Battery health state online detection method and system

Abstract

The invention relates to the technical field of battery detection, in particular to a battery health state online detection method and system, and the method comprises the steps: obtaining the current electric quantity of a target battery, the time node and the remaining electric quantity of the last charging end, and the time node and the remaining electric quantity of initial discharging; calculating a time difference value and an electric quantity difference value of the two stages to obtain first unit power consumption, judging whether the first unit power consumption exceeds a threshold value or not, and directly judging that the battery health is abnormal if the first unit power consumption exceeds the threshold value; if not, calculating the time and electric quantity difference between the current discharge stage and the initial discharge stage to obtain second unit power consumption, obtaining ideal unit power consumption in combination with the average output power of the stage, and judging the health state of the battery through the difference. According to the invention, the precision and real-time performance of the online detection of the health state of the battery are improved, and the problem that the traditional method depends on a single parameter and is weak in anti-interference capability is solved.

Description

Technical Field

[0001] This application relates to the field of battery testing technology, and in particular to an online method and system for detecting battery health status. Background Technology

[0002] With the rapid development of the new energy industry, lithium-ion batteries, lead-acid batteries, and other energy storage batteries are widely used in electric vehicles, portable electronic devices, energy storage power stations, and other scenarios. Battery State of Health (SOH) is a core indicator for measuring the degree of battery performance degradation and remaining lifespan; its detection accuracy directly affects the safety and economic efficiency of battery use.

[0003] Traditional battery health status testing methods are mostly offline, such as capacity testing and internal resistance testing. These methods require removing the battery from the equipment and placing it in a specialized testing environment, resulting in low testing efficiency, complex operation, and the inability to monitor in real time. While online testing methods can achieve real-time monitoring, existing technologies often suffer from reliance on a single parameter, weak anti-interference capabilities, and significant susceptibility to changes in operating conditions, making them unsuitable for complex and ever-changing real-world operating scenarios. Therefore, there is an urgent need for a multi-dimensional, high-precision, and interference-resistant online battery health status testing method and system. Summary of the Invention

[0004] To improve the accuracy and real-time performance of online battery health status detection and to address the problems of traditional methods relying on a single parameter and having weak anti-interference capabilities, this application provides an online battery health status detection method and system.

[0005] Firstly, this application provides an online battery health status detection method, which adopts the following technical solution: A method for online detection of battery health status, comprising: Obtain the target battery's current charge level, the first time point of the most recent charging end, and the first remaining charge level; Obtain the second time point of the initial discharge of the target battery and the second remaining charge; Calculate the first time difference between the second time node and the first time node, and the first power difference between the second remaining power and the first remaining power; Based on the first time difference and the first power difference, the first unit power consumption is obtained; Determine whether the first unit of power consumption exceeds the first power consumption threshold; If the value exceeds the limit, the target battery is directly determined to be in an abnormal health state. If the time limit is not exceeded, calculate the second time difference between the current time node and the second time node, and the second power difference between the current power level and the second remaining power level; The second unit power consumption is obtained based on the second time difference and the second power difference; By combining the average output power from the second time node to the current time node, the ideal unit power consumption can be obtained; The battery health status is determined by the difference between the second unit power consumption and the ideal unit power consumption.

[0006] Optional, also includes: If the most recent charging record of the target battery is not available, then the current voltage and internal resistance of the target battery are detected. Obtain the rated voltage and rated internal resistance range of the target battery; Determine whether the current voltage is within the rated voltage range and whether the current internal resistance is within the rated internal resistance range; If the current voltage exceeds the rated voltage range or the current internal resistance exceeds the rated internal resistance range, the battery health status is directly determined to be abnormal. If the current voltage and current internal resistance both meet the rated standards, then obtain the power fluctuation data of the target battery within the preset time threshold. Based on power fluctuation data, the number of power fluctuations and the maximum fluctuation amplitude are counted. If the number of fluctuations exceeds the set threshold or the maximum fluctuation amplitude exceeds the set threshold, the battery health status is determined to be abnormal. If neither the number of fluctuations nor the maximum fluctuation amplitude exceeds the threshold, the battery health status level is determined by combining the current power level with the rated total power level.

[0007] Optionally, based on power fluctuation data, the number of power fluctuations and the maximum fluctuation amplitude can be counted, including: Set a power fluctuation judgment range, which is the minimum effective range of power change within a single detection cycle; Based on a preset frequency, extract all power detection data within a preset time threshold; Compare the power data from two consecutive measurements and calculate the change in power consumption. If the change in target power exceeds the power fluctuation judgment range, it is recorded as a valid fluctuation. The total number of valid fluctuations is accumulated to obtain the number of power fluctuations; The largest change in electricity volume among all valid fluctuations is selected as the maximum fluctuation amplitude.

[0008] Optionally, by combining the average output power from the second time node to the current time node, the ideal unit power consumption can be obtained, including: Collect the real-time output power sequence of the target battery from the second time node to the current time node; Remove outliers from the real-time output power sequence that exceed the normal power range; Calculate the arithmetic mean of the remaining effective power data to obtain the average output power; Based on the battery's rated capacity and average output power, determine the ideal power consumption benchmark per unit time; Verify whether the ideal power consumption benchmark per unit time is within the ideal power consumption range allowed by the battery design; If it is within the ideal power consumption range, then the ideal power consumption per unit time benchmark shall be used as the ideal unit power consumption. If the power consumption exceeds the ideal power consumption range, it will be corrected based on the power-power mapping relationship specified by the battery manufacturer to obtain the ideal unit power consumption.

[0009] Optionally, outliers in the real-time output power sequence that fall outside the normal power range will be removed, including: Obtain the rated output power range of the target battery and determine the upper and lower power limits; Iterate through each data point in the real-time output power sequence and determine whether it exceeds the upper power limit or falls below the lower power limit. If any data point is out of range, it is marked as a suspected outlier. Obtain the power change trend of several consecutive data points near suspected outliers; If the trend of change is consistent, the data is considered valid and retained. If the trends of change are inconsistent, identify the causes of operational fluctuations corresponding to suspected outliers. Causes of operational fluctuations include external disturbances and changes in operating conditions. If the fluctuation is caused by external interference, then suspected outliers will be removed. If the cause of the operational fluctuation is a change in operating conditions, it is determined to be a reasonable operational fluctuation, and the suspected outlier is retained. If no clear cause of operational fluctuations is identified, several sets of power data near the detection time are re-collected, and the arithmetic mean is used to replace the original suspected outlier.

[0010] Optionally, identifying the causes of operational fluctuations corresponding to suspected outliers includes: Acquire real-time monitoring data of the target battery testing environment; Extract environmental monitoring data at the detection time points corresponding to suspected outliers and compare them with the preset normal environmental range; If at least one environmental monitoring data point exceeds the preset normal environmental range, the cause of the operational fluctuation is determined to be environmental interference. If all environmental monitoring data are within the preset normal environmental range, then query the operation log of the testing equipment and determine whether there are any abnormal records of the equipment during the corresponding testing time. If there are records of equipment malfunctions, the cause of the operational fluctuations is determined to be equipment interference; If no equipment abnormality record is found, the load operation log of the target battery is used to determine whether there is a change in the specified operating condition at the corresponding detection time. If a specified change in operating conditions exists, the cause of the operational fluctuation is determined to be the change in operating conditions; If no specific operating condition change is found, it is determined that no clear cause of operational fluctuation has been identified.

[0011] Optionally, the ideal power consumption benchmark per unit time, based on the battery's rated capacity and average output power, includes: Obtain the rated capacity of the target battery; Extract the power-capacity dynamic power consumption coefficient matrix calibrated at the battery factory. The power-capacity dynamic power consumption coefficient matrix is ​​a set of dynamic power consumption calibration values ​​per unit time corresponding to different rated capacity ranges and different output power ranges. Based on the rated capacity of the target battery, a submatrix corresponding to the capacity range is matched from the power-capacity dynamic power consumption coefficient matrix; Based on the average output power, the dynamic power consumption coefficient for the corresponding power range is matched from the sub-matrix; The product of average output power and dynamic power consumption coefficient is calculated to obtain the initial ideal power consumption benchmark per unit time. Obtain the reasonable range of dynamic power consumption corresponding to the rated capacity of the battery, and verify whether the initial ideal power consumption benchmark per unit time is within the reasonable range of dynamic power consumption; If the initial ideal power consumption benchmark per unit time exceeds the upper limit of the reasonable range of dynamic power consumption, the initial ideal power consumption benchmark per unit time will be adjusted to the upper limit of the range. If the initial ideal power consumption benchmark per unit time is lower than the lower limit of the reasonable range of dynamic power consumption, then the initial ideal power consumption benchmark per unit time will be adjusted to the lower limit of the range. After interval adjustment, the initial ideal power consumption benchmark per unit time will be used as the ideal power consumption benchmark per unit time.

[0012] Optionally, the power-power mapping relationship is corrected based on the battery's factory settings to obtain the ideal unit power consumption, which includes: Extract multiple sets of standard power and corresponding ideal power consumption calibration data recorded when the battery leaves the factory; Based on standard power and ideal power consumption, a linear fitting mapping model is constructed; Substituting the average output power into the linear fitting mapping model, we obtain the initial corrected ideal power consumption; Obtain the difference between the current cumulative cycle count and the initial cycle count at the time of manufacture of the target battery; The power consumption correction factor is determined based on the difference in the number of cycles. Based on the initial corrected ideal power consumption and power consumption correction coefficient, the intermediate correction value is obtained; Collect reference values ​​of actual power consumption of healthy batteries of the same model under the same average output power; Calculate the deviation rate between the intermediate correction value and the actual power consumption reference value; If the deviation rate is within the allowable deviation range, the intermediate correction value will be used as the ideal unit power consumption. If the deviation rate exceeds the allowable deviation range, the parameters of the linear fitting mapping model are fine-tuned based on the actual power consumption reference value, and the ideal unit power consumption is recalculated.

[0013] Secondly, this application also discloses an online battery health status detection system, which adopts the following technical solution: A battery health status online detection system, comprising: The first acquisition module is used to acquire the current battery level, the first time point of the most recent end of charging, and the first remaining battery level of the target battery. The second acquisition module is used to acquire the second time point of the initial discharge of the target battery and the second remaining charge. The first calculation module is used to calculate the first time difference between the second time node and the first time node, and the first power difference between the second remaining power and the first remaining power. The third acquisition module is used to acquire the first unit power consumption based on the first time difference and the first power difference; The judgment module is used to determine whether the first unit power consumption exceeds the first power consumption threshold. If the first determination module exceeds the limit, the first determination module is used to directly determine that the target battery is in an abnormal health state. If the time limit is not exceeded, the second calculation module is used to calculate the second time difference between the current time node and the second time node, and the second power difference between the current power and the second remaining power. The fourth acquisition module is used to acquire the second unit power consumption based on the second time difference and the second power difference; The fifth acquisition module is used to obtain the ideal unit power consumption by combining the average output power from the second time node to the current time node; The second determination module is used to determine the battery health status by the difference between the second unit power consumption and the ideal unit power consumption.

[0014] In summary, this application includes the following beneficial technical effects: This application abandons the single-parameter detection mode. It first collects the time and remaining power data of two key nodes: the end of charging and the initial discharge of the target battery. It calculates the first unit power consumption, which reflects the self-discharge characteristics under static conditions. By comparing it with a preset threshold, it completes the initial anomaly screening, quickly filtering out abnormal situations caused by serious faults such as internal short circuits, and avoiding subsequent invalid calculations. For batteries that have not triggered the threshold, it further collects the time and power data from the initial discharge to the current moment, calculates the second unit power consumption under load conditions, and obtains the ideal unit power consumption that matches the actual operating conditions by combining the average output power of this stage. Finally, it completes the accurate judgment by the difference between the actual and ideal unit power consumption. The dual-stage multi-parameter collaborative detection mode avoids the one-sidedness of a single parameter. The layered and progressive judgment logic improves the anti-interference ability of complex operating conditions. At the same time, it is based on real-time data collection and calculation during battery operation, without the need for offline disassembly and detection, effectively balancing the accuracy and real-time performance of the detection. Attached Figure Description

[0015] Figure 1 This is a main flowchart of an online battery health status detection method according to an embodiment of this application; Figure 2 This is a flowchart showing the steps to obtain the ideal unit power consumption; Figure 3 This is a flowchart of the steps to remove outliers from the real-time output power sequence that exceed the normal power range; Figure 4 This is a flowchart illustrating the steps involved in identifying the causes of operational fluctuations corresponding to suspected outliers. Figure 5 This is a flowchart illustrating the steps involved in determining the ideal power consumption benchmark per unit time. Figure 6 This is a block diagram of an online battery health status detection system according to an embodiment of this application.

[0016] Explanation of reference numerals in the attached figures: 1. First acquisition module; 2. Second acquisition module; 3. First calculation module; 4. Third acquisition module; 5. Judgment module; 6. First judgment module; 7. Second calculation module; 8. Fourth acquisition module; 9. Fifth acquisition module; 10. Second judgment module. Detailed Implementation

[0017] Firstly, this application discloses an online method for detecting the health status of a battery.

[0018] Reference Figure 1 A method for online detection of battery health status, comprising steps S101 to S110: Step S101: Obtain the current battery level, the first time point of the most recent charging end, and the first remaining battery level of the target battery.

[0019] Specifically, the target battery refers to the energy storage battery that needs to undergo health status testing, such as electric vehicle power batteries and mobile phone lithium-ion batteries; the current charge refers to the remaining charge of the battery at the time of testing, in ampere-hours (Ah) or watt-hours (Wh), which can be collected in real time by the Battery Management System (BMS). In this embodiment, to improve the visualization of the charge, a percentage can also be set according to the charge; the first time node refers to the time when the target battery last completed the charging process, which is recorded in the form of a timestamp; the first remaining charge refers to the remaining charge of the target battery at the time when the most recent charging ended. This value is usually close to the rated capacity of the battery. Due to the charging cutoff characteristics of the battery, it is generally 95% to 100% of the rated capacity.

[0020] Step S102: Obtain the second time point of the initial discharge of the target battery and the second remaining charge.

[0021] Specifically, in this embodiment, the initial discharge refers to the initial stage when the target battery transitions from the resting state after charging to the load discharge state; the second time node refers to the time point when the target battery begins to enter the discharge state, which can be determined by the current change signal detected by the BMS; the second remaining charge refers to the remaining charge of the target battery at the initial discharge moment, and the difference between this value and the first remaining charge is the self-discharge amount during the battery's resting stage.

[0022] Step S103: Calculate the first time difference between the second time node and the first time node, and the first power difference between the second remaining power and the first remaining power.

[0023] Specifically, in this embodiment, the formula for calculating the first time difference is Δt1=t2−t1, where t2 is the timestamp of the second time node and t1 is the timestamp of the first time node, with the unit being hours (h); the formula for calculating the first energy difference is ΔQ1=Q1−Q2, where Q1 is the first remaining energy and Q2 is the second remaining energy, with the unit being ampere-hours (Ah), which reflects the self-discharge loss of the battery during the resting stage.

[0024] Step S104: Based on the first time difference and the first power difference, obtain the first unit power consumption.

[0025] Specifically, in this embodiment, the first unit power consumption refers to the self-discharge amount of the battery per unit time during the resting stage. The calculation formula is P1=ΔQ1 / Δt1, and the unit is ampere-hours per hour (Ah / h). This parameter is used to evaluate the self-discharge performance of the battery. Excessive self-discharge will directly affect the battery's endurance and health status.

[0026] Step S105: Determine whether the first unit power consumption exceeds the first power consumption threshold.

[0027] Specifically, in this embodiment, the first power consumption threshold refers to the maximum allowable value of self-discharge per unit time during the battery's resting period. This threshold is calibrated according to parameters such as the battery's model, rated capacity, and service life. For example, the first power consumption threshold of a brand-new lithium-ion battery is usually set to 0.005 Ah / h. If the battery has been used for more than 2 years, the threshold can be appropriately relaxed to 0.01 Ah / h.

[0028] Step S106: If the condition exceeds the limit, the target battery is directly determined to be in an abnormal health state.

[0029] Specifically, in this embodiment, if the first unit power consumption exceeds the first power consumption threshold, it indicates that the battery's self-discharge performance has severely degraded and there are faults such as internal short circuits and electrolyte decomposition. At this time, there is no need to conduct subsequent testing, and the battery health status can be directly determined to be abnormal, and an early warning signal will be triggered.

[0030] Step S107: If not exceeded, calculate the second time difference between the current time node and the second time node, and the second power difference between the current power and the second remaining power.

[0031] Specifically, in this embodiment, the current time node refers to the time point at which this detection step is performed; the calculation formula for the second time difference is Δt2=t0-t2, where t0 is the timestamp of the current time node, and the unit is hours (h); the calculation formula for the second charge difference is ΔQ2=Q2-Q0, which reflects the charge loss of the battery during the discharge stage, where Q0 is the current charge.

[0032] Step S108: Obtain the second unit power consumption based on the second time difference and the second power difference.

[0033] Specifically, in this embodiment, the second unit power consumption refers to the amount of charge lost by the battery per unit time during the discharge phase. The calculation formula is P2=ΔQ2 / Δt2, and the unit is ampere-hours per hour (Ah / h). This parameter is used to evaluate the actual power consumption level of the battery under load conditions.

[0034] Step S109: Combine the average output power from the second time node to the current time node to obtain the ideal unit power consumption.

[0035] Specifically, in this embodiment, the average output power refers to the average value of the real-time output power of the battery during the discharge phase, and the unit is watts (W); the ideal unit power consumption refers to the theoretical power consumption per unit time of the battery under the same output power conditions in a healthy state. This parameter needs to be calculated in combination with the battery's rated capacity, power-power consumption characteristics and other parameters, and is the benchmark value for judging the battery's health status.

[0036] Step S110: Determine the battery health status by the difference between the second unit power consumption and the ideal unit power consumption.

[0037] Specifically, in this embodiment, the difference ΔP = |P² - Pi| is calculated, where Pi is the ideal unit power consumption; a health judgment threshold ΔPmax is set. If ΔP ≤ ΔPmax, the battery health is judged to be good; if ΔPmax < ΔP ≤ 2ΔPmax, the battery health is judged to be slightly degraded; if ΔP > 2ΔPmax, the battery health is judged to be severely degraded.

[0038] The online battery health status detection method provided in this embodiment abandons the single-parameter detection mode. It first collects the time and remaining power data of two key nodes: the end of charging and the initial discharge of the target battery. The first unit power consumption, which reflects the self-discharge characteristics under static conditions, is calculated. By comparing it with a preset threshold, a preliminary anomaly screening is completed, quickly filtering out abnormal situations caused by serious faults such as internal short circuits, and avoiding subsequent invalid calculations. For batteries that have not triggered the threshold, the time and power data from the initial discharge to the current moment are further collected to calculate the second unit power consumption under load conditions. The ideal unit power consumption that matches the actual operating conditions is obtained by combining the average output power of this stage. Finally, the difference between the actual and ideal unit power consumption is used to complete the accurate judgment. The dual-stage multi-parameter collaborative detection mode avoids the one-sidedness of a single parameter. The layered and progressive judgment logic improves the anti-interference ability of complex operating conditions. At the same time, it is based on real-time data collection and calculation during battery operation, without the need for offline disassembly and detection, effectively balancing the accuracy and real-time performance of the detection.

[0039] In one embodiment of this example, steps S201 to S208 are further included: Step S201: If the most recent charging record of the target battery is not obtained, then detect the current voltage and internal resistance of the target battery.

[0040] Specifically, in this embodiment, scenarios where charging records are missing include the first use of the battery and loss of BMS data; the current voltage refers to the terminal voltage of the battery at the detection time, in volts (V), which can be collected by a voltage sensor; the internal resistance refers to the internal impedance of the battery, including ohmic internal resistance and polarization internal resistance, in milliohms (mΩ), which can be measured by AC impedance method or DC discharge method.

[0041] Step S202: Obtain the rated voltage and rated internal resistance range of the target battery.

[0042] Specifically, in this embodiment, the rated voltage refers to the standard voltage specified by the battery manufacturer. For example, the rated voltage of a lithium-ion battery is usually 3.7V, and the rated voltage of a lead-acid battery is usually 2V / cell. The rated internal resistance range refers to the allowable range of internal resistance of the battery in a healthy state. This range is specified according to the battery model. For example, the rated internal resistance range of a lithium-ion power battery with a rated capacity of 100Ah is usually 5~15mΩ.

[0043] Step S203: Determine whether the current voltage is within the rated voltage range and whether the current internal resistance is within the rated internal resistance range.

[0044] Specifically, in this embodiment, the rated voltage range refers to the normal voltage range of the battery in a static state, which is usually 90% to 110% of the rated voltage. For example, for a lithium-ion battery with a rated voltage of 3.7V, the rated voltage range is 3.33 to 4.07V. If the current voltage and internal resistance are both within the corresponding range, the battery's basic performance is determined to be normal; otherwise, it is determined to be abnormal.

[0045] Step S204: If the current voltage exceeds the rated voltage range or the current internal resistance exceeds the rated internal resistance range, the battery health status is directly determined to be abnormal.

[0046] Specifically, in this embodiment, a current voltage that is too low may be caused by deep discharge of the battery or internal short circuit, while a current voltage that is too high may be caused by overcharging or electrolyte decomposition. If the internal resistance exceeds the range, it indicates that the battery has faults such as plate sulfation or separator aging. All of the above situations can directly determine that the battery health status is abnormal.

[0047] Step S205: If the current voltage and current internal resistance both meet the rated standards, then obtain the power fluctuation data of the target battery within the preset time threshold.

[0048] Specifically, in this embodiment, the preset time threshold refers to the set duration of power data collection, which is usually set to 24 hours and can be adjusted according to actual detection needs; the power fluctuation data refers to the sequence data of battery power changes over time within the preset time threshold, which reflects the stability of battery power.

[0049] Step S206: Based on the power fluctuation data, count the number of power fluctuations and the maximum fluctuation amplitude.

[0050] Specifically, in this embodiment, the number of power fluctuations refers to the number of valid fluctuations in battery power within a preset time threshold; the maximum fluctuation amplitude refers to the maximum value of a single power change among all valid fluctuations. This parameter is used to evaluate the stability of battery power.

[0051] Step S207: If the number of fluctuations exceeds the set number threshold or the maximum fluctuation amplitude exceeds the set amplitude threshold, the battery health status is determined to be abnormal.

[0052] Specifically, in this embodiment, the number threshold refers to the maximum allowed number of effective power fluctuations within a preset time threshold, for example, the number threshold is set to 5 times within 24 hours; the amplitude threshold refers to the maximum allowed value of a single power fluctuation, for example, it is set to 5% of the rated capacity; if any parameter exceeds the threshold, it indicates that the battery power stability is poor, and there is a charge / discharge control fault or internal performance degradation.

[0053] Step S208: If the number of fluctuations and the maximum fluctuation amplitude do not exceed the threshold, the battery health status level is determined by combining the current power level with the rated total power level.

[0054] Specifically, in this embodiment, the rated total capacity refers to the total charge quantity specified by the battery at the factory, in ampere-hours (Ah); the calculated percentage η = Q0 / Qr × 100%, where Qr is the rated total capacity; if η ≥ 80%, it is judged as health state A; if 60% ≤ η < 80%, it is judged as health state B; if η < 60%, it is judged as health state C, requiring timely maintenance or replacement.

[0055] The online battery health status detection method provided in this embodiment can determine the health status by combining voltage and internal resistance detection with power stability analysis when charging records are missing, thus filling the detection gap in scenarios without charging records.

[0056] In one embodiment of this example, step S206, based on power fluctuation data, counts the number of power fluctuations and the maximum fluctuation amplitude, including steps S301 to S306: Step S301: Set the power fluctuation judgment range.

[0057] Specifically, in this embodiment, the power fluctuation judgment range is the minimum effective range of power change within a single detection cycle, for example, set to 0.5% of the rated capacity. If the power change is less than this threshold, it is determined to be an invalid fluctuation caused by measurement error; otherwise, it is determined to be a valid fluctuation. Here, a single detection cycle refers to the power data collection interval, for example, set to 1 hour.

[0058] Step S302: Based on the preset frequency, extract all power detection data within the preset time threshold.

[0059] Specifically, the preset frequency refers to the frequency of power data collection, for example, set to 1 time / hour; all detection data within the preset time threshold are extracted to form a power-time series dataset {t1,Q1),(t2,Q2),...,(tn,Qn)}, where n is the number of data collections.

[0060] Step S303: Compare the power data from two adjacent detections sequentially and calculate the change in power.

[0061] Specifically, in this embodiment, the formula for calculating the change in power between two adjacent detections is ΔQi=∣Qi+1−Qi∣, where i=1,2,...,n−1, and the unit is ampere-hours (Ah). This change reflects the degree of power fluctuation within a single detection cycle.

[0062] Step S304: If the change in target power exceeds the power fluctuation judgment range, it is recorded as a valid fluctuation.

[0063] Specifically, in this embodiment, the target power change refers to the calculated ΔQi; if ΔQi > ΔQt (ΔQt is the threshold of the judgment interval), it is determined to be a valid fluctuation and counted.

[0064] Step S305: Accumulate the number of all valid fluctuations to obtain the number of power fluctuations.

[0065] Specifically, in this embodiment, the number of times that ΔQi>ΔQt is counted is the number of power fluctuations, which reflects the frequency of battery power fluctuations within a preset time threshold.

[0066] Step S306: Filter the largest change in electricity among all valid fluctuations as the maximum fluctuation amplitude.

[0067] Specifically, in this embodiment, the maximum value ΔQmax is selected from all valid fluctuations corresponding to ΔQi, which is the maximum fluctuation amplitude. This parameter reflects the severity of battery power fluctuations.

[0068] The online battery health status detection method provided in this embodiment improves the accuracy of power fluctuation data statistics by setting an effective fluctuation threshold to eliminate measurement errors, thus providing reliable data support for battery health status determination.

[0069] Reference Figure 2 In one embodiment of this example, step S109, which combines the average output power from the second time node to the current time node to obtain the ideal unit power consumption, includes steps S401 to S407: Step S401: Collect the real-time output power sequence of the target battery from the second time node to the current time node.

[0070] Specifically, in this embodiment, real-time output power refers to the instantaneous output power of the battery during the discharge phase, and the calculation formula is Pr(t)=U(t)×I(t), where Pr(t) is the real-time output power, U(t) is the instantaneous terminal voltage, and I(t) is the instantaneous discharge current; the real-time output power sequence refers to the power data set {P1,P2,...,Pm} acquired at a preset acquisition frequency within the time period Δt2.

[0071] Step S402: Remove outliers from the real-time output power sequence that are outside the normal power range.

[0072] Specifically, in this embodiment, the normal power range refers to the range of output power of the battery under safe operating conditions, which is determined by the rated output power of the battery. For example, for a battery with a rated output power of 1000W, the normal power range is set to 0~1200W. Power data exceeding this range is judged as abnormal values ​​and needs to be eliminated or corrected.

[0073] Step S403: Calculate the arithmetic mean of the remaining effective power data to obtain the average output power.

[0074] Specifically, in this embodiment, after removing outliers, the remaining effective power data is {P1′,P2′,...,Pk′}, and the formula for calculating the average output power is: .

[0075] Step S404: Determine the ideal power consumption benchmark per unit time based on the battery's rated capacity and average output power.

[0076] Specifically, in this embodiment, the ideal power consumption benchmark per unit time refers to the theoretical power consumption of a healthy battery per unit time under the corresponding rated capacity and average output power.

[0077] Step S405: Verify whether the ideal power consumption benchmark per unit time is within the ideal power consumption range allowed by the battery design.

[0078] Specifically, in this embodiment, the ideal power consumption range refers to the allowable range of power consumption per unit time specified by the battery manufacturer. This range is set according to the battery's chemical system and rated capacity. For example, for a lithium-ion battery with a rated capacity of 100Ah, the ideal power consumption range is set to 0.1~0.3Ah / h.

[0079] Step S406: If it is within the ideal power consumption range, then the ideal power consumption per unit time benchmark is taken as the ideal unit power consumption.

[0080] Specifically, in this embodiment, if the ideal power consumption benchmark per unit time is within the ideal power consumption range, it indicates that the benchmark value meets the battery design standards and can be directly used as the ideal unit power consumption for subsequent health status determination.

[0081] Step S407: If the power consumption exceeds the ideal power consumption range, the power-power consumption mapping relationship calibrated by the battery factory is corrected to obtain the ideal unit power consumption.

[0082] Specifically, in this embodiment, the power-power consumption mapping relationship refers to the curve or data table showing the correspondence between the output power calibrated at the battery factory and the power consumption per unit time. If the ideal power consumption benchmark per unit time exceeds the ideal power consumption range, the ideal power consumption benchmark per unit time needs to be corrected according to this mapping relationship to ensure the accuracy of the ideal power consumption per unit time.

[0083] The online battery health state detection method provided by this embodiment improves the accuracy of the ideal unit power consumption by removing power outliers and correcting the reference value in combination with the battery design standard, providing a reliable reference for determining the battery health state.

[0084] Referring to Figure 3 , in one implementation manner of this embodiment, removing the outliers exceeding the normal power range in the real-time output power sequence in step S402 includes steps S501 to S509: Step S501: Obtain the rated output power range of the target battery and determine the power upper limit value and the power lower limit value.

[0085] Specifically, in this embodiment, the rated output power range refers to the output power interval of the target battery under safe operating conditions, which is calibrated by the battery manufacturer; the power upper limit value refers to the maximum value of the rated output power, usually 120% of the rated power, and the power lower limit value refers to the minimum value of the rated output power, usually 0W (static state).

[0086] Step S502: Traverse each data point in the real-time output power sequence and determine whether it exceeds the power upper limit value or is lower than the power lower limit value.

[0087] Specifically, in this embodiment, traverse the power sequence {P1, P2,..., Pm}. If there exists a data point Pi > Pmax or Pi < Pmin (Pmax is the power upper limit value, Pmin is the power lower limit value), then mark this data point as a suspected outlier. A suspected outlier refers to a power data point that exceeds the rated output power range of the target battery when traversing each data point in the real-time output power sequence collected from the second time node to the current time node of the target battery.

[0088] Step S503: If there exists a data point exceeding the range, mark it as a suspected outlier.

[0089] Specifically, in this embodiment, the marking of the suspected outlier needs to be associated with the collection timestamp of the data point for subsequent analysis of the abnormal cause. The marking format is {(ti, Pi), abnormal marking}.

[0090] Step S504: Obtain the power change trend of several consecutive data points near the suspected outlier.

[0091] Specifically, in this embodiment, several consecutive data points nearby refer to 3 to 5 data points before and after the suspected outlier; the power change trend refers to the rising, falling or stable trend of the data points, which can be determined by linear fitting or slope calculation. For example, calculate the slope of adjacent data points ki = (Pi+1 - Pi) / (ti+1 - ti). If the slope signs are the same, it is determined that the trend is the same.

[0092] Step S505: If the trend of change is consistent, the data is determined to be valid and retained.

[0093] Specifically, in this embodiment, if the power change trends of data points near the suspected outlier are consistent, it indicates that the outlier is a normal fluctuation caused by changes in battery operating conditions, such as a sudden increase in load causing the power to briefly exceed the upper limit. In this case, the data point is determined to be valid data and is retained.

[0094] Step S506: If the trends of change are inconsistent, identify the causes of operational fluctuations corresponding to the suspected outliers.

[0095] Specifically, in this embodiment, the causes of operational fluctuations include external interference and changes in operating conditions. External interference includes sudden changes in ambient temperature and failure of detection equipment, while changes in operating conditions include sudden changes in load and switching of charging and discharging modes. The identification of causes requires a comprehensive judgment based on information such as environmental data and equipment logs.

[0096] Step S507: If the fluctuation is caused by external interference, then remove the suspected outliers.

[0097] Specifically, in this embodiment, if the cause is determined to be external interference, it means that the suspected outlier is caused by factors other than battery performance and does not reflect the actual output characteristics of the battery. Therefore, the data point needs to be removed from the power sequence.

[0098] Step S508: If the cause of the operational fluctuation is a change in operating conditions, it is determined to be a reasonable operational fluctuation, and the suspected outlier is retained.

[0099] Specifically, in this embodiment, if the cause is determined to be a change in operating conditions, it means that the suspected abnormal value is caused by fluctuations in operating conditions during the normal operation of the battery, such as a sudden power change when an electric vehicle accelerates. In this case, it is determined to be a reasonable fluctuation, and the data point is retained.

[0100] Step S509: If no clear cause of operational fluctuation is identified, several sets of power data near the detection time are collected again, and the arithmetic mean is used to replace the original suspected outlier.

[0101] Specifically, in this embodiment, if the cause of the anomaly cannot be clearly identified, 3 to 5 sets of power data need to be collected again near the collection time corresponding to the suspected anomaly value, the arithmetic mean is calculated, and the average value is used to replace the original suspected anomaly value to ensure the integrity of the power sequence.

[0102] The online battery health status detection method provided in this embodiment distinguishes between normal fluctuations and abnormal data through trend analysis and cause identification, avoiding the erroneous rejection of valid data and improving the accuracy of average output power calculation.

[0103] Reference Figure 4In one embodiment of this example, step S506, which identifies the cause of operational fluctuations corresponding to suspected outliers, includes steps S601 to S608: Step S601: Obtain real-time monitoring data of the target battery testing environment.

[0104] Specifically, in this embodiment, the real-time monitoring data of the detection environment includes parameters such as ambient temperature, humidity, and air pressure, which are collected by environmental sensors deployed near the battery.

[0105] Step S602: Extract environmental monitoring data at the detection time points corresponding to suspected outliers and compare them with the preset normal environmental range.

[0106] Specifically, in this embodiment, the preset normal environmental range refers to the range of environmental parameters for normal battery operation. For example, the normal temperature range of a lithium-ion battery is -20℃ to 60℃, and the normal humidity range is 20% to 80%RH. The environmental data corresponding to the suspected outlier time point ti is extracted and compared to see if it exceeds the normal range.

[0107] Step S603: If at least one environmental monitoring data exceeds the preset normal environmental range, the cause of the operational fluctuation is determined to be environmental interference.

[0108] Specifically, in this embodiment, a sudden increase in ambient temperature may cause changes in the battery's internal resistance, which in turn may cause fluctuations in output power; excessive humidity may cause leakage in the battery casing, affecting the accuracy of power detection. All of the above situations are judged as environmental interference.

[0109] Step S604: If all environmental monitoring data are within the preset normal environmental range, query the operation log of the detection equipment and determine whether there are any abnormal records of the equipment at the corresponding detection time.

[0110] Specifically, the operation log of the testing equipment includes the working status records of voltage sensors, current sensors, and data acquisition units; the equipment anomaly records include records of sensor failures, data transmission interruptions, and data acquisition unit crashes. By querying the logs, it can be determined whether the power data anomalies are caused by equipment failures.

[0111] Step S605: If there is a record of equipment abnormality, the cause of the operational fluctuation is determined to be equipment interference.

[0112] Specifically, in this embodiment, equipment malfunctions such as sensor calibration deviation and data acquisition delay can lead to distortion of power detection data, which is then determined to be equipment interference.

[0113] Step S606: If there is no equipment abnormality record, then combine the target battery's load operation log to determine whether there is a change in the specified operating condition at the corresponding detection time.

[0114] Specifically, in this embodiment, the load operation log includes the working status record of the battery load, such as the acceleration, deceleration, and hill climbing conditions of the electric vehicle; the specified operating condition change refers to the operating condition where the load power suddenly changes, such as the load suddenly increasing from 100W to 1000W.

[0115] Step S607: If there is a change in the specified operating condition, then the cause of the operational fluctuation is determined to be the change in operating condition.

[0116] Specifically, in this embodiment, a sudden change in load conditions will cause a synchronous change in battery output power, resulting in a situation that exceeds the normal power range. This is then determined to be a reasonable fluctuation caused by the change in operating conditions.

[0117] Step S608: If there is no change in the specified operating condition, it is determined that no clear cause of the operational fluctuation has been identified.

[0118] Specifically, if there are no abnormalities in environmental interference, equipment interference, or load conditions, it means that the cause of the abnormality cannot be identified through the existing data, and subsequent data re-collection and replacement operations need to be performed.

[0119] The online battery health status detection method provided in this embodiment solves the problem of handling suspected outliers caused by unknown factors by supplementing and replacing data. This avoids the loss of power sequence integrity caused by directly removing data and avoids the calculation deviation caused by retaining outliers, thus ensuring the accuracy of subsequent calculations of average output power and ideal unit power consumption.

[0120] Reference Figure 5 In one embodiment of this example, step S404, based on the battery's rated capacity and average output power, determines the ideal power consumption benchmark per unit time, including steps S701 to S709: Step S701: Obtain the rated capacity of the target battery.

[0121] Specifically, in this embodiment, the rated capacity refers to the total charge of the battery under standard charging and discharging conditions, which is specified by the manufacturer. For example, the rated capacity of electric vehicle power batteries is usually 50~200Ah.

[0122] Step S702: Extract the power-capacity dynamic power consumption coefficient matrix calibrated at the battery factory.

[0123] Specifically, in this embodiment, the power-capacity dynamic power consumption coefficient matrix is ​​a set of dynamic power consumption calibration values ​​per unit time corresponding to different rated capacity ranges and different output power ranges.

[0124] Step S703: Based on the rated capacity of the target battery, match the submatrix corresponding to the capacity range from the power-capacity dynamic power consumption coefficient matrix.

[0125] Specifically, in this embodiment, if the rated capacity Qr of the target battery is in the interval Qa ~ Qa+1, extract the row vector corresponding to this interval to form a capacity sub-matrix {Ka1, Ka2,..., Kan}.

[0126] Step S704: Based on the average output power, match the dynamic power consumption coefficient corresponding to the power interval from the sub-matrix.

[0127] Specifically, in this embodiment, if the average output power is in the interval Pb ~ Pb+1, extract the corresponding coefficient Kab from the capacity sub-matrix.

[0128] Step S705: Calculate the product of the average output power and the dynamic power consumption coefficient to obtain the initial ideal power consumption benchmark per unit time.

[0129] Specifically, in this embodiment, the initial ideal power consumption benchmark per unit time refers to the theoretical power consumption per unit time calculated based on the power-capacity dynamic power consumption coefficient calibrated at the factory for the target battery in a healthy state within the interval corresponding to its rated capacity and the interval corresponding to the current average output power. The calculation formula for the initial ideal power consumption benchmark per unit time is .

[0130] Step S706: Obtain the reasonable interval of dynamic power consumption corresponding to the battery rated capacity, and verify whether the initial ideal power consumption benchmark per unit time is within the reasonable interval of dynamic power consumption.

[0131] Specifically, in this embodiment, the reasonable interval of dynamic power consumption refers to the allowable range of power consumption per unit time corresponding to the rated capacity, which is calibrated by the manufacturer. For example, for a battery with a rated capacity of 100Ah, the reasonable interval of dynamic power consumption is 0.1 ~ 0.3Ah / h.

[0132] Step S707: If the initial ideal power consumption benchmark per unit time exceeds the upper limit of the reasonable interval of dynamic power consumption, adjust the initial ideal power consumption benchmark per unit time to the upper limit value of the interval.

[0133] Specifically, in this embodiment, if Pi0 > Pu (Pu is the upper limit value of the interval), then set Pi0 = Pu.

[0134] Step S708: If the initial ideal power consumption benchmark per unit time is lower than the lower limit of the reasonable interval of dynamic power consumption, adjust the initial ideal power consumption benchmark per unit time to the lower limit value of the interval.

[0135] Specifically, in this embodiment, if Pi0 < Pl (Pl is the lower limit value of the interval), then set Pi0 = Pl.

[0136] Step S709: After the interval adjustment, use the initial ideal power consumption benchmark per unit time as the ideal power consumption benchmark per unit time.

[0137] Specifically, in this embodiment, Pi0 after adjusting the upper and lower limits is the final ideal power consumption benchmark per unit time, which is used for subsequent determination of ideal unit power consumption.

[0138] The online battery health status detection method provided in this embodiment avoids the problem of benchmark value distortion caused by power-capacity dynamic power consumption coefficient matrix matching deviation or extreme operating condition data by setting a reasonable range of dynamic power consumption and imposing boundary constraints on the initial benchmark value, thus ensuring the rationality and rigor of the ideal power consumption benchmark per unit time.

[0139] In one embodiment of this example, step S407 involves correcting the power-power consumption mapping relationship based on the battery's factory calibration to obtain the ideal unit power consumption, including steps S801 to S810: Step S801: Extract multiple sets of standard power and corresponding ideal power consumption calibration data recorded when the battery leaves the factory.

[0140] Specifically, in this embodiment, the calibration data of standard power and corresponding ideal power consumption refers to the ideal power consumption data per unit time under different output power measured by the manufacturer through experiments before the battery leaves the factory, in the form of {(Ps1,Pi1),(Ps2,Pi2),...,(Psk,Pik)}.

[0141] Step S802: Construct a linear fitting mapping model based on standard power and ideal power consumption.

[0142] Specifically, in this embodiment, the expression of the linear fitting mapping model is Pi = a × Ps + b, where a is the slope and b is the intercept. The model parameters are obtained by fitting the calibration data using the least squares method.

[0143] Step S803: Substitute the average output power into the linear fitting mapping model to obtain the initial corrected ideal power consumption.

[0144] Step S804: Obtain the difference between the current cumulative number of cycles of the target battery and the initial number of cycles at the time of manufacture.

[0145] Specifically, in this embodiment, the number of cycles refers to the number of times the battery completes one full charge-discharge process; the formula for calculating the difference in the number of cycles is ΔN=Nc-Ni, where Nc is the current cumulative number of cycles and Ni is the initial number of cycles at the factory (usually 0).

[0146] Step S805: Determine the power consumption correction coefficient based on the difference in the number of cycles.

[0147] Specifically, in this embodiment, the power consumption correction factor refers to a factor set according to the battery cycle degradation characteristics. The more battery cycles, the larger the power consumption correction factor. The correction factor can be calibrated through the battery cycle degradation curve, for example... ,in This is the attenuation factor, provided by the manufacturer.

[0148] Step S806: Based on the initial corrected ideal power consumption and power consumption correction coefficient, obtain the intermediate correction value.

[0149] Specifically, in this embodiment, the intermediate correction value refers to the intermediate value of the ideal power consumption per unit time obtained after the initial corrected ideal power consumption is corrected by the battery cycle decay characteristics. It is a transitional parameter connecting the initial corrected ideal power consumption and the final ideal unit power consumption. The formula for calculating the intermediate correction value is Pi2=Pi1×k.

[0150] Step S807: Collect the reference value of the actual power consumption of healthy batteries of the same model under the same average output power.

[0151] Specifically, in this embodiment, a healthy battery of the same model refers to a battery with the same model as the target battery, a cycle count of less than 50, and a health status of A; the actual power consumption reference value refers to the actual power consumption of the healthy battery per unit time under the same average output power, which is measured experimentally.

[0152] Step S808: Calculate the deviation rate between the intermediate correction value and the actual power consumption reference value.

[0153] Specifically, in this embodiment, the deviation rate is a quantitative indicator used to measure the degree of difference between the intermediate correction value and the actual power consumption reference value of the same type of healthy battery. Its core function is to determine the accuracy of the intermediate correction value and then decide whether it is necessary to fine-tune the linear fitting mapping model. The formula for calculating the deviation rate is δ=|Pi2−Pre| / Pre×100%, where Pre is the actual power consumption reference value.

[0154] Step S809: If the deviation rate is within the allowable deviation range, then the intermediate correction value is taken as the ideal unit power consumption.

[0155] Specifically, in this embodiment, the allowable deviation range is usually set to ±5%; if δ≤5%, it means that the accuracy of the intermediate correction value meets the requirements and can be used as the ideal unit power consumption.

[0156] Step S810: If the deviation rate exceeds the allowable deviation range, fine-tune the parameters of the linear fitting mapping model based on the actual power consumption reference value, and recalculate the ideal unit power consumption.

[0157] Specifically, in this embodiment, the method for fine-tuning the parameters is as follows: taking Pre as the target value, adjust the slope a and intercept b of the linear model so that a × +b is close to Pre; after adjustment, the ideal unit power consumption is recalculated.

[0158] The online battery health status detection method provided in this embodiment solves the problem of deviation between the theoretical fitting model and the actual battery operating state through a benchmark reference-guided model parameter fine-tuning mechanism, further improving the calculation accuracy of ideal unit power consumption and ensuring the reliability of battery health status determination.

[0159] Secondly, this application also discloses an online battery health status detection system.

[0160] Reference Figure 6 A battery health status online detection system, comprising: The first acquisition module 1 is used to acquire the current battery level, the first time point of the most recent end of charging, and the first remaining battery level of the target battery; The second acquisition module 2 is used to acquire the second time point of the initial discharge of the target battery and the second remaining charge. The first calculation module 3 is used to calculate the first time difference between the second time node and the first time node, and the first power difference between the second remaining power and the first remaining power. The third acquisition module 4 is used to acquire the first unit power consumption based on the first time difference and the first power difference; Module 5 is used to determine whether the first unit power consumption exceeds the first power consumption threshold. If the first determination module 6 exceeds the limit, the first determination module 6 is used to directly determine that the target battery is in an abnormal health state. If the time limit is not exceeded, the second calculation module 7 is used to calculate the second time difference between the current time node and the second time node, and the second power difference between the current power and the second remaining power. The fourth acquisition module 8 is used to acquire the second unit power consumption based on the second time difference and the second power difference; The fifth acquisition module 9 is used to obtain the ideal unit power consumption by combining the average output power from the second time node to the current time node; The second determination module 10 is used to determine the battery health status by the difference between the second unit power consumption and the ideal unit power consumption.

[0161] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A method for online detection of battery health status, characterized in that, include: Obtain the target battery's current charge level, the first time point of the most recent charging end, and the first remaining charge level; Obtain the second time point of the initial discharge of the target battery and the second remaining charge; Calculate the first time difference between the second time node and the first time node, and the first power difference between the second remaining power and the first remaining power; Based on the first time difference and the first power difference, the first unit power consumption is obtained; Determine whether the first unit of power consumption exceeds the first power consumption threshold; If the value exceeds the limit, the target battery is directly determined to be in an abnormal health state. If the time limit is not exceeded, calculate the second time difference between the current time node and the second time node, and the second power difference between the current power level and the second remaining power level; The second unit power consumption is obtained based on the second time difference and the second power difference; By combining the average output power from the second time node to the current time node, the ideal unit power consumption can be obtained; The battery health status is determined by the difference between the second unit power consumption and the ideal unit power consumption.

2. The method for online detection of battery health status according to claim 1, characterized in that, Also includes: If the most recent charging record of the target battery is not available, then the current voltage and internal resistance of the target battery are detected. Obtain the rated voltage and rated internal resistance range of the target battery; Determine whether the current voltage is within the rated voltage range and whether the current internal resistance is within the rated internal resistance range; If the current voltage exceeds the rated voltage range or the current internal resistance exceeds the rated internal resistance range, the battery health status is directly determined to be abnormal. If the current voltage and current internal resistance both meet the rated standards, then obtain the power fluctuation data of the target battery within the preset time threshold. Based on power fluctuation data, the number of power fluctuations and the maximum fluctuation amplitude are counted. If the number of fluctuations exceeds the set threshold or the maximum fluctuation amplitude exceeds the set threshold, the battery health status is determined to be abnormal. If neither the number of fluctuations nor the maximum fluctuation amplitude exceeds the threshold, the battery health status level is determined by combining the current power level with the rated total power level.

3. The method for online detection of battery health status according to claim 2, characterized in that, Based on power fluctuation data, the number of power fluctuations and the maximum fluctuation amplitude are statistically analyzed, including: Set a power fluctuation judgment range, which is the minimum effective range of power change within a single detection cycle; Based on a preset frequency, extract all power detection data within a preset time threshold; Compare the power data from two consecutive measurements and calculate the change in power consumption. If the change in target power exceeds the power fluctuation judgment range, it is recorded as a valid fluctuation. The total number of valid fluctuations is accumulated to obtain the number of power fluctuations; The largest change in electricity volume among all valid fluctuations is selected as the maximum fluctuation amplitude.

4. The method for online detection of battery health status according to claim 1, characterized in that, Based on the average output power from the second time point to the current time point, the ideal unit power consumption is obtained as follows: Collect the real-time output power sequence of the target battery from the second time node to the current time node; Remove outliers from the real-time output power sequence that exceed the normal power range; Calculate the arithmetic mean of the remaining effective power data to obtain the average output power; Based on the battery's rated capacity and average output power, determine the ideal power consumption benchmark per unit time; Verify whether the ideal power consumption benchmark per unit time is within the ideal power consumption range allowed by the battery design; If it is within the ideal power consumption range, then the ideal power consumption per unit time benchmark shall be used as the ideal unit power consumption. If the power consumption exceeds the ideal power consumption range, it will be corrected based on the power-power mapping relationship specified by the battery manufacturer to obtain the ideal unit power consumption.

5. The method for online detection of battery health status according to claim 4, characterized in that, Outliers in the real-time output power sequence that exceed the normal power range include: Obtain the rated output power range of the target battery and determine the upper and lower power limits; Iterate through each data point in the real-time output power sequence and determine whether it exceeds the upper power limit or falls below the lower power limit. If any data point is out of range, it is marked as a suspected outlier. Obtain the power change trend of several consecutive data points near suspected outliers; If the trend of change is consistent, the data is considered valid and retained. If the trends of change are inconsistent, identify the causes of operational fluctuations corresponding to suspected outliers. Causes of operational fluctuations include external disturbances and changes in operating conditions. If the fluctuation is caused by external interference, then suspected outliers will be removed. If the cause of the operational fluctuation is a change in operating conditions, it is determined to be a reasonable operational fluctuation, and the suspected outlier is retained. If no clear cause of operational fluctuations is identified, several sets of power data near the detection time are re-collected, and the arithmetic mean is used to replace the original suspected outlier.

6. The method for online detection of battery health status according to claim 5, characterized in that, Identifying the causes of operational fluctuations corresponding to suspected outliers includes: Acquire real-time monitoring data of the target battery testing environment; Extract environmental monitoring data at the detection time points corresponding to suspected outliers and compare them with the preset normal environmental range; If at least one environmental monitoring data point exceeds the preset normal environmental range, the cause of the operational fluctuation is determined to be environmental interference. If all environmental monitoring data are within the preset normal environmental range, then query the operation log of the testing equipment and determine whether there are any abnormal records of the equipment during the corresponding testing time. If there are records of equipment malfunctions, the cause of the operational fluctuations is determined to be equipment interference; If no equipment abnormality record is found, the load operation log of the target battery is used to determine whether there is a change in the specified operating condition at the corresponding detection time. If a specified change in operating conditions exists, the cause of the operational fluctuation is determined to be the change in operating conditions; If no specific operating condition change is found, it is determined that no clear cause of operational fluctuation has been identified.

7. The method for online detection of battery health status according to claim 4, characterized in that, Based on the battery's rated capacity and average output power, the ideal power consumption benchmark per unit time includes: Obtain the rated capacity of the target battery; Extract the power-capacity dynamic power consumption coefficient matrix calibrated at the battery factory. The power-capacity dynamic power consumption coefficient matrix is ​​a set of dynamic power consumption calibration values ​​per unit time corresponding to different rated capacity ranges and different output power ranges. Based on the rated capacity of the target battery, a submatrix corresponding to the capacity range is matched from the power-capacity dynamic power consumption coefficient matrix; Based on the average output power, the dynamic power consumption coefficient for the corresponding power range is matched from the sub-matrix; The product of average output power and dynamic power consumption coefficient is calculated to obtain the initial ideal power consumption benchmark per unit time. Obtain the reasonable range of dynamic power consumption corresponding to the rated capacity of the battery, and verify whether the initial ideal power consumption benchmark per unit time is within the reasonable range of dynamic power consumption; If the initial ideal power consumption benchmark per unit time exceeds the upper limit of the reasonable range of dynamic power consumption, the initial ideal power consumption benchmark per unit time will be adjusted to the upper limit of the range. If the initial ideal power consumption benchmark per unit time is lower than the lower limit of the reasonable range of dynamic power consumption, then the initial ideal power consumption benchmark per unit time will be adjusted to the lower limit of the range. After interval adjustment, the initial ideal power consumption benchmark per unit time will be used as the ideal power consumption benchmark per unit time.

8. The method for online detection of battery health status according to claim 4, characterized in that, Based on the power-power mapping relationship specified at the battery manufacturer's factory, the ideal unit power consumption is obtained as follows: Extract multiple sets of standard power and corresponding ideal power consumption calibration data recorded when the battery leaves the factory; Based on standard power and ideal power consumption, a linear fitting mapping model is constructed; Substituting the average output power into the linear fitting mapping model, we obtain the initial corrected ideal power consumption; Obtain the difference between the current cumulative cycle count and the initial cycle count at the time of manufacture of the target battery; The power consumption correction factor is determined based on the difference in the number of cycles. Based on the initial corrected ideal power consumption and power consumption correction coefficient, the intermediate correction value is obtained; Collect reference values ​​of actual power consumption of healthy batteries of the same model under the same average output power; Calculate the deviation rate between the intermediate correction value and the actual power consumption reference value; If the deviation rate is within the allowable deviation range, the intermediate correction value will be used as the ideal unit power consumption. If the deviation rate exceeds the allowable deviation range, the parameters of the linear fitting mapping model are fine-tuned based on the actual power consumption reference value, and the ideal unit power consumption is recalculated.

9. A battery health status online detection system, characterized in that, include: The first acquisition module (1) is used to acquire the current power level of the target battery, the first time point of the most recent end of charging, and the first remaining power level. The second acquisition module (2) is used to acquire the second time point of the initial discharge of the target battery and the second remaining charge. The first calculation module (3) is used to calculate the first time difference between the second time node and the first time node, and the first power difference between the second remaining power and the first remaining power. The third acquisition module (4) is used to acquire the first unit power consumption based on the first time difference and the first power difference; The judgment module (5) is used to determine whether the first unit power consumption exceeds the first power consumption threshold; If the first determination module (6) exceeds the limit, the first determination module (6) is used to directly determine that the target battery is in an abnormal health state. If the second calculation module (7) does not exceed the limit, the second calculation module (7) is used to calculate the second time difference between the current time node and the second time node, and the second power difference between the current power and the second remaining power. The fourth acquisition module (8) is used to acquire the second unit power consumption based on the second time difference and the second power difference; The fifth acquisition module (9) is used to obtain the ideal unit power consumption by combining the average output power from the second time node to the current time node; The second determination module (10) is used to determine the battery health status by the difference between the second unit power consumption and the ideal unit power consumption.

Images