple locations, improves the detection efficiency and accuracy. Its working principle is to use network communication protocols and encryption technologies to ensure the security and integrity of the abnormal event evolution data during the transmission process. In the implementation manner, a wireless communication module and a data encryption chip are integrated in the security assessment device, and the abnormal event evolution data is sent to the remote server according to the specified format and protocol, and corresponding decryption and storage technologies are adopted on the server side for data management. Specifically, a wireless communication module, such as a 5G or Wi-Fi module, is integrated in the security assessment device, and the processed data is transmitted to the remote server through the encrypted wireless network; a secure network communication protocol, such as the TLS protocol, is adopted to ensure the integrity and security of the data during the transmission process. Cloud storage and management: The remote server adopts a distributed storage system to store the received data on multiple storage nodes to ensure that the data will not be lost due to a single point of failure; corresponding data management software is installed on the server side to classify, index, and back up the stored data, which facilitates the collaborative research of expert teams in multiple locations. Experts can log in to the server through remote access permissions, view and analyze the data, and perform secondary mining and analysis of the data to provide support for further research and decision-making. The functions of data remote transmission and cloud storage facilitate the collaboration and data sharing among expert teams, can make full use of resources from all parties, accelerate the research and solution process of battery safety issues, and improve the battery safety management level of the entire industry.

[0073] 203. Input the image data into a preset image analysis model for image feature extraction and classification prediction to obtain the physical structure change information and flame form change information of the lithium-ion battery to be evaluated during the abnormal event evolution process.

[0074] For the embodiments of the present invention, in order to improve the prediction accuracy of the preset image analysis model, it is first necessary to train and construct the preset image analysis model. Based on this, the method includes: constructing a preset initial image analysis model and obtaining a sample data set, where the sample data set includes image data of sample batteries with physical structure change labels and flame morphology change labels in abnormal states; dividing the sample data set into training data and test data, training the preset initial image analysis model with the training data, and testing the trained preset initial image analysis model with the test data. Finally, the trained preset initial image analysis model that meets the test conditions is determined as the preset image analysis model, where meeting the test conditions means that the number of training times meets the requirements or the test accuracy meets the requirements, etc. Further, the constructed preset image analysis model is used to analyze and process the image data. The specific processing process includes: the preset image analysis model includes a convolutional layer, a pooling layer, and a fully connected layer; the image data is input into the convolutional layer for image feature extraction to obtain image features, and the image features are input into the pooling layer for downsampling processing to obtain downsampled image features; the downsampled image features are input into the fully connected layer for image analysis to obtain the physical structure change information and flame morphology change information of the lithium-ion battery to be evaluated during the evolution of abnormal events. Specifically, for the image data, a preset image analysis model, such as a convolutional neural network, is used for analysis. The image data collected by the high-definition optical imaging sensor is input into the pre-trained preset image analysis model. This model has multiple convolutional layers, pooling layers, and fully connected layers. For example, the convolutional kernel size is 3x3, the stride is 1, and the pooling layer uses max pooling. The convolutional layer automatically extracts the image features of the lithium-ion battery to be evaluated during the evolution of abnormal events, such as the shape of the flame, the expansion and deformation of the battery, etc.; the pooling layer reduces the amount of data, and finally the fully connected layer is used for classification to identify the physical structure changes and flame morphology changes of the battery, etc.

[0075] 204. Input various numerical data into the preset numerical analysis model for regression analysis and classification prediction to obtain the energy release process information and chemical reaction path information of the lithium-ion battery to be evaluated during the evolution of abnormal events.

[0076] Among them, the numerical data includes gas concentration and type, temperature, battery temperature, current, voltage, etc.; the energy release process information includes the change trend of the gas concentration released by the lithium-ion battery to be evaluated during the evolution of the abnormal event, the rising trend of the temperature, etc.; the chemical reaction path information includes the decomposition process of the solid electrolyte interface film of the lithium-ion battery to be evaluated during the evolution of the abnormal event, releasing gas and heat; the decomposition information of the electrolyte, generating combustible gas; the decomposition process of the positive electrode material, releasing oxygen; the reaction process between the negative electrode and oxygen, further releasing heat; the external combustion process, the electrolyte reacts violently with the oxygen in the air, etc.

[0077] For the embodiments of the present invention, in order to improve the prediction accuracy of the preset numerical analysis model, it is first necessary to train and construct the preset numerical analysis model. Based on this, the method includes: constructing a preset initial numerical analysis model and obtaining a sample data set, where the sample data set includes sample data such as gas concentration and type, temperature, battery temperature, current, voltage, etc. of the sample battery with energy release process labels and chemical reaction path labels in an abnormal state; dividing the sample data set into training data and test data, using the training data to train the preset initial numerical analysis model, and using the test data to test the trained preset initial numerical analysis model, and finally determining the trained preset initial numerical analysis model that meets the test conditions as the preset numerical analysis model, where meeting the test conditions means that the number of training times meets the requirements or the test accuracy meets the requirements, etc. Further, the constructed preset numerical analysis model is used to analyze and process the numerical data. The specific processing process includes: the preset numerical analysis model includes multiple decision trees, and each decision tree includes a root node, child nodes, and leaf nodes; each numerical data is respectively input into the root node of each decision tree for feature extraction to obtain the features on the root node, and based on the splitting features and splitting rules of the root node, the features on the root node are divided into the corresponding child nodes of the root node for feature extraction to obtain the features on the child nodes; based on the splitting features and splitting rules of the child nodes, the features on the child nodes are divided into the corresponding leaf nodes of the child nodes for regression analysis and classification prediction to obtain the energy release process information and chemical reaction path information of the lithium-ion battery to be evaluated during the evolution of the abnormal event. Specifically, for numerical data such as temperature, gas concentration, and pressure data, a preset numerical analysis model such as a random forest algorithm is used for analysis. These data are input into the preset numerical analysis model. The preset numerical analysis model constructs multiple decision trees, and each tree is trained based on different features and randomly selected data subsets for regression analysis and classification prediction of the numerical data, such as predicting information such as the change trend of gas concentration, the rising rate of temperature, and the chemical reactions of various released gases.

[0078] 205. Based on the physical structure change information, flame morphology change information, energy release process information, and chemical reaction path information, construct an abnormal evolution mathematical model of the lithium-ion battery to be evaluated during the evolution process of abnormal events.

[0079] For the embodiments of the present invention, after obtaining the physical structure change information, flame morphology change information, energy release process information, and chemical reaction path information of the lithium-ion battery to be evaluated during the evolution of abnormal events, it is necessary to construct an abnormal evolution mathematical model based on the above information. Based on this, step 205 specifically includes: Based on elasticity or plasticity mechanics, combined with the stress changes caused by thermal expansion or phase change, the finite element method is used to simulate the physical structure deformation of the lithium-ion battery to be evaluated based on the physical structure change information, and a physical structure change model is obtained; Computational fluid dynamics is used based on the flame morphology change information to simulate flame propagation, temperature field, gas flow and other flame morphology deformations, and a flame morphology model is obtained; The energy release process information is substituted into the control equations such as heat conduction and thermal radiation, and an energy release process model is constructed based on the law of conservation of energy; Based on the chemical reaction path information, the reaction steps of the lithium-ion battery to be evaluated during the evolution data of abnormal events are determined, such as the process data of SEI decomposition, electrolyte decomposition, and cathode material decomposition, and substituted into the control equations such as reaction kinetics and chain reaction coupling, so as to obtain a chemical reaction path model; Finally, the physical structure change model, flame morphology model, energy release process model, and chemical reaction path model are coupled based on a physical field coupling software to obtain an abnormal evolution mathematical model of the lithium-ion battery to be evaluated during the evolution of abnormal events. Specifically, the physical structure change may involve mechanical deformation, rupture, etc. of the internal materials of the battery (such as electrodes, separators). A solid mechanics model, such as elasticity or plasticity mechanics, needs to be combined with the stress changes caused by thermal expansion or phase change. This part may require the finite element method to simulate the structural deformation. The flame morphology change belongs to the combustion process, involving fluid dynamics and combustion models. Computational fluid dynamics is needed to simulate flame propagation, temperature field, gas flow, etc. The combustion model may include premixed combustion, diffusion flame, or more complex turbulent combustion models. The energy release process may involve heat conduction, thermal radiation, convection, etc., and an energy conservation equation is needed. At the same time, the chemical reactions inside the battery (such as thermal runaway reactions) will release a large amount of heat, and this part needs to be combined with the chemical reaction kinetics model, considering the relationship between the heat generation rate and temperature, such as the Arrhenius equation (the equation for the effect of temperature on the reaction rate). The chemical reaction path information needs to clarify the reaction steps during battery thermal runaway, such as the decomposition of SEI (Solid Electrolyte Interphase), electrolyte decomposition, cathode material decomposition, etc. The kinetic parameters (activation energy, pre-exponential factor) of each reaction require experimental data support, and the mutual influence between reactions, such as chain reactions or competitive reactions, also needs to be considered. Next, these models are coupled using multi-physics simulation software such as COMSOL and ANSYS. For example, structural deformation will affect heat conduction and fluid flow, and the change in temperature will in turn affect material properties and reaction rates. This two-way coupling may require iterative solution until the fields reach equilibrium.Furthermore, considering the initial conditions and boundary conditions, for example, the initial temperature distribution of the battery, material properties, surrounding environmental conditions (such as ventilation), initial defects (such as internal short - circuit points), etc., the boundary conditions may include the convective heat transfer coefficient, radiation boundary conditions, etc., to obtain the abnormal evolution mathematical model. Through in - depth data analysis and processing using advanced artificial intelligence algorithms, the embodiments of the present invention can deeply explore the hidden information behind the data and construct a more detailed and accurate abnormal evolution mathematical model, which helps to comprehensively understand the process and mechanism of battery abnormal events evolution and provides strong support for formulating effective safety protection measures.

[0080] 206. Evaluate the safety of the lithium - ion battery to be evaluated based on the abnormal evolution mathematical model.

[0081] For the embodiments of the present invention, after constructing the abnormal evolution mathematical model of the lithium - ion battery to be evaluated during the evolution process of abnormal events, it is necessary to evaluate the safety of the lithium - ion battery to be evaluated based on the abnormal evolution mathematical model. Based on this, step 206 specifically includes: extracting at least one safety evaluation parameter of the lithium - ion battery to be evaluated from the abnormal evolution mathematical model, and evaluating the safety of the lithium - ion battery to be evaluated based on each safety evaluation parameter and its corresponding preset parameter threshold.

[0082] Among them, the safety evaluation parameters include: the appearance form evaluation parameter of the lithium - ion battery to be evaluated, gas production rate evaluation parameter, gas concentration evaluation parameter, temperature evaluation parameter, pressure evaluation parameter, etc. Each flatness parameter is set with a corresponding preset parameter threshold according to actual needs. Specifically, when conducting safety evaluation, it is respectively determined whether the appearance form evaluation parameter, gas production rate evaluation parameter, gas concentration evaluation parameter, temperature evaluation parameter, and pressure evaluation parameter are greater than the corresponding preset parameter thresholds, and the safety level of the lithium - ion battery to be evaluated is determined according to the types and quantities of the evaluation parameters greater than the preset parameter thresholds.

[0083] According to another method for evaluating the safety of a lithium-ion battery provided by the present invention, compared with the current method of evaluating the safety of a lithium-ion battery only by measuring the battery capacity, the present invention inputs energy to the lithium-ion battery to be evaluated through an energy input device in the safety evaluation device. During the energy input process, when it is detected that the lithium-ion battery to be evaluated meets the trigger condition for recording the evolution data of abnormal events, the evolution data of abnormal events of the lithium-ion battery to be evaluated is collected, and an abnormal evolution mathematical model is constructed based on the evolution data of abnormal events. Finally, the safety of the lithium-ion battery to be evaluated is evaluated according to the abnormal evolution mathematical model. Therefore, by starting to collect the evolution data of abnormal events only when the lithium-ion battery to be evaluated meets the trigger condition for recording the evolution data of abnormal events, the time and computing resources wasted in processing data when no abnormality occurs can be avoided, thereby improving the efficiency of evaluating the safety of lithium-ion batteries and reducing the cost of safety evaluation. At the same time, since the abnormal evolution mathematical model contains all the evolution process information of the lithium-ion battery in abnormal events, evaluating the safety of the lithium-ion battery through the abnormal evolution mathematical model can improve the accuracy of evaluating the safety of lithium-ion batteries.

[0084] Further, as Figure 1 a specific implementation of Figure 3 this, an embodiment of the present invention provides an evaluation device for the safety of a lithium-ion battery, which is applied to a safety evaluation device. The safety evaluation device includes a detection device and an energy input device. As

[0085] shown, the device includes: a control unit 31, a collection unit 32, and an evaluation unit 33.

[0086] The control unit 31 can be used to control the energy input device to perform step-by-step enhanced energy input on the lithium-ion battery to be evaluated in the safety evaluation device, and use the detection device to detect the real-time state data of the lithium-ion battery to be evaluated after the energy input.

[0087] The collection unit 32 can be used to determine whether the lithium-ion battery to be evaluated meets the trigger condition for recording the evolution data of abnormal events based on the real-time state data. If it meets the condition, the detection device is triggered to collect the evolution data of abnormal events of the lithium-ion battery to be evaluated within a preset time window in real time, where the preset time window refers to the time from the moment when the lithium-ion battery to be evaluated meets the trigger condition to the moment when the evolution of abnormal events of the lithium-ion battery to be evaluated is completed.

[0088] In a specific application scenario, in order to determine whether the lithium-ion battery to be evaluated meets the trigger condition for recording the evolution data of abnormal events, such as Figure 4 shown, the acquisition unit 32 includes a state recognition module 321 and a determination module 322.

[0089] The state recognition module 321 can be used to form time series data from the real-time state data at the current moment and each historical moment before the current moment, and input the time series data into a preset battery state recognition model for running state recognition, so as to obtain the running state of the lithium-ion battery to be evaluated at the current moment.

[0090] The determination module 322 can be used to determine that the lithium-ion battery to be evaluated does not meet the trigger condition for recording the evolution data of abnormal events if the running state is a healthy running state, and determine that the lithium-ion battery to be evaluated meets the trigger condition for recording the evolution data of abnormal events if the running state is an accident running state.

[0091] In a specific application scenario, the evolution data of abnormal events includes image data and various numerical data of the lithium-ion battery to be evaluated during the evolution of abnormal events; in order to construct an abnormal evolution mathematical model, the evaluation unit 33 includes an image processing module 331, a numerical processing module 332, and a construction module 333.

[0092] The image processing module 331 can be used to input the image data into a preset image analysis model for image feature extraction and classification prediction, so as to obtain the physical structure change information and flame form change information of the lithium-ion battery to be evaluated during the evolution of abnormal events.

[0093] The numerical processing module 332 can be used to input various numerical data into a preset numerical analysis model for regression analysis and classification prediction, so as to obtain the energy release process information and chemical reaction path information of the lithium-ion battery to be evaluated during the evolution of abnormal events.

[0094] The construction module 333 can be used to construct an abnormal evolution mathematical model of the lithium-ion battery to be evaluated during the evolution of abnormal events based on the physical structure change information, the flame form change information, the energy release process information, and the chemical reaction path information.

[0095] In a specific application scenario, the energy input device adopts a combination of a high-frequency pulse power supply and a microwave generator; the detection device includes a plurality of detection units, and each detection unit is composed of a variety of sensors; in order to control the energy input device to perform step-by-step enhanced energy input on the lithium-ion battery to be evaluated in the safety evaluation device, the control unit 31 includes a control module 311 and an arrangement module 312.

[0096] The control module 311 can be used to control the high-frequency pulse power supply to deliver a gradually increasing pulsed current with an adjustable amplitude and a frequency within a first preset frequency range to the lithium-ion battery to be evaluated in the safety evaluation device, and control the microwave generator to emit a gradually increasing microwave signal with a controllable power and a frequency within a second preset frequency range to the lithium-ion battery to be evaluated in the safety evaluation device.

[0097] The arrangement module 312 can be used to determine the position information of the lithium-ion battery to be evaluated in the safety evaluation device, and arrange each detection unit in a preset arrangement form based on the position information.

[0098] In a specific application scenario, multi-layer heat insulation and buffer materials are provided on the inner wall of the cabin of the safety evaluation device, a plurality of ventilation openings with different apertures and positions are provided on the cabin, and intelligent regulating valves are equipped at preset positions of the cabin.

[0099] In a specific application scenario, in order to store the abnormal event evolution data, the device further includes a storage unit 34.

[0100] The storage unit 34 can be used to perform at least one operation of data cleaning and normalization on the abnormal event evolution data to obtain the preprocessed abnormal event evolution data; classify the preprocessed abnormal event evolution data to obtain the abnormal event evolution data under different classification categories, and encrypt the abnormal event evolution data under each classification category to obtain the encrypted abnormal event evolution data under each classification category; store the encrypted abnormal event evolution data under each classification category distributively according to the classification categories to obtain the abnormal event evolution data under different storage nodes.

[0101] In a specific application scenario, in order to evaluate the safety of the lithium-ion battery to be evaluated, the evaluation unit 33 can specifically be used to extract at least one safety evaluation parameter of the lithium-ion battery to be evaluated from the abnormal evolution mathematical model, and evaluate the safety of the lithium-ion battery to be evaluated based on each safety evaluation parameter and its corresponding preset parameter threshold.

[0102] It should be noted that for other corresponding descriptions of each functional module involved in the lithium-ion battery safety evaluation device provided in the embodiments of the present invention, reference can be made to Figure 1 the corresponding description of the method shown, which will not be elaborated here.

[0103] Based on the above as Figure 1The method described above. Correspondingly, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the following steps are implemented: Applied to a security evaluation device, the security evaluation device includes a detection device and an energy input device, and includes: controlling the energy input device to perform step-by-step enhanced energy input on the lithium-ion battery to be evaluated in the security evaluation device, and using the detection device to detect the real-time state data of the lithium-ion battery to be evaluated after the energy input; based on the real-time state data, determining whether the lithium-ion battery to be evaluated meets the trigger condition for recording the evolution data of abnormal events. If it meets, triggering the detection device to collect the evolution data of abnormal events of the lithium-ion battery to be evaluated in a preset time window in real time, where the preset time window refers to the moment when the lithium-ion battery to be evaluated meets the trigger condition to the moment when the evolution of abnormal events of the lithium-ion battery to be evaluated is completed; based on the evolution data of abnormal events, constructing an abnormal evolution mathematical model of the lithium-ion battery to be evaluated during the evolution of abnormal events, and based on the abnormal evolution mathematical model, evaluating the security of the lithium-ion battery to be evaluated.

[0104] Based on the above as Figure 1 shown in the method and as Figure 3 shown in the embodiment of the device, an embodiment of the present invention further provides an entity structure diagram of a computer device, as Figure 5 shown. The computer device includes: a processor 41, a memory 42, and a computer program stored on the memory 42 and executable on the processor. The memory 42 and the processor 41 are both arranged on a bus 43. When the processor 41 executes the program, the following steps are implemented: Applied to a security evaluation device, the security evaluation device includes a detection device and an energy input device, and includes: controlling the energy input device to perform step-by-step enhanced energy input on the lithium-ion battery to be evaluated in the security evaluation device, and using the detection device to detect the real-time state data of the lithium-ion battery to be evaluated after the energy input; based on the real-time state data, determining whether the lithium-ion battery to be evaluated meets the trigger condition for recording the evolution data of abnormal events. If it meets, triggering the detection device to collect the evolution data of abnormal events of the lithium-ion battery to be evaluated in a preset time window in real time, where the preset time window refers to the moment when the lithium-ion battery to be evaluated meets the trigger condition to the moment when the evolution of abnormal events of the lithium-ion battery to be evaluated is completed; based on the evolution data of abnormal events, constructing an abnormal evolution mathematical model of the lithium-ion battery to be evaluated during the evolution of abnormal events, and based on the abnormal evolution mathematical model, evaluating the security of the lithium-ion battery to be evaluated.

[0105] Through the technical solution of the present invention, the energy input device in the safety evaluation device inputs energy to the lithium-ion battery to be evaluated. During the energy input process, when it is detected that the lithium-ion battery to be evaluated is in the trigger condition for recording the evolution data of abnormal events, the evolution data of abnormal events of the lithium-ion battery to be evaluated is collected, and an abnormal evolution mathematical model is constructed based on the evolution data of abnormal events. Finally, the safety of the lithium-ion battery to be evaluated is evaluated according to the abnormal evolution mathematical model. Thus, by starting to collect the evolution data of abnormal events only when the lithium-ion battery to be evaluated is in the trigger condition for recording the evolution data of abnormal events, the time and computing resources wasted in processing the data when no abnormality occurs can be avoided, thereby improving the safety evaluation efficiency of the lithium-ion battery and reducing the safety evaluation cost. At the same time, since the abnormal evolution mathematical model contains all the evolution process information of the lithium-ion battery in abnormal events, evaluating the safety of the lithium-ion battery through the abnormal evolution mathematical model can improve the accuracy of the safety evaluation of the lithium-ion battery.

[0106] Obviously, those skilled in the art should understand that the above-mentioned modules or steps of the present invention can be implemented by a general-purpose computing device. They can be concentrated on a single computing device or distributed on a network composed of multiple computing devices. Optionally, they can be implemented by program codes executable by the computing device. Thus, they can be stored in a storage device and executed by the computing device. And in some cases, the steps shown or described can be executed in a different order from here, or they can be separately made into individual integrated circuit modules, or multiple modules or steps among them can be made into a single integrated circuit module to implement. In this way, the present invention is not limited to any specific combination of hardware and software.

[0107] The above are only the preferred embodiments of the present invention and are not used to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for evaluating the safety of a lithium-ion battery, characterized in that: Applied to a safety assessment device, the safety assessment device includes a detection device, an energy input device, including: Controlling the energy input device to perform step-by-step energy input on the lithium-ion battery to be evaluated in the safety evaluation device, and using the detection device to detect real-time status data of the lithium-ion battery to be evaluated after the energy input; Based on the real-time status data, determine whether the lithium-ion battery to be evaluated meets the triggering condition for recording abnormal event evolution data, and if so, trigger the detection device to collect the abnormal event evolution data of the lithium-ion battery to be evaluated within a preset time window in real time, wherein the preset time window refers to the time when the lithium-ion battery to be evaluated meets the triggering condition to the time when the abnormal event evolution of the lithium-ion battery to be evaluated is completed; Based on the abnormal event evolution data, an abnormal evolution mathematical model of the lithium ion battery to be evaluated during the abnormal event evolution process is constructed, and based on the abnormal evolution mathematical model, the safety of the lithium ion battery to be evaluated is evaluated.

2. The method according to claim 1, characterized in that: The determining, based on the real-time status data, whether the lithium-ion battery to be evaluated meets the triggering condition for recording abnormal event evolution data includes: The real-time status data at the current moment and each historical moment before the current moment constitute time series data, and the time series data is input into a preset battery status recognition model to perform operation status recognition, so as to obtain the operation status of the lithium-ion battery to be evaluated at the current moment; If the operating state is a healthy operating state, it is determined that the lithium-ion battery to be evaluated does not meet the triggering condition for recording abnormal event evolution data; if the operating state is an accident operating state, it is determined that the lithium-ion battery to be evaluated meets the triggering condition for recording abnormal event evolution data.

3. The method according to claim 1, characterized in that: The abnormal event evolution data includes image data and multiple numerical data of the lithium-ion battery to be evaluated during the abnormal event evolution process; The abnormal event evolution mathematical model of the to-be-evaluated lithium-ion battery during the abnormal event evolution process is constructed based on the abnormal event evolution data, including: Inputting the image data into a preset image analysis model to extract image features and perform classification prediction to obtain physical structure change information and flame morphology change information of the lithium-ion battery to be evaluated during the evolution of the abnormal event; Input various numerical data into a preset numerical analysis model for regression analysis and classification prediction to obtain energy release process information and chemical reaction path information of the lithium-ion battery to be evaluated during the evolution of abnormal events; Based on the physical structure change information, the flame morphology change information, the energy release process information, and the chemical reaction path information, a mathematical model of abnormal evolution of the lithium-ion battery to be evaluated during the evolution of abnormal events is constructed.

4. The method according to claim 1, characterized in that The energy input device adopts a combination of a high-frequency pulse power supply and a microwave generator; The step of controlling the energy input device to perform a step-by-step energy input on the lithium-ion battery to be evaluated in the safety evaluation device comprises: Control the high-frequency pulse power supply to deliver a pulse current with a frequency in a first preset frequency range and an adjustable amplitude to the lithium-ion battery to be evaluated in the safety assessment device, and control the microwave generator to transmit a microwave signal with a frequency in a second preset frequency range and a controllable power to the lithium-ion battery to be evaluated in the safety assessment device; The detection device comprises a plurality of detection units, each of which is composed of a plurality of sensors; Before using the detection device to detect the real-time status data of the lithium-ion battery to be evaluated after energy input, the method further includes: The position information of the lithium-ion battery to be evaluated in the safety evaluation device is determined, and based on the position information, each of the detection units is arranged in a preset arrangement form.

5. The method according to claim 1, characterized in that The inner wall of the cabin of the safety assessment equipment is provided with multiple layers of heat insulation and buffer materials, a plurality of ventilation holes with different apertures and positions are provided on the cabin, and an intelligent regulating valve is provided at a preset position of the cabin.

6. The method according to claim 1, characterized in that After triggering the detection device to collect the abnormal event evolution data of the lithium-ion battery to be evaluated within a preset time window in real time, the method further includes: Performing at least one of data cleaning and normalization on the abnormal event evolution data to obtain the pre-processed abnormal event evolution data; Classifying the preprocessed abnormal event evolution data to obtain abnormal event evolution data under different classification categories, and encrypting the abnormal event evolution data under each classification category to obtain encrypted abnormal event evolution data under each classification category; The encrypted abnormal event evolution data under each classification category is distributedly stored according to the classification category to obtain the abnormal event evolution data under different storage nodes.

7. The method according to claim 1, characterized in that The step of evaluating the safety of the lithium-ion battery to be evaluated based on the abnormal evolution mathematical model includes: At least one safety assessment parameter of the lithium-ion battery to be assessed is extracted from the abnormal evolution mathematical model, and the safety of the lithium-ion battery to be assessed is assessed based on each of the safety assessment parameters and its corresponding preset parameter threshold.

8. A lithium-ion battery safety assessment device, characterized in that: Applied to a safety assessment device, the safety assessment device includes a detection device, an energy input device, including: A control unit, used to control the energy input device to perform a step-by-step energy input to the lithium-ion battery to be evaluated in the safety evaluation device, and use the detection device to detect real-time status data of the lithium-ion battery to be evaluated after the energy input; A collection unit, used to determine whether the lithium-ion battery to be evaluated meets the triggering condition for recording abnormal event evolution data based on the real-time status data, and if so, trigger the detection device to collect the abnormal event evolution data of the lithium-ion battery to be evaluated within a preset time window in real time, wherein the preset time window refers to the time when the lithium-ion battery to be evaluated meets the triggering condition to the time when the abnormal event evolution of the lithium-ion battery to be evaluated is completed; An evaluation unit is used to construct an abnormal evolution mathematical model of the lithium ion battery to be evaluated during the abnormal event evolution process based on the abnormal event evolution data, and to evaluate the safety of the lithium ion battery to be evaluated based on the abnormal evolution mathematical model.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images