A method and system for monitoring and early warning of charging thermal runaway of shared electric vehicle batteries

Abstract

The present application relates to the field of battery temperature control early warning, especially to a shared electric vehicle battery charging thermal runaway monitoring and early warning method and system, the method comprising the following steps: based on the temperature sensors of each battery position of the charging cabinet, the surface temperature of the electric vehicle battery being charged is continuously collected at a fixed sampling period, and the temperature monitoring curve of different battery positions is constructed; based on the temperature monitoring curve, double-index weighted calculation is performed to obtain the thermal runaway risk index sequence of each battery; the short-term risk index of the thermal runaway risk index sequence is predicted to obtain the risk index prediction value; and the abnormal position number is determined based on the risk index prediction value. The present application improves the early identification accuracy of thermal abnormalities during the charging of shared electric vehicle batteries, thereby reducing the probability of thermal runaway accidents and the misjudgment rate, and improving the operation safety of shared charging equipment.

Description

Technical Field

[0001] This invention relates to the field of battery temperature control and early warning, and in particular to a method and system for monitoring and early warning of thermal runaway during charging of shared electric vehicle batteries. Background Technology

[0002] With the continuous development of the sharing economy and the concept of green travel, shared electric vehicles have been widely used in various scenarios such as short-distance urban travel, on-demand delivery, and community commuting due to their convenience, low cost, and high flexibility. Especially with the rapid popularization of centralized battery swapping and charging models, a large number of shared electric vehicle batteries need to be centrally charged and managed within battery swapping or charging cabinets, thereby effectively improving vehicle operating efficiency and battery turnover efficiency. As a result, lithium-ion batteries, due to their advantages of high energy density, long cycle life, and high charging efficiency, are gradually becoming the main power source for shared electric vehicles.

[0003] Existing thermal safety management methods for shared electric vehicle batteries mostly rely on a single temperature threshold for detecting thermal anomalies. When the battery temperature exceeds a set threshold, protection is achieved by stopping charging or cutting off power. This type of method typically depends on a single temperature parameter, lacking the ability to comprehensively assess the rate of temperature rise, temperature trend, and continuous thermal changes, making it difficult to accurately identify abnormal characteristics in the early stages of thermal runaway. The charging environment for shared electric vehicles is complex, easily affected by changes in ambient temperature, fluctuations in charging current, and thermal coupling effects within the charging enclosure. This causes traditional fixed threshold methods to easily misjudge normal charging heat as abnormal thermal risks, resulting in a high false alarm rate; or they may fail to identify true thermal anomalies in a timely manner, leading to missed alarms and ultimately lower accuracy in overall thermal runaway analysis. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention proposes a method and system for monitoring and warning of thermal runaway during charging of shared electric vehicle batteries, thereby resolving at least one of the aforementioned technical issues.

[0005] To achieve the above objectives, the present invention provides a method for monitoring and early warning of thermal runaway during charging of shared electric vehicle batteries, comprising the following steps: Step S1: Based on the temperature sensors of each battery compartment in the charging cabinet, the surface temperature of the electric vehicle battery being charged is continuously collected at a fixed sampling period to construct temperature monitoring curves for different battery compartments. Step S2: Based on the temperature monitoring curve, perform a dual-index weighted calculation to obtain the thermal runaway risk index sequence for each battery; Step S3: Perform short-time risk index prediction on the thermal runaway risk index sequence to obtain the predicted risk index value; Step S4: Determine the abnormal position number based on the risk index prediction value; activate the multi-component gas detection module to perform real-time gas sampling according to the abnormal position number, and extract the current measured concentration values ​​of different gas components; Step S5: Analyze the content of abnormal gas components based on the current measured concentration value and output an abnormal gas analysis report; Step S6: Conduct a thermal runaway risk assessment based on the abnormal gas analysis report and risk index prediction value to obtain the thermal runaway risk level; perform adaptive thermal runaway early warning response processing based on the thermal runaway risk level.

[0006] This specification provides a charging thermal runaway monitoring and early warning system for shared electric vehicle batteries, used to execute the charging thermal runaway monitoring and early warning method for shared electric vehicle batteries as described above, including: The data acquisition module is used to continuously collect the surface temperature of the electric vehicle battery being charged at a fixed sampling period based on the temperature sensors of each battery compartment in the charging cabinet, and to construct temperature monitoring curves for different battery compartments. The risk calculation module is used to perform dual-index weighted calculation based on the temperature monitoring curve to obtain the thermal runaway risk index sequence of each battery. The prediction module is used to perform short-term risk index prediction on the thermal runaway risk index sequence to obtain the predicted risk index value; The gas sampling module is used to determine the abnormal warehouse number based on the risk index prediction value; and to activate the multi-component gas detection module to perform real-time gas sampling based on the abnormal warehouse number, and to extract the current measured concentration values ​​of different gas components. The gas analysis module is used to analyze the content of abnormal gas components in the current measured concentration value and output an abnormal gas analysis report; The early warning response module is used to assess the risk of thermal runaway based on the abnormal gas analysis report and the predicted risk index, and obtain the thermal runaway risk level; and to perform adaptive thermal runaway early warning response processing based on the thermal runaway risk level.

[0007] The specific benefits of this invention are as follows: By installing temperature sensors in each battery compartment of the charging cabinet and continuously collecting data at a fixed sampling period, parallel temperature sensing of multiple batteries is achieved, reducing the information loss problem caused by single-point monitoring; the constructed temperature monitoring curve can completely present the temperature rise trajectory during battery charging, providing continuous data support for identifying abnormal temperature rise, thereby improving the completeness and stability of temperature change capture. Based on the temperature monitoring curve, a weighted calculation of both temperature rise amplitude and temperature rise rate is introduced, enabling risk assessment to reflect the degree of heat accumulation and the intensity of temperature rise changes, avoiding misjudgments caused by a single parameter; the construction of a risk index sequence can realize the quantitative expression and horizontal comparison of battery thermal state, improving the identifiability and distinguishability of thermal risk differences between different batteries; short-term prediction processing of the risk index sequence can estimate the risk change trend in advance, improving the ability to perceive the direction of risk growth; through multi-timescale trend analysis, the ability to identify rapid and continuous temperature rise characteristics can be enhanced, making risk change performance more forward-looking and continuous. Based on the prediction results, abnormal compartments are identified, and multi-component gas detection modules are activated in a targeted manner to achieve targeted gas sampling and reduce the consumption of detection resources in irrelevant compartments. By acquiring information on multiple gas components in real time, the internal chemical decomposition characteristics of the battery can be quickly captured, improving the targeting and sensitivity of abnormal state identification. By comparing and analyzing the measured gas concentration with the background concentration baseline, the influence of environmental factors can be effectively eliminated, improving the accuracy of gas anomaly identification. Joint analysis of multiple gas components can enhance the comprehensive judgment ability of electrolyte decomposition and abnormal reactions, improving the reliability of abnormal feature identification. By integrating gas analysis results with risk prediction values ​​for risk assessment, a multi-dimensional comprehensive judgment of thermal runaway states can be achieved, improving the stability of risk level classification. Based on different risk levels, differentiated response processing can be implemented to achieve a graded control strategy from early warning to suppression, improving the timeliness and effectiveness of overall safety management. Attached Figure Description

[0008] Figure 1 This is a schematic flowchart of the steps of a method for monitoring and early warning of thermal runaway during charging of a shared electric vehicle battery according to the present invention. Figure 2 This is a detailed flowchart illustrating the implementation steps of step S1. Figure 3 This is a detailed flowchart illustrating the implementation steps of step S2; Figure 4 This is a schematic diagram of the temperature monitoring curve in step S1; Figure 5 This is a schematic diagram showing the baseline deviation multiples for different gas components. Detailed Implementation

[0009] It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the invention.

[0010] This application provides a method and system for monitoring and early warning of thermal runaway during charging of a shared electric vehicle battery. The implementing entities of the method and system include, but are not limited to, mechanical equipment, data processing platforms, cloud server nodes, and network upload devices that can be considered general computing nodes in this application. The data processing platform includes, but is not limited to, at least one of an audio / image management system, an information management system, and a cloud-based data management system.

[0011] Please see Figures 1 to 5 This invention provides a method for monitoring and early warning of thermal runaway during charging of shared electric vehicle batteries, comprising the following steps: Step S1: Based on the temperature sensors of each battery compartment in the charging cabinet, the surface temperature of the electric vehicle battery being charged is continuously collected at a fixed sampling period to construct temperature monitoring curves for different battery compartments. Step S2: Based on the temperature monitoring curve, perform a dual-index weighted calculation to obtain the thermal runaway risk index sequence for each battery; Step S3: Perform short-time risk index prediction on the thermal runaway risk index sequence to obtain the predicted risk index value; Step S4: Determine the abnormal position number based on the risk index prediction value; activate the multi-component gas detection module to perform real-time gas sampling according to the abnormal position number, and extract the current measured concentration values ​​of different gas components; Step S5: Analyze the content of abnormal gas components based on the current measured concentration value and output an abnormal gas analysis report; Step S6: Conduct a thermal runaway risk assessment based on the abnormal gas analysis report and risk index prediction value to obtain the thermal runaway risk level; perform adaptive thermal runaway early warning response processing based on the thermal runaway risk level.

[0012] In the embodiments of the present invention, see Figure 1 The diagram below illustrates the steps of a method for monitoring and warning of thermal runaway during charging of a shared electric vehicle battery according to the present invention. In this example, the steps of the method include: Step S1: Based on the temperature sensors of each battery compartment in the charging cabinet, the surface temperature of the electric vehicle battery being charged is continuously collected at a fixed sampling period to construct temperature monitoring curves for different battery compartments. Step S2: Based on the temperature monitoring curve, perform a dual-index weighted calculation to obtain the thermal runaway risk index sequence for each battery; Step S3: Perform short-time risk index prediction on the thermal runaway risk index sequence to obtain the predicted risk index value; Step S4: Determine the abnormal position number based on the risk index prediction value; activate the multi-component gas detection module to perform real-time gas sampling according to the abnormal position number, and extract the current measured concentration values ​​of different gas components; Step S5: Analyze the content of abnormal gas components based on the current measured concentration value and output an abnormal gas analysis report; Step S6: Conduct a thermal runaway risk assessment based on the abnormal gas analysis report and risk index prediction value to obtain the thermal runaway risk level; perform adaptive thermal runaway early warning response processing based on the thermal runaway risk level.

[0013] In one specific embodiment, all 12 battery compartments (B1 to B12) within the charging cabinet are monitored in parallel. The surface temperature of the batteries is continuously collected at a fixed sampling period of 5 seconds, forming their respective original temperature sequences. Taking battery B5 as an example, its original temperature sequence within a 5-minute temperature rise window is as follows: B5: T = {34.2, 34.6, 35.1, 35.8, 36.7, 37.6, 38.4, 39.6} (℃); The remaining normal cells, such as B1-B4 and B6-B12, have the following sequences: Bi: T = {34.0, 34.1, 34.2, 34.3, 34.4, 34.5, 34.6, 34.7} (℃); The original sequence is then subjected to moving average filtering. The filtering formula is as follows: n = (Tn + Tn-1 + Tn-2 + Tn-3 + Tn-4) / 5; After filtering, the B5 temperature sequence smoothed into a monotonically increasing curve, with a terminal temperature of 39.6℃ and an initial temperature of 34.2℃, a standard deviation of approximately 0.62℃. The remaining batteries exhibited fluctuations below 1℃, indicating a stable charging state. Based on the temperature monitoring curves, the temperature rise amplitude and rate were calculated for each battery, expressed as follows: ΔT = Tend – Tstart; v = ΔT / Δt (Δt = 5 min); Taking B5 as an example: ΔT = 39.6 34.2 = 5.4℃; v = 5.4 / 5 = 1.08 ℃ / min; According to the segmented scoring rules, v=1.08℃ / min corresponds to 25 points, ΔT=5.4℃ corresponds to 10 points, and the dual-indicator weighted risk index is expressed as: R = 0.4×SΔT + 0.6×Sv; Substituting the values, we get: R = 0.4 × 10 + 0.6 × 25 = 19; The remaining battery risk indices all ranged from 8 to 12. Subsequently, multi-timescale predictions (2 min, 8 min, 20 min) were performed on the risk index series, using the short-term growth rate calculation: k = (Rt Rt-1) / Δt; The B5 short-term series is 8→10→13, yielding k≈4.5 / min, mean μ≈10.33, and standard deviation σ≈2.05. Short-term forecasting uses a trend extrapolation model: Rpred (predicted risk index) = Rt + k × tfuture; Under the condition tfuture = 3 min: Rpred = 19 + 4.5 × 3 = 32.5; Since Rpred < 85 (safety threshold) at this time, the activation condition of the gas detection module was not triggered, and the system only continued to update the temperature and risk index.

[0014] In subsequent continuous monitoring, as the temperature rise of B5 battery continued to accelerate, its short-term growth rate increased from 4.5 / min to over 20 / min, causing the risk index prediction value to evolve further. For example, when Rt = 60, k = 10 / min, then Rpred = 60 + 10×3 = 90. At this point, the following condition must be met: Rpred ≥ 85 (safety threshold); Therefore, an abnormal storage location was determined: B5 was marked as an abnormal storage location, and the multi-component gas detection module for that location was activated for high-frequency sampling (sampling period 5 s). The subsequent measured gas values ​​were: H2 = 820 ppm, CO = 95 ppm, VOC = 210 ppm, and the deviation rate was calculated as: P = (Cmeas) Cbase) / Cbase × 100%; For example: H2 deviation rate = (820 40) / 40 = 1950%; Ultimately, due to "Rpred≥85 + multiple gas anomalies synchronously occurring", battery B5 was determined to be in a Level 3 risk state. During the response phase, the Level 3 risk trigger immediately executed a power-off control, causing the charging current of B5 to rapidly drop from its current value to 0A. The dry powder spraying device inside the cabinet was activated and put into release mode, providing localized coverage and suppression of the target compartment. Simultaneously, event information (cabinet number A01, compartment B5, Rpred=90, H2=820 ppm, etc.) was uploaded to the fire emergency response platform, enabling remote alarm and coordinated response. The remaining batteries B1-B4 and B6-B12 continued normal charging and independent risk monitoring, unaffected by this abnormal event.

[0015] In this embodiment, see Figure 2 The diagram below illustrates the detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include: Temperature sensors in each battery compartment of the charging cabinet continuously collect the surface temperature of the electric vehicle batteries being charged at a fixed sampling period, generating the original temperature sequence of each compartment. Sensor glitch noise is identified in the original temperature time series, and sliding window mean filtering is performed to generate a filtered temperature series. Determine the timestamp for temperature acquisition; map and label the filtered temperature sequence based on the timestamp, and perform discrete fitting to construct temperature monitoring curves for different battery compartments.

[0016] In this embodiment, the shared electric vehicle battery charging cabinet has multiple independent battery compartments, each equipped with a temperature sensor to collect real-time data on the battery surface temperature during charging. The temperature sensor is preferably a digital thermistor or an NTC thermistor, and the temperature measurement range is set to... The measurement accuracy is controlled within ±0.3℃ and the temperature resolution is not less than 0.1℃, ranging from 20℃ to 120℃. The sensor is installed in the middle of the battery casing or near the tab area, with the distance between it and the battery surface controlled within 3 mm to improve the ability to detect local heat accumulation areas. After the battery enters the charging state, continuous temperature acquisition is performed according to a fixed sampling period, which can be set to 2 seconds, 5 seconds, or 10 seconds. In this embodiment, sampling is preferably performed once every 5 seconds. After each sampling, the detected temperature value is converted into a digital signal and transmitted to the data processing unit through the RS485 communication bus or CAN communication bus. The data is cached and stored in chronological order to form the original temperature time series of the corresponding compartment. During the acquisition process, parameters such as compartment number, charging current, charging voltage, ambient temperature, and charging status are recorded simultaneously. Information such as whether the charging current exceeds 15A and whether the ambient temperature of the compartment is higher than 35℃ is also recorded to analyze the thermal change characteristics of the battery in conjunction with actual operating conditions.

[0017] Because charging cabinets are susceptible to electromagnetic interference, communication fluctuations, fan start / stop, and unstable sensor contact during operation, instantaneous anomalies may appear in the raw temperature data. A sudden jump in temperature from 36℃ to 52℃ within 5 seconds, followed by a return to the normal range, is typical of glitch noise. To avoid anomalies affecting thermal runaway assessment, glitch identification processing is performed on the raw temperature sequence. The temperature change amplitude between adjacent sampling points is compared, and a temperature abrupt change threshold is set. A temperature change exceeding 6℃ within a single sampling period is considered a suspected anomaly. Verification is then performed by considering the temperature change trend of 3 to 5 sampling points before and after the anomaly. If the temperature fluctuation range before and after the anomaly point is less than 2℃, the current data is identified as glitch noise and corrected using the average of nearby valid data. After anomaly identification, a sliding window mean filtering method is used to smooth the temperature sequence. The sliding window length is preferably set to 5 sampling points, which can be adjusted to 3 sampling points in fast charging mode and extended to 7 sampling points in slow charging mode. By calculating the continuous temperature average, random high-frequency fluctuations are weakened, making the temperature change trend more stable, thus forming a reliable filtered temperature sequence.

[0018] After obtaining the filtered temperature data, each temperature acquisition result is timestamped to establish a correspondence between temperature data and acquisition time. A high-precision real-time clock module is installed inside the charging cabinet, and time calibration is performed every 12 hours via network time synchronization to keep the overall time error within ±1 second. After each temperature sampling, the corresponding acquisition time information, including hour, minute, second, and millisecond-level time parameters, is recorded, and the time information is bound to the corresponding temperature value to form a temperature data record with time attributes. Since the actual acquired data is discrete time-point data, discrete fitting processing is performed on the filtered temperature sequence to improve the continuity of temperature change trends. A combination of piecewise linear fitting and low-order curve fitting is used to construct the temperature monitoring curve. When the temperature change amplitude is less than 3℃, linear interpolation is used for connection; when there is a continuous temperature rise for three consecutive sampling cycles with a temperature rise rate exceeding 1℃ / min, curve fitting is used to reconstruct the local temperature rise trajectory. The temperature rise rate per unit time is calculated in real time. When the battery surface temperature remains above 45°C for more than 60 seconds and the temperature rise rate exceeds 1.5°C / min, it is determined that there is a risk of thermal runaway and the corresponding warning state is triggered.

[0019] In this embodiment, see Figure 3 The diagram below illustrates the detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include: Set a temperature rise calculation window; divide the temperature monitoring curve based on the temperature rise calculation window and extract multiple temperature monitoring windows; The temperature rise characteristics of each battery are calculated based on multiple temperature monitoring windows to obtain the temperature rise amplitude and temperature rise rate of each battery. Based on the temperature rise amplitude and temperature rise rate, a dual-index weighted calculation is performed to obtain the real-time thermal runaway risk index for each battery. In this embodiment, a temperature rise calculation window is set based on the thermal change characteristics during battery charging. The window duration can be adjusted according to the charging mode, set to 3-5 minutes in normal charging and 1-3 minutes in fast charging. This embodiment preferably uses a 5-minute temperature rise calculation window, that is, the temperature change within 5 consecutive minutes is taken as an independent thermal analysis unit. The window sliding step size is preferably set to 30 seconds or 60 seconds to allow for partial overlap between adjacent windows, thereby improving the ability to capture thermal change trends. The temperature monitoring curve is continuously divided according to time sequence to form multiple temperature monitoring windows, each containing all temperature sampling points within the corresponding time period. Under a 5-second sampling period, a single 5-minute window can contain 60 sets of temperature data. Independent temperature monitoring windows are established for different battery compartments, and parameters such as the corresponding compartment number, window start time, window end time, ambient temperature, and charging current are recorded simultaneously. When the ambient temperature is higher than 35°C or the charging current exceeds 15A, the corresponding window is automatically marked as a high-load thermal analysis window to improve the sensitivity to abnormal temperature rise behavior, thereby providing a unified data analysis basis for subsequent temperature rise characteristic calculations.

[0020] The starting and ending temperature values ​​within each temperature monitoring window are extracted, and the overall temperature change within that time period is calculated. When a battery rises from 32℃ to 41℃ within a 5-minute window, the corresponding temperature rise is 9℃. The average temperature rise rate per unit time is calculated based on the window duration. When the temperature rises by 9℃ within 5 minutes, the corresponding temperature rise rate is 1.8℃ / min. To avoid the influence of local instantaneous fluctuations on the calculation results, the average temperature change trend of multiple consecutive sampling points within the window is prioritized during the calculation process, and the continuity of change between consecutive windows is used for verification. When the temperature rise rate deviation between two consecutive windows exceeds 30%, the data of the current window is subjected to secondary smoothing to improve calculation stability. For batteries in fast charging mode, the window duration can be appropriately shortened to 2 minutes to improve the response capability to rapid temperature rise phenomena; for normal slow charging mode, a 5-minute window is maintained to enhance trend stability. After the calculation is completed, the temperature rise amplitude, temperature rise rate, window time range and storage location number of each battery are associated and recorded to form a temperature rise feature dataset for the corresponding battery, providing basic parameters for thermal runaway risk assessment.

[0021] The temperature rise rate is mainly used to reflect the thermal runaway development trend of the battery over a short period of time, while the temperature rise amplitude reflects the overall heat accumulation degree of the battery. The temperature rise rate is first divided into intervals and scored. When the temperature rise rate is below 0.5℃ / min, it is judged as a stable thermal state, corresponding to a score of 0; when the temperature rise rate is between 0.5 and 1℃ / min, it is judged as a slight temperature rise, corresponding to a score of 10; when the temperature rise rate is between 1 and 2℃ / min, it is judged as a slow rise, corresponding to a score of 25; when the temperature rise rate reaches 2 to 3℃ / min, it is judged as a medium-speed rise, corresponding to a score of 40; when the temperature rise rate reaches 4 to 6℃ / min, it is judged as a rapid temperature rise state, corresponding to a score of 75; when the temperature rise rate exceeds 8℃ / min, it is directly judged as a critical state of thermal runaway, corresponding to a score of 100. The temperature rise amplitude was assigned a score based on different ranges: 0 points for a rise below 5°C, 35 points for a rise between 15-25°C, and 100 points for a rise exceeding 50°C. After assigning scores to both indicators, a weighted average was calculated with a weight of 0.4 for the temperature rise amplitude and 0.6 for the temperature rise rate. This weights the rate of thermal change in the risk assessment, enhancing the ability to identify the early stages of thermal runaway. The resulting real-time thermal runaway risk index for the corresponding battery was then quantified into a risk level ranging from 0 to 100. A risk index above 70 was considered high-risk, and above 85 was considered dangerous.

[0022] Each battery compartment is assigned a unique number, encoded using the format "cabinet number + compartment number," such as B01-03 representing battery compartment number 3 in charging cabinet B01. After risk calculation, parameters such as the corresponding compartment number, risk index, temperature rise rate, temperature rise amplitude, sampling time, and charging status are uniformly packaged and recorded to form a thermal risk data unit for the corresponding battery. The risk index data corresponding to each moment is stored continuously in chronological order to construct a thermal runaway risk index sequence for each battery.

[0023] In this embodiment, step S3 includes the following steps: Define three time analysis periods; Based on the three time analysis periods, the thermal runaway risk index sequence is subjected to multi-time-scale trend calculation to extract the risk index change trends at multiple time scales; the risk index change trends include risk growth rate, mean and standard deviation of volatility.

[0024] Based on the trend of the risk index change, a short-term risk index prediction is made to obtain the predicted risk index value. In this embodiment, three different time analysis periods are set for the thermal runaway risk index sequence to reflect the characteristics of rapid changes in the short term, stable changes in the medium term, and cumulative changes in the long term, respectively. The three time analysis periods are defined as short-term analysis period, medium-term analysis period, and long-term analysis period. The short-term analysis period is set to 1 min–3 min, preferably using a 2-minute analysis window, to identify the risk of rapid temperature rise of the battery in a short period; the medium-term analysis period is set to 5 min–10 min, preferably using an 8-minute analysis window, to analyze the stable thermal change trend of the battery during continuous charging; the long-term analysis period is set to 15 min–30 min, preferably using a 20-minute analysis window, to assess the overall thermal accumulation state of the battery. All three analysis periods are continuously updated using a sliding time window method, with the window sliding step size preferably set to 30 s or 60 s, so that there is continuous overlap between different scales, thereby improving the ability to capture risk change trends. In specific implementation, each time analysis period corresponds to an independent risk data cache area to store the thermal runaway risk index data for the corresponding time period. It also synchronously associates operating parameters such as the storage compartment number, charging current, ambient temperature, and current charging stage. When the ambient temperature exceeds 35°C or the battery is in the fast charging phase, the data refresh rate of the short-term analysis cycle should be appropriately increased to enhance the ability to identify rapid thermal anomalies. By setting analysis cycles of varying scales, both the identification of rapid changes in the early stages of thermal runaway and the analysis of long-term thermal accumulation trends can be considered, thereby improving the overall accuracy of thermal risk monitoring.

[0025] The changes in the risk index within each analysis period are statistically calculated, including the risk growth rate, the risk mean, and the standard deviation of fluctuation. The risk growth rate reflects the rate at which the risk index increases over time; if the risk index increases from 40 to 65 within a 2-minute short-term analysis period, a significant rapid upward trend is identified. The risk mean reflects the overall thermal risk level within the current time period; within an 8-minute medium-term analysis period, the average of all risk indices is calculated to determine whether the battery is currently in a medium-to-high risk state for an extended period. The standard deviation of fluctuation reflects the stability of the risk index changes; a high standard deviation indicates significant fluctuations in the risk index, potentially indicating intermittent abnormal heating. In practice, independent trend analysis results are established for different scales: short-term analysis focuses on identifying rapid heating, medium-term analysis focuses on judging continuous heating, and long-term analysis focuses on analyzing the degree of heat accumulation. When the short-term risk growth rate exceeds 20% and the standard deviation of fluctuation is greater than 15, a tendency for sudden thermal anomalies is identified; when the risk mean remains above 70 within a long-term analysis period, a risk of continuous heat accumulation is identified. It also cross-compares the changing trends across different scales. When the short-term and long-term risk trends show a continuous upward trend, the risk level of thermal runaway is increased, thereby enhancing the ability to identify complex thermal change behaviors.

[0026] By combining the risk growth rate, risk mean, and standard deviation of volatility within short-term, medium-term, and long-term analysis periods, the future trend of the risk index is predicted. A combination of trend extension and weighted change analysis is preferred to predict the risk index within the next 1 to 5 minutes; this embodiment preferably predicts risk changes within the next 3 minutes. During the prediction process, a high weight (0.5) is assigned to the short-term risk growth rate, a weight of 0.3 to the medium-term mean trend, and a weight of 0.2 to the long-term heat accumulation trend, thereby improving sensitivity to rapid thermal anomaly changes. When the short-term risk index grows rapidly for several consecutive periods, and the standard deviation of risk volatility continues to increase, it is determined that the risk has a further accelerating upward trend, and the predicted growth rate is increased. When the current risk index is 68, and the short-term growth rate exceeds 15% for 60 seconds consecutively, the risk index is predicted to exceed 85 within the next 3 minutes, thus triggering an early high-risk warning state. A dynamic correction mechanism is set up for the prediction results. When the deviation between the new real-time risk index and the prediction exceeds 10%, the prediction parameters are recalibrated to ensure that the prediction results are consistent with the actual thermal change state.

[0027] In this embodiment, step S4 includes the following steps: The risk index prediction value is compared and analyzed based on a preset safety threshold. When the risk index prediction value of the battery in the charging cabinet is not less than the preset safety threshold, the abnormal storage location number is extracted. Based on the abnormal compartment number, the corresponding compartment of the charging cabinet is activated to perform real-time gas sampling and extract the compartment gas signal of the abnormal battery. The gas signal in the storage compartment is quantitatively detected and the concentration of each component is calculated to extract the current measured concentration value of different gas components.

[0028] In this embodiment, a unified safety threshold parameter is pre-set to determine whether the battery has a risk of thermal runaway. Preferably, a risk index of 85 is set as the safety threshold. When the predicted risk index value for a battery reaches or exceeds 85, the corresponding battery is determined to have a high risk of thermal runaway. The predicted risk index values ​​for each compartment are cyclically detected in chronological order, and stability is judged based on multiple consecutive sampling cycles. When the predicted risk index value is not lower than 85 for three consecutive analysis cycles and the duration exceeds 60 seconds, it is considered a valid abnormal state to avoid false triggering due to short-term fluctuations or instantaneous anomalies. After completing the risk determination, the abnormal compartment number of the corresponding battery is immediately extracted, and the corresponding cabinet number, predicted risk index value, current temperature rise rate, ambient temperature, and charging current are recorded simultaneously. When the predicted risk index value of compartment 5 in charging cabinet A01 reaches 91 and the temperature rise rate exceeds 2℃ / min, "A01-05" is marked as an abnormal compartment number, and the corresponding compartment's gas detection process begins, thereby achieving rapid location and accurate identification of high-risk battery compartments.

[0029] Each battery compartment is equipped with an independent gas sampling channel and a multi-component gas detection module. The detection module includes a volatile organic compound (VOC) sensor, a hydrogen sensor, a carbon monoxide sensor, and an electrolyte characteristic gas sensor, used to identify abnormal gas components that may be released during the early stages of battery thermal runaway. The gas sampling channel is preferably located at the top of the compartment or near the direction of battery depressurization, with the distance between the sampling port and the battery surface preferably controlled within the range of 5 cm to 15 cm to improve gas sampling sensitivity. When an abnormal compartment is activated, the corresponding gas detection module immediately enters high-frequency sampling mode, shortening the sampling period from 30 s under normal conditions to 5 s, and activating a micro-vacuum device to actively extract gas from the compartment. The gas sampling duration is preferably set to 60 s to 120 s, during which time the gas change signal inside the compartment is continuously acquired. To avoid interference caused by gas diffusion from adjacent compartments, the airflow exchange channel near the abnormal compartment is simultaneously closed, and the fan speed is reduced to minimize the impact of airflow disturbance on the detection results. During the sampling process, the gas voltage signal, current signal and concentration change curve collected in real time are buffered and recorded to form a gas signal data sequence for the corresponding abnormal compartment, providing a data basis for subsequent gas component concentration analysis.

[0030] The raw voltage or current signals output by the gas detection module are standardized and converted into corresponding gas concentration values ​​based on preset calibration parameters. The preferred gas components for detection include hydrogen, carbon monoxide, volatile organic compounds (VOCs), electrolyte cracking gases, and combustible hydrocarbons. Specifically, the hydrogen concentration detection range is preferably set to 0–10000 ppm, the carbon monoxide range to 0–1000 ppm, and the VOCs range to 0–5000 ppm. During the detection process, each gas is calculated independently and corrected using temperature and humidity compensation parameters. When the humidity inside the chamber exceeds 80%, humidity drift correction is applied to the gas concentration results to reduce the impact of environmental factors on detection accuracy. After concentration calculation, the current measured concentration values ​​for each gas component are extracted and continuously recorded chronologically. When the detected hydrogen concentration exceeds 1500 ppm, the carbon monoxide concentration exceeds 120 ppm, and the VOC concentration continues to rise, it is determined that the battery may have entered the electrolyte decomposition stage. The study also conducted correlation analysis on the concentration changes of different gases, and further increased the risk level of thermal runaway when multiple hazardous gas components increased rapidly.

[0031] In this embodiment, step S5 includes the following steps: Extract the baseline of gas background concentration under normal charging conditions of the charging cabinet; The current measured concentration value is compared with the gas background concentration baseline item by item to calculate the deviation value of different gas components; Based on the deviation value, the deviation rate is calculated, and the content of abnormal gas components is analyzed, and an abnormal gas analysis report is output.

[0032] In this embodiment, under stable slow charging conditions with no abnormal batteries or overheating alarms in the charging cabinet, a continuous and stable operating period is selected for gas background sampling. The preferred sampling duration is 10-30 minutes; this embodiment uses 20 minutes as the baseline sampling period. During this time period, the gas detection module samples at a fixed frequency, with a sampling period that can be set to 5 or 10 seconds, recording concentration data for various components such as hydrogen, carbon monoxide, volatile organic compounds (VOCs), and electrolyte characteristic gases. To improve baseline stability, the sampled data is processed by removing extreme values ​​within the top and bottom 5% to eliminate the influence of occasional disturbances. The remaining data is averaged to obtain the baseline background concentration values ​​for various gases under normal charging conditions. The background concentration of hydrogen is typically in the range of 0-50 ppm, carbon monoxide is approximately 0-10 ppm, and VOCs are approximately 10-80 ppm. Corrections are made based on the current ambient temperature (e.g., 25℃-35℃) and humidity (e.g., 40%-70%RH). Ultimately, a multi-component gas background concentration baseline set is formed for subsequent anomaly deviation comparison analysis, thereby avoiding misjudgment caused by environmental factors.

[0033] Independent comparative calculations were performed for hydrogen, carbon monoxide, volatile organic compounds, and electrolyte cracking gases. When the measured hydrogen concentration at a given moment was 820 ppm and the background baseline was 40 ppm, the calculated hydrogen deviation was 780 ppm. Similarly, when the measured carbon monoxide concentration was 95 ppm and the background baseline was 8 ppm, the deviation was 87 ppm. To ensure data consistency, a time synchronization mechanism was introduced during the comparison process to match the background baseline with the current sampling data on different scales, avoiding errors caused by differences in sampling time periods. Dynamic corrections were applied to the deviation values. When the ambient temperature exceeded 40℃ or the humidity exceeded 80%, temperature and humidity correction factors were applied to adjust the deviation values ​​to reduce the impact of environmental disturbances.

[0034] The deviation rate is calculated based on the relative relationship between the deviation value and the background baseline to eliminate the influence of differences in gas concentrations. When the background hydrogen concentration is 40 ppm and the measured deviation is 780 ppm, the deviation rate is significantly higher than that of carbon monoxide or other gases, indicating a more significant abnormal hydrogen release. During the calculation, graded analysis standards are set for different gases: a deviation rate below 50% is considered a slight fluctuation, a deviation rate between 50% and 200% is considered a moderate anomaly, and a deviation rate exceeding 200% is considered a severe abnormal release. A comprehensive analysis of the anomaly levels of each gas is performed. When hydrogen and carbon monoxide are at moderate to severe anomaly levels, and volatile organic compounds continue to rise, it is determined that the battery may be in the early stages of electrolyte decomposition or thermal runaway. The deviation values, deviation rates, anomaly levels, warehouse numbers, sampling times, and other information for each gas are summarized to generate an abnormal gas analysis report.

[0035] In this embodiment, the adaptive thermal runaway early warning response processing specifically includes: The thermal runaway risk levels include Level 1 risk, Level 2 risk, and Level 3 risk; The Level 1 risk report pushes the charging compartment number, risk probability, and details of gas anomalies to the operation and maintenance management platform. Level 2 risk triggers the automatic current limiting and power reduction operation of the charging cabinet in response to the push of maintenance alarms, and sends location alerts to the handheld terminals of the on-site personnel. Level 3 risk immediately triggers the automatic power-off of the charging cabinet, activates the dry powder spraying pre-compression device inside the cabinet to standby mode, and sends an alarm message to the fire emergency response platform.

[0036] In this embodiment, key indicators from the abnormal gas analysis report are extracted in a structured manner, including hydrogen deviation rate, carbon monoxide deviation rate, volatile organic gas anomaly level, and electrolyte characteristic gas release intensity. These are then normalized in conjunction with the corresponding battery compartment's risk index prediction value (quantitative value of 0-100). During the fusion analysis, the gas anomaly intensity is comprehensively compared with the risk index prediction value. When the risk index prediction value exceeds 85 and the hydrogen deviation rate exceeds 200%, the risk is determined to be in a rapid evolution stage. When the risk index is in the 70-85 range and multiple gas groups show moderate anomalies, it is determined to be a state of continuous risk accumulation. Based on the above fusion results, the thermal runaway risk is divided into a three-level risk system, including Level 1 risk, Level 2 risk, and Level 3 risk. Level 1 risk indicates an abnormal trend but has not reached a critical state; Level 2 risk indicates that thermal runaway has entered the development stage and has the potential to spread; and Level 3 risk indicates that it has approached or reached the critical state of thermal runaway.

[0037] In a Level 1 risk scenario, only the corresponding charging compartment number, risk probability value (60%–75%), and details of gas anomalies, including abnormal increases in hydrogen, carbon monoxide, and VOCs, are pushed to the operation and maintenance management platform for remote monitoring and manual confirmation. In a Level 2 risk scenario, in addition to pushing operation and maintenance alarm information, the charging cabinet is automatically controlled to limit current and reduce power, gradually decreasing the charging current from 15A to 8A–10A. Location warning information, including the specific compartment number, risk level, and real-time temperature rise rate, is sent to the handheld terminal of the on-site personnel for rapid on-site response. In a Level 3 risk scenario, the highest level of safety response mechanism is triggered, immediately cutting off the charging circuit power. The dry powder spray pre-compression device inside the cabinet enters standby or triggered state to suppress potential thermal diffusion reactions. Simultaneously, an alarm message is sent to the fire emergency response platform, including a timestamp, cabinet number, compartment number, risk level, and a summary of gas anomaly data. This multi-level linkage rapid emergency response mechanism minimizes the risk of thermal runaway expansion.

[0038] In this embodiment, a charging thermal runaway monitoring and early warning system for shared electric vehicle batteries is provided, used to execute the charging thermal runaway monitoring and early warning method for shared electric vehicle batteries as described above, including: The data acquisition module is used to continuously collect the surface temperature of the electric vehicle battery being charged at a fixed sampling period based on the temperature sensors of each battery compartment in the charging cabinet, and to construct temperature monitoring curves for different battery compartments. The risk calculation module is used to perform dual-index weighted calculation based on the temperature monitoring curve to obtain the thermal runaway risk index sequence of each battery. The prediction module is used to perform short-term risk index prediction on the thermal runaway risk index sequence to obtain the predicted risk index value; The gas sampling module is used to determine the abnormal warehouse number based on the risk index prediction value; and to activate the multi-component gas detection module to perform real-time gas sampling based on the abnormal warehouse number, and to extract the current measured concentration values ​​of different gas components. The gas analysis module is used to analyze the content of abnormal gas components in the current measured concentration value and output an abnormal gas analysis report; The early warning response module is used to assess the risk of thermal runaway based on the abnormal gas analysis report and the predicted risk index, and obtain the thermal runaway risk level; and to perform adaptive thermal runaway early warning response processing based on the thermal runaway risk level.

[0039] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0040] In the embodiments provided in this application, it should be understood that the disclosed apparatus / device and method can be implemented in other ways. The apparatus / device embodiments described above are merely illustrative, and the division of modules or units is only a logical functional division. In actual implementation, there may be other division methods. Multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, and the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0041] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0042] Therefore, the embodiments should be considered as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the application are intended to be included within the invention.

[0043] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement it. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein are implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A method for monitoring and early warning of charging thermal runaway of a shared electric vehicle battery, characterized in that, Includes the following steps: Step S1: Based on the temperature sensors of each battery compartment in the charging cabinet, the surface temperature of the electric vehicle battery being charged is continuously collected at a fixed sampling period to construct temperature monitoring curves for different battery compartments. Step S2: Based on the temperature monitoring curve, perform a dual-index weighted calculation to obtain the thermal runaway risk index sequence for each battery; Step S3: Perform short-time risk index prediction on the thermal runaway risk index sequence to obtain the predicted risk index value; Step S4: Determine the abnormal position number based on the risk index prediction value; activate the multi-component gas detection module to perform real-time gas sampling according to the abnormal position number, and extract the current measured concentration values ​​of different gas components; Step S5: Analyze the content of abnormal gas components based on the current measured concentration value and output an abnormal gas analysis report; Step S6: Based on the abnormal gas analysis report and risk index prediction value, conduct a thermal runaway risk assessment to obtain the thermal runaway risk level; Adaptive thermal runaway early warning and response processing based on thermal runaway risk level.

2. The method of claim 1, wherein the method further comprises: The specific steps of step S1 are as follows: Temperature sensors in each battery compartment of the charging cabinet continuously collect the surface temperature of the electric vehicle batteries being charged at a fixed sampling period, generating the original temperature sequence of each compartment. Sensor glitch noise is identified in the original temperature time series, and sliding window mean filtering is performed to generate a filtered temperature series. Determine the timestamp for temperature acquisition; The filtered temperature sequence is mapped and labeled based on the timestamp, and discrete fitting is performed to construct temperature monitoring curves for different battery compartments.

3. The method of claim 2, wherein the method further comprises: The specific steps of step S2 are as follows: Set a temperature rise calculation window; divide the temperature monitoring curve based on the temperature rise calculation window and extract multiple temperature monitoring windows; The temperature rise characteristics of each battery are calculated based on multiple temperature monitoring windows to obtain the temperature rise amplitude and temperature rise rate of each battery. Based on the temperature rise amplitude and temperature rise rate, a dual-index weighted calculation is performed to obtain the real-time thermal runaway risk index for each battery. Extract the charging compartment numbers of all batteries, encapsulate and record the real-time thermal runaway risk index based on the charging compartment numbers, and construct a thermal runaway risk index sequence for each battery.

4. The method of claim 3, wherein the method further comprises: The weighted calculation of the two indicators is as follows: the calculation weight of the temperature rise amplitude is 0.4, and the calculation weight of the temperature rise rate is 0.

6.

5. The method for monitoring and early warning of thermal runaway during charging of shared electric vehicle batteries according to claim 3, characterized in that, Step S3 is as follows: Define three time analysis periods; Based on the three time analysis periods, the thermal runaway risk index sequence is subjected to multi-time-scale trend calculation to extract the risk index change trend at multiple time scales. Based on the trend of the risk index change, a short-term risk index prediction is made to obtain the predicted risk index value.

6. The method for monitoring and early warning of thermal runaway during charging of shared electric vehicle batteries according to claim 5, characterized in that, The trend of the risk index includes the risk growth rate, mean, and standard deviation of volatility.

7. The method for monitoring and early warning of thermal runaway during charging of shared electric vehicle batteries according to claim 5, characterized in that, The specific steps of step S4 are as follows: The risk index prediction value is compared and analyzed based on a preset safety threshold. When the risk index prediction value of the battery in the charging cabinet is not less than the preset safety threshold, the abnormal storage location number is extracted. Based on the abnormal compartment number, the corresponding compartment of the charging cabinet is activated to perform real-time gas sampling and extract the compartment gas signal of the abnormal battery. The gas signal in the storage compartment is quantitatively detected and the concentration of each component is calculated to extract the current measured concentration value of different gas components.

8. The method for monitoring and early warning of thermal runaway during charging of shared electric vehicle batteries according to claim 7, characterized in that, The specific steps of step S5 are as follows: Extract the baseline of gas background concentration under normal charging conditions of the charging cabinet; The current measured concentration value is compared with the gas background concentration baseline item by item to calculate the deviation value of different gas components; Based on the deviation value, the deviation rate is calculated, and the content of abnormal gas components is analyzed, and an abnormal gas analysis report is output.

9. The method for monitoring and early warning of thermal runaway during charging of shared electric vehicle batteries according to claim 8, characterized in that, The adaptive thermal runaway early warning response processing is specifically as follows: The thermal runaway risk levels include Level 1 risk, Level 2 risk, and Level 3 risk; At the first-level risk level, the charging compartment number, risk probability, and details of gas anomalies are pushed to the operation and maintenance management platform. Level 2 risk triggers the automatic current limiting and power reduction operation of the charging cabinet in response to the push of maintenance alarms, and sends location alerts to the handheld terminals of the on-site personnel. Level 3 risk immediately triggers the automatic power-off of the charging cabinet, activates the dry powder spraying pre-compression device inside the cabinet to standby mode, and sends an alarm message to the fire emergency response platform.

10. A charging thermal runaway monitoring and early warning system for shared electric vehicle batteries, characterized in that, The method for monitoring and warning of thermal runaway during charging of a shared electric vehicle battery as described in claim 1 includes: The data acquisition module is used to continuously collect the surface temperature of the electric vehicle battery being charged at a fixed sampling period based on the temperature sensors of each battery compartment in the charging cabinet, and to construct temperature monitoring curves for different battery compartments. The risk calculation module is used to perform dual-index weighted calculation based on the temperature monitoring curve to obtain the thermal runaway risk index sequence of each battery. The prediction module is used to perform short-term risk index prediction on the thermal runaway risk index sequence to obtain the predicted risk index value; The gas sampling module is used to determine the abnormal warehouse number based on the risk index prediction value; and to activate the multi-component gas detection module to perform real-time gas sampling according to the abnormal warehouse number, and extract the current measured concentration values ​​of different gas components. The gas analysis module is used to analyze the content of abnormal gas components in the current measured concentration value and output an abnormal gas analysis report; The early warning response module is used to assess the risk of thermal runaway based on the abnormal gas analysis report and the predicted risk index, and obtain the thermal runaway risk level; and to perform adaptive thermal runaway early warning response processing based on the thermal runaway risk level.

Images