A Novel Dynamic Verification Method for Security Boundaries of Energy Storage Devices Based on Multi-Source Data

By generating incremental capacity curves and analyzing acoustic emission signals, the safety boundary of energy storage devices is dynamically corrected, solving the problem of lag in battery safety boundary assessment in existing technologies and improving the safety and adaptability of the charging process.

CN123066601APending Publication Date: 2026-11-10STATE GRID GANSU ELECTRIC POWER RESEARCH INSTITUTE
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611463574.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-20
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

Existing energy storage devices suffer from lagging safety boundary assessments and insufficient adaptability to adjusting operating parameters, making it difficult to accurately identify the dynamic migration characteristics of battery safety operating boundaries. This results in charging control strategies failing to match the current battery state in a timely manner.

Method used

By collecting the terminal voltage and charging current of the battery cluster, an incremental capacity curve is generated, the peak voltage value and half width at half maximum (WHM) of the main peak are extracted, and the structural fatigue equivalent is calculated by combining the time-frequency analysis of the acoustic emission signal. The charging limit value is dynamically corrected, and the terminal voltage response curve is monitored during power-limited operation to achieve dynamic verification of the safety boundary.

Benefits of technology

It improves the accuracy of safety boundary identification during the charging process of energy storage devices, reduces the charging safety risks caused by the lag in safety boundary judgment, and enhances the adaptability and reliability of battery operation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123066601A_ABST
    Figure CN123066601A_ABST
Patent Text Reader

Abstract

This invention discloses a novel dynamic verification method for the safety boundary of energy storage devices based on multi-source data, relating to the field of safety boundary technology. It addresses the problem of delayed safety boundary assessment in energy storage devices due to the inability of charging control strategies to match the current battery state in a timely manner. The method collects terminal voltage and charging current during the charging process of the energy storage system, generates an incremental capacity curve, and extracts the peak voltage value and half-width at half-maximum (WHM). It then calculates the peak offset velocity and peak width expansion acceleration to trigger a safety boundary correction procedure. Furthermore, it collects acoustic emission signals from the battery casing surface and performs time-frequency analysis, extracting ringing counts and energy accumulation rate to calculate the structural fatigue equivalent. Based on the structural fatigue equivalent, it derating the charging limit value and calculates the polarization voltage recovery slope based on the terminal voltage response curve during power-limited operation to verify the safety boundary recovery state. This improves the accuracy of safety boundary identification and dynamic verification for energy storage devices.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of security boundary technology, and more specifically, to a novel dynamic verification method for the security boundary of an energy storage device based on multi-source data. Background Technology

[0002] As the application scale of new energy storage devices in power regulation, renewable energy consumption and energy storage power stations continues to expand, energy storage devices need to maintain high operational reliability during frequent charging and discharging. Existing energy storage safety management technologies usually collect external operating parameters such as voltage, current and temperature during battery operation and combine them with preset safety thresholds to constrain the charging and discharging process in order to avoid the equipment being in an abnormal operating state for a long time.

[0003] The existing technology has the following shortcomings: Currently, in actual operation, the electrochemical reaction state and structural performance of energy storage devices continuously evolve with the cyclic charging and discharging process. Existing management methods based on fixed safety thresholds are difficult to reflect changes in the actual load-bearing capacity of the battery and cannot accurately identify the dynamic migration characteristics of the battery's safe operating boundary as the state changes. This results in the charging control strategy being unable to match the current state of the battery in a timely manner, causing the safety boundary assessment of energy storage devices to be lagging and the adaptability of operating parameter adjustments to be insufficient. Therefore, a novel dynamic verification method for the safety boundary of energy storage devices based on multi-source data is proposed.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a novel dynamic verification method for the safety boundary of energy storage devices based on multi-source data. This method utilizes a dynamic correction and recovery verification technique for the safety boundary, which integrates electrochemical state characteristics and structural state characteristics, to address the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a novel dynamic verification method for the security boundary of an energy storage device based on multi-source data, comprising the following steps: Step S1: When the energy storage system enters the charging state, the terminal voltage and charging current of the battery cluster are collected, differential operation is performed on the terminal voltage and charging current to generate an incremental capacity curve, the main reaction peak is identified in the incremental capacity curve and the peak voltage value and the full width at half maximum of the main peak are extracted. Step S2: Analyze the peak position offset speed based on the peak position voltage value of the main peak, analyze the peak width expansion acceleration based on the full width at half maximum of the main peak, and determine whether the safety boundary correction procedure is triggered by combining the peak position offset speed and the peak width expansion acceleration. Step S3: In the safety boundary correction procedure, the acoustic emission signal on the surface of the battery casing is collected, the time-frequency analysis of the acoustic emission signal is performed to extract the ringing count and energy accumulation rate, and the structural fatigue equivalent is calculated. The original charging limit value of the battery is derating corrected using the structural fatigue equivalent, and the corrected charging limit value is output. Step S4: Control the battery to perform power-limited operation according to the corrected charging limit value. During the power-limited operation, monitor the battery's terminal voltage response curve, calculate the polarization voltage recovery slope based on the terminal voltage response curve, and determine whether the safety boundary has been recovered.

[0007] In a preferred embodiment, in step S1, when the energy storage system receives a charging control command and puts the battery cluster into a charging state, the terminal voltage and charging current of the battery cluster are synchronously collected according to a preset sampling period. Terminal voltage is the actual potential difference between the positive and negative output terminals of the battery cluster. Charging current is the amount of charge transferred into the battery cluster per unit time. For two consecutive sampling times, the terminal voltage at the previous sampling time and the terminal voltage at the current sampling time are read and subtracted to obtain the terminal voltage increment; The charge increment of the input battery cluster is calculated based on the charging current at adjacent sampling times; The ratio of charge increment to terminal voltage increment is used as the incremental capacity value at each sampling time.

[0008] In a preferred embodiment, in step S1, the terminal voltage is used as the horizontal axis and the corresponding incremental capacity value is used as the vertical axis to establish an incremental capacity curve in order of terminal voltage from low to high. Local peak search is performed on the incremental capacity curve. The incremental capacity value corresponding to a certain sampling time is compared with the incremental capacity values ​​corresponding to the previous and next sampling times to obtain candidate response peaks. Compare the incremental capacity values ​​corresponding to all candidate reaction peaks, and determine the candidate reaction peak with the largest incremental capacity value as the main reaction peak. The terminal voltage corresponding to the main reaction peak is determined as the peak position voltage value of the main peak. Read the maximum incremental capacity value corresponding to the main reaction peak and obtain the left half-height position and the right half-height position; Subtract the terminal voltage corresponding to the left half-height position from the terminal voltage corresponding to the right half-height position to obtain the full width at half-height of the main peak.

[0009] In a preferred embodiment, in step S2, the difference between the peak voltage value of the main peak in the current analysis period and the peak voltage value of the main peak in the previous analysis period is calculated and divided by the time interval between the two analysis periods to obtain the peak position shift speed. The rate of change of the previous peak width and the rate of change of the current peak width are determined based on the full width at half maximum (FWHM) of the main peak corresponding to three consecutive analysis periods. Subtract the previous peak width change rate from the current peak width change rate, and divide by the time interval between the two peak width change rates to obtain the peak width broadening acceleration. The peak position shift velocity and peak width broadening acceleration were normalized to obtain peak position shift characterization values ​​and peak width broadening characterization values, respectively.

[0010] In a preferred embodiment, in step S2, the peak position shift characterization value and the peak width broadening characterization value are multiplied by a preset weighting coefficient and then weighted and summed to obtain the safety boundary change correlation value. When the associated value of the change in the security boundary reaches the preset security boundary correction trigger threshold, the security boundary correction procedure is triggered. The safety boundary correction procedure will not be triggered if the associated value of the safety boundary change does not reach the preset safety boundary correction trigger threshold.

[0011] In a preferred embodiment, in step S3, after the safety boundary correction procedure is triggered, the acoustic emission signal on the surface of the battery casing is collected, and the continuously recorded acoustic emission signal is segmented according to a preset time window. Acoustic emission signals are transient elastic wave signals generated and propagated to the surface of the battery casing when stress is released in a localized part of the battery casing. Perform a short-time Fourier transform on the acoustic emission signal within each time window to convert the acoustic emission signal in the time domain into the frequency amplitude in the frequency domain; The frequency energy is obtained by squaring the frequency amplitude corresponding to each frequency. The total frequency energy corresponding to each frequency within the same time window is obtained by summing the frequency energies of each frequency within the same time window.

[0012] In a preferred embodiment, in step S3, when the total frequency energy corresponding to the time window is greater than the preset energy judgment benchmark value, the corresponding time window is marked as the acoustic emission activity time window; Otherwise, the corresponding time window is not marked; The time windows that are consecutively marked as acoustic emission activity time windows are merged into a single acoustic emission event, and the start and end times of each acoustic emission event are determined. Ring counts are extracted based on acoustic emission events. The ring counts corresponding to all acoustic emission events within the same analysis time interval are accumulated to obtain the ring count corresponding to the analysis time interval. The ringing count reflects the cumulative number of effective acoustic emission oscillations generated by the battery casing within the analysis time interval.

[0013] In a preferred embodiment, in step S3, the acoustic emission energy is calculated based on the acoustic emission signal corresponding to the acoustic emission event, and the acoustic emission energy corresponding to all acoustic emission events within the same analysis time interval is accumulated to obtain the cumulative acoustic emission energy corresponding to the analysis time interval. The energy accumulation rate is obtained by subtracting the cumulative acoustic emission energy of the previous analysis time interval from the cumulative acoustic emission energy of the current analysis time interval, and dividing by the time interval between the current analysis time interval and the previous analysis time interval. Calculate the structural fatigue equivalent based on ring count and energy accumulation rate; Calculate the derating factor based on the structural fatigue equivalent, and read the original charging limit value of the battery. The original charging limit value is the maximum charging power that the battery is allowed to use before the safety boundary correction procedure is executed. Multiply the original charging limit value by the derating factor to obtain the corrected charging limit value.

[0014] In a preferred embodiment, in step S4, the charging converter is controlled to adjust the charging power of the input battery pack so that the battery pack performs power-limited operation at a charging power not exceeding the modified charging limit value. During power-limited operation, the terminal voltage of the battery cluster is continuously collected, and the continuously collected terminal voltages are arranged according to the sampling time to form a terminal voltage response curve. The terminal voltage at the moment of entering power-limited operation is taken as the initial terminal voltage of the terminal voltage response curve, and the difference between the initial terminal voltage and the current terminal voltage is determined as the current polarization voltage recovery amount.

[0015] In a preferred embodiment, in step S4, the polarization voltage recovery amount corresponding to adjacent sampling times in the terminal voltage response curve is read in chronological order, and the polarization voltage recovery slope between adjacent sampling times is calculated. When the absolute value of the polarization voltage recovery slope continues to decrease within the continuous analysis time interval, and the current absolute value of the polarization voltage recovery slope is less than the polarization voltage recovery judgment threshold, the safety boundary is judged to be restored, and the safety boundary correction procedure ends. Otherwise, if the security boundary is determined not to have been restored, the security boundary correction procedure will continue to be executed.

[0016] The technical effects and advantages of this invention are as follows: This invention improves the accuracy of safety boundary identification and dynamic verification during the charging process of an energy storage system by collecting the terminal voltage and charging current of the battery cluster, generating an incremental capacity curve based on differential operations, and extracting the peak voltage value and full width at half maximum (FWHM) of the main peak. It calculates the peak offset velocity and peak width broadening acceleration to determine whether a safety boundary correction procedure is triggered. After the safety boundary correction procedure is triggered, it collects the acoustic emission signal from the battery casing surface and performs time-frequency analysis, extracting ringing counts and energy accumulation rate to calculate the structural fatigue equivalent. Based on the structural fatigue equivalent, it derating the original charging limit value. During power-limited operation, it calculates the polarization voltage recovery slope based on the terminal voltage response curve to determine the safety boundary recovery state. This improves the accuracy of safety boundary identification and dynamic verification during the charging process of energy storage devices, enables dynamic correction of the charging limit value according to changes in the battery's electrochemical and structural states, reduces the charging safety risk caused by delayed safety boundary judgment, and improves the safe operation capability of energy storage devices. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the implementation of the novel dynamic verification method for the security boundary of an energy storage device based on multi-source data, as described in this invention.

[0018] Figure 2 This is a schematic diagram illustrating the steps of the novel dynamic verification method for the security boundary of an energy storage device based on multi-source data, as described in this invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] This invention improves the accuracy of safety boundary identification and dynamic verification during the charging process of an energy storage system by collecting the terminal voltage and charging current of the battery cluster during the charging process, generating an incremental capacity curve based on differential operations, and extracting the peak voltage value and full width at half maximum (FWHM) of the main peak. It calculates the peak offset velocity and peak width broadening acceleration to determine whether a safety boundary correction procedure is triggered. After the safety boundary correction procedure is triggered, it collects the acoustic emission signal on the surface of the battery casing and performs time-frequency analysis, extracts the ringing count and energy accumulation rate to calculate the structural fatigue equivalent, and performs derating correction on the original charging limit value based on the structural fatigue equivalent. During power-limited operation, it calculates the polarization voltage recovery slope based on the terminal voltage response curve to determine the safety boundary recovery state. This invention enables dynamic correction of the charging limit value as the battery electrochemical and structural states change.

[0021] Example 1, as Figures 1 to 2As shown, a novel dynamic verification method for the security boundary of an energy storage device based on multi-source data includes the following steps: Step S1: When the energy storage system enters the charging state, the terminal voltage and charging current of the battery cluster are collected, differential operation is performed on the terminal voltage and charging current to generate an incremental capacity curve, the main reaction peak is identified in the incremental capacity curve and the peak voltage value and the full width at half maximum of the main peak are extracted. Step S2: Analyze the peak position offset speed based on the peak position voltage value of the main peak, analyze the peak width expansion acceleration based on the full width at half maximum of the main peak, and determine whether the safety boundary correction procedure is triggered by combining the peak position offset speed and the peak width expansion acceleration. Step S3: In the safety boundary correction procedure, the acoustic emission signal on the surface of the battery casing is collected, the time-frequency analysis of the acoustic emission signal is performed to extract the ringing count and energy accumulation rate, and the structural fatigue equivalent is calculated. The original charging limit value of the battery is derating corrected using the structural fatigue equivalent, and the corrected charging limit value is output. Step S4: Control the battery to perform power-limited operation according to the corrected charging limit value. During the power-limited operation, monitor the battery's terminal voltage response curve, calculate the polarization voltage recovery slope based on the terminal voltage response curve, and determine whether the safety boundary has been recovered.

[0022] The specific implementation is as follows: In step S1, after the energy storage system enters the charging state, it continuously collects the terminal voltage and charging current of the battery cluster, and generates an incremental capacity curve based on the relationship between the terminal voltage and charging current at adjacent sampling times. The main reaction peak is identified from the incremental capacity curve, and the peak voltage value and full width at half maximum (FWHM) of the main peak are extracted.

[0023] Specifically, when the energy storage system receives a charging control command and puts the battery cluster into a charging state, it synchronously collects the terminal voltage and charging current of the battery cluster according to a preset sampling period. The terminal voltage is the actual potential difference between the positive and negative output terminals of the battery cluster, which is used to reflect the voltage response of the battery cluster under the current charging state. The greater the change in terminal voltage between adjacent sampling times, the greater the voltage change of the battery cluster in the corresponding time period. The charging current is the amount of charge transferred into the battery cluster per unit time, which is used to reflect the charge input rate in the current charging process. The larger the charging current value, the more charge is input into the battery cluster per unit time.

[0024] It should be noted that the charging control command is a control command used to enable the battery cluster to perform electrical energy input according to a preset charging power or preset charging current; the energy storage system applies charging current to the battery cluster according to the charging control command, so that the battery cluster is in a charging state of continuously receiving electrical energy. The energy storage system includes the battery cluster, the charging converter, and the data acquisition device for collecting the operating electrical parameters of the battery cluster; the battery cluster is formed by combining multiple battery cells in series and parallel.

[0025] For two consecutive sampling times, read the terminal voltage at the previous sampling time and the terminal voltage at the current sampling time, and subtract the previous terminal voltage from the current terminal voltage to obtain the terminal voltage increment for that sampling interval. The charge increment of the input battery cluster is calculated based on the charging current at adjacent sampling times. Specifically, the sampling time interval between adjacent sampling times is obtained. The sampling time interval is the actual time difference between the current sampling time and the previous sampling time. When a fixed sampling period is used for acquisition, the sampling time interval is equal to the preset sampling period. When a non-fixed time interval is used for acquisition, the sampling time interval is directly calculated based on two actual sampling timestamps. Then, the charging current corresponding to two adjacent sampling times is read, the two charging currents are added together and divided by two to obtain the average charging current of the sampling time interval. Finally, the average charging current is multiplied by the sampling time interval to obtain the charge increment of the input battery cluster between adjacent sampling times. ; in, Indicates the charge increment. and This represents the charging current collected at two adjacent sampling times. Indicates the sampling time interval. This represents the index value of the sampling time, which allows the continuously collected charging current to be converted into the amount of charge change within the corresponding time period.

[0026] The ratio of charge increment to terminal voltage increment is used as the incremental capacity value at each sampling time. The incremental capacity value represents the charge change corresponding to a unit change in terminal voltage. The larger the absolute value, the larger the charge input corresponding to a smaller change in terminal voltage within the corresponding voltage change range. That is, the charge reception change of the battery cluster is more concentrated within this voltage range. The smaller the value, the larger the terminal voltage change corresponding to the same change in charge.

[0027] After obtaining the incremental capacity value corresponding to each sampling time, the terminal voltage is used as the abscissa and the corresponding incremental capacity value is used as the ordinate to establish an incremental capacity curve in order of terminal voltage from low to high. The incremental capacity curve is used to characterize the charge change characteristics corresponding to the unit terminal voltage change in different terminal voltage ranges of the battery cluster. When there is a corresponding electrochemical reaction process inside the battery cluster, the proportional relationship between the terminal voltage change and the charge change changes, causing the incremental capacity curve to form a local peak region in the corresponding terminal voltage region.

[0028] A local peak search is performed on the incremental capacity curve. The incremental capacity value corresponding to a certain sampling time is compared with the incremental capacity values ​​corresponding to the previous sampling time and the next sampling time. When the incremental capacity value at the current sampling time is greater than the incremental capacity value at the previous sampling time and is greater than or equal to the incremental capacity value at the next sampling time, the current sampling time is marked as a candidate response peak. The candidate response peak is a peak-shaped region in the incremental capacity curve that has a local maximum incremental capacity value.

[0029] Compare the incremental capacity values ​​corresponding to all candidate reaction peaks, and determine the candidate reaction peak with the largest incremental capacity value as the main reaction peak. The main reaction peak represents the local reaction peak with the highest peak value in the current incremental capacity curve. The terminal voltage corresponding to the main reaction peak is determined as the main peak position voltage value. The main peak position voltage value indicates the specific position of the main reaction peak in the terminal voltage direction. Its value corresponds to the voltage position of the current main electrochemical reaction feature. When the main peak position voltage value changes, it indicates that the voltage position corresponding to the main electrochemical reaction feature changes. This parameter is used to calculate the peak position shift rate of the main reaction peak in subsequent calculations.

[0030] Read the maximum incremental capacity value corresponding to the main reaction peak, and determine half of this maximum incremental capacity value as the half-peak height. Use this half-peak height as the unified measurement benchmark for the half-peak width of the main peak. Search from the peak position corresponding to the main reaction peak towards the side with lower terminal voltage to determine the position where the incremental capacity curve drops to the half-peak height, and read the terminal voltage corresponding to this position as the left half-peak position. Then search from the peak position corresponding to the main reaction peak towards the side with higher terminal voltage to determine the position where the incremental capacity curve drops to the half-peak height, and read the terminal voltage corresponding to this position as the right half-peak position. When the half-peak height is located between two actual sampling data points, determine the terminal voltage position corresponding to the half-peak height based on the relationship between the terminal voltage position and the incremental capacity value change at two adjacent sampling times, so as to avoid measurement position deviation caused by directly using adjacent sampling times.

[0031] Subtracting the terminal voltage corresponding to the left half-height position from the terminal voltage corresponding to the right half-height position yields the main peak half-height width. The main peak half-height width represents the range of terminal voltages covered by the main reaction peak at half the peak height. The larger the value, the wider the range of terminal voltages covered by the main reaction peak, indicating a greater degree of peak width broadening of the main reaction peak. This main peak half-height width serves as an input parameter for subsequent calculation of peak width broadening acceleration, characterizing the degree of change in the width of the main reaction peak during the charging process.

[0032] Through the above processing, the energy storage system obtains continuous corresponding terminal voltage and charging current during the execution of charging control commands, and generates an incremental capacity curve by utilizing the correspondence between the terminal voltage change and the charge input. The main reaction peak with the largest peak value is determined in the incremental capacity curve, and the peak position voltage value and the full width at half maximum (FWHM) of the main reaction peak are further extracted from the main reaction peak. The peak position voltage value is used to describe the voltage position of the main reaction peak, and the FWHM of the main peak is used to describe the voltage distribution width of the main reaction peak, providing a direct data basis for the calculation of peak position offset velocity and peak width broadening acceleration in step S2.

[0033] In step S2, the peak voltage value and half-width at half maximum (WHM) of the main peak obtained during continuous charging sampling are read, and the peak voltage value and WHM of the main peak corresponding to different charging analysis periods are sorted according to the sampling time. Two adjacent analysis periods are taken as a comparison interval, and the rate of change of the peak voltage value and the acceleration of change of the WHM of the main peak are calculated respectively to determine the degree of change of the position of the main reaction peak and the degree of change of the peak width. Based on the comprehensive results of the two change characteristics, it is determined whether the safety boundary correction procedure is triggered.

[0034] Specifically, the peak voltage value of the main peak corresponding to the current analysis period is recorded as the current main peak voltage value, and the peak voltage value of the main peak corresponding to the previous analysis period is recorded as the previous main peak voltage value. At the same time, the sampling time corresponding to the current analysis period and the previous analysis period is read, the time interval between the two analysis periods is calculated, the difference between the current main peak voltage value and the previous main peak voltage value is calculated, and the difference is divided by the time interval between the two analysis periods to obtain the peak position shift rate. The peak position shift rate reflects the rate at which the peak voltage of the main reaction peak changes position with charging time. The larger the absolute value, the more obvious the change in the voltage position of the main reaction peak per unit time, indicating that the characteristic voltage position corresponding to the main reaction process has shifted faster.

[0035] When calculating the peak position shift rate, the direction of peak position voltage change is preserved. When the current main peak voltage value is higher than the main peak voltage value of the previous analysis period, the peak position shift rate is determined as a positive value. When the current main peak voltage value is lower than the main peak voltage value of the previous analysis period, the peak position shift rate is determined as a negative value. The sign of the peak position shift rate is used to indicate whether the main reaction peak moves towards a higher or lower voltage direction. The absolute value of the peak position shift rate is used to indicate the rate of movement, thereby avoiding the loss of the peak position movement direction by only using the amount of peak position change.

[0036] Subsequently, the half-width at half-maximum (WHM) of the main peak corresponding to the current analysis period and the previous analysis period are read. The difference between the current WHM of the main peak and the WHM of the main peak in the previous analysis period is calculated to obtain the change in peak width between adjacent analysis periods. Then, the change in peak width of the current peak is compared with the change in peak width obtained in the previous analysis interval, and the peak width broadening acceleration is calculated in combination with the time interval between adjacent analysis intervals. The specific process is as follows: first, a continuous sequence of WHM of the main peak is formed according to the time order of the analysis periods. The change rate of the previous peak width and the change rate of the current peak width are determined based on the WHM of the main peak corresponding to three consecutive analysis periods. Then, the change rate of the current peak width is subtracted from the change rate of the previous peak width, and divided by the time interval between the two change rates of peak width to obtain the peak width broadening acceleration.

[0037] The peak width broadening acceleration reflects the degree of acceleration of the broadening change of the main peak half-width during the charging process. The larger the value, the faster the growth rate of the main peak half-width increases, that is, the broadening process of the main reaction peak is accelerating. When the peak width broadening acceleration is positive, it means that the growth rate of the main peak half-width is increasing. When the peak width broadening acceleration is close to zero, it means that the growth rate of the main peak half-width is changing little. When the peak width broadening acceleration is negative, it means that the growth rate of the main peak half-width is decreasing. By retaining the directional information of the peak width broadening acceleration, it is possible to distinguish between different states of continuous acceleration of broadening and decrease in broadening rate.

[0038] After obtaining the peak position shift rate and peak width broadening acceleration, the two parameters are dimensionless to enable a comprehensive judgment on the peak position shift rate and peak width broadening acceleration under a unified dimension. Specifically, the corresponding normal operation characteristic range in the pre-established historical charging data of the battery cluster is read, and the reference scales of the peak position shift rate and peak width broadening acceleration are determined respectively. The current peak position shift rate is divided by the corresponding peak position shift reference scale to obtain the peak position shift characterization value, and the current peak width broadening acceleration is divided by the corresponding peak width broadening reference scale to obtain the peak width broadening characterization value. The two characterization values ​​reflect the degree of change of the current peak position change rate and peak width broadening acceleration relative to the corresponding reference state, respectively.

[0039] A comprehensive calculation is performed based on the peak position shift characterization value and the peak width broadening characterization value. The peak position shift characterization value and the peak width broadening characterization value are multiplied by preset weighting coefficients and then weighted and summed to obtain the safety boundary change correlation value. The weighting coefficient corresponding to the peak position shift characterization value represents the influence of peak position change characteristics in safety boundary judgment, and the weighting coefficient corresponding to the peak width broadening characterization value represents the influence of peak width broadening characteristics in safety boundary judgment. Both weighting coefficients are determined based on historical battery cluster operating data or experimental calibration results. A larger safety boundary change correlation value indicates a higher degree of rapid shift of the main reaction peak accompanied by accelerated peak width broadening under the current charging state.

[0040] The safety boundary change correlation value is compared with the safety boundary correction trigger threshold. When the safety boundary change correlation value reaches the safety boundary correction trigger threshold, it is determined that the electrochemical characteristics of the current battery cluster have reached the state that requires charging limit correction, and the safety boundary correction procedure is triggered. When the associated value of the safety boundary change does not reach the safety boundary correction trigger threshold, the safety boundary correction procedure is not triggered. Instead, the terminal voltage and charging current during the subsequent charging process are obtained, the peak position voltage value and the full width at half maximum (FWHM) value of the main peak are updated, and the peak position offset velocity and peak width expansion acceleration are recalculated.

[0041] The safety boundary correction trigger threshold is determined based on the historical charging operation data and safety test data corresponding to the battery cluster. It is used to limit the extent to which the peak position shift and the peak widening acceleration need to be corrected to the original charging limit value.

[0042] In step S3, after the safety boundary correction procedure is triggered, the acoustic emission signal on the surface of the battery casing is collected and continuously recorded according to the collection time. The continuously recorded acoustic emission signal is segmented according to a preset time window. A short-time Fourier transform is performed on the acoustic emission signal in each time window to convert the time domain acoustic emission signal into the frequency domain frequency amplitude. The frequency amplitude corresponding to each frequency is squared to obtain the frequency energy. Then, the frequency energies corresponding to each frequency in the same time window are summed to obtain the total frequency energy corresponding to that time window.

[0043] The acoustic emission signal is a transient elastic wave signal generated and propagated to the surface of the battery casing when stress is released in the local structure inside the battery casing. It reflects the mechanical fluctuations generated by the stress release in the internal structure of the battery casing. The larger the amplitude, the stronger the transient elastic wave response generated by the corresponding acoustic emission activity. The total frequency energy reflects the overall energy level of the acoustic emission signal within the corresponding time window. The larger the value, the higher the energy released by the acoustic emission activity within the corresponding time window.

[0044] The energy judgment benchmark value is determined based on the acoustic emission signals collected during normal battery charging. Specifically, the acoustic emission signals of normal charging under normal charging conditions without structural fatigue are obtained. The normal charging acoustic emission signals are segmented according to the same time window as the current acoustic emission signal, and the total frequency energy corresponding to each normal charging time window is calculated using the same short-time Fourier transform process. The total frequency energy corresponding to all normal charging time windows is arranged in ascending order of value. The preset quantile value of the total frequency energy of the normal charging time window is determined as the energy judgment benchmark value. The preset quantile value is determined based on the degree of overlap between the distribution of background energy and abnormal acoustic emission energy in the normal charging acoustic emission data, so that the energy judgment benchmark value can distinguish normal background acoustic emission activity from acoustic emission activity with significantly increased energy.

[0045] The total frequency energy corresponding to each time window is compared with the energy judgment benchmark value. When the total frequency energy corresponding to the time window is greater than the energy judgment benchmark value, the time window is marked as an acoustic emission activity time window. When the total frequency energy corresponding to the time window is less than or equal to the energy judgment benchmark value, the time window is not marked. The marking status of adjacent time windows is checked in chronological order. Time windows that are consecutively marked as acoustic emission activity time windows are merged into one acoustic emission event. When there are unmarked time windows between two acoustic emission activity time windows, the two acoustic emission activity time windows are divided into different acoustic emission events. The start time and end time of each acoustic emission event are determined in this way, and the duration of the acoustic emission event is obtained by subtracting the start time from the end time.

[0046] Ring counts are extracted based on acoustic emission events. Within the continuous data corresponding to each acoustic emission event, the amplitude of the acoustic emission signal is compared sample by sample. The process of the acoustic emission signal amplitude rising from no higher than the detection amplitude reference value to higher than the detection amplitude reference value is defined as one ring count. When the acoustic emission signal amplitude drops again to no higher than the detection amplitude reference value and then rises again to higher than the detection amplitude reference value, it is included in the next ring count. The ring counts corresponding to all acoustic emission events within the same analysis time interval are accumulated to obtain the ring count corresponding to that analysis time interval.

[0047] The ringing count reflects the cumulative number of effective acoustic emission oscillations generated by the battery casing within the analysis time interval. The larger the value, the more frequently the acoustic emission oscillations occur within the corresponding analysis time interval.

[0048] It should be noted that the detection amplitude reference value is determined based on the background acoustic emission signal amplitude collected from the surface of the battery casing under normal charging conditions. This is obtained by statistically analyzing the amplitude distribution of the acoustic emission signal under normal charging conditions and determining the statistical upper limit of the background amplitude. The detection amplitude reference value is used to distinguish between normal background oscillations and acoustic emission oscillations with amplitudes higher than the normal background level. The acoustic emission energy is calculated based on the acoustic emission signal corresponding to the acoustic emission event. The amplitude of the acoustic emission signal at each sampling time within each acoustic emission event is squared, and the squared amplitude is multiplied by the corresponding sampling time interval and accumulated to obtain the acoustic emission energy corresponding to the acoustic emission event. The acoustic emission energies corresponding to all acoustic emission events within the same analysis time interval are summed to obtain the cumulative acoustic emission energy corresponding to that analysis time interval. The cumulative acoustic emission energy corresponding to the current analysis time interval is subtracted from the cumulative acoustic emission energy corresponding to the previous analysis time interval, and then divided by the time interval between the current analysis time interval and the previous analysis time interval to obtain the energy accumulation rate.

[0049] The structural fatigue equivalent is calculated based on the ringing count and energy accumulation rate. The ringing count corresponding to the current analysis time interval is divided by the ringing count baseline value to obtain the relative value of the ringing count. The energy accumulation rate corresponding to the current analysis time interval is divided by the energy accumulation rate baseline value to obtain the relative value of the energy accumulation rate. The structural fatigue equivalent is then calculated using a product coupling method. The specific calculation process is as follows: 1 is added to the relative value of the ringing count to obtain the first coupling factor. 1 is added to the relative value of the energy accumulation rate to obtain the second coupling factor. The first coupling factor and the second coupling factor are multiplied and then 1 is subtracted to obtain the structural fatigue equivalent.

[0050] It should be noted that the ringing count benchmark value is determined based on the ringing count statistics of multiple analysis time intervals under normal battery charging conditions; the energy accumulation rate benchmark value is determined based on the energy accumulation rate statistics of multiple analysis time intervals under normal battery charging conditions.

[0051] After obtaining the structural fatigue equivalent, the original charging limit value of the battery is read. The derating factor is calculated based on the structural fatigue equivalent. The structural fatigue equivalent is divided by the sum of one and the structural fatigue equivalent to obtain the structural fatigue influence coefficient. Then, the maximum derating ratio determined based on the battery structural fatigue test is read. The structural fatigue influence coefficient is multiplied by the maximum derating ratio to obtain the derating ratio. Finally, the derating factor is obtained by subtracting the derating ratio from one.

[0052] The original charging limit value is the maximum charging power allowed by the battery before the safety boundary correction procedure is executed. It is derived from the original charging control parameters corresponding to the current charging condition. The larger the value, the higher the maximum charging power allowed by the battery when no derating correction is performed. The maximum derating ratio is determined according to the maximum charging power reduction ratio allowed in the battery structure fatigue test. It is used to limit the maximum reduction ratio of the original charging limit value under the structural fatigue state. The larger the value, the higher the maximum charging power derating degree allowed.

[0053] The original charging limit value is multiplied by the derating factor to obtain the corrected charging limit value. The corrected charging limit value reflects the maximum charging power allowed by the battery under the current structural fatigue state. The smaller the value, the higher the degree of restriction on the battery charging power under the current structural fatigue state. The corrected charging limit value is output to step S4 as the upper limit of the charging power for controlling the battery to perform power-limited operation in step S4.

[0054] In step S4, the corrected charging limit value output in step S3 is read, and the corrected charging limit value is used as the upper limit of the current charging power of the battery. The charging converter is controlled to adjust the charging power of the input battery cluster, so that the battery cluster performs power-limited operation at a charging power not exceeding the corrected charging limit value.

[0055] During power-limited operation, the terminal voltage of the battery cluster is continuously collected, and the continuously collected terminal voltages are arranged according to the sampling time to form a terminal voltage response curve. The terminal voltage response curve is plotted with the sampling time on the horizontal axis and the terminal voltage at the corresponding sampling time on the vertical axis, which is used to reflect the voltage response process of the battery cluster after the charging power is reduced.

[0056] The terminal voltage at the moment of entering power-limited operation is taken as the initial terminal voltage of the terminal voltage response curve. During the power-limited operation, the terminal voltage is continuously read. When the collected terminal voltage is lower than the initial terminal voltage, the difference between the initial terminal voltage and the current terminal voltage is determined as the current polarization voltage recovery amount. The polarization voltage recovery amount reflects the drop in voltage after the battery cluster transitions from the original charging state to the power-limited operation state due to the weakening of the polarization effect. The larger the value, the more polarization voltage is released after the power-limited operation, indicating that the degree of polarization formed in the original charging state is higher.

[0057] The polarization voltage recovery values ​​at adjacent sampling times in the terminal voltage response curve are read sequentially, and the polarization voltage recovery slope between adjacent sampling times is calculated. Specifically, the polarization voltage recovery value at the current sampling time is subtracted from the polarization voltage recovery value at the previous sampling time, and then divided by the time interval between the two sampling times to obtain the polarization voltage recovery slope corresponding to the current sampling time.

[0058] The polarization voltage recovery slope reflects the rate at which the polarization voltage recovery changes over time during power-limited operation. The larger the value, the faster the polarization voltage recovery increases per unit time, indicating a more pronounced voltage drop. As power-limited operation continues, when the polarization voltage recovery slope gradually decreases and approaches zero, it indicates that the increase in polarization voltage recovery per unit time is gradually decreasing, and the voltage response curve becomes flatter, indicating that the polarization voltage formed by the battery cluster from its original charging state has been basically recovered.

[0059] Time series analysis is performed on the polarization voltage recovery slope obtained by continuous sampling. The polarization voltage recovery slope in the current analysis time interval is compared with the polarization voltage recovery slope in the previous analysis time interval. When the absolute value of the polarization voltage recovery slope in the continuous analysis time interval continues to decrease, and the absolute value of the current polarization voltage recovery slope is less than the polarization voltage recovery judgment threshold, it is determined that the terminal voltage response curve has entered the polarization voltage recovery state, and then the safety boundary recovery is judged, and the safety boundary correction procedure ends. Otherwise, if the terminal voltage is still in a state of significant change, it is determined that the safety boundary has not been restored, and the safety boundary correction procedure continues to be executed.

[0060] It should be noted that the polarization voltage recovery judgment threshold is determined based on historical test data of the battery cluster after normal charging and power-limited operation. Specifically, the terminal voltage response curve of the battery cluster is collected when the safety boundary is confirmed to be restored. The polarization voltage recovery slope corresponding to each sampling time is calculated according to the same sampling method as the current terminal voltage response curve. The polarization voltage recovery judgment threshold is determined based on the distribution of polarization voltage recovery slope when the safety boundary has been restored.

[0061] Once the safety boundary is determined to be restored, the safety boundary restoration status is output, and the current battery operating status is maintained. When the safety boundary is determined not to be restored, the corrected charging limit value is maintained as the current charging power upper limit, the battery continues to be controlled to perform power-limited operation, and the terminal voltage is continuously collected to update the terminal voltage response curve and polarization voltage recovery slope until the polarization voltage recovery slope meets the safety boundary restoration judgment condition, thereby completing the dynamic verification of the effect of the corrected charging limit value.

[0062] Finally, it should be noted that in this paper, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0063] Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0064] In this document, the singular forms “a,” “an,” and “the” may also include the plural forms unless the context clearly indicates otherwise. It should also be understood that terms such as “comprising / including” or “having” specify the presence of the stated features, integrals, steps, operations, components, parts, or combinations thereof, but do not preclude the possibility of the presence or addition of one or more other features, integrals, steps, operations, components, parts, or combinations thereof. Meanwhile, the term “and / or” as used in this specification includes any and all combinations of the associated listed items.

[0065] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referred to each other.

[0066] The above description of the disclosed embodiments will enable those skilled in the art to make or use various modifications to these embodiments. It will be readily apparent to those skilled in the art that the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A novel dynamic verification method for the security boundary of an energy storage device based on multi-source data, characterized in that: Includes the following steps: Step S1: When the energy storage system enters the charging state, the terminal voltage and charging current of the battery cluster are collected, differential operation is performed on the terminal voltage and charging current to generate an incremental capacity curve, the main reaction peak is identified in the incremental capacity curve and the peak voltage value and the full width at half maximum of the main peak are extracted. Step S2: Analyze the peak position offset speed based on the peak position voltage value of the main peak, analyze the peak width expansion acceleration based on the full width at half maximum of the main peak, and determine whether the safety boundary correction procedure is triggered by combining the peak position offset speed and the peak width expansion acceleration. Step S3: In the safety boundary correction procedure, the acoustic emission signal on the surface of the battery casing is collected, the time-frequency analysis of the acoustic emission signal is performed to extract the ringing count and energy accumulation rate, and the structural fatigue equivalent is calculated. The original charging limit value of the battery is derating corrected using the structural fatigue equivalent, and the corrected charging limit value is output. Step S4: Control the battery to perform power-limited operation according to the corrected charging limit value. During the power-limited operation, monitor the battery's terminal voltage response curve, calculate the polarization voltage recovery slope based on the terminal voltage response curve, and determine whether the safety boundary has been recovered.

2. The novel dynamic verification method for the security boundary of an energy storage device based on multi-source data according to claim 1, characterized in that: In step S1, when the energy storage system receives a charging control command and puts the battery cluster into a charging state, the terminal voltage and charging current of the battery cluster are synchronously collected according to a preset sampling period. Terminal voltage is the actual potential difference between the positive and negative output terminals of the battery cluster. Charging current is the amount of charge transferred into the battery cluster per unit time. For two consecutive sampling times, the terminal voltage at the previous sampling time and the terminal voltage at the current sampling time are read and subtracted to obtain the terminal voltage increment; The charge increment of the input battery cluster is calculated based on the charging current at adjacent sampling times; The ratio of charge increment to terminal voltage increment is used as the incremental capacity value at each sampling time.

3. The novel dynamic verification method for the security boundary of an energy storage device based on multi-source data according to claim 2, characterized in that: In step S1, the terminal voltage is used as the horizontal axis and the corresponding incremental capacity value is used as the vertical axis to establish an incremental capacity curve in order of terminal voltage from low to high. Local peak search is performed on the incremental capacity curve. The incremental capacity value corresponding to a certain sampling time is compared with the incremental capacity values ​​corresponding to the previous and next sampling times to obtain candidate response peaks. Compare the incremental capacity values ​​corresponding to all candidate reaction peaks, and determine the candidate reaction peak with the largest incremental capacity value as the main reaction peak. The terminal voltage corresponding to the main reaction peak is determined as the peak position voltage value of the main peak. Read the maximum incremental capacity value corresponding to the main reaction peak and obtain the left half-height position and the right half-height position; Subtract the terminal voltage corresponding to the left half-height position from the terminal voltage corresponding to the right half-height position to obtain the full width at half-height of the main peak.

4. The novel dynamic verification method for the security boundary of an energy storage device based on multi-source data according to claim 3, characterized in that: In step S2, the difference between the peak voltage value of the main peak in the current analysis period and the peak voltage value of the main peak in the previous analysis period is calculated and divided by the time interval between the two analysis periods to obtain the peak position shift rate. The rate of change of the previous peak width and the rate of change of the current peak width are determined based on the full width at half maximum (FWHM) of the main peak corresponding to three consecutive analysis periods. Subtract the previous peak width change rate from the current peak width change rate, and divide by the time interval between the two peak width change rates to obtain the peak width broadening acceleration. The peak position shift velocity and peak width broadening acceleration were normalized to obtain peak position shift characterization values ​​and peak width broadening characterization values, respectively.

5. The novel dynamic verification method for the security boundary of an energy storage device based on multi-source data according to claim 4, characterized in that: In step S2, the peak position shift characterization value and the peak width broadening characterization value are multiplied by preset weighting coefficients and then weighted and summed to obtain the safety boundary change correlation value. When the associated value of the change in the security boundary reaches the preset security boundary correction trigger threshold, the security boundary correction procedure is triggered. The safety boundary correction procedure will not be triggered if the associated value of the safety boundary change does not reach the preset safety boundary correction trigger threshold.

6. The novel dynamic verification method for the security boundary of an energy storage device based on multi-source data according to claim 1, characterized in that: In step S3, after the safety boundary correction procedure is triggered, the acoustic emission signal on the surface of the battery casing is collected, and the continuously recorded acoustic emission signal is segmented according to a preset time window. Acoustic emission signals are transient elastic wave signals generated and propagated to the surface of the battery casing when stress is released in a localized part of the battery casing. Perform a short-time Fourier transform on the acoustic emission signal within each time window to convert the acoustic emission signal in the time domain into the frequency amplitude in the frequency domain; The frequency energy is obtained by squaring the frequency amplitude corresponding to each frequency. The total frequency energy corresponding to each frequency within the same time window is obtained by summing the frequency energies of each frequency within the same time window.

7. The novel dynamic verification method for the security boundary of an energy storage device based on multi-source data according to claim 6, characterized in that: In step S3, when the total frequency energy corresponding to the time window is greater than the preset energy judgment benchmark value, the corresponding time window is marked as the acoustic emission activity time window; Otherwise, the corresponding time window is not marked; The time windows that are consecutively marked as acoustic emission activity time windows are merged into a single acoustic emission event, and the start and end times of each acoustic emission event are determined. Ring counts are extracted based on acoustic emission events. The ring counts corresponding to all acoustic emission events within the same analysis time interval are accumulated to obtain the ring count corresponding to the analysis time interval. The ringing count reflects the cumulative number of effective acoustic emission oscillations generated by the battery casing within the analysis time interval.

8. The novel dynamic verification method for the security boundary of an energy storage device based on multi-source data according to claim 6, characterized in that: In step S3, the acoustic emission energy is calculated based on the acoustic emission signal corresponding to the acoustic emission event, and the acoustic emission energy corresponding to all acoustic emission events within the same analysis time interval is accumulated to obtain the cumulative acoustic emission energy corresponding to the analysis time interval. The energy accumulation rate is obtained by subtracting the cumulative acoustic emission energy of the previous analysis time interval from the cumulative acoustic emission energy of the current analysis time interval, and dividing by the time interval between the current analysis time interval and the previous analysis time interval. Calculate the structural fatigue equivalent based on ring count and energy accumulation rate; Calculate the derating factor based on the structural fatigue equivalent, and read the original charging limit value of the battery. The original charging limit value is the maximum charging power that the battery is allowed to use before the safety boundary correction procedure is executed. Multiply the original charging limit value by the derating factor to obtain the corrected charging limit value.

9. The novel dynamic verification method for the security boundary of an energy storage device based on multi-source data according to claim 8, characterized in that: In step S4, the charging converter is controlled to adjust the charging power of the input battery pack so that the battery pack operates at a power limit not exceeding the modified charging limit value. During power-limited operation, the terminal voltage of the battery cluster is continuously collected, and the continuously collected terminal voltages are arranged according to the sampling time to form a terminal voltage response curve. The terminal voltage at the moment of entering power-limited operation is taken as the initial terminal voltage of the terminal voltage response curve, and the difference between the initial terminal voltage and the current terminal voltage is determined as the current polarization voltage recovery amount.

10. The novel dynamic verification method for the security boundary of an energy storage device based on multi-source data according to claim 9, characterized in that: In step S4, the polarization voltage recovery amount corresponding to adjacent sampling times in the terminal voltage response curve is read in chronological order, and the polarization voltage recovery slope between adjacent sampling times is calculated. When the absolute value of the polarization voltage recovery slope continues to decrease within the continuous analysis time interval, and the current absolute value of the polarization voltage recovery slope is less than the polarization voltage recovery judgment threshold, the safety boundary is judged to be restored, and the safety boundary correction procedure ends. Otherwise, if the security boundary is determined not to have been restored, the security boundary correction procedure will continue to be executed.