Method for on-line grouping of battery capacity

Abstract

The application discloses a kind of methods for grouping battery online nuclear capacity, it is related to the field of intelligent management of energy storage system.The method comprises the following steps: S1: collecting battery fast multidimensional state characteristics, construct physical response intensity;S2: construct spectrum stability based on multidimensional state characteristics, combine physical response intensity and spectrum stability to construct battery dynamic pressure function;S3: calculate battery activation energy and reinforcement gain factor, construct dynamic deterioration propulsion;S4: combine battery dynamic pressure function and dynamic deterioration propulsion to calculate the multidimensional health risk index of battery;S5: realize grouping mark and state of charge estimation based on fuzzy membership degree.Through realizing the fine modeling of battery state under complex environment and the forward-looking prediction of deterioration trend, multidimensional health risk index is constructed to support intelligent dynamic grouping, which significantly improves system operation safety and battery resource utilization efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of intelligent management of energy storage systems, in particular to a method for online grouping and capacity determination of storage batteries. BACKGROUND

[0002] In special environments such as islands, offshore platforms, and polar research stations far from the main power grid, key infrastructure such as lighthouses often rely on renewable energy sources such as wind and solar power as the main power supply. To cope with adverse weather or insufficient light / wind, such systems are usually equipped with battery packs as backup energy and energy storage units. The operating state of the battery directly affects the continuous operation and emergency response capabilities of basic equipment such as navigation lights, automatic control systems, and remote communication and identification devices, so there are high requirements for online monitoring and health management of battery packs.

[0003] Existing battery management systems mostly focus on collecting basic electrical parameters such as voltage, current, and temperature of individual batteries and making threshold judgments, lacking deep diagnostic mechanisms for battery operating conditions in marine or highly corrosive environments. Especially in complex working conditions such as salt spray corrosion, sea wind impact, and extreme temperature differences, the aging process of the battery often exhibits highly nonlinear and strongly coupled characteristics, relying solely on voltage and SOC estimation can easily lead to misjudgment and fail to timely reflect multi-dimensional degradation factors such as structural degradation, electrolyte abnormalities, and shell damage.

[0004] In addition, the grouping strategy of some current battery systems is often based on factory capacity or manual preset standards to perform static grouping, which lacks real-time flexibility and is difficult to intelligently adjust according to actual working conditions, health levels, and environmental disturbances, leading to overuse of some batteries and redundant backup of others, thereby affecting overall energy utilization efficiency and battery service life.

[0005] On the other hand, some literature has attempted to combine environmental data for battery aging analysis, but mostly focuses on laboratory modeling or offline simulation, and has not established an online analysis and dynamic grouping mechanism that can operate long-term in complex outdoor environments. In addition, the comprehensive utilization of battery spectrum signals, electrochemical dynamic responses, and physical disturbance coupling is still relatively scattered, lacking systematic modeling and engineering-based solutions.

[0006] Therefore, there is an urgent need for an online method that can adapt to complex environments, integrate multi-dimensional feature information, support dynamic evaluation of battery state and intelligent grouping and capacity determination, to promote the development of battery pack management from traditional static strategies to dynamic, efficient, safe, and adaptive directions, and to improve the energy security and operational stability of key systems such as lighthouses in remote environments. SUMMARY

[0007] Based on the above-mentioned disadvantages of the prior art, the purpose of the present application is to provide a method for online grouping and capacity checking of a storage battery, so as to solve the above-mentioned technical problems.

[0008] To achieve the above-mentioned purpose, the present application provides the following technical solution: a method for online grouping and capacity checking of a storage battery, comprising:

[0009] S1: collecting a fast multi-dimensional state feature of a battery, extracting a physical feature according to the multi-dimensional state feature, and constructing a physical response strength;

[0010] S2: extracting a frequency spectrum signal feature according to the multi-dimensional state feature, constructing a frequency spectrum stability, and combining the physical response strength and the frequency spectrum stability to construct a battery dynamic pressure function;

[0011] S3: calculating a battery activation energy and a reinforcement gain factor according to the multi-dimensional state feature, and constructing a dynamic deterioration propulsion amount of the storage battery;

[0012] S4: combining the battery dynamic pressure function and the dynamic deterioration propulsion amount to calculate a multi-dimensional health risk index of the storage battery;

[0013] S5: based on the multi-dimensional health risk index and the multi-dimensional state feature, calculating a membership degree of the storage battery among a use group, a buffer group and a standby group through a fuzzy membership function, and completing grouping marking and state of charge estimation of the storage battery group according to the maximum membership degree.

[0014] The present application is further provided that the S1 comprises:

[0015] The multi-dimensional environmental features and the battery number of each battery in the lighthouse storage battery group are collected by using a sensor network;

[0016] The multi-dimensional state feature comprises a structural feature, an electrochemical feature and an environmental feature;

[0017] The structural feature comprises a battery shell temperature, a battery internal temperature, a micro-strain signal, a vibration intensity signal and a gas emission rate;

[0018] The electrochemical feature comprises a current, a voltage, an internal resistance and an electrolyte conductivity;

[0019] The environmental feature comprises a salt mist crystallization rate, an adherent conductivity and a sea wind infrasound wave intensity.

[0020] The present application is further provided that the multi-dimensional features collected in a unit time are segmented according to a preset fixed time to form a continuous time window;

[0021] In each time window, the change trend of the battery shell temperature is extracted, and a voltage signal acquired synchronously is normalized;

[0022] The processing results in all time windows are summarized to generate a temperature excitation index for measuring the temperature change intensity under a unit voltage response.

[0023] The application is further configured to extract the change characteristics of the micro-strain signal according to the set sampling time window, and perform coupling analysis on the vibration intensity signal acquired synchronously, combine the analysis results in each time window, and construct a structure vibration coupling response index based on the micro-strain signal and the vibration intensity signal in the multi-dimensional state characteristics.

[0024] The temperature excitation index and the structure vibration coupling response index are exponentially fused according to a preset weight parameter to construct a physical response intensity.

[0025] The application is further configured to include S2.

[0026] The current and salt spray crystallization rate in the multi-dimensional state characteristics are combined to construct a current environment coupling signal.

[0027] The current environment coupling signal is subjected to Fourier transform and power spectral density normalization processing to obtain a frequency spectrum energy distribution.

[0028] The frequency spectrum energy distribution is used to define a frequency spectrum barycenter, and the frequency spectrum barycenter difference between adjacent time windows is calculated to obtain a frequency jump rate.

[0029] The application is further configured to calculate a power spectrum entropy for measuring the energy concentration degree on the basis of the frequency spectrum energy distribution.

[0030] The power spectrum entropy and the frequency jump rate are combined to construct a frequency spectrum stability index.

[0031] The physical response intensity and the frequency spectrum stability index are combined to construct a battery dynamic pressure function, and the battery dynamic pressure is calculated.

[0032] The application is further configured to include S3.

[0033] A multi-dimensional battery state tensor is constructed by feature fusion according to the multi-dimensional state characteristics.

[0034] A ring state disturbance aggregation factor is constructed by combining the multi-dimensional battery state tensor and the salt spray crystallization rate in the multi-dimensional state characteristics.

[0035] A reinforcement gain factor is constructed by combining the ring state disturbance aggregation factor, the intensity signal in the multi-dimensional characteristics, and the salt spray crystallization rate.

[0036] The application is further provided that the battery activation energy is calculated by Arrhenius empirical fitting method according to the chemical reaction characteristics of the battery material, and the chemical reaction characteristics are calculated by multi-dimensional state characteristics, including electrolyte decomposition rate, SEI film growth amount, and metal corrosion rate.

[0037] The dynamic degradation driving amount is constructed by nonlinearly weighting and fusing the battery activation energy, the reinforcement gain factor, and the battery shell temperature in the multi-dimensional state characteristics.

[0038] The application is further provided that the S4 comprises:

[0039] The micro-strain change rate and the electrolyte saturation degree change amplitude are calculated based on the multi-dimensional state characteristics, and the structural disturbance factor is constructed by combining and extracting the micro-strain change rate and the electrolyte saturation degree change amplitude.

[0040] The voltage transient change rate and the internal resistance dynamic response characteristics are calculated based on the multi-dimensional state characteristics, and the electrochemical disturbance factor is constructed by combining the voltage transient change rate and the internal resistance dynamic response characteristics.

[0041] The environmental excitation potential energy factor is calculated according to the combined action of the salt mist crystallization rate and the sea breeze infrasound wave intensity in the multi-dimensional state characteristics.

[0042] The multi-dimensional health risk index is constructed by nonlinearly coupling the battery dynamic pressure, the dynamic degradation driving amount, the structural disturbance factor, and the environmental excitation potential energy factor, and introducing the electrochemical disturbance factor as an adjustment term.

[0043] The application is further provided that the S5 comprises:

[0044] The multi-factor state of charge estimation model is constructed based on the factory preset battery rated capacity, by fusing the multi-dimensional health risk index, the battery activation energy, and the salt mist crystallization rate in the multi-dimensional state characteristics, to obtain a current state of charge estimation value.

[0045] The core group membership function is constructed based on the multi-dimensional health risk index, the current state of charge estimation value, and the battery shell temperature in the multi-dimensional state characteristics.

[0046] The standby group membership function is constructed based on the multi-dimensional state characteristics, in combination with the multi-dimensional health risk index, the internal resistance, and the sea breeze infrasound wave intensity.

[0047] The buffer group membership is constructed according to the difference between the core group and the standby group memberships.

[0048] The battery numbers of the core group, the standby group, and the buffer group are obtained by comprehensively comparing the membership function results of each battery.

[0049] The current state of charge estimation values of the battery blocks in the core group, the standby group and the buffer group are respectively aggregated to obtain the state of charge of the core group, the standby group and the buffer group.

[0050] The application provides a method for online grouping and capacity determination of a battery, which comprises the following steps: S1, collecting a multi-dimensional state feature of the battery and constructing a physical response strength according to the multi-dimensional state feature; S2, constructing a frequency spectrum stability according to a frequency spectrum signal feature extracted from the multi-dimensional state feature, and combining the physical response strength and the frequency spectrum stability to construct a battery dynamic pressure function; S3, calculating a battery activation energy and a reinforcement gain factor according to the multi-dimensional state feature, and constructing a dynamic deterioration promotion amount of the battery; S4, combining the battery dynamic pressure function and the dynamic deterioration promotion amount to calculate a multi-dimensional health risk index of the battery; and S5, calculating a membership degree of the battery among a use group, a buffer group and a standby group based on the multi-dimensional health risk index and the multi-dimensional state feature by using a fuzzy membership function, and completing grouping marking and state of charge estimation of the battery group according to the maximum membership degree, and the beneficial effects of the method include:

[0051] The method can realize fine modeling of the battery state in a complex environment, can capture dynamic response characteristics of the battery under extreme conditions such as high humidity, high salt and strong vibration, and can realize health evaluation and risk discrimination in a complex environment by constructing the physical response strength by using a temperature excitation index and a structure vibration coupling response index, and comprehensively forming the battery dynamic pressure function in combination with the frequency spectrum stability index.

[0052] The method can enhance the predictability of the battery deterioration trend and the forward-looking control ability, can effectively depict the potential accelerated aging mechanism of the battery by constructing the reinforcement gain factor and the battery activation energy and calculating the dynamic deterioration promotion amount, can provide a forward-looking judgment basis for health management and maintenance strategies, and can avoid sudden failures caused by accumulation of deterioration hazards.

[0053] The method can support intelligent dynamic grouping based on the health risk, can divide the use group, the buffer group and the standby group by using the multi-dimensional health risk index as a core decision basis and the fuzzy membership function, can break away from the traditional static preset grouping mode, can improve the battery resource allocation efficiency, and can guarantee the energy stability of a key load and the maximization of the overall life of the battery group.

[0054] The above description is only a summary of the technical scheme of the application, in order to more clearly understand the technical means of the application, the application can be implemented according to the content of the description, and in order to make the above and other purposes, characteristics and advantages of the application more obvious and easy to understand, the following specific embodiments of the application are described. BRIEF DESCRIPTION OF DRAWINGS

[0055] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiments will be briefly introduced. Obviously, the drawings in the following description only represent some embodiments of the present application, and for those skilled in the art, other drawings can be obtained from these drawings without any creative effort. In the drawings:

[0056] Figure 1 A flow chart of a method for online grouping and capacity calculation of a battery according to an exemplary embodiment of the present application is shown. DETAILED DESCRIPTION

[0057] The embodiments of the present application will be described hereinafter with reference to the drawings and preferred embodiments, and other advantages and effects of the present application can be easily understood by those skilled in the art from the contents disclosed in the present specification. The present application can also be implemented or applied in other different embodiments, and the details in the present specification can be modified or changed in various ways based on different views and applications without departing from the spirit of the present application. It should be understood that the preferred embodiments are only for illustrating the present application, but not for limiting the protection scope of the present application.

[0058] It should be noted that the drawings provided in the following embodiments only schematically illustrate the basic concept of the present application, and only the components related to the present application are shown in the drawings, but not drawn according to the number, shape and size of the components in actual implementation. The type, number and ratio of each component in actual implementation can be arbitrarily changed, and the layout type of the components can also be more complex.

[0059] In the following description, a large number of details are discussed to provide a more thorough explanation of the embodiments of the present application, however, it is obvious for those skilled in the art that the embodiments of the present application can be implemented without these specific details, and in other embodiments, the known structures and devices are shown in the form of block diagrams instead of details, to avoid making the embodiments of the present application difficult to understand.

[0060] Embodiment:

[0061] A method for online grouping and capacity calculation of a battery, as shown in Figure 1 , comprises:

[0062] S1: collecting a fast multi-dimensional state feature of a battery, extracting a physical feature according to the multi-dimensional state feature, and constructing a physical response strength;

[0063] S2: extracting a frequency spectrum signal feature according to the multi-dimensional state feature, constructing a frequency spectrum stability, and combining the physical response strength and the frequency spectrum stability to construct a battery dynamic pressure function;

[0064] S3: calculating the battery activation energy and reinforcement gain factor according to the multi-dimensional state characteristics, and constructing the dynamic degradation propulsion of the battery;

[0065] S4: calculating the multi-dimensional health risk index of the battery by combining the dynamic pressure function of the battery and the dynamic degradation propulsion;

[0066] S5: calculating the membership of the battery between the use group, the buffer group and the standby group by the fuzzy membership function based on the multi-dimensional health risk index and the multi-dimensional state characteristics, and completing the grouping marking and the state of charge estimation of the battery group according to the maximum membership.

[0067] The application further provides that the S1 comprises:

[0068] The multi-dimensional environmental characteristics and the battery number of each battery in the lighthouse battery group are collected by a sensor network;

[0069] The multi-dimensional state characteristics comprise structural characteristics, electrochemical characteristics and environmental characteristics;

[0070] The structural characteristics comprise the battery shell temperature, the battery internal temperature, the micro-strain signal, the vibration intensity signal and the gas emission rate.

[0071] The electrochemical characteristics comprise the current, the voltage, the internal resistance and the electrolyte conductivity.

[0072] The environmental characteristics comprise the salt mist crystallization rate, the attached object conductivity and the sea wind infrasound wave intensity. Specifically, each battery in the lighthouse battery group is equipped with independent physical, electrochemical and environmental sensors, and each battery is bound to a unique number; the system collects 12 characteristic data of each battery through various sensors at a set time frequency, and the collected data is uniformly uploaded to an edge computing unit or a remote control center through an embedded system; in order to eliminate sensor time delay and noise interference, first, all characteristic signals are subjected to time alignment and filtering processing, invalid or abnormal data is cleaned by using sliding window smoothing, edge elimination and abnormal value identification algorithm to complete data denoising and cleaning to obtain clear sensor data, all clear sensor data of the same battery are uniformly encoded to form the state vector of the single battery, and then the multi-dimensional state characteristic set of the battery is constructed.

[0073] The application further provides that the multi-dimensional characteristics collected in a unit time are segmented according to a preset fixed time to form a continuous time window.

[0074] In each time window, the change trend of the battery shell temperature is extracted, and the voltage signal obtained synchronously is subjected to normalization processing.

[0075] The processing results in all time windows are summarized to generate a temperature excitation index for measuring the degree of temperature change under unit voltage response. Specifically, according to the set time length, the specific value of the time length is not limited, which is set according to actual needs during system deployment, such as 5 minutes, and is divided into continuous time windows. The temperature excitation index is used to measure the intensity of temperature fluctuation under voltage normalization, and the larger the value is, the more intense the temperature change under unit voltage is, which represents the potential thermal runaway risk; the temperature excitation index calculation logic is: wherein, is the temperature excitation index; is the total number of time windows, which is the total number of time windows divided according to the time window width in the unit time during system planning; is the battery shell temperature, which is one of the multi-dimensional state characteristics, and is data collected by a sensor and processed by a system; is the voltage, which is one of the multi-dimensional state characteristics, and is data collected by a sensor and processed by a system; is the time window width, which is a fixed time window interval set by the system, such as 5 minutes; is the time point, which is a variable for integration, representing each time from the start to the end in the current integration formula time window; is a small constant, which is a preset fixed value, and is set to by default.

[0076] The application further provides that, based on the micro-strain signal and the vibration intensity signal in the multi-dimensional state characteristics, the change characteristics of the micro-strain signal are extracted according to the set sampling time window, and the vibration intensity signal obtained synchronously is coupled and analyzed, and the structural vibration coupling response index is constructed combined with the analysis results in each time window;

[0077] The temperature excitation index and the structural vibration coupling response index are exponentially fused according to a preset weight parameter to construct a physical response intensity. Specifically, the structural vibration coupling response index is used to measure whether vibration process induces unstable deformation of the structure, indicating the mechanical stress risk, starting from the joint relationship of "strain change rate" and "vibration intensity", reflecting the structural response intensity of the battery in the physical disturbance environment, so as to evaluate the potential mechanical stress risk and structural aging trend; the structural vibration coupling response index calculation logic is: wherein, is the structural vibration coupling response index, and the larger the value is, the more likely it is that there is a vibration-induced structural damage risk; is the micro-strain signal, representing the micro-deformation of the battery shell or structural member per unit length, which is one of the multi-dimensional state characteristics, and is measured by a strain gauge or a fiber Bragg grating sensor, and then processed by a system. is the vibration intensity signal, representing the first The acceleration amplitude of the block battery current time in the three-dimensional direction of space is one of the multi-dimensional state characteristics, is measured by a three-axis MEMS vibration sensor, and is data processed by the system. The physical response intensity is a unified index by coupling the temperature excitation index representing thermal stress and the structure vibration coupling response index representing mechanical stress, and reflects the current physical operation pressure of the battery; the physical response intensity calculation logic is: wherein, is the physical response intensity, and the larger the value is, the greater the physical disturbance of the battery block under the influence of heat and structure; is the thermal excitation weight, used for adjusting the proportion of thermal excitation in the overall physical pressure, and the value range is between 0.5-1.5, and the value is recommended to be 1.2 in the high-temperature and humid environment of summer islands, so as to strengthen the influence of the heat factor, and the value can be appropriately adjusted with the change of the environment temperature; is the structure disturbance weight, used for adjusting the proportion of the structure disturbance influence, and the value range is between 0.5-1.5, and in the area where vibration occurs frequently, such as a wind-facing lighthouse, the value can be set to 1.3-1.5, so as to reflect the importance of the vibration caused by the sea wind.

[0078] The application further sets that the S2 comprises:

[0079] The current environment coupling signal is constructed based on the current and the salt mist crystallization rate in the multi-dimensional state characteristics;

[0080] The power spectrum density normalization processing is performed on the current environment coupling signal after Fourier transform, so as to obtain the frequency spectrum energy distribution;

[0081] The spectral center is defined based on the spectral energy distribution, and the frequency hopping rate is obtained by calculating the difference between the spectral centers of adjacent time windows. Specifically, by combining the current signal with the salt spray crystallization rate at the corresponding time point in each divided time window, the application does not limit the combination method, which can be one-to-one association, and can use point-by-point multiplication, normalized weighting or non-linear amplification such as exponential modulation to construct the current environment coupling signal. The constructed current environment coupling signal is subjected to fast Fourier transform in each time window to convert the time domain signal into a frequency domain signal. The amplitude square of each frequency component, i.e. the power spectrum, can be obtained by Fourier transform, and then the power is normalized to make the energy sum equal to 1, thereby obtaining the spectral energy distribution, which can be compared jointly between different time windows. The normalized spectral energy distribution is regarded as a "mass distribution", and the frequency axis is regarded as a "position axis"; the spectral center of the current window is obtained by calculating the weighted sum of the frequency multiplied by the corresponding energy; the center indicates that the main frequency of the signal fluctuation in this time period is concentrated in which frequency region. The difference between the spectral centers of adjacent two time windows is obtained, and the center change amount is obtained. In order to obtain dynamic characteristics, the center change amount is divided by the time interval between windows to obtain the frequency hopping rate, which is used to reflect the degree of change of the spectral center in adjacent two time periods, and is an important indicator for measuring the stability of the running frequency of the battery. Frequent hopping represents abnormal running fluctuation.

[0082] The application further provides that the power spectrum entropy used to measure the energy concentration degree is calculated based on the spectral energy distribution;

[0083] The power spectrum entropy and the frequency hopping rate are combined to construct a spectral stability index;

[0084] The battery dynamic pressure function is constructed by combining the physical response intensity and the spectral stability index, and the battery dynamic pressure is calculated. Specifically, the power spectrum entropy is used to measure whether the energy is concentrated or dispersed in the frequency domain, which is an information entropy calculated based on the normalized spectral energy distribution. The lower the power spectrum entropy value, the more concentrated the energy, and the better the system stability. The higher the power spectrum entropy value, the more widely the energy is distributed, and there may be strong interference or fault signs. The power spectrum entropy calculation logic is as follows: wherein, is the power spectrum entropy, which represents the dispersion degree of the spectral energy; and are the upper and lower limits of Fourier frequency domain integration, which are determined by the system sampling frequency range; is the spectral energy distribution, which is obtained by normalizing the power obtained by performing Fourier transform on the current environment coupling signal; is a small constant, which is a preset fixed value, and is set to by default; The greater the spectral power density is, the smaller the contribution to the spectral entropy is, because The greater the negative value is, the smaller the product is, reflecting the more concentrated energy, wherein the logarithmic function The base is 2 by default, and the unit of the calculation result is bit, and in actual engineering applications, such as implementation using embedded systems, MATLAB or Python, the natural logarithm ln is more convenient, and the unit is nat, and the conversion relationship between the two is: The modification of the base will not affect the relative numerical distribution of the spectral entropy and its effectiveness in the judgment of spectral stability analysis. The spectral stability index represents the fusion of the power spectrum entropy and the frequency jump rate into a normalized expression to represent whether the system frequency characteristics are stable. Wherein, The spectral stability index is closer to 1, indicating that it is more stable, and the smaller the value is, the more unstable it is. The spectral barycenter jump rate represents the difference between the frequency centers of the current window and the previous window, and is used to reflect whether the spectrum changes dramatically. The barycenter change amount is obtained by subtracting the barycenter of the adjacent two time windows; in order to obtain dynamic characteristics, the barycenter change amount is divided by the window interval, and the frequency jump rate is obtained. The spectral entropy weight factor is used to control the influence degree of the spectral entropy on the total disturbance, and the value range is between 1 and 2. The frequency jump weight factor is used to control the influence degree of the jump rate on the disturbance, and the value range is between 1 and 2. The battery dynamic pressure function is used to calculate the comprehensive operation pressure of the battery block at the current moment, and integrates the evaluation of the thermal disturbance, mechanical stress and frequency domain disturbance level; the calculation result of the battery dynamic pressure function is used as the key input quantity of subsequent health risk assessment, aging trend modeling and failure prediction, and the higher the value is, the more unstable the battery operating state is, and the higher the potential risk is; in the construction of the dynamic pressure function, the spectral stability index is used as a negative feedback item to participate in the adjustment, and when the spectral stability is higher, the external disturbance is smaller, and the dynamic pressure is more inhibited, avoiding false judgment caused by short-time abnormity. The battery dynamic pressure function is coupled by calculating the physical response intensity and the reverse regulation effect of the spectral stability index, and outputs the battery dynamic pressure representing the current comprehensive dynamic pressure level of the battery.

[0085] The application further provides that the S3 comprises:

[0086] A multi-dimensional battery state tensor is constructed by fusing the multi-dimensional state characteristics;

[0087] An ambient state disturbance aggregation factor is constructed in combination with the multi-dimensional battery state tensor and the salt spray crystallization rate in the multi-dimensional state characteristics.

[0088] Based on the ring state disturbance aggregation factor, the intensity signal in the multi-dimensional feature and the salt mist crystallization rate, a reinforcement gain factor is constructed. Specifically, the multi-dimensional battery state tensor is a high-dimensional representation body integrating structural features, electrochemical features and environmental features, used to describe the comprehensive operation state of the battery at the current time and historical stage; wherein the multi-dimensional state feature includes: structural features, electrochemical features and environmental features; the structural features include: battery shell temperature, battery internal temperature, micro-strain signal, vibration intensity signal, gas emission rate; the electrochemical features include: current, voltage, internal resistance and electrolyte conductivity; the environmental features include: salt mist crystallization rate, adherend conductivity, sea wind infrasound wave intensity; the invention adopts the way of splicing and merging item by item, unifies the modal and time alignment of various feature data, constructs the multi-dimensional battery state tensor, and realizes the unified expression of the battery health state and the stress response level. The ring state disturbance aggregation factor is used to describe the aggregation response ability of multi-modal state to environmental factors, by using the multi-dimensional battery state tensor of the current time window, the overall disturbance degree of the tensor in the time window (which can be understood as the fluctuation intensity of the tensor) is calculated first; then the salt mist crystallization rate of the corresponding period is introduced, and the above tensor disturbance intensity is weighted by the environment; if the salt mist crystallization rate is high, it means that the external corrosion is strong, and the tensor disturbance is more sensitive; otherwise, the weighting factor is reduced, and finally the ring state disturbance aggregation factor is formed; the ring state disturbance aggregation factor calculation logic is as follows: Wherein, is the ring state disturbance aggregation factor; is the multi-dimensional battery state tensor; is the salt mist crystallization rate, which is one of the multi-dimensional state features, collected based on the surface deposition sensor and the conductivity estimation method; is the L2 norm of the multi-dimensional state tensor, used to measure the cooperative disturbance intensity of the multi-modal signals such as voltage, conductivity and temperature. The reinforcement gain factor is an amplification gain term after the aggregation of the ring state disturbance aggregation factor, the vibration intensity signal and the salt mist crystallization rate, used for the "stress response gain" in the nonlinear enhancement dynamic aging rate model, to strengthen the sensitivity evaluation of the influence of extreme environments such as high salt, high vibration and high disturbance; the reinforcement gain factor calculation logic is as follows: Wherein, is the reinforcement gain factor; is the vibration weight, controlling the weight of vibration intensity in degradation, the recommended value range is between 0.5 and 3, and the specific data needs to be determined by the actual demand of actual deployment scene; is the corrosion weight, used to control the weight of corrosion on the amplification effect of aging, the recommended value range is between 0.1 and 2; is the disturbance weight, used to control the coupled disturbance intensity, the recommended value range is between 0.1 and 1; is the vibration intensity signal; is the salt mist crystallization rate.

[0089] The application is further configured to calculate the battery activation energy according to the chemical reaction characteristics of the battery material by the Arrhenius empirical fitting method, and the chemical reaction characteristics are calculated by multi-dimensional state characteristics, including electrolyte decomposition rate, SEI film growth amount, and metal corrosion rate;

[0090] The dynamic degradation promotion amount is constructed by nonlinearly weighting and fusing the battery activation energy, the reinforcement gain factor, and the battery shell temperature in the multi-dimensional state characteristics. Specifically, the electrolyte decomposition rate can be estimated by the change of conductivity, gas detection, air pressure sensor, or multiple discharge voltage platform changes; the SEI film growth amount is indirectly estimated by the increase of internal resistance, voltage hysteresis characteristics, and material modeling; and the metal corrosion rate is obtained according to the salt fog concentration, electrochemical noise change, and contact resistance rising trend. The calculation of the electrolyte decomposition rate, the SEI film growth amount, and the metal corrosion rate is the prior art, and the application does not limit the way of calculating the electrolyte decomposition rate, the SEI film growth amount, and the metal corrosion rate. The application uses the Arrhenius empirical fitting method to construct the relationship curve of the rate and the temperature. The function relationship of the reaction rate to the inverse temperature is fitted in the semi-logarithmic coordinates, the linear fitting slope is extracted therefrom, and the battery activation energy of the battery material reaction is indirectly estimated by combining the physical constant. The battery activation energy is used to quantify the starting threshold of the chemical reaction under the current thermal state and reflects the activity of the reaction. The method is based on the existing electrochemical theory and reliable test method, has good realizability and reproducibility, and the related technical details are not described here. , wherein, is the dynamic degradation promotion amount; is the battery intrinsic aging constant, and the value range is between 0.01 and 1, which is set according to the actual deployment scene; is the battery activation energy; is the Boltzmann constant, which is a fixed value and takes the value of ; is the battery shell temperature; is the reinforcement gain factor.

[0091] The application is further configured to include the S4.

[0092] The micro-strain change rate and the electrolyte saturation degree change amplitude are calculated based on the multi-dimensional state characteristics, and the structural disturbance factor is constructed by combining the micro-strain change rate and the electrolyte saturation degree change amplitude;

[0093] The voltage transient change rate and the internal resistance dynamic response characteristics are calculated based on the multi-dimensional state characteristics, and the electrochemical disturbance factor is constructed by combining the voltage transient change rate and the internal resistance dynamic response characteristics;

[0094] According to the combined effect of the salt mist crystallization rate and the sea wind infrasound wave intensity in the multi-dimensional state characteristics, an environmental excitation potential energy factor is calculated;

[0095] The battery dynamic pressure, the dynamic degradation propulsion amount, the structure disturbance factor and the environmental excitation potential energy factor are nonlinearly coupled, and an electrochemical disturbance factor is introduced as an adjustment term to construct a multi-dimensional health risk index. Specifically, the structure disturbance factor is used to capture the dynamic risk at the structure level, especially the micro-deformation (micro-strain) and the "accumulation-consumption" change of the electrolyte in the structure cavity; the structure disturbance factor calculation logic is as follows: , wherein, is the structure disturbance factor; is the micro-strain change rate, which is used to represent the change rate of the micro-strain with respect to time at a specific time point. Here, the numerical difference of the continuously collected strain signals is obtained, but the present application is not limited thereto, and other ways of calculation can be used; is the electrolyte saturation degree. The electrolyte decomposition is often accompanied by gas generation, and the loss degree of the electrolyte can be indirectly judged by monitoring the pressure change. Based on the gas generation rate in the multi-dimensional state characteristics, the loss amount is estimated in combination with the theoretical decomposition gas generation ratio of the electrolyte, the remaining electrolyte proportion is obtained, and the corresponding saturation degree is obtained. The specific calculation process is known in the art, and will not be described in detail here; is the reference baseline saturation degree, which is the saturation degree under the ideal state. It is set according to historical experience and defaults to 0.85. It can be adaptively modified according to the historical data of the actual deployment environment; is the reference saturation deviation, which is used to reflect the possible local supersaturation or dry state. The electrochemical disturbance factor is used to measure the stress intensity of the battery due to the change of the electrochemical characteristics at a specific time, and is an adjustment term in the multi-dimensional health risk model, which is used to quantify the electrochemical stability of the battery under the condition of load change or aging. The electrochemical disturbance factor calculation logic is as follows: , wherein, is the electrochemical disturbance factor. The larger the value, the more intense the voltage fluctuation and the higher the impedance, reflecting the instability of the current electrochemical state; is the voltage; is the internal resistance, which is used to reflect the impedance effect caused by aging or material structure change; is the voltage transient change rate, which is used to reflect the load responsiveness or polarization degree. The environmental excitation potential energy factor is used to measure the degree of implicit excitation effect of the battery caused by external harsh environmental factors, such as salt mist corrosion and sea wind infrasound. It is a bridge for environmental disturbance to enter the physical model. The environmental excitation potential energy factor calculation logic is as follows: , wherein, is the environmental excitation potential energy factor. The larger the value, the stronger the excitation of the current environmental factors on the battery, and the potential risk increases; is the salt mist crystallization rate; is the sea breeze infrasound wave intensity; is the electrochemical impedance amplitude, which refers to the total impedance modulus value of the battery system to the alternating excitation at a given frequency or frequency range, and is used to reflect: the internal polarization impedance of the battery, such as: charge transfer, double layer, electrolyte conductivity, electrode / electrolyte interface reaction activity, the anti-interference ability of the battery in the frequency domain, such as corrosion resistance and fluctuation resistance, non-ideal behaviors such as battery aging, corrosion, lithium precipitation, and the like. The greater the impedance amplitude, generally represents: internal ion migration is not smooth; SEI film thickening; corrosion layer increases; or environmental interference causes medium polarization and other phenomena to be aggravated; the specific value is that in the scene with EIS equipment, the modulus value at 10Hz is selected as the electrochemical impedance amplitude by collecting Nyquist data every period; in the absence of special equipment, the charge transfer resistance or polarization impedance value is extracted as the electrochemical impedance amplitude by constructing an equivalent circuit to fit the transient response data; both methods are prior art, and will not be described here. The multi-dimensional health risk index is used to evaluate the current comprehensive health risk state of the battery in real time, and integrates the mechanical, chemical, environmental and spectral disturbance factors of the battery; the calculation logic is: , wherein, is the multi-dimensional health risk index; is the battery dynamic pressure; is the dynamic deterioration promotion amount; is the structure disturbance factor; is the environmental excitation potential energy factor; is the electrochemical disturbance factor.

[0096] The application further provides that the S5 comprises:

[0097] Based on the factory preset battery rated capacity, a multi-factor state of charge estimation model is constructed by integrating the multi-dimensional health risk index, the battery activation energy and the salt spray crystallization rate in the multi-dimensional state characteristics, to obtain a current state of charge estimation value;

[0098] Based on the multi-dimensional health risk index, the current state of charge estimation value and the battery shell temperature of the multi-dimensional state characteristics, a core group membership function is constructed;

[0099] Based on the multi-dimensional state characteristics, a standby group membership function is constructed in combination with the multi-dimensional health risk index, the internal resistance and the sea breeze infrasound wave intensity;

[0100] A buffer group membership is constructed according to the difference between the core group and the standby group membership;

[0101] The battery numbers of the core group, the standby group and the buffer group are obtained by comprehensively comparing the membership function results of each battery;

[0102] The current state of charge estimation values of the battery blocks in the core group, the standby group and the buffer group are respectively aggregated to obtain the state of charge of the core group, the standby group and the buffer group. Specifically, the current state of charge estimation value represents the current effective electric quantity / rated electric quantity, which is used to judge the remaining available energy and support the scheduling strategy; and the current state of charge estimation value calculation logic is as follows: wherein, is the current state of charge estimation value; is the battery rated capacity, which is the nominal standard capacity when the battery is manufactured by the manufacturer; is a health risk weight coefficient, representing the intensity of the influence of the risk score on the SOC, and the value range is between 0.2 and 0.4; is a multi-dimensional health risk index; is a thermal attenuation weight coefficient, used to control the influence of temperature on the SOC, and the value range is between 0.05 and 0.15; is the battery activation energy; is the shell temperature; is the Boltzmann constant; is a corrosion influence weight coefficient, used to represent the influence degree of salt mist corrosion on the SOC, and the value range is between 0.1 and 0.1; is the salt mist crystallization rate; is a corrosion influence index, used to control the nonlinear enhancement of the salt mist crystallization rate in the model, and the value range is between 1.5 and 3.0. The core group membership function calculation logic is as follows: wherein, is the core group membership; is the temperature rise amplitude, representing the temperature rise amplitude of the battery shell per unit time, used to reflect abnormal heating; , and are core group discrimination weights, and the value range is between 1 and 5, and the default is 1, used to adjust the control effect of different coefficients. The standby group membership function calculation logic is as follows: wherein, is the standby group membership; is the internal resistance; is the sea breeze infrasound wave intensity; , and are standby group discrimination weights, and the value range is between 0.5 and 5, and the design is complementary to the core group. The buffer group membership is obtained by calculating the value obtained by subtracting the sum of the core group membership and the standby group membership from 1. By calculating the membership of all battery blocks in the battery pack, grouping according to the membership, and then calculating the average of the state of charge estimation values of the battery blocks in all groups, the state of charge of the core group, the standby group and the buffer group is obtained.

[0103] The above-described embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented by software, the above-described embodiments can be implemented in whole or in part in the form of 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, the processes or functions described in the embodiments of the present application are wholly or partially generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another, for example, the computer instructions can be transferred from one website, computer, server, or data center to another via wired (for example, infrared, wireless, microwave, etc.) or wireless means. The computer-readable storage medium can be any available medium accessible by a computer or a data storage device such as a server, data center, or the like, which includes one or more available medium collections. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state disk.

[0104] It should be understood that the term "and / or" herein merely describes an association relationship of associated objects, which means that there can be three relationships, for example, A and / or B can represent the following three cases: A exists alone, A and B exist together, and B exists alone, where A and B can be singular or plural. In addition, the character " / " herein generally represents an "or" relationship between the front and rear associated objects, but can also represent an "and / or" relationship, which can be understood in the context before and after.

[0105] In this application, "at least one" means one or more, and "multiple" means two or more. "At least one of the following" or the like means any combination of the items, including any combination of single or multiple items. For example, at least one of a, b, or c can represent a, b, c, a-b, a-c, b-c, or a-b-c, where a, b, and c can be single or multiple.

[0106] It should be understood that in various embodiments of the present application, the size of the sequence number of the above-described processes does not mean the order of execution, and the execution order of the processes should be determined by their functions and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0107] Those skilled in the art can clearly understand that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are performed 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 the present application.

[0108] Those skilled in the art can clearly understand that, for the convenience and brevity of the description, the specific working processes of the above-described system, device and unit can refer to the corresponding processes in the foregoing method embodiments, which will not be repeated here.

[0109] In several embodiments provided in the present application, it should be understood that the disclosed system can be implemented in other ways. For example, the above-described device embodiments are only schematic, for example, the division of the units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interface, device or unit, and can be electrical, mechanical or other forms.

[0110] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place, or can be distributed on a plurality of network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0111] In addition, each functional unit in each embodiment of the present application can be integrated into a processing unit, or each unit can exist physically independently, or two or more units can be integrated into one unit.

[0112] If the functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the parts that contribute to the prior art or parts of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0113] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for on-line grouping of battery capacity, characterized by, Comprise: S1: Collecting the multi-dimensional state characteristics of the battery blocks, extracting physical characteristics according to the multi-dimensional state characteristics to construct the physical response strength; S2: Extracting the frequency spectrum signal characteristics according to the multi-dimensional state characteristics to construct the frequency spectrum stability, combining the physical response strength and the frequency spectrum stability to construct the battery dynamic pressure function; S3: Calculating the battery activation energy and the reinforcement gain factor according to the multi-dimensional state characteristics, and constructing the dynamic degradation propulsion of the battery; S4: Combining the battery dynamic pressure function and the dynamic degradation propulsion to calculate the multi-dimensional health risk index of the battery; S5: Based on the multi-dimensional health risk index and the multi-dimensional state characteristics, calculating the membership of the battery among the use group, the buffer group and the standby group through the fuzzy membership function, and completing the grouping marking and the state of charge estimation of the battery group according to the maximum membership; The S5 comprises: based on the factory preset battery rated capacity, fusing the multi-dimensional health risk index, the battery activation energy and the salt fog crystallization rate in the multi-dimensional state characteristics, constructing a multi-factor state of charge estimation model to obtain a current state of charge estimation value; Based on the multi-dimensional health risk index, the current state of charge estimation value and the battery shell temperature of the multi-dimensional state characteristics, a core group membership function is constructed; Based on the multi-dimensional state characteristics, combining the multi-dimensional health risk index, the internal resistance and the sea breeze infrasound wave intensity, a standby group membership function is constructed; According to the difference between the memberships of the core group and the standby group, a buffer group membership is constructed; Comprehensively comparing the membership function results of each battery, the battery numbers of the core group, the standby group and the buffer group are obtained respectively; The current state of charge estimation values of the battery blocks in the core group, the standby group and the buffer group are respectively summarized to obtain the state of charge of the core group, the standby group and the buffer group.

2. The method of claim 1, wherein, The S1 comprises: Using a sensor network to collect the multi-dimensional environmental characteristics and the battery number of each battery in the lighthouse battery group; The multi-dimensional state characteristics include: structural characteristics, electrochemical characteristics and environmental characteristics; The structural characteristics include: battery shell temperature, battery internal temperature, micro-strain signal, vibration intensity signal and gas emission rate; The electrochemical characteristics include: current, voltage, internal resistance and electrolyte conductivity; The environmental characteristics include: salt fog crystallization rate, attached object conductivity and sea breeze infrasound wave intensity.

3. The method of claim 2, wherein: The multi-dimensional characteristics collected in a unit time are segmented according to a preset fixed time to form a continuous time window; In each time window, the change trend of the battery shell temperature is extracted and normalized with the synchronous acquired voltage signal; The processing results in all time windows are summarized to generate a temperature excitation index for measuring the degree of temperature change under a unit voltage response.

4. The method of claim 3, wherein: Based on the micro-strain signal and the vibration intensity signal in the multi-dimensional state characteristics, the change characteristics of the micro-strain signal are extracted according to a set sampling time window, coupled with the synchronous acquired vibration intensity signal, and the analysis results in each time window are combined to construct a structural vibration coupling response index; The temperature excitation index and the structure vibration coupling response index are exponentially fused according to a preset weight parameter to construct a physical response strength.

5. The method of claim 1, wherein, The S2 includes: A current environment coupling signal is constructed based on a combination of the current and the salt spray crystallization rate in the multi-dimensional state feature; After Fourier transform is performed on the current environment coupling signal, power spectrum density normalization processing is performed to obtain a frequency spectrum energy distribution; A frequency spectrum gravity center is defined based on the frequency spectrum energy distribution, and a frequency jump rate is obtained by calculating the difference between the frequency spectrum gravity centers of adjacent time windows.

6. The method of claim 5, wherein, A power spectrum entropy for measuring the energy concentration degree is calculated based on the frequency spectrum energy distribution; A frequency spectrum stability index is constructed by combining the power spectrum entropy and the frequency jump rate; A battery dynamic stress function is constructed by combining the physical response strength and the frequency spectrum stability index, and the battery dynamic stress is calculated.

7. The method of claim 1, wherein the step of grouping the batteries is performed on-line. The S3 includes: A multi-dimensional battery state tensor is constructed by feature fusion based on the multi-dimensional state feature; A ring state disturbance aggregation factor is constructed by combining the multi-dimensional battery state tensor and the salt spray crystallization rate in the multi-dimensional state feature; An intensification gain factor is constructed based on the ring state disturbance aggregation factor, the intensity signal in the multi-dimensional feature, and the salt spray crystallization rate.

8. The method of claim 7, wherein, A battery activation energy is calculated based on the chemical reaction characteristics of the battery material by the Arrhenius empirical fitting method, and the chemical reaction characteristics are calculated based on the multi-dimensional state feature, including: electrolyte decomposition rate, SEI film growth amount, and metal corrosion rate; A dynamic degradation promotion amount is constructed by nonlinearly weighting and fusing the battery activation energy, the intensification gain factor, and the battery shell temperature in the multi-dimensional state feature.

9. The method of claim 1, wherein, The S4 includes: A structure disturbance factor is constructed by combining the micro-strain change rate and the electrolyte saturation degree change amplitude calculated based on the multi-dimensional state feature; An electrochemical disturbance factor is constructed by combining the voltage transient change rate and the internal resistance dynamic response feature calculated based on the multi-dimensional state feature; An environmental excitation potential energy factor is calculated based on the combined effect of the salt spray crystallization rate and the sea breeze infrasound intensity in the multi-dimensional state feature; A multi-dimensional health risk index is constructed by nonlinearly coupling the battery dynamic stress, the dynamic degradation promotion amount, the structure disturbance factor, and the environmental excitation potential energy factor, and introducing the electrochemical disturbance factor as an adjustment term.

Images