A method for periodic reliability testing of storage batteries

Abstract

This invention discloses a method for periodic reliability testing of storage batteries, relating to the field of storage battery testing. The method includes: acquiring battery operating status data within a testing cycle, extracting features by dividing continuous sub-intervals, and generating a non-steady-state mutation response rate sequence; constructing a behavioral perturbation flux function, calculating health indicators, and forming a set of indicators with a single physical meaning; extracting health indicators from adjacent cycles and calculating a covariant structure factor matrix; constructing a degradation potential energy function by combining health indicators, covariant structure factors, and perturbation flux to measure state evolution; using the potential energy functions of three consecutive cycles to extrapolate future reliability intervals; generating a feedback correction factor, dynamically updating parameter paths, and achieving closed-loop adaptive control. By constructing a periodic behavioral perturbation flux function, fusing health indicators and covariant structure factors to construct a degradation potential energy function, and introducing a closed-loop feedback correction mechanism, the method achieves enhanced perception of battery structural perturbations, characterization of state evolution intensity, and adaptive updating of detection parameters.

Description

Technical Field

[0001] This invention relates to the field of battery testing, specifically a method for periodically testing the reliability of batteries. Background Technology

[0002] As a critical energy storage and power supply unit, batteries are widely used in electric vehicles, energy storage systems, uninterruptible power supplies (UPS), and industrial control, among other fields. Their operating status directly affects the stability, safety, and lifespan of the system. Therefore, periodic reliability testing of batteries has become an important means of ensuring their long-term operational performance.

[0003] Current battery testing methods primarily rely on periodic data collection of basic parameters such as voltage, current, internal resistance, and temperature, combined with simple empirical models or statistical algorithms to estimate health status. Common approaches include determining the degree of degradation through open-circuit voltage and static internal resistance, or predicting lifespan based on capacity decline trends. However, these methods generally suffer from the following problems:

[0004] First, the lack of structured modeling for non-steady-state operation processes makes it impossible to effectively capture the abrupt response and abnormal evolution characteristics in the dynamic charging and discharging process, resulting in detection results that are easily affected by transient disturbances and have low accuracy.

[0005] Secondly, some methods simply average multiple health parameters, failing to reflect the coupling relationship between physical quantities at different evolution stages, lacking structural co-modeling between parameters, and easily ignoring deep-seated performance degradation characteristics.

[0006] Furthermore, most current methods use static judgment criteria, which cannot combine historical trends with current fluctuations to predict future states. They lack a forward-looking reliability assessment mechanism, thus limiting the strategy optimization capabilities of battery management systems.

[0007] Furthermore, the lack of a complete periodic feedback adjustment mechanism makes it impossible to correct key model parameters based on detection errors, resulting in the detection system struggling to maintain stability and adaptability during long-term operation.

[0008] Therefore, there is an urgent need to propose a reliability periodic testing method that can perform multi-dimensional modeling of the unsteady evolution process of batteries, integrate mutation response analysis and structural co-reasoning, and has prediction and feedback capabilities, thereby improving the integrity, adaptability and evolution capability of the testing process. Summary of the Invention

[0009] Based on the shortcomings of the prior art described above, the purpose of this invention is to provide a method for periodically testing the reliability of storage batteries, so as to solve the above-mentioned technical problems.

[0010] To achieve the above objectives, the present invention provides the following technical solution: a method for periodically testing the reliability of a storage battery, comprising:

[0011] The battery's operating status data is acquired during the detection period. The time axis is divided into a preset number of continuous sub-intervals. State features are extracted from each sub-interval and fused to generate a non-steady-state mutation response rate sequence.

[0012] A perturbation flux function characterizing the fluctuation intensity and dissipative structure of the battery's operating state within a cycle is constructed based on the non-steady-state mutation response rate sequence.

[0013] Health indicators are calculated based on behavioral perturbation flux functions, and a set of health indicators with a single physical meaning is constructed to characterize the state characteristics within the detection period.

[0014] Based on the set of health indicators, health indicators from the previous and current testing periods are extracted, and the health indicator covariance structure factor of each pair of health indicators in adjacent periods is calculated to construct a health indicator covariance structure factor matrix.

[0015] Based on the current set of health indicators, the covariant structure factor of health indicators, and the behavioral perturbation flux function, a degradation potential function is constructed to measure the overall structural characteristics of the cyclical state evolution.

[0016] Based on the degradation potential energy function values ​​of three adjacent detection cycles, calculate the lower and upper bounds of reliability for future detection cycles, and deduce the range of reliability changes in future detection cycles.

[0017] The difference between the reliability upper bound of future detection cycles and the critical health indicators triggered in the current cycle is calculated to generate a feedback correction factor. Based on the feedback correction factor, the parameter paths of the covariance structure factor and the degradation potential function of the health indicators are updated.

[0018] The present invention is further configured such that generating the non-steady-state mutation response rate sequence includes:

[0019] During the detection period, the battery's voltage, current, temperature, internal resistance, and corresponding time stamp data are acquired, and the time axis within the detection period is divided into a preset number of continuous sub-intervals.

[0020] For each sub-interval, extract the voltage boundary value, current boundary value, internal resistance boundary value, and temperature sequence data within that interval. Calculate the voltage boundary difference, current boundary difference, internal resistance boundary difference, and temperature change rate based on the various boundary values.

[0021] Based on the voltage boundary difference, current boundary difference, internal resistance boundary difference, and temperature change rate, a single numerical representation of the catastrophe structure quantity is constructed. The catastrophe structure quantities corresponding to each sub-interval are arranged in sequence to generate a non-steady-state catastrophe response rate sequence, which characterizes the segmented state fluctuation characteristics within the detection period.

[0022] The present invention is further configured such that the construction of the behavioral perturbation flux function includes:

[0023] Based on the non-steady-state abrupt change response rate sequence within the detection period, and combined with the internal resistance change rate, current disturbance amplitude, and temperature change rate of each sub-interval, a sub-interval disturbance expansion coefficient sequence is constructed.

[0024] Based on the voltage boundary difference, current boundary difference, and internal resistance boundary of each sub-interval, construct the energy consumption coupling structure quantity sequence;

[0025] By combining the perturbation expansion coefficient sequence with the energy dissipation coupled structural quantity sequence, a behavioral perturbation flux function for the detection period is generated, which characterizes the local mutation amplification features and the full-cycle structural dissipation intensity within the detection period.

[0026] The present invention is further configured such that the construction of the health indicator set includes:

[0027] Based on the behavioral perturbation flux function, and combined with five physical quantities during the detection period, namely, capacity change, internal resistance evolution rate, temperature extreme value shift, voltage hysteresis behavior and hysteresis energy loss, a set of health indicators with corresponding single physical meanings is constructed.

[0028] The set of health indicators includes indicators of capacity degradation, internal resistance increase, temperature deviation, voltage response hysteresis, and hysteresis loss, which correspond to the state characteristics within the detection period.

[0029] The present invention is further configured such that constructing the covariant structural factor matrix of health indicators includes:

[0030] Extract the corresponding health indicators from the previous detection period and the current detection period, perform difference processing on the evolution and changes of the same type of health indicators in adjacent periods, and form a set of periodic indicator differences;

[0031] For any two different types of indicators in the health indicator set, calculate the evolutionary response dissipation term between any two different types of indicators in the current cycle, and describe the asymmetric response relationship between indicators in the cycle evolution process;

[0032] Based on the combination relationship between the periodic index difference set and the response dissipation term, the health index covariance structure factor, which measures the intensity of the co-evolution between any two indicators in adjacent detection periods, is calculated, and the health index covariance structure factor matrix is ​​constructed.

[0033] The present invention is further configured such that the construction of the degradation potential function includes:

[0034] Based on the health indicators in the health indicator set, and combined with the covariance structure factor of the health indicators, a weighted coupling degree term is constructed between the indicators to reflect the intensity of the interaction influence of multidimensional health indicators.

[0035] Based on the weighted coupling degree term, combined with the higher-order nonlinear deviation terms of each health indicator, a higher-order driving component reflecting the abnormal amplitude of a single indicator is constructed.

[0036] By coupling the weighted coupling degree term with the higher-order driving component through a nonlinear function, a comprehensive expression for the health index is obtained.

[0037] By introducing a behavioral perturbation flux function as a regulating factor, the comprehensive expression of health indicators is nonlinearly adjusted to achieve the overall construction of the degradation potential function.

[0038] The present invention is further configured such that the reliability change range in the projected future preset period includes:

[0039] The degradation potential energy function values ​​within three consecutive detection cycles are extracted. Based on this sequence, a high-order nonlinear dynamic prediction model is constructed to deduce the predicted degradation potential energy values ​​for future preset cycles.

[0040] By combining the fluctuation characteristics of predicted values ​​and historical degradation potential functions, a reliability boundary adjustment factor is generated to reflect the range of uncertainty in the evolution of system state.

[0041] Based on the predicted degradation potential energy and the reliability boundary adjustment factor, the lower and upper limits of reliability for future periods are calculated to form the reliability variation range.

[0042] The present invention is further configured such that the parameter path for updating the covariant structure factor and degradation potential function of the health indicator includes:

[0043] Based on the reliability upper bound of future testing cycles and the critical health indicators triggered in the current testing cycle, a health indicator difference matrix is ​​constructed.

[0044] Based on the health indicator difference matrix and the current period health indicator covariance structure factor matrix, a feedback correction factor is generated;

[0045] The parameter paths of the covariant structure factor matrix and the degradation potential function of the current cycle health indicators are adjusted based on the feedback correction factor.

[0046] The detection process is adjusted to achieve periodic closed-loop feedback updates and dynamic adaptive evolution.

[0047] The present invention is further configured such that the generation of the feedback correction factor includes:

[0048] A nonlinear adjustment function is constructed based on the covariant structural factor matrix of health indicators and the difference matrix of health indicators.

[0049] By combining the product mapping of covariant structure factors and difference matrices with exponential decay weighting, asymmetric feedback correction with proximity structure is achieved.

[0050] The feedback correction factor, as a multidimensional parameter path adjustment factor, dynamically adjusts the parameter evolution path of the covariant structural factor matrix and the degradation potential function of health indicators.

[0051] The present invention is further configured such that the periodic closed-loop feedback update and dynamic adaptive evolution of the detection process by adjusting includes:

[0052] The change path of the off-diagonal elements of the covariant structure factor matrix of health indicators is defined as the first type of parameter path, and the change path of the higher-order nonlinear driving components corresponding to each health indicator in the degradation potential function is defined as the second type of parameter path.

[0053] Based on the magnitude and directionality of the feedback correction factor, asymmetric perturbation gain adjustment is performed on the first type of parameter path;

[0054] The feedback correction factor is introduced into the second type of parameter path, and a feedback modulation weight is applied to the nonlinear offset term corresponding to each health indicator.

[0055] By utilizing the joint regulation results of the first type of parameter path and the second type of parameter path, the co-evolutionary relationship of health indicators and the expression structure of the degradation potential function are synchronously reconstructed, thereby realizing the parameter stability domain of the detection process in the multi-period structural evolution.

[0056] This invention provides a method for periodic reliability testing of a storage battery. The method involves acquiring the battery's operating status data within a testing period, dividing the time axis into a predetermined number of continuous sub-intervals, extracting state features from each sub-interval, and fusing these features to generate a non-steady-state mutation response rate sequence. Based on the non-steady-state mutation response rate sequence, a behavioral perturbation flux function characterizing the fluctuation intensity and dissipative structure of the battery's operating status within the period is constructed. Health indicators are calculated based on the behavioral perturbation flux function, constructing a set of health indicators with a single physical meaning to characterize the state features within the testing period. Based on the health indicator set, health indicators from the previous testing period and the current testing period are extracted, and the correlation between each pair of health indicators in adjacent periods is calculated. The covariant structure factor of health indicators is used to construct a covariant structure factor matrix of health indicators. Based on the set of health indicators, the covariant structure factor of health indicators, and the behavioral perturbation flux function of the current period, a degradation potential function is constructed to measure the overall structural characteristics of the periodic state evolution. According to the degradation potential function values ​​of three adjacent detection periods, the lower and upper bounds of reliability for future detection periods are calculated, and the reliability change range in future detection periods is deduced. The difference between the upper bound of reliability for future detection periods and the critical health indicator triggered in the current period is calculated to generate a feedback correction factor. Based on the feedback correction factor, the parameter paths of the covariant structure factor of health indicators and the degradation potential function are updated. The beneficial effects include:

[0057] 1. Construct a multidimensional perturbation flux function to enhance structural perception capability: By combining multiple structural parameters such as abrupt response rate, current perturbation, internal resistance rate and energy consumption coupling, a periodic behavior perturbation flux function is established to reflect the intensity of perturbation and dissipation mechanism within the period, providing a structural dynamic basis for health status measurement.

[0058] 2. Constructing a degradation potential function to characterize the overall evolution intensity of the system: By integrating the set of health indicators, covariant structural factors and behavioral perturbation fluxes, a high-order nonlinear degradation potential function is formed, which can measure the degree of comprehensive structural shift of the system within the detection period, laying the foundation for reliability prediction.

[0059] 3. Introduce a closed-loop feedback mechanism to improve adaptability: By analyzing the difference between the upper bound of future periodic reliability and the current periodic critical health index, a feedback correction factor is generated, the covariant structure factor and degradation potential energy parameter path are dynamically updated, and a periodic evolution closed loop is constructed to realize the dynamic self-adjustment and robustness enhancement of the detection model.

[0060] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0061] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0062] Figure 1 The flowchart illustrates a method for periodically testing the reliability of a storage battery, as shown in an exemplary embodiment of the present invention. Detailed Implementation

[0063] The embodiments of the present invention will be described below with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention and not for limiting the scope of protection of the present invention.

[0064] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0065] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the invention. However, it will be apparent to those skilled in the art that embodiments of the invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the invention.

[0066] Example 1:

[0067] A method for periodically testing the reliability of a storage battery, such as Figure 1 As shown, it includes:

[0068] The battery's operating status data is acquired during the detection period. The time axis is divided into a preset number of continuous sub-intervals. State features are extracted from each sub-interval and fused to generate a non-steady-state mutation response rate sequence.

[0069] A perturbation flux function characterizing the fluctuation intensity and dissipative structure of the battery's operating state within a cycle is constructed based on the non-steady-state mutation response rate sequence.

[0070] Health indicators are calculated based on behavioral perturbation flux functions, and a set of health indicators with a single physical meaning is constructed to characterize the state characteristics within the detection period.

[0071] Based on the set of health indicators, health indicators from the previous and current testing periods are extracted, and the health indicator covariance structure factor of each pair of health indicators in adjacent periods is calculated to construct a health indicator covariance structure factor matrix.

[0072] Based on the current set of health indicators, the covariant structure factor of health indicators, and the behavioral perturbation flux function, a degradation potential function is constructed to measure the overall structural characteristics of the cyclical state evolution.

[0073] Based on the degradation potential energy function values ​​of three adjacent detection cycles, calculate the lower and upper bounds of reliability for future detection cycles, and deduce the range of reliability changes in future detection cycles.

[0074] The difference between the reliability upper bound of future detection cycles and the critical health indicators triggered in the current cycle is calculated to generate a feedback correction factor. Based on the feedback correction factor, the parameter paths of the covariance structure factor and the degradation potential function of the health indicators are updated.

[0075] The present invention is further configured such that generating the non-steady-state mutation response rate sequence includes:

[0076] Within the detection cycle, the battery's voltage, current, temperature, and internal resistance data, along with their corresponding time stamp data, are acquired. The time axis within the detection cycle is divided into a preset number of continuous sub-intervals. Specifically, within one detection cycle... Within this scope, time series data acquisition includes: voltage series. Current sequence Temperature sequence internal resistance sequence time stamp Divide the period into There are n continuous subintervals, each subinterval denoted as . ,satisfy: , ;

[0077] For each sub-interval, extract voltage boundary values, current boundary values, internal resistance boundary values, and temperature sequence data within that interval. Calculate the voltage boundary difference, current boundary difference, internal resistance boundary difference, and temperature change rate based on these boundary values. Specifically, for each sub-interval... Calculate the voltage boundary difference, current boundary difference, internal resistance boundary difference, and temperature change rate, where the voltage boundary difference is: Current boundary difference: Internal resistance boundary difference: Temperature change rate: ,in, and These are the upper and lower boundary values ​​of the voltage. and These are the upper and lower boundary values ​​of the current. and These are the upper and lower boundary values ​​of the internal resistance. and sub-interval The highest and lowest temperatures;

[0078] A single numerical representation of abrupt structural quantities is constructed based on voltage boundary difference, current boundary difference, internal resistance boundary difference, and temperature change rate. The abrupt structural quantities corresponding to each sub-interval are sequentially arranged to generate a non-steady-state abrupt response rate sequence, characterizing the segmented state fluctuation characteristics within the detection period. Specifically, the first... The number of mutation structures in each sub-region is ,in, To adjust the abrupt coupling complexity of sub-intervals.

[0079] The mutation structure quantity of each sub-interval Arranged in chronological order, this constitutes the non-steady-state mutation response rate sequence for the detection period: The number of mutant structures constructed It possesses a sensitive ability to detect high-frequency disturbances and multi-variable coordinated jumps, making it suitable for dynamic modeling of unsteady systems.

[0080] The present invention is further configured such that the construction of the behavioral perturbation flux function includes:

[0081] Based on the non-steady-state abrupt change response rate sequence within the detection period, and combined with the internal resistance change rate, current disturbance amplitude, and temperature change rate of each sub-interval, a sub-interval disturbance spread coefficient sequence is constructed; specifically, the sub-interval disturbance spread coefficient sequence is defined as follows: ,in The number of sub-intervals. sub-interval The internal perturbation spread coefficient, ,in, The rate of change of internal resistance, The internal resistance boundary difference represents the dynamic variation of the internal resistance within the sub-interval. The amplitude of the current disturbance represents the maximum fluctuation in the absolute value of the current. For the rate of temperature change, This is a perturbation nonlinear amplification factor used to enhance response sensitivity in high-fluctuation ranges. , The start and end times of the sub-interval. For a moment The current value;

[0082] Based on the voltage boundary difference, current boundary difference, and internal resistance boundary of each sub-interval, a sequence of energy-consumption coupled structural quantities is constructed; specifically, the sequence of energy-consumption coupled structural quantities is defined. , For the energy consumption coupling structure of the sub-interval, , Let the internal resistance boundary of the subinterval be , The energy response index determines the energy sensitivity within the abrupt change range. This is a regulating factor used to suppress the risk of the denominator being zero;

[0083] By combining the perturbation spread coefficient sequence with the energy dissipation coupled structural quantity sequence, a behavioral perturbation flux function for the detection period is generated. This function characterizes the local amplification features of sudden changes within the detection period and the overall structural dissipation intensity throughout the period. Specifically, and One-to-one coupling generates behavioral perturbation flux functions: ,in, The exponential coupling factor is used to control the nonlinear cooperative mode between disturbance and structure. The perturbation structure proportion coefficient can be set as the normalized result of the perturbation structure amount in each sub-interval. The final behavior disturbance flux function is in scalar form, measuring the intensity of the comprehensive structural disturbance flux within the period. By introducing boundary features of voltage, current, temperature, and internal resistance, it fully captures the unsteady-state disturbance behavior within the period, avoiding the use of traditional statistical average indices and enhancing the dynamic response capability to abrupt change segments. By utilizing a nonlinear coupling structure to construct the disturbance flux, it effectively improves the sensitivity to sensing signals of abnormal system energy consumption and health degradation.

[0084] The present invention is further configured such that the construction of the health indicator set includes:

[0085] Based on the behavioral perturbation flux function, and combined with five physical quantities during the detection period, namely, capacity change, internal resistance evolution rate, temperature extreme value shift, voltage hysteresis behavior and hysteresis energy loss, a set of health indicators with corresponding single physical meanings is constructed.

[0086] The health indicator set includes indicators of capacity degradation magnitude, internal resistance increase, temperature deviation index, voltage response hysteresis, and hysteresis loss, each corresponding to the state characteristics within the detection period. Specifically, the formula for calculating the capacity degradation magnitude indicator is as follows: , This represents the instantaneous effective output capacity for the current cycle. For nominal reference capacity, Represents the flux function of behavioral perturbation The time-series channel response extracted along the capacity direction. To adjust the exponent parameters, This is the capacity-based scaling factor. This refers to the time interval of the current detection cycle.

[0087] The formula for calculating the degree of increase in internal resistance is: ,in, The rate of internal resistance evolution. This refers to the energy consumption coupling term at the current moment within the cycle. Represents the flux function of behavioral perturbation The response function value for an internal resistance path. , This is the structural adjustment coefficient. This is the normalization constant;

[0088] The formula for calculating the temperature deviation index is: ,in, These are the highest and lowest temperature values ​​within the cycle. The target reference temperature set for the system. This represents the perturbation flux response at the corresponding temperature extreme. The power exponent. This is the temperature difference offset normalization term;

[0089] The formula for calculating the voltage response hysteresis index is: ,in, The rate of change of voltage. For instantaneous current, Let be the channel response value of the disturbance flux function in the voltage trajectory direction. To adjust the parameters, Normalization factor;

[0090] The formula for calculating the hysteresis loss index is: ,in, These are instantaneous voltage and current signals within a period. For theoretical power input, This represents the hysteresis energy loss channel value in the flux function. The power exponent. This is the normalized quantity of the power difference integral;

[0091] Construct a set of health indicators with corresponding single physical meaning .

[0092] The present invention is further configured such that constructing the covariant structural factor matrix of health indicators includes:

[0093] Health indicators corresponding to the previous and current testing periods are extracted. The evolution and changes of the same type of health indicator in adjacent periods are then analyzed using difference processing to form a set of periodic indicator differences. Specifically, in the current testing period... Compared with the previous cycle In the process, the corresponding sets of health indicators are extracted respectively. , Define the set of periodic indicator differences as This indicates the absolute change behavior of each type of health indicator between two adjacent periods.

[0094] For any two different categories of health indicators, calculate the evolutionary response dissipation term between them in the current period, describing the asymmetric response relationship between the indicators during the cyclical evolution process; specifically, for any two different health indicators... , We need to describe the response asymmetry that exists during their coordinated changes in the current period and construct the dissipation term: ,in, The degenerate potential function corresponds to respectively function terms, To measure right The influence rate of the function path reflects the coupling asymmetry. For the first During the cycle and The corresponding perturbation flux modulation factor. The exponential parameter is used to enhance the control over asymmetric strength. This formula constructs a term for energy dissipation behavior caused by cross-response between indices.

[0095] Based on the combinational relationship between the periodic index difference set and the response dissipation term, the health index covariance structure factor, which measures the co-evolution intensity between any two indicators in adjacent detection periods, is calculated, and a health index covariance structure factor matrix is ​​constructed. Specifically, a nonlinear interaction model is established between the periodic index difference set and the evolutionary response dissipation term. ,in, For health indicators Within the current cycle's range of change, For dissipation terms, characterize and The asymmetric coupling strength between them The term represents the coupling of flux between periodic index pairs and disturbances, indicating the coupled response path of flux in the periodic disturbance structure. The parameters are adjusted to control the amplification or suppression of the covariant structure strength, ultimately resulting in... Constructing the covariant structural factor matrix of health indicators .

[0096] The present invention is further configured such that the construction of the degradation potential function includes:

[0097] Based on the health indicators in the health indicator set, and combined with the covariance structure factor of the health indicators, a weighted coupling degree term is constructed among the indicators to reflect the strength of the interaction influence of multidimensional health indicators; specifically, based on the first... Health indicator covariance structure factor matrix during the detection period Extracting indicators Cooperative response strength And introduce a disturbance amplitude adjustment coefficient. Construct a weighted coupling term among the indicators: ,in, This is a fixed coupling gain factor, reflecting the impact of this index on the existence of long-term physical coupling. [4.5] To control the nonlinear response exponent, The first in the current testing cycle Health index values Reflection indicators , The strength of the coupling effect within the current cycle;

[0098] Based on the weighted coupling term and combined with the higher-order nonlinear deviation terms of each health indicator, a higher-order driving component reflecting the abnormal amplitude of a single indicator is constructed. Specifically, to characterize the local abnormal offset amplitude of a certain indicator in the current period, based on its offset path relative to the nearest neighbor structure, the following is defined: ,in, The anti-dynamic adjustment weight represents the health indicator. Mapping coefficients in behaviorally resistive flux structures Adjusting for the nonlinearity of the influence of neighboring indicators, Enhance the local offset amplification effect. Indicators of health The nonlinear driving response components;

[0099] By coupling the weighted coupling term with the higher-order driving components through a nonlinear function, a comprehensive expression for the health index is obtained; specifically, the comprehensive expression for the health index is: ,in, For the number of health indicators, As an index adjustment factor, it reflects the indicator The relative driving weights of the overall degradation structure This represents the comprehensive state term determined by the coupling between health indicators and local biases.

[0100] By introducing a behavioral perturbation flux function as a regulating factor, the comprehensive expression of health indicators is nonlinearly adjusted to achieve the overall construction of the degradation potential function. Specifically, the behavioral perturbation flux function within the detection period is used. As a regulator, it is applied to the synthetic expression to control the modulation intensity of the perturbation structure on the degradation trend within the period: , For the first The periodic behavior perturbation flux function value, and the final degenerate potential energy function value. It serves as a comprehensive measure of the overall structural health of the current cycle system, and as a core input variable for predicting the reliability range of future cycles and triggering feedback correction mechanisms.

[0101] The present invention is further configured such that the reliability change range in the projected future preset period includes:

[0102] The degradation potential energy function values ​​within three consecutive detection periods are extracted. Based on this sequence, a high-order nonlinear dynamic prediction model is constructed to deduce the predicted degradation potential energy values ​​for future preset periods. Specifically, let the current detection period be... Extract the degradation potential function values ​​for the most recent three periods: , , A three-dimensional structure mapping vector is constructed based on the historical degradation potential energy values ​​of the three periods: This leads to the prediction cycle. Predicted value of the degenerate potential function: ,in, For the first Value of the periodic degenerate potential energy function. This is a perturbation correction term to prevent structural degradation from stopping. These are the exponential control parameters for each evolutionary path. This is the periodic state adjustment coefficient. For the nonlinear mapping parameters of the future periodic structure, The degenerate potential energy function for predicting the period;

[0103] By combining the fluctuation characteristics of predicted values ​​and historical degradation potential functions, a reliability boundary adjustment factor is generated to reflect the range of uncertainty in the system state evolution. Specifically, to reflect the range of uncertainty in predicted values, a reliability boundary adjustment factor is constructed based on the historical fluctuation structure. ,in, This is the second-order potential energy jump term, which measures the oscillatory acceleration of the system. For the evolutionary offset amplification structure, For perturbation terms, to prevent division by zero, The offset intensity index, It is a non-linear adjustment parameter. To predict uncertain structural quantities, a scaling factor is defined for the future interval range;

[0104] Based on the predicted degradation potential energy and the reliability boundary adjustment factor, the lower and upper bounds of reliability for future periods are calculated, forming a reliability variation range. Specifically, the boundaries of the reliability range for future periods are calculated based on the predicted potential energy function value and the boundary adjustment factor. , ,in, This represents the lower bound of future periodic reliability. The above formula represents the upper bound of future periodic reliability. It shows that the higher the degradation potential energy value, the lower the reliability and the tighter the boundary.

[0105] The present invention is further configured such that the parameter path for updating the covariant structure factor and degradation potential function of the health indicator includes:

[0106] A health indicator difference matrix is ​​constructed based on the reliability upper bound of future testing cycles and the critical health indicators triggered within the current testing cycle; specifically, based on the reliability upper bound of future testing cycles... The set of critical health indicators triggered within the current detection cycle Construct a matrix of differences in health indicators ,in, This indicates that the reliability upper bound is subjected to the first... Dimension index mapping transformation function, To adjust the exponent and decay factor and ensure the nonlinear sensitivity of the difference matrix to the distance between indicators, this matrix reflects the errors between indicators and their structural dependence, avoiding simple linear differences and highlighting the influence of local proximity.

[0107] Based on the health indicator difference matrix and the current period health indicator covariance structure factor matrix, a feedback correction factor is generated. ;

[0108] The parameter paths of the covariant structure factor matrix and the degradation potential function of the current cycle health indicators are adjusted based on the feedback correction factor.

[0109] The detection process is adjusted to achieve periodic closed-loop feedback updates and dynamic adaptive evolution.

[0110] The present invention is further configured such that the generation of the feedback correction factor includes:

[0111] A nonlinear adjustment function is constructed based on the covariant structural factor matrix of health indicators and the difference matrix of health indicators.

[0112] By combining the product mapping of covariant structure factors and difference matrices with exponential decay weighting, asymmetric feedback correction with proximity structure is achieved.

[0113] The feedback correction factor, as a multidimensional parameter path regulating factor, dynamically adjusts the parameter evolution path of the covariant structure factor matrix and the degradation potential function of health indicators. Specifically, the calculation formula for the feedback correction factor is as follows: ,in, As a covariant structural factor for health indicators, A matrix of differences in health indicators. It is an exponential decay adjustment factor. The amplitude is controlled to form a nonlinear feedback correction factor with neighboring exponential weights, ensuring its complex and orderly distribution and avoiding simple linear combinations.

[0114] The present invention is further configured such that the periodic closed-loop feedback update and dynamic adaptive evolution of the detection process by adjusting includes:

[0115] The change path of the off-diagonal elements of the covariant structure factor matrix of health indicators is defined as the first type of parameter path, and the change path of the higher-order nonlinear driving components corresponding to each health indicator in the degradation potential function is defined as the second type of parameter path.

[0116] Based on the amplitude and directionality of the feedback correction factor, asymmetric perturbation gain adjustment is performed on the first type of parameter path. Specifically, this adjustment process dynamically adjusts the intensity of the co-evolution among indicators by introducing the modulation effect of the feedback correction factor, reflecting the non-equilibrium changes in the interaction between indicators. Asymmetry ensures that the adjustment process considers the directionality and strength differences of the mutual influence between indicators, avoiding system state ambiguity caused by simple equilibrium adjustment. Based on the generated feedback correction factor, its amplitude is extracted. With directional vector Among them, amplitude Indicates the corrected intensity, directional vector. , For health indicators, the distribution tendency of feedback corrections across various indicator dimensions is characterized, and the covariant structure factor matrix of health indicators is analyzed. Central African diagonal elements Apply asymmetric perturbation gain adjustment: ,in, The adjustment coefficient controls the intensity of the effect of the amplitude and directional weights. The disturbance attenuation factor is used to limit the range of disturbance influence. This formula reflects the asymmetric influence of feedback correction on the covariant relationship between indicators, ensuring that the disturbance gain depends on the distribution difference of the feedback direction, and avoiding the simple accumulation of linear superposition through exponential mapping.

[0117] A feedback correction factor is introduced into the second type of parameter path, applying feedback modulation weights to the nonlinear offset term corresponding to each health indicator. Specifically, this achieves dynamic response adjustment to the abnormal amplitude of a single indicator. The feedback modulation weights reflect the mapping relationship between the current system state and the prediction reliability deviation, promoting adaptive updating of the degradation potential function parameters and maintaining the model's accuracy and robustness. The feedback correction factor also incorporates a higher-order driving component adjustment factor path of the degradation potential function, applying feedback modulation weights to the nonlinear offset term corresponding to each health indicator. Apply feedback modulation weights: ,in, To adjust the strength parameters, Used for nonlinear amplification feedback effects Indicates the feedback direction vector corresponding to the first The structure of the components of each indicator ensures that the feedback correction can generate weighted adjustments for the nonlinear shifts of different health indicators, thereby improving the model's fine-grained adaptive capability.

[0118] By utilizing the joint regulation results of the first and second type of parameter paths, the co-evolutionary relationship of health indicators and the expression structure of the degradation potential function are simultaneously reconstructed, achieving parameter stability of the detection process in multi-period structural evolution. Specifically, the adjusted covariant structure factor matrix of health indicators is... With higher-order driving components of the degenerate potential function These parameters are applied together to the model to perform synchronous reconstruction. Synchronous reconstruction is achieved through a nonlinear mapping function. accomplish: ,in, The function represents the overall parameter state of the current period. The design aims to satisfy the parameter stability domain constraints, ensure the stability and robustness of the updated model parameters in multi-period structural evolution, and avoid instability caused by over-adjustment.

[0119] It should be noted that the specific operation methods of each module and unit in the battery reliability periodic testing method provided in the above embodiments have been described in detail in the method embodiments, and will not be repeated here. In practical applications, the battery reliability periodic testing method provided in the above embodiments can be assigned to different functional modules as needed, that is, the internal structure of the system can be divided into different functional modules to complete all or part of the functions described above, and this is not a limitation.

[0120] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0121] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. A and B can be singular or plural. Additionally, the character " / " in this article generally indicates an "or" relationship between the preceding and following related objects, but it can also represent an "and / or" relationship. Please refer to the context for a more accurate understanding.

[0122] In this application, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or multiple items. For example, at least one of a, b, or c can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0123] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0124] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0125] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0126] In the several embodiments provided in this application, it should be understood that the disclosed system can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0127] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0128] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0129] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0130] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for periodically testing the reliability of a storage battery, characterized in that, include: The battery's operating status data is acquired during the detection period. The time axis is divided into a preset number of continuous sub-intervals. State features are extracted from each sub-interval and fused to generate a non-steady-state mutation response rate sequence. A perturbation flux function characterizing the fluctuation intensity and dissipative structure of the battery's operating state within a cycle is constructed based on the non-steady-state mutation response rate sequence. Health indicators are calculated based on behavioral perturbation flux functions, and a set of health indicators with a single physical meaning is constructed to characterize the state characteristics within the detection period. Based on the set of health indicators, health indicators from the previous and current testing periods are extracted, and the health indicator covariance structure factor of each pair of health indicators in adjacent periods is calculated to construct a health indicator covariance structure factor matrix. Based on the current set of health indicators, the covariant structure factor of health indicators, and the behavioral perturbation flux function, a degradation potential function is constructed to measure the overall structural characteristics of the cyclical state evolution. Based on the degradation potential energy function values ​​of three adjacent detection cycles, calculate the lower and upper bounds of reliability for future detection cycles, and deduce the range of reliability changes in future detection cycles. The difference between the reliability upper bound of future detection cycles and the critical health indicators triggered in the current cycle is calculated to generate a feedback correction factor. Based on the feedback correction factor, the parameter paths of the covariance structure factor and the degradation potential function of the health indicators are updated.

2. The method for periodically testing the reliability of a storage battery according to claim 1, characterized in that, Generating nonsteady-state mutation response rate sequences includes: During the detection period, the battery's voltage, current, temperature, internal resistance, and corresponding time stamp data are acquired, and the time axis within the detection period is divided into a preset number of continuous sub-intervals. For each sub-interval, extract the voltage boundary value, current boundary value, internal resistance boundary value, and temperature sequence data within that interval. Calculate the voltage boundary difference, current boundary difference, internal resistance boundary difference, and temperature change rate based on the various boundary values. Based on the voltage boundary difference, current boundary difference, internal resistance boundary difference, and temperature change rate, a single numerical representation of the catastrophe structure quantity is constructed. The catastrophe structure quantities corresponding to each sub-interval are arranged in sequence to generate a non-steady-state catastrophe response rate sequence, which characterizes the segmented state fluctuation characteristics within the detection period.

3. The method for periodically testing the reliability of a storage battery according to claim 2, characterized in that, Constructing the behavioral perturbation flux function includes: Based on the non-steady-state abrupt change response rate sequence within the detection period, and combined with the internal resistance change rate, current disturbance amplitude, and temperature change rate of each sub-interval, a sub-interval disturbance expansion coefficient sequence is constructed. Based on the voltage boundary difference, current boundary difference, and internal resistance boundary of each sub-interval, construct the energy consumption coupling structure quantity sequence; By combining the perturbation expansion coefficient sequence with the energy dissipation coupled structural quantity sequence, a behavioral perturbation flux function for the detection period is generated, which characterizes the local mutation amplification features and the full-cycle structural dissipation intensity within the detection period.

4. The method for periodically testing the reliability of a storage battery according to claim 1, characterized in that, Building a set of health indicators includes: Based on the behavioral perturbation flux function, and combined with five physical quantities during the detection period, namely, capacity change, internal resistance evolution rate, temperature extreme value shift, voltage hysteresis behavior and hysteresis energy loss, a set of health indicators with corresponding single physical meanings is constructed. The set of health indicators includes indicators of capacity degradation, internal resistance increase, temperature deviation, voltage response hysteresis, and hysteresis loss, which correspond to the state characteristics within the detection period.

5. The method for periodically testing the reliability of a storage battery according to claim 4, characterized in that, Constructing the covariant structural factor matrix of health indicators includes: Extract the corresponding health indicators from the previous detection period and the current detection period, perform difference processing on the evolution and changes of the same type of health indicators in adjacent periods, and form a set of periodic indicator differences; For any two different types of indicators in the health indicator set, calculate the evolutionary response dissipation term between any two different types of indicators in the current cycle, and describe the asymmetric response relationship between indicators in the cycle evolution process; Based on the combination relationship between the periodic index difference set and the response dissipation term, the health index covariance structure factor, which measures the intensity of the co-evolution between any two indicators in adjacent detection periods, is calculated, and the health index covariance structure factor matrix is ​​constructed.

6. The method for periodically testing the reliability of a storage battery according to claim 5, characterized in that, Constructing the degenerate potential function includes: Based on the health indicators in the health indicator set, and combined with the covariance structure factor of the health indicators, a weighted coupling degree term is constructed between the indicators to reflect the intensity of the interaction influence of multidimensional health indicators. Based on the weighted coupling degree term, combined with the higher-order nonlinear deviation terms of each health indicator, a higher-order driving component reflecting the abnormal amplitude of a single indicator is constructed. By coupling the weighted coupling degree term with the higher-order driving component through a nonlinear function, a comprehensive expression for the health index is obtained. By introducing a behavioral perturbation flux function as a regulating factor, the comprehensive expression of health indicators is nonlinearly adjusted to achieve the overall construction of the degradation potential function.

7. The method for periodically testing the reliability of a storage battery according to claim 1, characterized in that, The range of reliability changes in the projected future preset period includes: The degradation potential energy function values ​​within three consecutive detection cycles are extracted. Based on this sequence, a high-order nonlinear dynamic prediction model is constructed to deduce the predicted degradation potential energy values ​​for future preset cycles. By combining the fluctuation characteristics of predicted values ​​and historical degradation potential functions, a reliability boundary adjustment factor is generated to reflect the range of uncertainty in the evolution of system state. Based on the predicted degradation potential energy and the reliability boundary adjustment factor, the lower and upper limits of reliability for future periods are calculated to form the reliability variation range.

8. The method for periodically testing the reliability of a storage battery according to claim 1, characterized in that, The parameter paths for updating the covariant structure factor and degradation potential function of health indicators include: Based on the reliability upper bound of future testing cycles and the critical health indicators triggered in the current testing cycle, a health indicator difference matrix is ​​constructed. Based on the health indicator difference matrix and the current period health indicator covariance structure factor matrix, a feedback correction factor is generated; The parameter paths of the covariant structure factor matrix and the degradation potential function of the current cycle health indicators are adjusted based on the feedback correction factor. The detection process is adjusted to achieve periodic closed-loop feedback updates and dynamic adaptive evolution.

9. The method for periodically testing the reliability of a storage battery according to claim 8, characterized in that, The generated feedback correction factors include: A nonlinear adjustment function is constructed based on the covariant structural factor matrix of health indicators and the difference matrix of health indicators. By combining the product mapping of covariant structure factors and difference matrices with exponential decay weighting, asymmetric feedback correction with proximity structure is achieved. The feedback correction factor, as a multidimensional parameter path adjustment factor, dynamically adjusts the parameter evolution path of the covariant structural factor matrix and the degradation potential function of health indicators.

10. A method for periodically testing the reliability of a storage battery according to claim 8, characterized in that, The periodic closed-loop feedback update and dynamic adaptive evolution of the detection process are achieved through adjustments, including: The change paths of the off-diagonal elements of the covariant structure factor matrix of health indicators are defined as the first type of parameter paths, and the change paths of the higher-order nonlinear driving components corresponding to each health indicator in the degradation potential function are defined as the second type of parameter paths. Based on the magnitude and directionality of the feedback correction factor, asymmetric perturbation gain adjustment is performed on the first type of parameter path; The feedback correction factor is introduced into the second type of parameter path, and a feedback modulation weight is applied to the nonlinear offset term corresponding to each health indicator. By utilizing the joint regulation results of the first type of parameter path and the second type of parameter path, the co-evolutionary relationship of health indicators and the expression structure of the degradation potential function are synchronously reconstructed, thereby realizing the parameter stability domain of the detection process in the multi-period structural evolution.

Images