AI-based early warning method for thermal runaway of electric scooter battery

Abstract

The application discloses an AI-based early warning method for battery thermal runaway of an electric scooter, and relates to the technical field of battery warning, which comprises the following steps: obtaining a first historical function according to a historical temperature difference and a historical radiation average value; obtaining a second historical function according to a historical current value and a historical working temperature change value; obtaining a real-time battery temperature, a real-time temperature difference and a real-time current value according to real-time operation data of the electric scooter battery; constructing a real-time target line segment according to the real-time battery temperature; obtaining a real-time prediction difference according to the real-time temperature difference, the real-time current value, the real-time target line segment, the first historical function and the second historical function; obtaining a real-time prediction final temperature value according to the real-time target line segment and the real-time prediction difference, and performing early warning for battery thermal runaway according to the real-time prediction final temperature value; and the application is used to solve the problem that there is no accurate prediction method for thermal runaway in the prior art, resulting in untimely and inaccurate prediction of battery thermal runaway.

Description

Technical Field

[0001] This invention relates to the field of battery early warning technology, specifically to an AI-based early warning method for thermal runaway of electric scooter batteries. Background Technology

[0002] With the rapid popularization of the short-distance micro-mobility industry, electric scooters are widely used in urban commuting and shared mobility due to their portability and low cost. However, due to the abuse caused by various scenarios such as vehicle bumps and vibrations, outdoor high and low temperature environments, frequent fast charging, long-term heavy-load riding, and battery cell aging, electric scooter batteries are prone to thermal runaway. Thermal runaway of batteries can lead to fire and explosion safety accidents, so it is necessary to detect battery thermal runaway in a timely manner. Existing battery thermal runaway detection methods rely on battery operating data or battery temperature gauges to determine if an abnormality has occurred. These methods often use fixed thresholds to predict thermal runaway, making them unpredictable. Furthermore, data affecting thermal runaway detection can lead to inaccurate assessments, such as ambient temperature. For example, patent application CN111370784A discloses a battery thermal runaway early warning method, but this method cannot accurately predict thermal runaway. The lack of an accurate method for predicting thermal runaway in existing technologies results in untimely and inaccurate predictions. Summary of the Invention

[0003] This invention aims to at least partially solve one of the technical problems in the prior art. It obtains historical temperature differences and corresponding historical average radiation values ​​based on normal operating data of an electric scooter battery; it then obtains a first historical function based on the historical temperature differences and historical average radiation values; it further obtains historical current values ​​and historical operating temperature changes based on the normal operating data of the electric scooter battery, and a second historical function based on these values; it acquires real-time battery temperature, real-time temperature difference, and real-time current values ​​based on real-time operating data of the electric scooter battery; it constructs a real-time target line segment based on the real-time battery temperature; it obtains a real-time predicted difference based on the real-time temperature difference, real-time current value, real-time target line segment, first historical function, and second historical function; it obtains a real-time predicted final temperature value based on the real-time target line segment and real-time predicted difference; and it provides early warning of battery thermal runaway based on the real-time predicted final temperature value. This addresses the problem in the prior art where the lack of an accurate method for predicting thermal runaway leads to untimely and inaccurate predictions of battery thermal runaway.

[0004] To achieve the above objectives, this application provides an AI-based early warning method for thermal runaway of electric scooter batteries, comprising the following steps: S100: Obtain historical temperature difference values ​​based on normal operating data of electric scooter batteries; obtain historical average radiation values ​​corresponding to historical temperature difference values; obtain the first historical function based on historical temperature difference values ​​and historical average radiation values. S200: Based on the normal operation data of the electric scooter battery, obtain historical current value and historical operating temperature change value, and obtain a second historical function based on the historical current value and historical operating temperature change value. S300 obtains real-time battery temperature, real-time temperature difference, and real-time current value based on real-time electric scooter battery operation data; constructs a real-time target line segment based on the real-time battery temperature; and obtains the real-time predicted difference based on the real-time temperature difference, real-time current value, real-time target line segment, first historical function, and second historical function. The S400 obtains the real-time predicted final temperature value based on the difference between the real-time target line segment and the real-time prediction, and provides early warning of battery thermal runaway based on the real-time predicted final temperature value.

[0005] Furthermore, S100 includes the following sub-steps: Obtain the temperature range of the electric scooter battery during normal operation and mark it as the historical battery temperature range; Obtain the range of ambient temperature where the electric scooter is located, and mark it as the historical ambient temperature range; Obtain any temperature within the historical battery temperature range and label it as the historical battery temperature; Obtain any temperature within the historical ambient temperature range and label it as the historical ambient temperature; Obtain the difference between historical ambient temperature and historical battery temperature, and label it as historical temperature difference.

[0006] Furthermore, S100 also includes the following sub-steps: Set a duration and label it as a unit duration; When the electric scooter battery is not running, the change in battery temperature after acquiring different historical temperature difference values ​​per unit time is identified as historical temperature radiation value. Obtain the first number of historical temperature radiation values ​​under different historical temperature differences; Obtain the average of historical temperature radiation values ​​under the same historical temperature difference, and label it as the historical radiation mean.

[0007] Furthermore, S100 also includes the following sub-steps: A Cartesian coordinate system is established with historical temperature difference as the horizontal axis and historical average radiation as the vertical axis, and this system is designated as the first historical coordinate system. The historical temperature difference and the corresponding historical mean radiation are used as coordinate points on the x-axis and y-axis, respectively, and are marked as the first historical coordinate point. Obtain all first historical coordinate points, plot all first historical coordinate points in the first historical coordinate system to obtain a scatter plot, and label it as the first historical scatter plot; The function obtained by fitting the first historical coordinate points in the first historical scatter plot is identified as the first historical function.

[0008] Furthermore, S200 includes the following sub-steps: Obtain the operating current range of the electric scooter battery and label it as the historical current range; Obtain any current within the historical current range and label it as a historical current value; Set a specific temperature range to keep the temperature of the electric scooter battery and the ambient temperature of the electric scooter within that specific temperature range; Within a specific temperature range and under the same historical current value, the temperature change value of the electric scooter battery after a unit of time is obtained and identified as the historical operating temperature change value. Obtain the historical operating temperature variation values ​​corresponding to different historical current values.

[0009] Furthermore, S200 also includes the following sub-steps: A Cartesian coordinate system is established with historical current values ​​on the horizontal axis and historical operating temperature changes on the vertical axis, and this system is designated as the first historical coordinate system. The historical current value and the corresponding historical operating temperature change value are obtained and used as coordinate points on the horizontal and vertical axes, respectively, and are marked as the second historical coordinate points; Obtain all the second historical coordinate points, plot all the second historical coordinate points in the first historical coordinate system to obtain a scatter plot, and label it as the second historical scatter plot; The function obtained by fitting the second historical coordinate points in the second historical scatter plot is labeled as the second historical function.

[0010] Furthermore, S300 includes the following sub-steps: Obtain the temperature of the electric scooter battery and label it as the real-time battery temperature; Get the ambient temperature where the electric scooter is located and label it as the real-time ambient temperature; Obtain the difference between the real-time ambient temperature and the real-time battery temperature, and label it as the real-time temperature difference; Obtain the current flow of the electric scooter battery and label it as a real-time current value.

[0011] Furthermore, S300 also includes the following sub-steps: The real-time battery temperature is acquired once every first interval and marked as the real-time interval temperature. A Cartesian coordinate system is established with time as the horizontal axis and real-time interval temperature as the vertical axis, and it is labeled as the third real-time coordinate system. The acquisition time and the corresponding real-time interval temperature are used as coordinate points on the horizontal and vertical axes, respectively, and are marked as the third real-time coordinate point; Obtain all the third real-time coordinate points, plot all the third real-time coordinate points in the third real-time coordinate system to obtain a scatter plot, and label it as the third real-time scatter plot; Only the latest third real-time coordinate points within the second time period are retained in the third real-time coordinate system; In the third real-time coordinate system, adjacent third real-time coordinate points are connected to obtain line segments, which are then identified as real-time connection line segments. Retrieve the latest real-time connection segments and mark them as real-time target segments.

[0012] Furthermore, S300 also includes the following sub-steps: Obtain the left and right endpoints of the real-time connecting line segments and label them as the first line segment endpoint and the second line segment endpoint; Obtain the real-time battery temperature, real-time temperature difference, and real-time current value corresponding to the endpoint of the first line segment, and identify them as the first battery temperature, the first temperature difference, and the first current value. Substitute the first temperature difference value as the horizontal axis value into the first historical function to obtain the vertical axis value, which is then identified as the first real-time predicted value. Substitute the first current value as the horizontal axis value into the second historical function to obtain the vertical axis value, which is then identified as the second real-time predicted value. The sum of the first battery temperature, the first real-time predicted value, and the second real-time predicted value is obtained and identified as the first predicted temperature value. Get the real-time battery temperature corresponding to the endpoint of the second line segment and mark it as the second battery temperature; The difference between the first predicted temperature value and the second battery temperature is obtained and identified as the real-time predicted difference.

[0013] Furthermore, S400 includes the following sub-steps: Obtain the slope of the real-time target line segment and label it as the real-time target slope; The segment whose endpoints are extended by a unit time length with the real-time target slope is identified as the real-time predicted segment. Obtain the temperature at the right end of the real-time predicted line segment and label it as the real-time temperature prediction value; The sum of the real-time prediction difference and the real-time temperature prediction value is obtained and identified as the final real-time predicted temperature value. Determine whether the real-time predicted final temperature value is within the normal operating temperature range of the battery. If not, immediately issue a battery thermal runaway warning.

[0014] The beneficial effects of this invention are as follows: This invention obtains historical temperature differences and corresponding historical average radiation values ​​based on normal operating data of electric scooter batteries; it obtains a first historical function based on the historical temperature differences and historical average radiation values; it obtains historical current values ​​and historical operating temperature change values ​​based on normal operating data of electric scooter batteries, and obtains a second historical function based on the historical current values ​​and historical operating temperature change values; it obtains real-time battery temperature, real-time temperature difference, and real-time current values ​​based on real-time electric scooter battery operating data; it constructs a real-time target line segment based on the real-time battery temperature; it obtains a real-time predicted difference based on the real-time temperature difference, real-time current value, real-time target line segment, first historical function, and second historical function; it obtains a real-time predicted final temperature value based on the real-time target line segment and real-time predicted difference; and it provides early warning of battery thermal runaway based on the real-time predicted final temperature value. The advantages are that it adapts to different battery and ambient temperatures, while improving the timeliness and accuracy of battery thermal runaway prediction. This invention obtains the final predicted temperature value in real time based on the difference between the real-time target line segment and the real-time prediction. Its advantage lies in obtaining the adjustment value for temperature changes under different conditions, thereby improving the accuracy of battery thermal runaway prediction. Attached Figure Description

[0015] Figure 1 This is a flowchart of the steps of the method of the present invention; Figure 2 This is a schematic diagram of the first history function of the present invention; Figure 3 This is a schematic diagram of the second history function of the present invention; Figure 4 This is a schematic diagram of the real-time predicted line segment of the present invention. Detailed Implementation

[0016] 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 some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] Example 1, please refer to Figure 1 As shown, this application provides an AI-based early warning method for thermal runaway of electric scooter batteries, including the following steps: Step S100: Obtain historical temperature difference values ​​based on normal operating data of the electric scooter battery; obtain the historical average radiation value corresponding to the historical temperature difference values; obtain the first historical function based on the historical temperature difference values ​​and the historical average radiation value; Step S100 includes the following sub-steps: Step S101: Obtain the temperature range of the electric scooter battery during normal operation and mark it as the historical battery temperature range; obtain the temperature range of the ambient temperature where the electric scooter is located and mark it as the historical ambient temperature range. Step S102: Obtain any temperature within the historical battery temperature range and mark it as the historical battery temperature; obtain any temperature within the historical ambient temperature range and mark it as the historical ambient temperature; obtain the difference between the historical ambient temperature and the historical battery temperature and mark it as the historical temperature difference; this facilitates obtaining the difference between different battery temperatures and ambient temperatures. When the battery and ambient temperatures are different, temperature will be transferred, so the historical temperature difference is obtained. Step S103: Set a duration, labeled as unit duration; when the electric scooter battery stops running, the change in battery temperature after acquiring different historical temperature differences within a unit duration is labeled as historical temperature radiation value; for example, the unit duration is 10s, the unit duration is the data acquisition time interval, and the same unit duration is the predicted battery temperature change after that time. Step S104: Obtain a first number of historical temperature radiation values ​​under different historical temperature differences; obtain the average of historical temperature radiation values ​​under the same historical temperature difference, and mark it as the historical radiation average; the historical radiation average is the accurate historical temperature radiation value corresponding to the historical temperature difference, so the first number should not be set too small, for example, the first number is 100.

[0018] Step S105: Establish a Cartesian coordinate system with historical temperature difference as the horizontal axis and historical average radiation as the vertical axis, and label it as the first historical coordinate system; Step S106: Obtain the historical temperature difference and the corresponding historical average radiation value as the coordinate points of the horizontal and vertical axes, respectively, and mark them as the first historical coordinate points. Step S107: Obtain all the first historical coordinate points, plot all the first historical coordinate points in the first historical coordinate system to obtain a scatter plot, and label it as the first historical scatter plot. Step S108: Fit the first historical coordinate points in the first historical scatter plot to obtain a function, and mark it as the first historical function; For practical applications, please refer to Figure 2 As shown, the first historical function is obtained, which is the historical radiation mean value corresponding to different historical temperature differences, that is, the heat transfer relationship between the battery and the environment within 10 seconds.

[0019] Step S200 involves obtaining historical current values ​​and historical operating temperature change values ​​based on the normal operation data of the electric scooter battery, and obtaining a second historical function based on the historical current values ​​and historical operating temperature change values. Step S200 includes the following sub-steps: Step S201: Obtain the operating current range of the electric scooter battery and mark it as the historical current range; the historical current value will be different when the electric scooter is running on different slopes, which will result in different amounts of heat generated by the battery. Step S202: Obtain any current within the historical current range and mark it as a historical current value; Step S203: Set a specific temperature space to keep the temperature of the electric scooter battery and the ambient temperature of the electric scooter consistent within the specific temperature space; setting a specific temperature space makes it easier to obtain the heat generated by the battery under different currents. Step S204: Within a specific temperature space and under the same historical current value, obtain the temperature change value of the electric scooter battery after a unit time, and mark it as the historical operating temperature change value; obtain the historical operating temperature change value corresponding to different historical current values.

[0020] Step S205: Establish a Cartesian coordinate system with historical current values ​​on the horizontal axis and historical operating temperature changes on the vertical axis, and label it as the first historical coordinate system; Step S206: Obtain the historical current value and the corresponding historical operating temperature change value as the coordinate points of the horizontal and vertical axes, respectively, and mark them as the second historical coordinate points; Step S207: Obtain all the second historical coordinate points, plot all the second historical coordinate points in the first historical coordinate system to obtain a scatter plot, and label it as the second historical scatter plot; Step S208: Fit the second historical coordinate points in the second historical scatter plot to obtain a function, and label it as the second historical function; For practical applications, please refer to Figure 3 As shown, the second historical function is obtained. The first historical function is the historical operating temperature change value corresponding to different historical current values, that is, the change value of battery heat under different current values ​​within 10 seconds.

[0021] Step S300 involves obtaining the real-time battery temperature, real-time temperature difference, and real-time current value based on the real-time operating data of the electric scooter battery; constructing a real-time target line segment based on the real-time battery temperature; and obtaining the real-time prediction difference based on the real-time temperature difference, real-time current value, real-time target line segment, first historical function, and second historical function. Step S300 includes the following sub-steps: Step S301: Obtain the temperature of the electric scooter battery and label it as the real-time battery temperature; obtain the ambient temperature of the electric scooter and label it as the real-time ambient temperature; obtain the difference between the real-time ambient temperature and the real-time battery temperature and label it as the real-time temperature difference. Step S302: Obtain the current flow of the electric scooter battery and label it as a real-time current value.

[0022] Step S303: Obtain the real-time battery temperature once every first time interval and mark it as the real-time interval temperature; Step S304: Establish a Cartesian coordinate system with time as the horizontal axis and real-time interval temperature as the vertical axis, and label it as the third real-time coordinate system; Step S305: Obtain the time and the corresponding real-time interval temperature as the coordinate points of the horizontal and vertical axes, respectively, and mark them as the third real-time coordinate points; Step S306: Obtain all the third real-time coordinate points, plot all the third real-time coordinate points in the third real-time coordinate system to obtain a scatter plot, and label it as the third real-time scatter plot. Step S307: In the third real-time coordinate system, only the third real-time coordinate point within the latest unit time period is retained; Step S308: In the third real-time coordinate system, connect adjacent third real-time coordinate points to obtain line segments, which are marked as real-time connection line segments; obtain the latest real-time connection line segment, which is marked as the real-time target line segment; for example, the real-time target line segment is for short-term battery temperature changes. Step S309: Obtain the left and right endpoints of the real-time connecting line segments, and label them as the first line segment endpoint and the second line segment endpoint; please refer to... Figure 4 As shown, the endpoints of the first and second line segments correspond to the real-time interval temperatures of 30s and 40s, respectively. Step S310: Obtain the real-time battery temperature, real-time temperature difference, and real-time current value corresponding to the endpoint of the first line segment, and label them as the first battery temperature, the first temperature difference, and the first current value. Step S311: Substitute the first temperature difference value as the horizontal axis value into the first historical function to obtain the vertical axis value, and mark it as the first real-time predicted value. Step S312: Substitute the first current value as the horizontal axis value into the second historical function to obtain the vertical axis value, and mark it as the second real-time predicted value; Step S313: Obtain the sum of the first battery temperature, the first real-time predicted value, and the second real-time predicted value, and label it as the first predicted temperature value; Step S314: Obtain the real-time battery temperature corresponding to the endpoint of the second line segment and mark it as the second battery temperature; Step S315: Obtain the difference between the first predicted temperature value and the second battery temperature, and mark it as the real-time predicted difference; that is, the difference between the predicted temperature value and the battery temperature. If battery thermal runaway occurs, the higher the second battery temperature, the smaller the real-time predicted difference will be; therefore, the smaller the negative value of the real-time predicted difference, the greater the probability of battery thermal runaway. In practical applications, the first temperature difference is 0℃. Substituting 0℃ as the horizontal axis value into the first historical function yields a vertical axis value of 0℃, thus the first real-time predicted value is 0℃. The first current value is 10.5A. Substituting the first current value of 10.5A as the horizontal axis value into the second historical function yields a vertical axis value of 5℃, thus the second real-time predicted value is 5℃. Therefore, the first predicted temperature value is: 24℃ + 5℃ + 0℃ = 29℃. The second battery temperature is 29℃, so the real-time predicted difference is 0℃.

[0023] Step S400: Obtain the real-time predicted final temperature value based on the difference between the real-time target line segment and the real-time prediction; perform early warning of battery thermal runaway based on the real-time predicted final temperature value; Step S400 includes the following sub-steps: Step S401: Obtain the slope of the real-time target line segment and mark it as the real-time target slope; Step S402: The line segment whose endpoints are extended by the real-time target slope for a unit time is marked as the real-time prediction line segment; that is, the change in battery temperature within 10 seconds. Step S403: Obtain the temperature of the right end point of the real-time predicted line segment and mark it as the real-time temperature prediction value; Step S404: Obtain the sum of the real-time prediction difference and the real-time temperature prediction value, and mark it as the real-time predicted final temperature value; to make the real-time predicted final temperature value more accurate. Step S405: Determine whether the real-time predicted final temperature value is within the normal operating temperature range of the battery. If not, immediately issue a battery thermal runaway warning; that is, if the predicted battery temperature will exceed the normal operating temperature range of the battery, it is considered that the battery is in thermal runaway. For practical applications, please refer to Figure 4 As shown, the obtained real-time temperature prediction value is 34℃, and the real-time prediction difference is 0℃, so the real-time predicted final temperature value is 34℃; for example, if the normal operating temperature of the battery is less than 50℃, then the real-time predicted final temperature value of 34℃ is judged to be less than 50℃, indicating that the battery is normal.

[0024] Example 2: This application also provides an electronic device, which may include: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus. The memory stores computer-readable instructions, and the processor can call these instructions. When the processor executes these computer-readable instructions, it performs steps such as those in the AI-based early warning method for thermal runaway of electric scooter batteries to achieve the following functions: obtaining historical temperature differences and corresponding historical average radiation values ​​based on normal operating data of the electric scooter battery; obtaining a first historical function based on the historical temperature differences and historical average radiation values; obtaining historical current values ​​and historical operating temperature change values ​​based on normal operating data of the electric scooter battery, and obtaining a second historical function based on these values; obtaining real-time battery temperature, real-time temperature difference, and real-time current value based on real-time operating data of the electric scooter battery; constructing a real-time target line segment based on the real-time battery temperature; obtaining a real-time predicted difference based on the real-time temperature difference, real-time current value, real-time target line segment, first historical function, and second historical function; obtaining a real-time predicted final temperature value based on the real-time target line segment and real-time predicted difference; and providing an early warning of battery thermal runaway based on the real-time predicted final temperature value.

[0025] Furthermore, when the logical instructions in the aforementioned memory can be implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0026] Example 3: This application also provides a computer program product, which includes a computer program stored on a computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the AI-based early warning method for thermal runaway of electric scooter batteries provided by the above methods. The method includes: obtaining historical temperature differences based on normal operating data of the electric scooter battery, and obtaining historical average radiation values ​​corresponding to the historical temperature differences; obtaining a first historical function based on the historical temperature differences and historical average radiation values; obtaining historical current values ​​and historical operating temperature change values ​​based on normal operating data of the electric scooter battery, and obtaining a second historical function based on the historical current values ​​and historical operating temperature change values; obtaining real-time battery temperature, real-time temperature differences, and real-time current values ​​based on real-time operating data of the electric scooter battery; constructing a real-time target line segment based on the real-time battery temperature; obtaining a real-time predicted difference based on the real-time temperature difference, real-time current value, real-time target line segment, first historical function, and second historical function; obtaining a real-time predicted final temperature value based on the real-time target line segment and real-time predicted difference; and performing early warning of battery thermal runaway based on the real-time predicted final temperature value.

[0027] Example 4: This application also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it performs the steps of the AI-based early warning method for thermal runaway of electric scooter batteries described above to achieve the following functions: obtaining historical temperature differences and corresponding historical average radiation values ​​based on normal operating data of the electric scooter battery; obtaining a first historical function based on the historical temperature differences and historical average radiation values; obtaining historical current values ​​and historical operating temperature change values ​​based on normal operating data of the electric scooter battery, and obtaining a second historical function based on the historical current values ​​and historical operating temperature change values; obtaining real-time battery temperature, real-time temperature differences, and real-time current values ​​based on real-time operating data of the electric scooter battery; constructing a real-time target line segment based on the real-time battery temperature; obtaining a real-time predicted difference based on the real-time temperature difference, real-time current value, real-time target line segment, first historical function, and second historical function; obtaining a real-time predicted final temperature value based on the real-time target line segment and real-time predicted difference; and performing early warning of battery thermal runaway based on the real-time predicted final temperature value.

[0028] Based on the above description of the embodiments, the embodiments of the present invention can be provided as methods, systems, or computer program products. Based on this understanding, the above technical solutions, in essence or in terms of their contribution to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or certain parts of the embodiments.

[0029] In the embodiments provided in this application, it should be understood that the disclosed system or method can be implemented in other ways. The embodiments described above are merely illustrative. For example, the division of modules or units is only a logical functional division, and there may be other division methods in actual implementation. Furthermore, multiple modules or units may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the coupling or direct coupling or communication connection shown or discussed may be through some communication interfaces. The indirect coupling or communication connection between systems, modules, and units may be electrical, mechanical, or other forms.

[0030] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. An AI-based early warning method for thermal runaway in electric scooter batteries, characterized in that, Includes the following steps: S100: Obtain historical temperature difference values ​​based on normal operating data of electric scooter batteries; obtain historical average radiation values ​​corresponding to historical temperature difference values; obtain the first historical function based on historical temperature difference values ​​and historical average radiation values. S200: Based on the normal operation data of the electric scooter battery, obtain historical current value and historical operating temperature change value, and obtain a second historical function based on the historical current value and historical operating temperature change value. S300 obtains real-time battery temperature, real-time temperature difference, and real-time current value based on real-time electric scooter battery operation data; constructs a real-time target line segment based on the real-time battery temperature; and obtains the real-time predicted difference based on the real-time temperature difference, real-time current value, real-time target line segment, first historical function, and second historical function. The S400 obtains the real-time predicted final temperature value based on the difference between the real-time target line segment and the real-time prediction, and provides early warning of battery thermal runaway based on the real-time predicted final temperature value.

2. The AI-based early warning method for thermal runaway of electric scooter batteries according to claim 1, characterized in that, S100 includes the following sub-steps: Obtain the temperature range of the electric scooter battery during normal operation and mark it as the historical battery temperature range; Obtain the range of ambient temperature where the electric scooter is located, and mark it as the historical ambient temperature range; Obtain any temperature within the historical battery temperature range and label it as the historical battery temperature; Obtain any temperature within the historical ambient temperature range and label it as the historical ambient temperature; Obtain the difference between historical ambient temperature and historical battery temperature, and label it as historical temperature difference.

3. The AI-based early warning method for thermal runaway of electric scooter batteries according to claim 2, characterized in that, S100 also includes the following sub-steps: Set a duration and label it as a unit duration; When the electric scooter battery is not running, the change in battery temperature after acquiring different historical temperature difference values ​​per unit time is identified as historical temperature radiation value. Obtain the first number of historical temperature radiation values ​​under different historical temperature differences; Obtain the average of historical temperature radiation values ​​under the same historical temperature difference, and label it as the historical radiation mean.

4. The AI-based early warning method for thermal runaway of electric scooter batteries according to claim 3, characterized in that, S100 also includes the following sub-steps: A Cartesian coordinate system is established with historical temperature difference as the horizontal axis and historical average radiation as the vertical axis, and this system is designated as the first historical coordinate system. The historical temperature difference and the corresponding historical mean radiation are used as coordinate points on the x-axis and y-axis, respectively, and are marked as the first historical coordinate point. Obtain all first historical coordinate points, plot all first historical coordinate points in the first historical coordinate system to obtain a scatter plot, and label it as the first historical scatter plot; The function obtained by fitting the first historical coordinate points in the first historical scatter plot is identified as the first historical function.

5. The AI-based early warning method for thermal runaway of electric scooter batteries according to claim 4, characterized in that, S200 includes the following sub-steps: Obtain the operating current range of the electric scooter battery and label it as the historical current range; Obtain any current within the historical current range and label it as a historical current value; Set a specific temperature range to keep the temperature of the electric scooter battery and the ambient temperature of the electric scooter within that specific temperature range; Within a specific temperature range and under the same historical current value, the temperature change value of the electric scooter battery after a unit of time is obtained and identified as the historical operating temperature change value. Obtain the historical operating temperature variation values ​​corresponding to different historical current values.

6. The AI-based early warning method for thermal runaway of electric scooter batteries according to claim 5, characterized in that, S200 also includes the following sub-steps: A Cartesian coordinate system is established with historical current values ​​on the horizontal axis and historical operating temperature changes on the vertical axis, and this system is designated as the first historical coordinate system. The historical current value and the corresponding historical operating temperature change value are obtained and used as coordinate points on the horizontal and vertical axes, respectively, and are marked as the second historical coordinate points; Obtain all the second historical coordinate points, plot all the second historical coordinate points in the first historical coordinate system to obtain a scatter plot, and label it as the second historical scatter plot; The function obtained by fitting the second historical coordinate points in the second historical scatter plot is labeled as the second historical function.

7. The AI-based early warning method for thermal runaway of electric scooter batteries according to claim 6, characterized in that, S300 includes the following sub-steps: Obtain the temperature of the electric scooter battery and label it as the real-time battery temperature; Get the ambient temperature where the electric scooter is located and label it as the real-time ambient temperature; Obtain the difference between the real-time ambient temperature and the real-time battery temperature, and label it as the real-time temperature difference; Obtain the current flow of the electric scooter battery and label it as a real-time current value.

8. The AI-based early warning method for thermal runaway of electric scooter batteries according to claim 7, characterized in that, S300 also includes the following sub-steps: The real-time battery temperature is acquired once every first interval and marked as the real-time interval temperature. A Cartesian coordinate system is established with time as the horizontal axis and real-time interval temperature as the vertical axis, and it is labeled as the third real-time coordinate system. The acquisition time and the corresponding real-time interval temperature are used as coordinate points on the horizontal and vertical axes, respectively, and are marked as the third real-time coordinate point; Obtain all the third real-time coordinate points, plot all the third real-time coordinate points in the third real-time coordinate system to obtain a scatter plot, and label it as the third real-time scatter plot; Only the latest third real-time coordinate points within the second time period are retained in the third real-time coordinate system; In the third real-time coordinate system, adjacent third real-time coordinate points are connected to obtain line segments, which are then identified as real-time connection line segments. Retrieve the latest real-time connection segments and mark them as real-time target segments.

9. The AI-based early warning method for thermal runaway of electric scooter batteries according to claim 8, characterized in that, S300 also includes the following sub-steps: Obtain the left and right endpoints of the real-time connecting line segments and label them as the first line segment endpoint and the second line segment endpoint; Obtain the real-time battery temperature, real-time temperature difference, and real-time current value corresponding to the endpoint of the first line segment, and identify them as the first battery temperature, the first temperature difference, and the first current value. Substitute the first temperature difference value as the horizontal axis value into the first historical function to obtain the vertical axis value, which is then identified as the first real-time predicted value. Substitute the first current value as the horizontal axis value into the second historical function to obtain the vertical axis value, which is then identified as the second real-time predicted value. The sum of the first battery temperature, the first real-time predicted value, and the second real-time predicted value is obtained and identified as the first predicted temperature value. Get the real-time battery temperature corresponding to the endpoint of the second line segment and mark it as the second battery temperature; The difference between the first predicted temperature value and the second battery temperature is obtained and identified as the real-time predicted difference.

10. The AI-based early warning method for thermal runaway of electric scooter batteries according to claim 9, characterized in that, S400 includes the following sub-steps: Obtain the slope of the real-time target line segment and label it as the real-time target slope; The segment whose endpoints are extended by a unit time length with the real-time target slope is identified as the real-time predicted segment. Obtain the temperature at the right end of the real-time predicted line segment and label it as the real-time temperature prediction value; The sum of the real-time prediction difference and the real-time temperature prediction value is obtained and identified as the final real-time predicted temperature value. Determine whether the real-time predicted final temperature value is within the normal operating temperature range of the battery. If not, immediately issue a battery thermal runaway warning.

Images