High-rate lithium battery flash charging process thermal runaway early warning and current limiting control method

Abstract

The application discloses a high-rate lithium battery flash charging process thermal runaway early warning and current limiting control method and relates to the technical field of charge control. By executing initial current limiting after identifying thermal risk precursors, and combining risk response results before current limiting with state back-off results after current limiting, it is determined whether parameter back-off is caused by current limiting. The method avoids determining that the risk is removed only according to temperature, temperature rise rate or voltage pressure difference drop. When there is a back-off, the method further detects charging action to obtain response changes, forms a hidden risk judgment result, and indirectly verifies whether lithium precipitation, gas production, side reaction or micro-short circuit precursors still exist in the battery cell. According to the result, the method determines the strategy of maintaining current limiting, suspending charging or grading current recovery, thereby reducing the possibility of thermal risk rebound after recovering a large current.

Description

Technical Field

[0001] This invention relates to the field of charging control technology, specifically to a method for early warning and current limiting control of thermal runaway during high-rate lithium battery flash charging. Background Technology

[0002] Lithium-ion batteries are widely used in electric vehicles, energy storage devices, and high-power electrical equipment. As users' demands for charging efficiency continue to increase, high-rate flash charging is gradually becoming an important way to improve ease of use. Compared with ordinary charging, high-rate flash charging inputs a larger current to the battery in a shorter period of time, making the ion migration, electrode reactions, and heat generation processes inside the battery more intense. If the heat during the charging process cannot be dissipated in time, or if the battery cell itself is under unfavorable conditions such as low temperature, high charge state, or inconsistent aging, abnormal phenomena such as excessively rapid temperature rise, increased polarization, and widening of the single-cell voltage difference may occur, further increasing the risk of thermal runaway. To mitigate the aforementioned risks, existing technologies typically incorporate thermal runaway early warning and current limiting control mechanisms into the battery management system. This involves real-time monitoring of operating parameters during flash charging, such as cell temperature, voltage, current, temperature rise rate, individual cell voltage difference, and state of charge. Based on preset thresholds or risk assessment models, the system determines if any battery anomalies exist. When excessively rapid temperature rise, abnormal voltage difference, or increased thermal risk is detected, the system reduces the charging current, slows the current ramp-up, or enhances cooling to reduce heat generation and reaction load. Once the externally measurable parameters return to a safe range, the risk is assessed as mitigated, and charging is allowed to continue or gradually resume at a higher rate, thus achieving safety protection during the flash charging process.

[0003] However, the above methods mainly rely on changes in external parameters such as temperature, temperature rise rate, and voltage difference to determine whether the risk is under control. The basis for this judgment is still "current limiting if parameters are abnormal, and risk mitigation if parameters fall back." In special cases such as low-temperature flash charging, high-charge flash charging, inconsistent cell aging, or battery packs using strong cooling packaging, current limiting may only temporarily reduce surface temperature, temperature rise rate, or voltage difference. This does not mean that lithium plating, gas generation, side reactions, or micro-short circuit precursors inside the cell have been eliminated simultaneously. In other words, the system sees that the external signal has become stable, but the internal risk may still be in a slow accumulation state. If the existing control method directly restores a large charging current after the external parameters fall back, it is easy to misjudge the temporary suppression of abnormal signals by current limiting and cooling as the actual dissipation of thermal runaway risk. This could lead to a secondary increase in thermal risk during subsequent flash charging, or even a risk rebound, affecting the reliability of high-rate flash charging safety control. Summary of the Invention

[0004] The purpose of this invention is to solve the problem mentioned in the background art that the existing control method directly restores a large charging current after the external parameters fall back, which causes a secondary increase in thermal risk during the subsequent flash charging process, or even leads to a rebound in risk. The invention proposes a method for thermal runaway early warning and current limiting control during the flash charging process of high-rate lithium batteries.

[0005] In terms of implementation, this invention provides a method for early warning and current limiting control of thermal runaway during high-rate lithium battery flash charging, the method comprising: S1: Acquire the operating status of the target lithium battery during high-rate flash charging, and perform initial current limiting when thermal risk precursors are identified to obtain the risk response results before current limiting; S2: Obtain the state fallback result after the initial rate limiting, and based on the risk response result before rate limiting and the state fallback result, determine whether the state fallback result belongs to the occlusion fallback caused by the initial rate limiting, and obtain the fallback attribute result; the fallback attribute result includes occlusion fallback and non-occlusion fallback; S3: When the fallback attribute result is characterized as a shadow fallback, a detection charging action is applied to the target lithium battery, and the hidden risk judgment result is obtained based on the response change under the detection charging action. S4: Determine the subsequent charging strategy based on the results of the hidden risk assessment; if hidden risks still exist, maintain current limiting or suspend charging; if hidden risks are mitigated, restore charging current in stages, and continuously verify the risk rebound status during the restoration process.

[0006] Optionally, S1: The steps for acquiring the operating status of the target lithium battery during high-rate flash charging, performing initial current limiting when thermal risk precursors are identified, and obtaining the risk response results before current limiting are as follows: After the target lithium battery enters the high-rate flash charging mode, the single cell voltage, charging current, single cell surface temperature, inter-cell temperature difference, state of charge and cooling intensity of the target lithium battery are collected according to the preset sampling period to obtain the flash charging operation data sequence. The flash charging operation data sequence is divided into continuous windows to obtain multiple flash charging status windows, and the temperature rise rate, individual cell pressure difference, temperature difference expansion and voltage offset are determined based on each flash charging status window. The temperature rise rate, single-cell pressure difference, temperature difference expansion and voltage deviation are compared with the corresponding preset safety limits. When at least two of the parameters exceed the corresponding preset safety limits in the same flash charging state window, it is determined that the target lithium battery has a thermal risk precursor. When a thermal risk precursor is detected, the end time of the corresponding flash charging state window is determined as the initial current limiting trigger time, and an initial current limiting command is sent to the charging control unit according to the initial current limiting trigger time, so that the charging current of the target lithium battery is reduced to the first current limiting current. Multiple consecutive flash charging status windows before the initial current limiting trigger time are extracted as abnormal sections before current limiting. The changes in temperature rise rate, single-cell pressure difference, temperature difference expansion, voltage deviation, and charging current within the abnormal sections before current limiting are extracted in chronological order to obtain the risk response results before current limiting.

[0007] Optionally, S2: The steps for obtaining the state fallback result after the initial rate limiting, and determining whether the state fallback result belongs to the occlusion fallback caused by the initial rate limiting based on the risk response result before rate limiting and the state fallback result, to obtain the fallback attribute result are as follows: After the initial current limiting is implemented, the operating data of the target lithium battery is collected again at the same sampling period as before the initial current limiting, and the current limiting fallback response result is generated according to the same window division method as the risk response result before the current limiting. The temperature rise rate, individual unit pressure difference, temperature difference expansion, and voltage deviation in the fallback response results after current limiting are compared with the corresponding parameters before the initial current limiting. If at least two of the temperature rise rate, individual unit pressure difference, temperature difference expansion, and voltage deviation are lower than the corresponding parameters before the initial current limiting after the initial current limiting, then there is an explicit risk of fallback after the initial current limiting. When there is a decline in the apparent risk, the time period corresponding to the decline in apparent risk is matched with the time period of the decrease in charging current and the time period of the increase in cooling execution intensity to determine whether the decline in apparent risk occurs synchronously with the decrease in charging current or the increase in cooling execution intensity. If the apparent risk decline occurs simultaneously with the decrease in charging current or the increase in cooling execution intensity, and does not continue to decline after the charging current and cooling execution intensity remain stable, then the decline attribute result will be determined as masking decline. If the apparent risk falls back even after the charging current and cooling execution intensity remain stable, the fallback attribute result will be determined as unmasked fallback.

[0008] Optionally, a detection charging action is applied to the target lithium battery, and the result of the hidden risk assessment is obtained based on the response change of the detection charging action; When the fallback attribute result is characterized as a shadow fallback, the target lithium battery is kept under the first current limit for charging, and the charging current, state of charge, cell surface temperature and cell voltage at the current sampling time are obtained to obtain the reference state before detection. Based on the baseline state before detection, the charging current is increased from the first current limiting current to the detection current. The detection current is greater than the first current limiting current and less than the flash charging current before the initial current limiting is triggered. The target lithium battery is continuously charged under the detection current for a preset detection time, and the single cell voltage, single cell surface temperature, inter-cell temperature difference and charging current of the target lithium battery are continuously collected within the preset detection time to obtain the detection process response data. After the preset detection time is reached, the charging current is restored from the detection current to the first current limiting current, and the individual cell voltage, individual cell surface temperature and inter-cell temperature difference are collected again within the preset fallback time to obtain the fallback data after detection. The baseline state before detection, the response data during detection, and the fallback data after detection are bound in chronological order to form response data corresponding to the detection charging action. The response data is then analyzed and the residual response value and the rise and fall current response asymmetry value are calculated. The residual response value and the rise and fall current response asymmetry value are added together to obtain the hidden risk value. The hidden risk value is compared with the preset threshold, and the existence of hidden risk is determined based on the comparison result.

[0009] Optionally, the calculation steps for the residual response value are as follows: Extract residual judgment parameters from the response data corresponding to the detection charging action. For any residual judgment parameter, determine the baseline value of the residual judgment parameter in the baseline state before detection, the deviation peak value in the response data during the detection process, and the fallback end value at the end of the fallback data after detection. The deviation peak value is compared with the benchmark value, and the abnormal deviation direction of the residual judgment parameter is determined based on the comparison result. When the deviation peak value is greater than the benchmark value, the abnormal deviation direction is determined to be positive deviation. When the deviation peak value is less than the benchmark value, the abnormal deviation direction is determined to be negative deviation. When the deviation peak value is equal to the benchmark value, the residual judgment parameter is determined to be in a state of no deviation. Based on the direction of abnormal deviation, the final value of the pullback is compared with the benchmark value for residual value in the same direction. When the abnormal deviation is positive, the portion of the final value exceeding the benchmark value is taken as the residual value in the same direction; if the final value does not exceed the benchmark value, the residual value in the same direction is set to zero. When the abnormal deviation is negative, the portion of the benchmark value exceeding the final value of the pullback is taken as the residual value in the same direction; if the final value is not lower than the benchmark value, the residual value in the same direction is set to zero. When the residual determination parameter is in a state of no deviation, the residual value in the same direction is set to zero. The residual retention ratio of the residual judgment parameter is calculated based on the absolute value of the difference between the deviation peak and the benchmark value and the residual amount in the same direction. Calculate the retention ratio of all residue determination parameters separately, and subtract each retention ratio from a given value to obtain the degree of non-residue corresponding to each residue determination parameter; The overall non-residue level is obtained by continuously multiplying the non-residue levels corresponding to all residue determination parameters. Subtract the overall non-residual level from 1 to obtain the residual response value.

[0010] Optionally, the calculation steps for the asymmetric value of the rise and fall flow response are as follows: The current rise / fall determination parameters are extracted from the response data corresponding to the detection charging action. The current rise / fall determination parameters include the temperature rise rate, the individual cell pressure difference and the inter-cell temperature difference. The data corresponding to the application of the detection current is determined as the current rise response segment, and the data after the detection current is restored to the first current limiting current is determined as the current fall back segment. For any rise / fall rate determination parameter, determine the reference value of the rise / fall rate determination parameter under the reference state before detection, and determine the maximum deviation value of the rise / fall rate determination parameter in the rise response segment; The maximum deviation value is compared with the reference value, and the abnormal propulsion direction of the rise and fall flow determination parameter is determined based on the comparison result. When the maximum deviation value is greater than the reference value, the abnormal propulsion direction is determined to be positive propulsion. When the maximum deviation value is less than the reference value, the abnormal propulsion direction is determined to be negative propulsion. When the maximum deviation value is equal to the reference value, the rise and fall flow determination parameter is determined to be in a no-propulsion state. Based on the abnormal propagation direction, the parameter values ​​at two adjacent sampling times within the upflow response segment are compared one by one. When the abnormal propagation direction is positive, the increase in parameter value at the later sampling time compared to the previous sampling time is taken as the abnormal propagation component, and all abnormal propagation components are accumulated to obtain the abnormal propagation amount. When the abnormal propagation direction is negative, the decrease in parameter value at the previous sampling time compared to the later sampling time is taken as the abnormal propagation component, and all abnormal propagation components are accumulated to obtain the abnormal propagation amount. When the upflow / downflow determination parameter is in a no-propagation state, the abnormal propagation amount is determined to be zero. Calculate the asymmetric value of the rise and fall flow response based on the abnormal propulsion direction.

[0011] Optionally, the steps for calculating the asymmetric value of the rise and fall flow response based on the abnormal propulsion direction are as follows: Based on the abnormal advance direction, the parameter values ​​at two adjacent sampling times within the downcurrent backoff segment are compared one by one. When the abnormal advance direction is positive, the portion of the parameter value at the previous sampling time that is greater than the decrease in the parameter value at the next sampling time is taken as the effective backoff component, and all effective backoff components are accumulated to obtain the effective backoff amount. When the abnormal advance direction is negative, the portion of the parameter value at the next sampling time that is greater than the increase in the parameter value at the previous sampling time is taken as the effective backoff component, and all effective backoff components are accumulated to obtain the effective backoff amount. When the upcurrent / downcurrent determination parameter is in a no-advance state, the effective backoff amount is determined to be zero. Based on the abnormal advance and effective retreat, the closure gap ratio of the rise and fall flow determination parameters is calculated. Specifically, when the sum of the abnormal advance and effective retreat is not zero, the abnormal advance is first subtracted from the effective retreat. If the difference is greater than zero, the difference is used as the non-retreat amount; if the difference is not greater than zero, the non-retreat amount is determined to be zero. Then, the non-retreat amount is divided by the sum of the abnormal advance and effective retreat to obtain the closure gap ratio. When the sum of the abnormal advance and effective retreat is zero, the closure gap ratio is determined to be zero. Calculate the closed gap ratio of all rise and fall flow determination parameters respectively, and add each closed gap ratio to one to obtain the closed gap enhancement term corresponding to each rise and fall flow determination parameter; The closed gap enhancement terms corresponding to all rise and fall flow determination parameters are multiplied continuously to obtain the comprehensive closed gap enhancement value; Subtract one from the comprehensive gap closure enhancement value to obtain the normalized numerator, and add one to the comprehensive gap closure enhancement value to obtain the normalized denominator; Dividing the normalized numerator by the normalized denominator yields the asymmetric values ​​of the rise and fall flow responses.

[0012] Optionally, the steps for determining whether there are hidden risks based on the comparison results are as follows: By comparing the hidden risk value with the preset threshold, if the hidden risk value is not less than the preset threshold, it indicates that there is still a hidden thermal risk inside the corresponding target lithium battery.

[0013] Optionally, the steps for determining the subsequent charging strategy based on the results of the implicit risk assessment are as follows: When the latent risk assessment results indicate that there is still a latent thermal risk, the target lithium battery is controlled to maintain the target current-limiting charge, and the temperature rise rate, single-cell voltage difference, temperature difference expansion and voltage deviation are continuously monitored. If at least two of the following parameters—temperature rise rate, single-cell pressure difference, temperature difference expansion, and voltage deviation—fail to meet the preset recovery conditions within the preset monitoring period, then the charging of the target lithium battery will be suspended. When the latent risk assessment result indicates that the latent thermal risk has been mitigated, the charging current is gradually increased according to the preset recovery step size, starting from the first current limiting current, and the preset duration is maintained at each recovery current level. At each recovery current level, the operating data of the target lithium battery is collected, and it is determined whether there is a risk of rebound in the temperature rise rate, single-cell voltage difference, temperature difference expansion, and voltage deviation. If a risk rebound occurs, the charging current will be reduced back to the previous stage recovery current; If no risk rebound occurs, continue increasing the recovery current to the next level until the target flash charging current is reached. The beneficial effects of this invention are: This invention proposes a method for early warning and current limiting control of thermal runaway during high-rate lithium battery flash charging. After identifying early signs of thermal risk, an initial current limit is executed, and the risk response before and the state decline after the initial current limit are obtained. The method determines whether the state decline after the current limit is a shadowing decline caused by the current limiting action. This prevents the system from simply relying on the decline of external parameters such as temperature, temperature rise rate, or voltage difference to directly determine that the risk has been eliminated. Instead, it first distinguishes whether the decline in external parameters corresponds to a genuine risk mitigation or temporary suppression by current limiting. When a shadowing decline is detected, the charging action is detected to obtain response changes and form a hidden risk judgment result, thereby controlling the cell's performance. The system indirectly verifies potential internal lithium plating, gas generation, side reactions, or micro-short circuit precursors. Based on the assessment of hidden risks, it determines subsequent charging strategies such as maintaining current limiting, pausing charging, or gradually restoring the charging current. During the recovery process, it continuously verifies the risk rebound status. This avoids misjudging "temporary suppression of external parameters by current limiting and cooling" as "real dissipation of thermal runaway risk" under special circumstances such as low-temperature flash charging, high-charge flash charging, inconsistent cell aging, or strong cooling packaging. This reduces the possibility of a secondary increase or rebound in thermal risk after restoring a large charging current, and improves the reliability and safety of thermal runaway warning and current limiting control during high-rate flash charging. Attached Figure Description

[0014] Figure 1 The flowchart illustrates the method for early warning and current limiting control of thermal runaway during high-rate lithium battery flash charging, as provided in this embodiment of the invention. Detailed Implementation

[0015] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0016] The present invention provides a method for early warning and current limiting control of thermal runaway during high-rate lithium battery flash charging. See also: Figure 1 , Figure 1 A flowchart illustrating a method for early warning and current limiting control of thermal runaway during high-rate lithium battery flash charging, provided in an embodiment of the present invention. The method includes the following steps: S1: Acquire the operating status of the target lithium battery during high-rate flash charging, and perform initial current limiting when thermal risk precursors are identified to obtain the risk response results before current limiting; S2: Obtain the state fallback result after the initial rate limiting, and based on the risk response result before rate limiting and the state fallback result, determine whether the state fallback result belongs to the occlusion fallback caused by the initial rate limiting, and obtain the fallback attribute result; the fallback attribute result includes occlusion fallback and non-occlusion fallback; S3: When the fallback attribute result is characterized as a shadow fallback, a detection charging action is applied to the target lithium battery, and the hidden risk judgment result is obtained based on the response change under the detection charging action. S4: Determine the subsequent charging strategy based on the results of the hidden risk assessment; if hidden risks still exist, maintain current limiting or suspend charging; if hidden risks are mitigated, restore charging current in stages, and continuously verify the risk rebound status during the restoration process.

[0017] Based on the high-rate lithium battery flash charging process thermal runaway early warning and current limiting control method provided in this invention embodiment, after identifying the precursory thermal risk, an initial current limiting is performed, and the risk response result before current limiting and the state fallback result after the initial current limiting are obtained. The system then determines whether the state fallback after current limiting is a shadow fallback caused by the current limiting action. This prevents the system from simply relying on the fallback of external parameters such as temperature, temperature rise rate, or voltage difference to directly determine that the risk has been eliminated. Instead, it first distinguishes whether the fallback of external parameters corresponds to a genuine risk mitigation or a temporary suppression by current limiting. When a shadow fallback is determined to exist, the system further detects the charging action to obtain response changes and forms a hidden risk judgment result, thereby controlling the risk. Indirect verification is performed on potential lithium plating, gas generation, side reactions, or micro-short circuit precursors that may still exist inside the battery cell. Based on the results of the implicit risk assessment, subsequent charging strategies such as maintaining current limiting, pausing charging, or restoring charging current in stages are determined. During the recovery process, the risk rebound status is continuously verified. This can avoid misjudging "external parameters being temporarily suppressed by current limiting and cooling" as "the risk of thermal runaway has truly dissipated" under special circumstances such as low-temperature flash charging, high-charge flash charging, inconsistent cell aging, or strong cooling packaging. This reduces the possibility of a secondary increase or rebound in thermal risk after restoring a large charging current, and improves the reliability and safety of thermal runaway warning and current limiting control during high-rate flash charging.

[0018] In one embodiment, S1: The steps of acquiring the operating status of the target lithium battery during high-rate flash charging, performing initial current limiting when thermal risk precursors are identified, and obtaining the risk response result before current limiting are as follows: After the target lithium battery enters the high-rate flash charging mode, the individual cell voltage, charging current, individual cell surface temperature, inter-cell temperature difference, state of charge, and cooling intensity of the target lithium battery are continuously collected according to the preset sampling period. The above data are arranged into a flash charging operation data sequence according to the sampling time. Among them, the individual cell voltage is used to characterize the voltage response state of each cell during the flash charging process, the charging current is used to characterize the current input rate, the individual cell surface temperature is used to characterize the external thermal response state of each cell, the inter-cell temperature difference is used to characterize whether the internal heat distribution of the battery is balanced, the state of charge is used to characterize the current rechargeability margin of the target lithium battery, and the cooling intensity is used to characterize the degree of heat dissipation intervention of the cooling system on the target lithium battery. The flash charging operation data sequence is divided into continuous windows to obtain multiple flash charging state windows. Each flash charging state window includes the individual cell voltage, charging current, individual cell surface temperature, inter-cell temperature difference, and state of charge within several consecutive sampling periods. For each flash charging state window, the maximum temperature rise rate, maximum individual cell voltage difference, temperature difference expansion, and voltage offset within that window are calculated. The maximum temperature rise rate is the maximum rate of temperature increase of any individual cell within that window relative to the temperature at the previous sampling time; the maximum individual cell voltage difference is the difference between the highest and lowest individual cell voltages within that window; the temperature difference expansion is the difference between the inter-cell temperature difference at the end of the window and the inter-cell temperature difference at the beginning of the window; and the voltage offset is the maximum deviation of each individual cell voltage within that window relative to the average voltage of the individual cells within the same window. The maximum temperature rise rate, maximum single-cell voltage difference, temperature difference expansion, and voltage deviation are compared with their corresponding preset safety limits. If at least two of the following conditions are met simultaneously in any flash-charge state window, the target lithium battery is deemed to have a thermal risk precursor: the maximum temperature rise rate exceeds the preset temperature rise rate limit, the maximum single-cell voltage difference exceeds the preset voltage difference limit, the temperature difference expansion exceeds the preset temperature difference expansion limit, and the voltage deviation exceeds the preset voltage deviation limit; or the maximum temperature rise rate continuously increases under high charge conditions. This method ensures that thermal risk precursors are not determined directly based on a single temperature exceeding the limit, but rather based on a combination of temperature changes, voltage consistency, and the thermal distribution expansion state.

[0019] When it is determined that the target lithium battery has signs of thermal risk, the end time of the corresponding flash charge state window is recorded as the initial current limiting trigger time, and several consecutive flash charge state windows before the initial current limiting trigger time are extracted as the abnormal section before current limiting. The abnormal section before current limiting includes at least the window where the signs of thermal risk first appear and the state windows before and after it, in order to preserve the abnormal formation process of the target lithium battery before it is suppressed by current limiting.

[0020] After determining the initial current limiting trigger time, an initial current limiting command is sent to the charging control unit, reducing the charging current from the current flash charging current to the first current limiting current. The first current limiting current is less than the current flash charging current and not less than the minimum charging current required to maintain continuous battery charging monitoring. In this way, the continued heating intensity of the target lithium battery is reduced, while retaining the charging response conditions required for subsequent observation of the battery state decline process.

[0021] The operational data within the abnormal section before current limiting is processed, and the temperature rise rate change sequence, single cell pressure difference change sequence, temperature difference expansion change sequence, voltage offset change sequence, and corresponding charging current change sequence are extracted. The above change sequences are then bound according to the sampling time to form the risk response result before current limiting. The risk response result before current limiting is used to characterize the process of the apparent risk parameters of the target lithium battery gradually changing with high-rate charging input before the initial current limiting is implemented.

[0022] It should be noted that after the target lithium battery enters the high-rate flash charging mode, the individual cell voltage can be acquired through the voltage acquisition circuit connected to the sampling terminal of each individual cell in the battery management system; the charging current can be acquired through the Hall current sensor, shunt resistor or current sampling module set on the main charging circuit; the individual cell surface temperature can be acquired through the temperature sensor attached to the surface of the individual cell or key position of the module, and the temperature sensor can be a thermistor, thermocouple or digital temperature sensor; the temperature difference between individual cells can be obtained by the battery management system calculating the difference between the highest individual cell surface temperature and the lowest individual cell surface temperature at the same sampling time after acquiring the surface temperature of each individual cell; the state of charge can be estimated by the battery management system based on the integrated result of the charging current, combined with the mapping relationship of individual cell voltage, open circuit voltage and battery capacity parameters; the cooling execution intensity can be obtained by the battery thermal management control unit reading the current operating quantity of the cooling system, which includes at least one of the following: coolant pump speed, cooling medium flow rate, cooling valve opening, fan duty cycle, compressor speed or cooling power, used to characterize the degree of heat dissipation intervention of the cooling system on the target lithium battery at the current sampling time.

[0023] In one embodiment, S2: Obtaining the state fallback result after the initial rate limiting, and determining whether the state fallback result belongs to the occlusion fallback caused by the initial rate limiting based on the risk response result before rate limiting and the state fallback result, and obtaining the fallback attribute result are the following steps: After the initial current limiting is performed, the single cell voltage, charging current, single cell surface temperature, inter-cell temperature difference, state of charge, and cooling intensity of the target lithium battery are collected again according to the same sampling period as before the initial current limiting. The data collected continuously after the initial current limiting trigger time are arranged in the order of sampling time to obtain the operating data after current limiting. Based on the operational data after rate limiting, the data after the initial rate limiting is divided into windows according to the same window length and window movement method as the risk response results before rate limiting are generated, resulting in multiple post-rate limiting status windows. Calculate the temperature rise rate, unit pressure difference, temperature difference expansion, voltage offset, charging current change, and cooling intensity change corresponding to each current-limited state window, and arrange them in the time order of the current-limited state windows to obtain the current-limited fallback response results. Extract the direction of temperature rise rate, individual pressure difference, temperature difference expansion and voltage shift from the risk response results before the initial current limiting; and extract the direction of temperature rise rate, individual pressure difference, temperature difference expansion and voltage shift from the fallback response results after the initial current limiting. The temperature rise rate, single-cell pressure difference, temperature difference expansion, and voltage deviation after the initial current limiting are compared with the corresponding parameters before the initial current limiting to determine whether the apparent risk parameters have fallen back. When at least two of the temperature rise rate, single-cell pressure difference, temperature difference expansion, and voltage deviation are lower than the corresponding parameters before the initial current limiting after the initial current limiting, it is determined that the target lithium battery has an apparent risk decline. After confirming the existence of an external risk decline, the decline period of the external risk parameter is further correlated with the charging current decline period and the cooling execution intensity enhancement period. When the start time of the decline of the external risk parameter is within the preset response time range after the charging current decline period, or within the preset response time range after the cooling execution intensity enhancement period, it is determined that there is a time correlation between the external risk decline and the initial current limiting action or cooling enhancement action. After confirming the existence of a time correlation, the direction of change between the decline of the apparent risk parameters and the decline of the charging current in each state window after current limiting is compared; when the charging current continues to decline or remains at the first current limiting current, and the temperature rise rate, individual cell voltage difference, temperature difference expansion and voltage deviation decrease synchronously, it is determined that the decline of apparent risk is affected by the initial current limiting. Determine whether there is an independent drop-off segment in each state window after current limiting. An independent drop-off segment refers to the segment in a continuous state window where the charging current remains unchanged and the cooling intensity does not continue to increase, the temperature rise rate, individual unit voltage difference, temperature difference expansion, and voltage offset continue to decrease. If there is no independent fallback segment, or the number of continuous windows of the independent fallback segment is less than the preset number, the fallback result is determined to be a shading fallback caused by the initial rate limiting, and the fallback attribute result is determined to be a shading fallback. If there is an independent fallback segment, and the number of continuous windows of the independent fallback segment is not less than the preset number, then the fallback result is determined not to be a shading fallback caused by the initial rate limiting, and the fallback attribute result is determined to be a non-shading fallback.

[0024] It should be noted that the core of using the above method to determine whether the state decline result belongs to the shaded decline caused by the initial current limiting is that it no longer directly assumes that the decrease in the temperature rise rate, single-cell voltage difference, temperature difference expansion, and voltage deviation after current limiting is a true mitigation of thermal risk. Instead, it further distinguishes and judges based on the charging current decline period, the cooling intensity enhancement period, and whether the external risk parameters themselves have an independent and continuous decline. In this way, it is possible to identify whether the decline in external risk parameters is caused by the natural decay of internal risks in the cell or by the external suppression of abnormal performance by the initial current limiting and enhanced cooling. This avoids misjudging the changes in parameters temporarily flattened by current limiting as risk relief in low temperature, high charge state, inconsistent aging, or strong cooling packaging scenarios. At the same time, by setting an independent decline segment, it is required that the battery still shows a continuous decline trend after the current and cooling conditions stabilize before it is considered a non-shading decline. This improves the reliability of the judgment on whether a large charging current will be restored later and reduces the possibility of risk rebound and secondary thermal risk increase.

[0025] In one embodiment, S3: When the fallback attribute result is characterized as a shadow fallback, a probe charging action is applied to the target lithium battery, and the hidden risk judgment result is obtained based on the response change under the probe charging action. When the fallback attribute result is characterized as a shadow fallback, the target lithium battery is kept under the first current limit for charging, and the charging current, state of charge, cell surface temperature and cell voltage at the current sampling time are obtained to obtain the reference state before detection. Based on the baseline state before detection, the charging current is increased from the first current limiting current to the detection current. The detection current is greater than the first current limiting current and less than the flash charging current before the initial current limiting is triggered. The target lithium battery is continuously charged under the detection current for a preset detection time, and the single cell voltage, single cell surface temperature, inter-cell temperature difference and charging current of the target lithium battery are continuously collected within the preset detection time to obtain the detection process response data. After the preset detection time is reached, the charging current is restored from the detection current to the first current limiting current, and the individual cell voltage, individual cell surface temperature and inter-cell temperature difference are collected again within the preset fallback time to obtain the fallback data after detection. The baseline state before detection, the response data during detection, and the fallback data after detection are bound in chronological order to form response data corresponding to the detection charging action. The response data is then analyzed and the residual response value and the rise and fall current response asymmetry value are calculated. The residual response value and the rise and fall current response asymmetry value are added together to obtain the hidden risk value. The hidden risk value is compared with the preset threshold, and the existence of hidden risk is determined based on the comparison result.

[0026] It should be noted that after determining that "the state decline after current limiting may be a shadowing decline," high-rate flash charging is not immediately resumed. Instead, a controlled-amplitude and time-controlled probe charging action is used to test whether there are still hidden risks within the battery cell that were masked by current limiting. Specifically, the system first keeps the target lithium battery charging under the first current limiting current and records the charging current, state of charge, cell surface temperature, and cell voltage at this time as a baseline state for subsequent comparison before the probe, which is equivalent to determining "what level the battery was at before the probe began." Then, without restoring the original high-rate flash charging current, the charging current is slightly increased from the first current limiting current to the probe current, so that the battery is subjected to a controllable current stimulus. The probe current is higher than the current limiting current but lower than the flash charging current before the initial current limiting. The purpose is to slightly remove the current limiting shadow, rather than re-applying a high-risk shock. During the preset detection period of continuous current measurement, the system continuously collects cell voltage, cell surface temperature, inter-cell temperature difference, and charging current to obtain response data during the detection process. This data is used to observe whether the battery exhibits abnormal sensitive reactions after a small increase in current, such as a sudden increase in temperature, an expansion of the cell voltage difference, or a significant shift in the voltage of a particular cell. When the detection period ends, the system immediately restores the charging current to the first current limit and continues to collect cell voltage, cell surface temperature, and inter-cell temperature difference within the preset fallback period to obtain fallback data after detection. This data is used to observe whether these responses can return to near the level before detection after the detection current is removed. For example, if a battery was originally being flash-charged at four times the current rate, but was then current-limited to one time the current rate after triggering a thermal risk, the detection current can be set to 1.5 times or 2 times the current rate, and then returned to one time the current rate after a short period of time. If the temperature, voltage, and differential pressure only change slightly during this short-term detection, and can fall back smoothly after the detection current is withdrawn, it indicates that the internal risk may have been mitigated. Conversely, if a slight increase in current causes a significant response, or if the abnormality remains after the current is reduced back to the current-limiting current, it indicates that the previous stable parameters may have only been caused by the current-limiting shielding, and there may still be lithium plating, side reactions, or local abnormal precursors inside the cell.

[0027] In one implementation, the steps for calculating the residual response value are as follows: Residual determination parameters are extracted from the response data corresponding to the detection charging action. The residual determination parameters include the temperature rise rate, the individual cell pressure difference, and the inter-cell temperature difference. For any residual judgment parameter, determine the baseline value of the residual judgment parameter under the reference state before detection, the peak deviation in the response data during the detection process, and the final value of the fallback data after detection. The deviation peak value is compared with the benchmark value, and the abnormal deviation direction of the residual judgment parameter is determined according to the comparison result. When the deviation peak value is greater than the benchmark value, the abnormal deviation direction is determined to be positive deviation. When the deviation peak value is less than the benchmark value, the abnormal deviation direction is determined to be negative deviation. When the deviation peak value is equal to the benchmark value, the residual judgment parameter is determined to be in a state of no deviation. Based on the direction of the abnormal deviation, the final value of the pullback is compared with the baseline value for residual values ​​in the same direction. When the abnormal deviation is positive, the portion of the final value exceeding the baseline value is taken as the residual amount in the same direction; if the final value does not exceed the baseline value, the residual amount in the same direction is set to zero. When the abnormal deviation is negative, the portion of the baseline value exceeding the final value of the pullback is taken as the residual amount in the same direction; if the final value is not lower than the baseline value, the residual amount in the same direction is set to zero. When the residual determination parameter is in a state of no deviation, the residual amount in the same direction is set to zero. The residue retention ratio of this residue determination parameter is calculated based on the absolute value of the difference between the peak deviation and the benchmark value, and the amount of residue in the same direction. The residue retention ratio is the amount of residue in the same direction divided by the sum of the absolute value of the difference and the amount of residue in the same direction. When the sum of the absolute value of the difference and the amount of residue in the same direction is zero, the residue retention ratio is set to zero. The specific formula is as follows: In the formula, This represents the residual retention ratio of the j-th residual determination parameter. , , These represent the deviation peak value, baseline value, and same-direction residual amount of the j-th residual determination parameter, respectively. Calculate the retention ratio of all residue determination parameters separately, and subtract each retention ratio from a given value to obtain the degree of non-residue corresponding to each residue determination parameter; The overall non-residue level is obtained by continuously multiplying the non-residue levels corresponding to all residue determination parameters. Subtracting the overall non-residual degree from 1 yields the residual response value; the residual response value ranges from zero to one, and the larger the residual response value, the more obvious the response remains along the abnormal deviation direction after the probe current is withdrawn.

[0028] It should be noted that the residual response value is a quantitative indicator used to characterize whether the response of the target lithium battery, such as the rate of temperature rise, the differential pressure between individual cells, and the temperature difference between individual cells, remains in the direction of the abnormal deviation formed during the detection process after the detection charging action ends. In essence, it measures whether the risk traces have not dissipated after the detection current is withdrawn. Under normal circumstances, the detection charging action is only a short-term, small-amplitude current increase. When the detection current returns to the first current-limiting current, the increase in the rate of temperature rise, the expansion of the differential pressure between individual cells, or the increase in the temperature difference between individual cells caused by the detection current should fall back and gradually approach the baseline state before detection. At this time, the residual response value is small, indicating that the aforementioned changes are mainly reversible current responses. Conversely, if the rate of temperature rise is still higher than the level before detection, the differential pressure between individual cells continues to expand, and the temperature difference between individual cells has not converged after the detection current is withdrawn, it indicates that the abnormal response triggered by the detection action has not disappeared synchronously with the current withdrawal, but has left a continuous response trace inside the cell. This trace often means that there may be hidden risks inside the battery, such as lithium plating, continuous exothermic side reactions, abnormal local impedance, micro-short circuit precursors, or uneven heat distribution. A larger residual response value indicates that at least one or more of the residual judgment parameters maintain a high residual value along the abnormal deviation direction after detection. For example, if the temperature difference between cells was 2 degrees Celsius before detection, increased to 5 degrees Celsius during detection, and remained around 4 degrees Celsius after detection, it indicates that the temperature difference did not effectively recover after the detection current was withdrawn. Similarly, if the cell pressure difference was increased during detection and remained higher than the pre-detection level after the detection current was withdrawn, it indicates that the electrochemical response of some cells has continuously deviated from that of other cells. Therefore, a larger residual response value suggests that the anomaly of the target lithium battery is not simply caused by the instantaneous stimulation of the detection current, but may be the result of internal risks being triggered and then retained or extended. When this value is used as a component of the latent risk value, a larger value naturally increases the latent risk value and indicates a higher probability that latent thermal risks still exist inside the target lithium battery.

[0029] The advantage of calculating the residual response value in the above way is that it does not simply determine whether the temperature, pressure difference or temperature difference exceeds the threshold after detection. Instead, it first uses the baseline state before detection as a reference, then determines the abnormal deviation direction of each parameter during the detection process, and finally extracts only the part that is still retained along the abnormal direction after the detection current is withdrawn as the residue. Therefore, it can more accurately distinguish between "normal reversible fluctuations caused by detection action" and "abnormal traces that have not dissipated after the detection is completed". If only the absolute value of the parameter after detection or the difference before and after detection is used for judgment, it is easily affected by the initial temperature of the battery, the ambient temperature, the state of charge (SOC), the consistency of the cells, and the cooling conditions. For example, some batteries may have a high temperature but return to normal, or some batteries may have a low value after detection but still retain obvious abnormalities relative to their own benchmark, which can lead to misjudgment. The above method is based on a chain calculation of "deviation direction - same direction residue - residue retention ratio - comprehensive non-residue degree", which can extract the response traces that are really stuck in the abnormal direction. By continuously multiplying, the obvious residue of any key parameter can be reflected in the final value, which is more suitable for judging whether the internal risk of the target lithium battery has really dissipated after the detection current is withdrawn, thus improving the reliability of hidden thermal risk identification.

[0030] In one implementation, the steps for calculating the asymmetric value of the rise and fall flow response are as follows: The current rise / fall determination parameters are extracted from the response data corresponding to the detection charging action. The current rise / fall determination parameters include the temperature rise rate, the individual cell pressure difference and the inter-cell temperature difference. The data corresponding to the application of the detection current is determined as the current rise response segment, and the data after the detection current is restored to the first current limiting current is determined as the current fall back segment. For any rise / fall rate determination parameter, determine the reference value of the rise / fall rate determination parameter under the reference state before detection, and determine the maximum deviation value of the rise / fall rate determination parameter in the rise response segment; The maximum deviation value is compared with the reference value, and the abnormal propulsion direction of the rise / fall flow determination parameter is determined based on the comparison result. When the maximum deviation value is greater than the reference value, the abnormal propulsion direction is determined to be positive; when the maximum deviation value is less than the reference value, the abnormal propulsion direction is determined to be negative; when the maximum deviation value is equal to the reference value, the rise / fall flow determination parameter is determined to be in a no-propulsion state. Based on the abnormal propagation direction, the parameter values ​​at two adjacent sampling times within the upflow response segment are compared one by one. When the abnormal propagation direction is positive, the increase in parameter value at the later sampling time compared to the previous sampling time is taken as the abnormal propagation component, and all abnormal propagation components are accumulated to obtain the abnormal propagation amount. When the abnormal propagation direction is negative, the decrease in parameter value at the previous sampling time compared to the later sampling time is taken as the abnormal propagation component, and all abnormal propagation components are accumulated to obtain the abnormal propagation amount. When the upflow / downflow determination parameter is in a no-propagation state, the abnormal propagation amount is determined to be zero. Based on the abnormal propagation direction, the parameter values ​​at two adjacent sampling times within the downcurrent retreat segment are compared one by one. When the abnormal propagation direction is positive, the portion of the parameter value at the previous sampling time that is greater than the decrease in the parameter value at the next sampling time is taken as the effective retreat component, and all effective retreat components are accumulated to obtain the effective retreat amount. When the abnormal propagation direction is negative, the portion of the parameter value at the next sampling time that is greater than the increase in the parameter value at the previous sampling time is taken as the effective retreat component, and all effective retreat components are accumulated to obtain the effective retreat amount. When the upcurrent / downcurrent determination parameter is in a no-propagation state, the effective retreat amount is determined to be zero. Based on the abnormal advance and effective retreat, the closure gap ratio of the rise / fall determination parameter is calculated. Specifically, when the sum of the abnormal advance and effective retreat is not zero, the abnormal advance is first subtracted from the effective retreat. If the difference is greater than zero, this difference is used as the non-retreat amount; if the difference is not greater than zero, the non-retreat amount is determined to be zero. Then, the non-retreat amount is divided by the sum of the abnormal advance and effective retreat to obtain the closure gap ratio. When the sum of the abnormal advance and effective retreat is zero, the closure gap ratio is determined to be zero. Calculate the closed gap ratio of all rise and fall flow determination parameters respectively, and add each closed gap ratio to one to obtain the closed gap enhancement term corresponding to each rise and fall flow determination parameter; The closed gap enhancement terms corresponding to all rise and fall flow determination parameters are multiplied continuously to obtain the comprehensive closed gap enhancement value; Subtract one from the comprehensive gap closure enhancement value to obtain the normalized numerator, and add one to the comprehensive gap closure enhancement value to obtain the normalized denominator; Dividing the normalized numerator by the normalized denominator yields the rise and fall current response asymmetry value. The rise and fall current response asymmetry value ranges from zero to one, and the larger the rise and fall current response asymmetry value, the more difficult it is for the abnormal advance formed during the rise of the probe current to be retracted after the probe current is withdrawn.

[0031] It should be noted that the rise and fall current response asymmetry value is a quantitative indicator used to measure the reversibility of the response of a target lithium battery between the "rise phase" and the "retreat phase" of the probe current. Its essence is not to see whether the temperature, pressure difference or temperature difference exceeds the limit at a certain moment, but to see whether the response amount pushed in an abnormal direction when the probe current rises can be pulled back accordingly after the probe current is withdrawn. Under normal circumstances, the probe current only increases briefly from the first current-limiting current to the probe current. If the internal state of the cell is stable, even if the temperature rise rate, the single-cell voltage difference, and the inter-cell temperature difference increase or deviate to a certain extent, they should revert accordingly after the current returns to the first current-limiting current. The abnormal advance amount in the current-up phase and the effective retreat amount in the current-down phase should be relatively matched. At this time, the asymmetry value of the current-up and current-down response is small, indicating that the response is mainly a reversible current-stimulated response. Conversely, if the temperature rise rate increases rapidly, the single-cell voltage difference widens, and the inter-cell temperature difference expands when the probe current increases, and these parameters do not retreat sufficiently or very slowly after the current is withdrawn, it indicates that the battery exhibits an asymmetric characteristic of "easily being pushed to abnormality but not easily recovering" in response to current-up stimulation. This often means that there may be hidden risks such as increased polarization induced by lithium plating, persistent local side reactions, abnormal single-cell impedance, local heat accumulation, or micro-short circuit precursors. For example, after the probe current increases, the differential voltage between cells expands from 20 millivolts to 60 millivolts, but only drops back to 50 millivolts after the current is restored, indicating that most of the abnormal advance caused by the current rise was not offset by the current fall. Similarly, the temperature difference between cells expands rapidly during the probe and remains high after the probe current is withdrawn, indicating that the heat distribution has become abnormal and difficult to dissipate synchronously with the current recovery. Since a larger asymmetry value in the rise / fall response indicates a larger unclosed portion of the abnormal advance relative to the effective retreat, meaning the abnormal response caused by the probe action is more difficult to eliminate by the current withdrawal action, when it is included as a component of the latent risk value, a larger value will increase the latent risk value and indicate a higher probability that latent thermal risks still exist inside the target lithium battery.

[0032] The advantage of calculating the asymmetry value of the current rise and fall response using the above method is that it does not simply compare the peak or average rate of change of parameters in the current rise and fall phases. Instead, it first determines the specific direction in which the probe current pushes the temperature rise rate, cell pressure difference, and inter-cell temperature difference towards the anomaly. Then, it extracts the "abnormal advance amount" accumulated along the abnormal direction in the current rise phase and the "effective retreat amount" actually generated along the recovery direction in the current fall phase. This allows for a direct determination of whether the abnormal response caused by the probe current can be offset by the withdrawal current action. Compared to simply looking at the difference before and after detection or the value at a single moment, this method can eliminate the influence of short-term fluctuations and local sampling noise, and more accurately identify the irreversible response characteristics of "current rise is easy to deteriorate and current fall is difficult to recover". At the same time, it reflects the abnormal part that has not been offset by the retreat through the closure gap ratio, and then uses normalization processing after continuous multiplication to obtain the final value, avoiding the problem of artificially setting weights. It also ensures that any significant asymmetry in any key parameter is reflected in the result, making it more suitable for judging whether there is a hidden thermal risk hidden by current limiting inside the target lithium battery.

[0033] In one embodiment, the step of comparing the hidden risk value with a preset threshold and determining whether there is a hidden risk based on the comparison result is as follows: Compare the hidden risk value with the preset threshold. If the hidden risk value is not less than the preset threshold, it means that there is still a hidden thermal risk inside the target lithium battery. If the hidden risk value is less than the preset threshold, it means that there is no hidden thermal risk inside the corresponding target lithium battery.

[0034] It should be noted that converting the aforementioned calculated hidden risk value into a clear control judgment result means determining whether the target lithium battery still has internal thermal risks that are not directly reflected by external parameters after experiencing the shielding fall-off and detection charging actions. Specifically, the hidden risk value is obtained by combining the residual response value and the rise-fall current response asymmetry value. The larger the value, the more obvious the abnormal traces after the detection current withdrawal, the more asymmetric the rise and fall currents, and the higher the possibility of lithium plating, persistent side reactions, local heat accumulation, or micro-short circuit precursors inside the target lithium battery. Therefore, the hidden risk value is compared with a preset threshold. If the hidden risk value is not less than the preset threshold, it means that the degree of abnormality exposed by the detection action has reached the risk judgment condition, and it should be determined that there is still a hidden thermal risk inside the target lithium battery. It is not advisable to directly restore a large charging current. If the hidden risk value is less than the preset threshold, it means that the residual response after detection is weak, the rise-fall current response can basically close, and the target lithium battery does not show obvious hidden thermal risks in the current state. This can be used as a basis for judgment on subsequent graded restoration of charging current. For example, if the preset threshold is 0.6, and the calculated hidden risk value is 0.72, then it is determined that there is still a hidden thermal risk; if the calculated result is 0.35, then it is determined that the hidden thermal risk has not reached the trigger condition.

[0035] In one embodiment, the following steps are taken: determining the subsequent charging strategy based on the latent risk assessment result; if latent risks still exist, maintaining current limiting or suspending charging; if latent risks are mitigated, gradually restoring the charging current, and continuously verifying the risk rebound status during the restoration process: When the latent risk assessment results indicate that there is still a latent thermal risk, the target lithium battery is controlled to maintain the target current-limiting charge, and the temperature rise rate, single-cell voltage difference, temperature difference expansion and voltage deviation are continuously monitored. If at least two of the following parameters—temperature rise rate, single-cell pressure difference, temperature difference expansion, and voltage deviation—fail to meet the preset recovery conditions within the preset monitoring period, then the charging of the target lithium battery will be suspended. When the latent risk assessment result indicates that the latent thermal risk has been mitigated, the charging current is gradually increased according to the preset recovery step size, starting from the first current limiting current, and the preset duration is maintained at each recovery current level. At each recovery current level, the operating data of the target lithium battery is collected, and it is determined whether there is a risk of rebound in the temperature rise rate, single-cell voltage difference, temperature difference expansion, and voltage deviation. If a risk rebound occurs, the charging current will be reduced back to the previous stage recovery current; If no risk rebound occurs, continue to increase to the next level of recovery current until the target flash charging current is reached.

[0036] It should be noted that the previously obtained implicit risk assessment results are translated into specific charging control actions: If it is determined that there is still an implicit thermal risk inside the target lithium battery, it means that the detection charging action has exposed residual response or asymmetric current rise / fall response. At this time, high-rate flash charging cannot be resumed directly. Instead, the first current-limiting current is maintained and observation continues. During the observation process, if at least two of the temperature rise rate, single-cell voltage difference, temperature difference expansion, and voltage deviation still cannot be restored to the preset safety conditions, it means that the risk has not been effectively mitigated under the current-limiting state. In this case, charging is further suspended to avoid further input of energy that could amplify the internal risk. If it is determined that the implicit thermal risk has been mitigated, the original target flash charging current is not immediately restored. Instead, starting from the first current-limiting current, the current is gradually increased according to the preset recovery step size and maintained at each current level for a period of time. At the same time, the temperature rise rate, single-cell voltage difference, temperature difference expansion, and voltage deviation are re-monitored to see if they deteriorate again. The advantage of this approach is that the recovery charging process itself serves as a continuous safety check. If a risk rebound occurs again at a certain current level, the current immediately reverts to the relatively safe recovery current of the previous level. If no rebound occurs, the current is increased to the next level until the target flash charging current is reached. For example, if the target flash charging current is four times the rated current, and after the initial current limiting, it is reduced to one times the rated current, and if the latent risk is mitigated, the current can be restored step by step at 1.5 times, 2 times, 2.5 times, etc., observing whether the temperature rise and voltage difference increase again at each level. Once a significant rebound occurs at two times the rated current, the current is no longer increased but instead reverted to one and a half times the rated current, thus avoiding a secondary increase in thermal risk due to a large current recovery at once.

[0037] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention should still fall within the scope of the claims of the present invention.

Claims

1. A method for early warning and current limiting control of thermal runaway during flash charging of high-rate lithium batteries, characterized in that, Includes the following steps: During high-rate flash charging, the operating status of the target lithium battery is acquired, and the initial current limiting is performed when thermal risk precursors are identified, so as to obtain the risk response results before current limiting. Obtain the status fallback result after the initial rate limiting, and based on the risk response result before rate limiting and the status fallback result, determine whether the status fallback result belongs to the occlusion fallback caused by the initial rate limiting, and obtain the fallback attribute result; When the fallback attribute result is characterized as a shadow fallback, a probe charging action is applied to the target lithium battery, and the hidden risk judgment result is obtained based on the response change under the probe charging action. The subsequent charging strategy will be determined based on the results of the assessment of hidden risks; If hidden risks still exist, current limiting or charging will be maintained; if hidden risks are mitigated, charging current will be restored in stages, and the risk rebound status will be continuously verified during the restoration process.

2. The method for early warning and current limiting control of thermal runaway during high-rate lithium battery flash charging as described in claim 1, characterized in that, The steps for acquiring the operating status of the target lithium battery during high-rate flash charging, performing initial current limiting when thermal risk precursors are identified, and obtaining the risk response results before current limiting are as follows: After the target lithium battery enters the high-rate flash charging mode, the single cell voltage, charging current, single cell surface temperature, inter-cell temperature difference, state of charge and cooling intensity of the target lithium battery are collected according to the preset sampling period to obtain the flash charging operation data sequence. The flash charging operation data sequence is divided into continuous windows to obtain multiple flash charging status windows, and the temperature rise rate, individual cell pressure difference, temperature difference expansion and voltage offset are determined based on each flash charging status window. The temperature rise rate, single-cell pressure difference, temperature difference expansion and voltage deviation are compared with the corresponding preset safety limits. When at least two of the parameters exceed the corresponding preset safety limits in the same flash charging state window, it is determined that the target lithium battery has a thermal risk precursor. When a thermal risk precursor is detected, the end time of the corresponding flash charging state window is determined as the initial current limiting trigger time, and an initial current limiting command is sent to the charging control unit according to the initial current limiting trigger time, so that the charging current of the target lithium battery is reduced to the first current limiting current. Multiple consecutive flash charging status windows before the initial current limiting trigger time are extracted as abnormal sections before current limiting. The changes in temperature rise rate, single-cell pressure difference, temperature difference expansion, voltage deviation, and charging current within the abnormal sections before current limiting are extracted in chronological order to obtain the risk response results before current limiting.

3. The method for early warning and current limiting control of thermal runaway during high-rate lithium battery flash charging as described in claim 1, characterized in that, The steps to determine whether the fallback result is due to occlusion fallback caused by the initial rate limiting and to obtain the fallback attribute result are as follows: After the initial current limiting is implemented, the operating data of the target lithium battery is collected again at the same sampling period as before the initial current limiting, and the current limiting fallback response result is generated according to the same window division method as the risk response result before the current limiting. The temperature rise rate, individual unit pressure difference, temperature difference expansion, and voltage deviation in the fallback response results after current limiting are compared with the corresponding parameters before the initial current limiting. If at least two of the temperature rise rate, individual unit pressure difference, temperature difference expansion, and voltage deviation are lower than the corresponding parameters before the initial current limiting after the initial current limiting, then there is an explicit risk of fallback after the initial current limiting. When there is a decline in the apparent risk, the time period corresponding to the decline in apparent risk is matched with the time period of the decrease in charging current and the time period of the increase in cooling execution intensity to determine whether the decline in apparent risk occurs synchronously with the decrease in charging current or the increase in cooling execution intensity. If the apparent risk decline occurs simultaneously with the decrease in charging current or the increase in cooling execution intensity, and does not continue to decline after the charging current and cooling execution intensity remain stable, then the decline attribute result is determined as masked decline.

4. The method for early warning and current limiting control of thermal runaway during high-rate lithium battery flash charging as described in claim 1, characterized in that, A detection charging action is applied to the target lithium battery, and the hidden risk assessment result is obtained based on the response change under the detection charging action; When the fallback attribute result is characterized as a shadow fallback, the target lithium battery is kept under the first current limit for charging, and the charging current, state of charge, cell surface temperature and cell voltage at the current sampling time are obtained to obtain the reference state before detection. Based on the baseline state before detection, the charging current is increased from the first current limiting current to the detection current. The detection current is greater than the first current limiting current and less than the flash charging current before the initial current limiting is triggered. The target lithium battery is continuously charged under the detection current for a preset detection time, and the single cell voltage, single cell surface temperature, inter-cell temperature difference and charging current of the target lithium battery are continuously collected within the preset detection time to obtain the detection process response data. After the preset detection time is reached, the charging current is restored from the detection current to the first current limiting current, and the individual cell voltage, individual cell surface temperature and inter-cell temperature difference are collected again within the preset fallback time to obtain the fallback data after detection. The baseline state before detection, the response data during detection, and the fallback data after detection are bound in chronological order to form response data corresponding to the detection charging action. The response data is then analyzed and the residual response value and the rise and fall current response asymmetry value are calculated. The residual response value and the rise and fall current response asymmetry value are added together to obtain the hidden risk value. The hidden risk value is compared with the preset threshold, and the existence of hidden risk is determined based on the comparison result.

5. The method for early warning and current limiting control of thermal runaway during high-rate lithium battery flash charging as described in claim 4, characterized in that, The steps for calculating the residual response value are as follows: Extract residual judgment parameters from the response data corresponding to the detection charging action. For any residual judgment parameter, determine the baseline value of the residual judgment parameter in the baseline state before detection, the deviation peak value in the response data during the detection process, and the fallback end value at the end of the fallback data after detection. The deviation peak value is compared with the benchmark value, and the abnormal deviation direction of the residual judgment parameter is determined based on the comparison result; The residual parameters in the same direction are determined based on the abnormal deviation direction, the final value of the fallback, and the baseline value. The residual retention ratio of the residual judgment parameter is calculated based on the absolute value of the difference between the deviation peak and the benchmark value and the residual amount in the same direction. Calculate the retention ratio of all residue determination parameters separately, and subtract each retention ratio from a given value to obtain the degree of non-residue corresponding to each residue determination parameter; The overall non-residue level is obtained by continuously multiplying the non-residue levels corresponding to all residue determination parameters. Subtract the overall non-residual level from 1 to obtain the residual response value.

6. The method for early warning and current limiting control of thermal runaway during high-rate lithium battery flash charging as described in claim 4, characterized in that, The steps for calculating the asymmetry value of the rise and fall response are as follows: Extract the current rise / fall determination parameters from the response data corresponding to the detection charging action, determine the data corresponding to the period when the detection current is applied as the current rise response segment, and determine the data after the detection current is restored to the first current limit current as the current fall back segment. For any rise / fall rate determination parameter, determine the reference value of the rise / fall rate determination parameter under the reference state before detection, and determine the maximum deviation value of the rise / fall rate determination parameter in the rise response segment; The maximum deviation value is compared with the benchmark value, and the abnormal propulsion direction of the rise and fall flow determination parameters is determined based on the comparison result. Based on the abnormal propulsion direction, the parameter values ​​at two adjacent sampling times within the upflow response segment are compared one by one to determine the abnormal propulsion amount; Calculate the asymmetric value of the rise and fall flow response based on the abnormal propulsion direction.

7. The method for early warning and current limiting control of thermal runaway during high-rate lithium battery flash charging as described in claim 4, characterized in that, The steps for calculating the asymmetric value of the upflow and downflow response based on the abnormal propulsion direction are as follows: Based on the abnormal advance direction, the parameter value changes of two adjacent sampling times within the downflow back-off segment are compared one by one to determine the effective back-off amount; Calculate the closure gap ratio of the rise and fall flow determination parameters based on the abnormal advance and effective retreat amounts; Calculate the closed gap ratio of all rise and fall flow determination parameters respectively, and add each closed gap ratio to one to obtain the closed gap enhancement term corresponding to each rise and fall flow determination parameter; The closed gap enhancement terms corresponding to all rise and fall flow determination parameters are multiplied continuously to obtain the comprehensive closed gap enhancement value; Subtract one from the comprehensive gap closure enhancement value to obtain the normalized numerator, and add one to the comprehensive gap closure enhancement value to obtain the normalized denominator; Dividing the normalized numerator by the normalized denominator yields the asymmetric values ​​of the rise and fall flow responses.

8. The method for early warning and current limiting control of thermal runaway during high-rate lithium battery flash charging as described in claim 4, characterized in that, The steps for determining whether there are hidden risks based on the comparison results are as follows: By comparing the hidden risk value with the preset threshold, if the hidden risk value is not less than the preset threshold, it indicates that there is still a hidden thermal risk inside the corresponding target lithium battery.

9. The method for early warning and current limiting control of thermal runaway during high-rate lithium battery flash charging as described in claim 1, characterized in that, The steps for determining the subsequent charging strategy based on the results of the implicit risk assessment are as follows: When the latent risk assessment results indicate that there is still a latent thermal risk, the target lithium battery is controlled to maintain the target current-limiting charge, and the temperature rise rate, single-cell voltage difference, temperature difference expansion and voltage deviation are continuously monitored. If at least two of the following parameters—temperature rise rate, single-cell pressure difference, temperature difference expansion, and voltage deviation—fail to meet the preset recovery conditions within the preset monitoring period, then the charging of the target lithium battery will be suspended. When the latent risk assessment result indicates that the latent thermal risk has been mitigated, the charging current is gradually increased according to the preset recovery step size, starting from the first current limiting current, and the preset duration is maintained at each recovery current level. At each recovery current level, the operating data of the target lithium battery is collected, and it is determined whether there is a risk of rebound in the temperature rise rate, single-cell voltage difference, temperature difference expansion, and voltage deviation. If a risk rebound occurs, the charging current will be reduced back to the previous stage recovery current; If no risk rebound occurs, continue to increase to the next level of recovery current until the target flash charging current is reached.

Images