CN114879043A - Lithium ion battery lithium analysis diagnosis method, device, equipment and medium - Google Patents
Lithium ion battery lithium analysis diagnosis method, device, equipment and medium Download PDFInfo
- Publication number
- CN114879043A CN114879043A CN202210044751.8A CN202210044751A CN114879043A CN 114879043 A CN114879043 A CN 114879043A CN 202210044751 A CN202210044751 A CN 202210044751A CN 114879043 A CN114879043 A CN 114879043A
- Authority
- CN
- China
- Prior art keywords
- data
- lithium ion
- ion battery
- battery
- lithium
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/367—Software therefor, e.g. for battery testing using modelling or look-up tables
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/392—Determining battery ageing or deterioration, e.g. state of health
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01M—PROCESSES OR MEANS, e.g. BATTERIES, FOR THE DIRECT CONVERSION OF CHEMICAL ENERGY INTO ELECTRICAL ENERGY
- H01M10/00—Secondary cells; Manufacture thereof
- H01M10/42—Methods or arrangements for servicing or maintenance of secondary cells or secondary half-cells
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
- Y02E60/10—Energy storage using batteries
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Manufacturing & Machinery (AREA)
- Chemical & Material Sciences (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Electrochemistry (AREA)
- General Chemical & Material Sciences (AREA)
- Secondary Cells (AREA)
Abstract
The application relates to a lithium ion battery lithium analysis diagnosis method, a lithium ion battery lithium analysis diagnosis device, lithium ion battery lithium analysis equipment and a lithium ion battery lithium analysis diagnosis medium, wherein the method comprises the following steps: calling a trained probabilistic machine learning model; the probability machine learning model is obtained by training on the basis of off-line battery aging data obtained by off-line testing, wherein the off-line battery aging data comprises capacity degradation track data, alternating current impedance degradation track data and label data of battery lithium analysis in the lithium ion battery aging process; acquiring online monitoring capacity degradation track data and online monitoring characteristic impedance degradation track data of a lithium ion battery to be diagnosed; and according to the online monitoring capacity degradation track data and the online monitoring characteristic impedance degradation track data, online estimation is carried out on the lithium analysis probability of the lithium ion battery to be diagnosed through a probability machine learning model, and the online lithium analysis diagnosis result of the lithium ion battery to be diagnosed is obtained. The accuracy of the diagnosis result can be greatly improved, and the method has better usability.
Description
Technical Field
The present disclosure relates to the field of battery management technologies, and in particular, to a lithium analysis diagnostic method, device, apparatus, and medium for a lithium ion battery.
Background
Lithium deposition is one of the dangerous side reactions occurring inside lithium ion batteries, and it often occurs in cases of low-temperature charging, large-current charging, and the like of the batteries. The nature of lithium deposition is that lithium ions in the electrolyte are not intercalated into the negative active particles during charging, but are directly reduced to metallic lithium on the particle surface. As the amount of lithium deposition accumulates during the aging process, the lithium metal on the surface of the negative electrode gradually grows into lithium dendrites, which then pierce the separator and cause short circuits in the battery. The internal short circuit induced by the large-scale lithium deposition inside the battery is one of the most important reasons for the thermal runaway accident of the battery itself under the condition of no external collision or heating and the like. Therefore, the diagnosis of whether the lithium ion battery is subjected to lithium analysis or not has important significance for improving the safety of the lithium ion battery system.
The current methods for diagnosing lithium ion batteries mainly fall into two categories: one is a post mortem (Postmortem) diagnostic method and the other is a non-destructive diagnostic method. However, in the process of implementing the invention, the inventor finds that the traditional nondestructive diagnosis method has the technical problem of low diagnosis accuracy.
Disclosure of Invention
In view of the above, it is desirable to provide a lithium ion battery lithium analysis diagnostic method, a lithium ion battery lithium analysis diagnostic apparatus, a battery diagnostic device, and a computer-readable storage medium, which can greatly improve the accuracy of a lithium analysis diagnostic result and improve usability.
In order to achieve the above purpose, the embodiment of the invention adopts the following technical scheme:
in one aspect, an embodiment of the present invention provides a lithium ion battery lithium analysis diagnostic method, including:
calling a trained probabilistic machine learning model; the probability machine learning model is obtained by training on the basis of off-line battery aging data obtained by off-line testing, wherein the off-line battery aging data comprises capacity degradation track data, alternating current impedance degradation track data and label data of battery lithium analysis in the lithium ion battery aging process;
acquiring online monitoring capacity degradation track data and online monitoring characteristic impedance degradation track data of a lithium ion battery to be diagnosed;
and according to the online monitoring capacity degradation track data and the online monitoring characteristic impedance degradation track data, online estimation is carried out on the lithium analysis probability of the lithium ion battery to be diagnosed through a probability machine learning model, and the online lithium analysis diagnosis result of the lithium ion battery to be diagnosed is obtained.
In another aspect, a lithium ion battery lithium analysis diagnostic apparatus is provided, including:
the learning calling module is used for calling the trained probability machine learning model; the probability machine learning model is obtained by training on the basis of off-line battery aging data obtained by off-line testing, wherein the off-line battery aging data comprises capacity degradation track data, alternating current impedance degradation track data and label data of battery lithium analysis in the lithium ion battery aging process;
the online data module is used for acquiring online monitoring capacity degradation track data and online monitoring characteristic impedance degradation track data of the lithium ion battery to be diagnosed;
and the online diagnosis module is used for online estimating the lithium analysis probability of the lithium ion battery to be diagnosed through the probability machine learning model according to the online monitoring capacity degradation track data and the online monitoring characteristic impedance degradation track data to obtain the online lithium analysis diagnosis result of the lithium ion battery to be diagnosed.
In still another aspect, a battery diagnosis device is further provided, which includes a memory and a processor, the memory stores a computer program, and the processor implements the steps of any one of the above lithium ion battery lithium analysis diagnosis methods when executing the computer program.
In still another aspect, a computer-readable storage medium is provided, on which a computer program is stored, and the computer program, when executed by a processor, implements the steps of any one of the above lithium ion battery lithium analysis diagnosis methods.
One of the above technical solutions has the following advantages and beneficial effects:
compared with the traditional diagnosis method, the lithium ion battery lithium analysis diagnosis method, the device, the equipment and the medium, the method, the device and the medium have the advantages that various aging information is fused, the subjectivity of manually judging the lithium analysis interface is effectively reduced, the accuracy of a diagnosis result can be greatly improved, and the method, the device, the equipment and the medium are easy to use. Specifically, the method comprises the following steps: 1) the problem of judging whether the lithium ion battery is subjected to lithium analysis aging or not is converted into soft classification on the basis of design thinking, and the lithium analysis probability is used as an index for battery lithium analysis diagnosis, so that the thinking is novel and has strong operability; 2) training a probabilistic machine learning model based on battery capacity degradation track data, alternating current impedance degradation track data and battery lithium analysis label data obtained by offline testing, wherein the trained model can integrate various aging information of the battery; 3) according to the online monitoring capacity degradation track data and the online monitoring characteristic impedance degradation track data of the lithium ion battery to be diagnosed, the probability machine learning model trained offline is called to perform online estimation on the lithium analysis aging probability of the lithium ion battery, online estimation of the lithium analysis aging probability index is used for achieving online diagnosis of the lithium analysis aging of the battery, and the method has good usability and can greatly improve the accuracy of a diagnosis result.
Drawings
FIG. 1 is a schematic flow chart of a lithium ion battery lithium analysis diagnostic method according to an embodiment;
FIG. 2 is a schematic diagram of a process for offline aging testing and model training of a battery in accordance with an embodiment;
FIG. 3 is a schematic diagram of a process for an embodiment of an offline burn-in test of a battery;
FIG. 4 is a graph illustrating the capacity fade trend of the battery with respect to lithium deposition aging and normal aging in one embodiment;
FIG. 5 is a schematic diagram of an alternate current impedance characteristic selection of a battery in one embodiment;
FIG. 6 is a graph illustrating the trend of AC impedance data between a lithium-aged battery and a normally aged battery in one embodiment; wherein, (a) is the development trend of the AC impedance data of the normally aged battery, and (b) is the development trend of the AC impedance data of the lithium-analyzed aged battery;
FIG. 7 is a schematic flow chart illustrating the on-line estimation of the lithium analysis probability of the battery according to one embodiment;
fig. 8 is a schematic block diagram of a lithium ion battery lithium analysis diagnostic apparatus according to an embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein in the description of the present application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application.
In addition, the technical solutions in the embodiments of the present invention may be combined with each other, but it must be based on the realization of those skilled in the art, and when the technical solutions are contradictory or cannot be realized, such a combination of technical solutions should be considered to be absent and not within the protection scope of the present invention.
Because the lithium-separation side reaction occurs in the lithium ion battery, and the lithium ion battery is used as a closed electrochemical system, the internal physical and chemical processes are very complicated, and the accurate diagnosis of whether the lithium separation occurs in the battery has certain difficulty.
In the conventional cadaver (Postmortem) diagnosis method, the lithium ion battery needs to be disassembled first, the negative electrode plate of the lithium ion battery is taken out, and then the diagnosis is carried out by using microscopic observation or using physicochemical detection means such as X-ray diffraction, nuclear magnetic resonance, neutron diffraction and the like.
In the conventional nondestructive diagnosis method, performance parameters such as the capacity and the internal resistance of the lithium ion battery are estimated on line by measuring signals such as the voltage, the current and the temperature of the lithium ion battery, and then whether lithium analysis occurs in the lithium ion battery is judged by combining with a relevant electrochemical mechanism inside the lithium ion battery. Such methods mainly include a Differential Voltage Analysis (DVA), an Incremental Capacity Analysis (ICA), and an aging behavior method. The DVA method and the ICA method are mainly used for judging whether lithium analysis occurs to the lithium ion battery in the current charging process, and the aging behavior method is mainly used for judging whether large-scale lithium analysis occurs to the lithium ion battery in the use history.
The invention aims to judge whether the lithium ion battery has excessive lithium analysis in use history or not, and further can be used for evaluating the safety of the lithium ion battery. Therefore, the invention mainly utilizes the aging characteristic data of the lithium ion battery to carry out lithium analysis diagnosis on the lithium ion battery, and can be regarded as an aging behavior method.
The traditional aging behavior method mainly diagnoses whether lithium analysis occurs to the lithium ion battery according to the capacity degradation condition, the internal resistance increase condition or the alternating current impedance spectrum change condition of the lithium ion battery, for example, the lithium analysis is diagnosed by using a battery capacity degradation track and an internal resistance degradation track, or the lithium analysis is diagnosed by using the current actual capacity value and the internal resistance value of the lithium ion battery, or the lithium analysis is diagnosed by using the impedance change condition of a characteristic frequency point, and the like. Although some of the above-mentioned research results of aging behavior methods exist at present, they are often difficult to use or have poor accuracy in practical applications, and for this reason, the inventors have found that there are two main reasons: 1. the aging characteristic data is relatively single and not comprehensive; 2. for parameters such as battery capacity and internal resistance, a threshold value or a boundary for judging whether lithium analysis occurs in the lithium ion battery needs to be manually determined.
In summary, the present invention provides a new lithium ion battery lithium analysis diagnostic method, which aims at the technical problem that the conventional lithium ion battery lithium analysis diagnostic method still has low diagnostic accuracy, and estimates the probability of lithium analysis aging of the lithium ion battery by using battery capacity degradation trajectory data and characteristic impedance degradation trajectory data as aging characteristic data of the battery and further using a probabilistic machine learning model, thereby implementing high-precision lithium analysis diagnosis of the lithium ion battery. The technical scheme of the invention integrates various aging information, reduces the subjectivity of manual judgment, can greatly improve the accuracy of a diagnosis result, and has better usability.
Referring to fig. 1, in one aspect, the present invention provides a lithium ion battery lithium deposition diagnosis method, including the following steps S12 to S16:
s12, calling the trained probability machine learning model; the probability machine learning model is obtained by training on the basis of off-line battery aging data obtained by off-line testing, and the off-line battery aging data comprises capacity degradation track data, alternating current impedance degradation track data and label data of battery lithium analysis in the lithium ion battery aging process.
It can be understood that the design idea of the invention is as follows: (1) converting the judgment of whether the lithium analysis aging of the lithium ion battery occurs into a soft classification problem, and using the lithium analysis probability as an index for battery lithium analysis diagnosis; (2) the soft classification is based on battery aging characteristic data, and mainly comprises capacity degradation track data and alternating current impedance degradation track data in the battery aging process; (3) according to the battery aging characteristic data, the soft classification problem is solved by using a probability machine learning model, so that the on-line estimation of the lithium analysis aging probability of the battery is realized.
The probabilistic machine learning model may be, but is not limited to, a related vector machine model, logistic regression, gaussian process, etc. in the prior art. For example, the label data for lithium analysis of the battery may be label data "1" for lithium analysis, label data "0" for no lithium analysis, or label data in the form of letters or other data, as long as it can effectively indicate whether lithium analysis occurs in the battery.
In some embodiments, the relationship of the offline battery aging data obtained with respect to the offline test to the probabilistic machine learning model is illustrated as follows: the off-line battery aging data in the lithium ion battery aging process is obtained by off-line testing of the lithium ion battery, and comprises capacity degradation track data, alternating current impedance degradation track data and label data of whether lithium analysis occurs in the battery aging process. Furthermore, the probability machine learning models such as a correlation vector machine can be trained by using the capacity degradation trajectory data and the alternating-current impedance degradation trajectory data as the input of the models and the label data of whether the battery separates lithium or not as the output, so that the trained probability machine learning models can be obtained in advance and loaded into the battery diagnosis equipment for online use. The training process and the mode of the probabilistic machine learning model can be understood by referring to the training process and the mode of the model in the prior art, and are not repeated in this specification.
And S14, acquiring the online monitoring capacity degradation track data and the online monitoring characteristic impedance degradation track data of the lithium ion battery to be diagnosed.
It is understood that the lithium ion battery to be diagnosed is an online lithium ion battery, which is the same type of battery as the lithium ion battery for which the offline test in the aforementioned step S12 is performed. The method comprises the steps of estimating the lithium analysis probability of the battery on line, firstly obtaining aging characteristic data such as a capacity degradation track and an alternating current impedance degradation track of the lithium ion battery to be diagnosed in the actual operation process, and then estimating the lithium analysis probability of the battery on line based on a probability machine learning model obtained by early off-line training, so as to judge whether the lithium analysis occurs to the battery. For example, in a set monitoring period, online monitoring capacity degradation trajectory data and online monitoring characteristic impedance degradation trajectory data obtained by online monitoring of the lithium ion battery to be diagnosed are input into the probabilistic machine learning model, that is, the probability of whether the lithium ion battery to be diagnosed analyzes lithium can be output, so that the lithium analysis online diagnosis result of the lithium ion battery to be diagnosed is obtained, such as whether lithium is analyzed, how high the lithium analysis probability is, and the like.
The capacity degradation track data and the characteristic impedance degradation track data are monitored on line and can be monitored and output on line by a battery management system of the lithium ion battery to be diagnosed.
And S16, according to the online monitoring capacity degradation track data and the online monitoring characteristic impedance degradation track data, performing online estimation on the lithium analysis probability of the lithium ion battery to be diagnosed through a probability machine learning model to obtain the online lithium analysis diagnosis result of the lithium ion battery to be diagnosed.
It can be understood that after the online monitoring capacity degradation trajectory data and the online monitoring characteristic impedance degradation trajectory data are obtained, the data are input to the trained probability machine learning model, and then the corresponding online diagnosis result can be output, that is, the online estimation of the lithium analysis probability of the lithium ion battery to be diagnosed is realized, so that the lithium analysis diagnosis of the lithium ion battery is realized.
Compared with the traditional diagnosis method, the lithium ion battery lithium analysis diagnosis method disclosed by the invention integrates various aging information, effectively reduces the subjectivity of manually judging the lithium analysis interface, can greatly improve the accuracy of a diagnosis result, and has better usability. Specifically, the method comprises the following steps: 1) the problem of judging whether the lithium ion battery is subjected to lithium analysis aging or not is converted into soft classification on the basis of design thinking, and the lithium analysis probability is used as an index for battery lithium analysis diagnosis, so that the thinking is novel and has strong operability; 2) training a probabilistic machine learning model based on battery capacity degradation track data, alternating current impedance degradation track data and battery lithium analysis label data obtained by offline testing, wherein the trained model can integrate various aging information of the battery; 3) according to the online monitoring capacity degradation track data and the online monitoring characteristic impedance degradation track data of the lithium ion battery to be diagnosed, the probability machine learning model trained offline is called to perform online estimation on the lithium analysis aging probability of the lithium ion battery, online estimation of the lithium analysis probability index is used for achieving online diagnosis of the lithium analysis aging of the battery, and the method has good usability and can greatly improve the accuracy of a diagnosis result.
Referring to fig. 2 and 3, in an embodiment, the capacity degradation trajectory data during the aging process of the lithium ion battery is obtained by performing a battery capacity test during an offline cyclic aging test of the lithium ion battery under different temperature stress levels and charging current stress levels.
It can be understood that, regarding the offline aging test and model training part of the battery, fig. 2 is a schematic flow chart of the offline aging test and model training of the battery. The method aims to obtain capacity degradation track data, alternating current impedance degradation track data and label data for judging whether lithium is analyzed or not for a battery offline aging test, wherein the capacity degradation track data, the alternating current impedance degradation track data and the label data are used for training a probabilistic machine learning model such as a correlation vector machine. Fig. 3 is a schematic flow chart of the battery offline aging test.
Since lithium separation from the battery occurs only during cyclic aging, and whether lithium separation from the battery occurs is determined by both the charging current and the battery temperature, this aging test will take into account both the battery temperature and the charging current stress during cyclic aging. Specifically, a plurality of temperature stress levels and charging current stress levels can be respectively arranged according to the characteristics of the battery, and then a plurality of (same) lithium ion batteries are respectively distributed to each group of stress levels to carry out the cyclic aging test. In addition, each time the calibration cycle number M is reached, a capacity test (for obtaining capacity degradation trajectory data of the battery) and an alternating current impedance spectrum test (for obtaining alternating current impedance degradation trajectory data of the battery) can be arranged, so that the capacity degradation trajectory data and the alternating current impedance degradation trajectory data of the battery can be obtained.
And finally, when the accumulated cycle number of a certain battery reaches the set total preset cycle number N, performing a disassembly experiment on the certain battery, and determining whether a lithium analysis side reaction occurs by using a related physicochemical detection means such as X-ray diffraction, so that label data of whether the lithium analysis occurs in the battery can be directly obtained. It should be noted that different predetermined cycle times may be set for the cells in the same set of stress levels to obtain the aging data of the cells at different aging stages.
In one embodiment, the capacity test in fig. 3 refers to a capacity (calibration) test of the battery at a normal temperature of 25 degrees celsius using a standard charging and discharging test condition recommended by a battery manufacturer. And taking the average value of the three capacity test values as the current actual capacity value of the battery. Through multiple capacity calibration tests, the capacity degradation track data of the battery can be obtained in an accumulated mode. Here, it should be noted that the battery capacity referred to in the present application is a capacity value measured at 25 degrees celsius by a standard charge/discharge condition test.
In one embodiment, as shown in fig. 2 and 3, the ac impedance degradation trajectory data during the aging process of the lithium ion battery is obtained by performing an ac impedance spectrum test during an offline cyclic aging test of the lithium ion battery under different temperature stress levels and charging current stress levels.
It can be understood that, for the acquisition of the data of the ac impedance degradation trajectory in the offline aging test of the battery, reference may be made specifically to the description of the ac impedance spectrum test described above. The impedance spectrum (calibration) test in fig. 3 is to obtain the ac impedance degradation trajectory data of the battery at 25 degrees celsius, and the impedance spectrum of the battery is tested at 25 degrees celsius at normal temperature by using the existing devices such as the electrochemical workstation, and the voltage or current disturbance test method can be used in the actual operation.
In one embodiment, the AC impedance degeneration trace data is impedance data obtained by performing AC impedance spectrum test in a frequency band of 1kHz-1 Hz.
It can be understood that for the selection of the battery aging characteristics, the capacity and the alternating current impedance of the lithium ion battery are selected as the battery aging characteristics, and the capacity degradation track data and the alternating current impedance degradation track data are used as the basis for diagnosing whether the lithium analysis aging occurs in the battery.
In the normal aging process of the lithium ion battery, the battery capacity gradually and slowly decays; and if the lithium ion battery is subjected to lithium precipitation aging, the capacity of the lithium ion battery is rapidly reduced. FIG. 4 is a graph showing the capacity fade trend of the battery with respect to lithium deposition aging and normal aging.
The frequency range of a general alternating current impedance test is wide, and impedance data of a proper frequency range needs to be specifically selected as characteristic quantity for diagnosing the lithium precipitation aging of the battery. The inventor finds in research that because lithium analysis reaction occurs at a solid-liquid interface of a battery cathode and is related to a middle-high frequency range polarization process such as charge transfer, impedance data in a middle-high frequency range of 1kHz-1Hz can be selected as a characteristic quantity. In practical operation, only impedance values of a plurality of frequency points may be specifically selected as a representative of the impedance data, such as but not limited to 1kHz frequency point impedance, 100Hz frequency point impedance, 10Hz frequency point impedance, and 1Hz frequency point impedance. FIG. 5 shows battery AC impedance data and selectable characteristic frequency points in the form of a Nyquist plotImpedance value, wherein, Z Im Representing the imaginary part of the AC impedance, Z Re Representing the real part of the ac impedance.
In the following description, for convenience, the characteristic frequency point impedance selected from the middle and high frequency range of 1kHz to 1Hz will be represented by "characteristic impedance". In practical operation, different specific characteristic frequency points can be selected from the frequency band.
In some embodiments, the impedance value of the lithium ion battery in the full frequency band gradually increases during the normal aging process of the lithium ion battery. If the lithium ion battery is subjected to lithium precipitation aging, the impedance value of the medium-high frequency band is not obviously increased due to the strong conductivity of lithium metal, and even part of the characteristic impedance value is possibly reduced to a certain extent. Fig. 6 shows a schematic diagram of the development trend of the ac impedance data of the aged lithium battery and the aged normal battery, wherein (a) is the development trend of the ac impedance data of the aged normal battery, and (b) is the development trend of the ac impedance data of the aged lithium battery.
As shown in fig. 3, after the tested lithium ion battery is aged to a certain degree, the lithium ion battery is disassembled and whether lithium analysis occurs is determined by using a necropsy means such as X-ray diffraction, so as to directly obtain the classification label data of whether lithium analysis occurs in the battery.
In one embodiment, with respect to the offline training of the probabilistic machine learning model described above: and (3) using the capacity degradation track data and the alternating-current impedance degradation track data as a basis for diagnosing whether the battery is subjected to lithium analysis aging, and judging the probability of the battery in lithium analysis by using a probability machine learning model such as a correlation vector machine. Specifically, in the model offline training part, capacity degradation trajectory data and characteristic impedance degradation trajectory data are used as input, classification label data are used as output, and an equiprobable machine learning model such as a relevance vector machine is subjected to offline training, wherein the goal of model optimization is to enable the classification accuracy of whether lithium analysis occurs to be highest.
Referring to fig. 7, in an embodiment, the process of acquiring the online monitoring capacity degradation trajectory data may specifically include the following processing steps:
acquiring battery voltage data and current data of the lithium ion battery to be diagnosed, and performing data preprocessing on the battery voltage data and the current data;
and performing online estimation on the battery capacity of the lithium ion battery to be diagnosed in a monitoring period based on an online capacity estimation method according to the preprocessed battery voltage data and current data to obtain online monitoring capacity degradation track data.
In an embodiment, as shown in fig. 7, the process of acquiring the online monitoring characteristic impedance degradation trajectory data may specifically include the following processing steps:
acquiring a characteristic impedance value of a lithium ion battery to be diagnosed measured on line and battery temperature data measured on line under a set frequency band (characteristic frequency);
performing impedance conversion according to the characteristic impedance value and the battery temperature data to obtain a characteristic impedance value at normal temperature;
and acquiring the characteristic impedance value at normal temperature in the monitoring period to obtain the online monitoring characteristic impedance degradation track data.
In some embodiments, a specific flow of the battery lithium analysis probability online estimation is shown in fig. 7. On the one hand, firstly, for the lithium ion battery to be diagnosed, the battery voltage monitored on line can be measured by the battery management systemV t Current ofI t And temperatureT t The signals are preprocessed. Then, the capacity value of the lithium ion battery can be estimated online based on the existing mature capacity estimation method (such as a model-based estimation method, a capacity increment method and the like). After accumulation of a period of time (i.e., a monitoring period, which can be flexibly set according to monitoring requirements), online monitoring capacity degradation trajectory data of the lithium ion battery can be obtained.
On the other hand, the existing special circuit in the field can be used for carrying out on-line measurement on the characteristic impedance of the lithium ion battery under a specific frequency point, or the existing machine learning method can be used for estimating the value of the characteristic impedance of the lithium ion battery on line. In addition, since the actual impedance value of the battery has a strong correlation with the temperature, and the battery temperature may fluctuate in actual operation, the value of the characteristic impedance measured (online) or estimated online actually needs to be converted according to an empirical formula given in the art, so as to obtain the impedance value of the lithium ion battery corresponding to the normal temperature of 25 ℃. After accumulation of a period of time (namely a monitoring period), online monitoring characteristic impedance degradation track data can be obtained.
And finally, according to the capacity degradation trajectory data and the characteristic impedance degradation trajectory data, performing online estimation on the lithium analysis probability of the battery by using a probability machine learning model trained offline, thereby realizing the lithium analysis diagnosis of the battery.
It should be noted that the contents of the online capacity estimation method, the online impedance measurement method, and the impedance conversion method at different temperatures involved in the flow shown in fig. 7 are well-known in the art and will not be described in detail in this specification.
It should be understood that although the various steps in the flowcharts of fig. 1-3 and 7 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps of fig. 1-3 and 7 may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performing the sub-steps or stages is not necessarily sequential, but may be performed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.
Referring to fig. 8, in an embodiment, there is further provided a lithium ion battery lithium analysis diagnostic apparatus 100, which includes a learning calling module 11, an online data module 13, and an online diagnostic module 15. The learning calling module 11 is configured to call a trained probabilistic machine learning model; the probabilistic machine learning model is obtained by training based on battery aging data obtained by offline testing, wherein the battery aging data comprises capacity degradation track data, alternating current impedance degradation track data and label data of battery lithium analysis in the lithium ion battery aging process. The online data module 13 is configured to obtain battery capacity degradation trajectory data and characteristic impedance degradation trajectory data for online monitoring of the lithium ion battery. The online diagnosis module 15 is configured to perform online estimation on the lithium analysis probability of the lithium ion battery through a probability machine learning model according to the online monitored battery capacity degradation trajectory data and characteristic impedance degradation trajectory data, so as to obtain an online diagnosis result of lithium analysis of the lithium ion battery.
Compared with the traditional diagnosis method, the lithium ion battery lithium analysis diagnosis device 100 integrates various aging information, effectively reduces subjectivity of manually judging a lithium analysis interface, can greatly improve accuracy of a diagnosis result, and has better usability. Specifically, the method comprises the following steps: 1) the problem of judging whether the lithium ion battery is subjected to lithium analysis aging or not is converted into soft classification on the basis of design thinking, and the lithium analysis probability is used as an index for battery lithium analysis diagnosis, so that the thinking is novel and has strong operability; 2) training a probabilistic machine learning model based on battery capacity degradation track data, alternating current impedance degradation track data and battery lithium analysis label data obtained by offline testing, wherein the trained model can integrate various aging information of the battery; 3) according to the online monitoring capacity degradation track data and the online monitoring characteristic impedance degradation track data of the lithium ion battery to be diagnosed, the probability machine learning model trained offline is called to perform online estimation on the lithium analysis aging probability of the lithium ion battery, online estimation of the lithium analysis aging probability index is used for achieving online diagnosis of the lithium analysis aging of the battery, and the method has good usability and can greatly improve the accuracy of a diagnosis result.
In one embodiment, the online data module described above may include a pre-processing sub-module and an estimation sub-module. The preprocessing submodule is used for acquiring battery voltage data and current data of the lithium ion battery to be diagnosed, and performing data preprocessing on the battery voltage data and the current data. And the estimation submodule is used for carrying out online estimation on the battery capacity of the lithium ion battery to be diagnosed in a monitoring time period based on an online capacity estimation method according to the preprocessed battery voltage data and the preprocessed current data to obtain online monitoring capacity degradation track data.
For specific limitations of the lithium ion battery lithium analysis diagnostic apparatus 100, reference may be made to the corresponding limitations of the lithium ion battery lithium analysis diagnostic method above, and details are not repeated here. The respective modules in the lithium ion battery lithium analysis diagnostic apparatus 100 described above may be entirely or partially implemented by software, hardware, or a combination thereof. The modules may be embedded in a hardware form or a device independent of a specific data processing function, or may be stored in a memory of the device in a software form, so that a processor may invoke and execute operations corresponding to the modules, where the device may be, but is not limited to, various computer devices, battery management devices, or monitoring devices in the art.
In still another aspect, a battery diagnosis device is further provided, which includes a memory and a processor, the memory stores a computer program, and the processor executes the computer program to implement the following steps: calling a trained probabilistic machine learning model; the probability machine learning model is obtained by training on the basis of off-line battery aging data obtained by off-line testing, wherein the off-line battery aging data comprises capacity degradation track data, alternating current impedance degradation track data and label data of battery lithium analysis in the lithium ion battery aging process; acquiring online monitoring capacity degradation track data and online monitoring characteristic impedance degradation track data of a lithium ion battery to be diagnosed; and according to the online monitoring capacity degradation track data and the online monitoring characteristic impedance degradation track data, online estimation is carried out on the lithium analysis probability of the lithium ion battery to be diagnosed through a probability machine learning model, and the online lithium analysis diagnosis result of the lithium ion battery to be diagnosed is obtained.
It should be noted that the battery diagnosis device may be, but not limited to, a battery management device, a battery monitoring computer or a microprocessor device in the art, and in addition to the memory and the processor, it also includes other necessary components not listed in detail in this specification, depending on the specific model of the battery diagnosis device.
In one embodiment, the processor, when executing the computer program, may further implement the additional steps or substeps in the embodiments of the lithium ion battery lithium analysis diagnostic method described above.
In yet another aspect, there is also provided a computer readable storage medium having a computer program stored thereon, the computer program when executed by a processor implementing the steps of: calling a trained probabilistic machine learning model; the probability machine learning model is obtained by training on the basis of off-line battery aging data obtained by off-line testing, wherein the off-line battery aging data comprises capacity degradation track data, alternating current impedance degradation track data and label data of battery lithium analysis in the lithium ion battery aging process; acquiring online monitoring capacity degradation track data and online monitoring characteristic impedance degradation track data of a lithium ion battery to be diagnosed; and according to the online monitoring capacity degradation track data and the online monitoring characteristic impedance degradation track data, online estimation is carried out on the lithium analysis probability of the lithium ion battery to be diagnosed through a probability machine learning model, and the online lithium analysis diagnosis result of the lithium ion battery to be diagnosed is obtained.
In one embodiment, when being executed by a processor, the computer program may further implement the additional steps or sub-steps in each embodiment of the lithium ion battery lithium analysis diagnostic method.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware related to instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), synchronous Link (Synchlink) DRAM (SLDRAM), Rambus DRAM (RDRAM), and interface DRAM (DRDRAM).
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above examples only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for those skilled in the art, various changes and modifications can be made without departing from the spirit of the present application, and all of them fall within the scope of the present application. Therefore, the protection scope of the present patent should be subject to the appended claims.
Claims (10)
1. A lithium ion battery lithium analysis diagnosis method is characterized by comprising the following steps:
calling a trained probabilistic machine learning model; the probability machine learning model is obtained by training on the basis of off-line battery aging data obtained by off-line testing, wherein the off-line battery aging data comprises capacity degradation track data, alternating current impedance degradation track data and label data of battery lithium analysis in the lithium ion battery aging process;
acquiring online monitoring capacity degradation track data and online monitoring characteristic impedance degradation track data of a lithium ion battery to be diagnosed;
and according to the online monitoring capacity degradation track data and the online monitoring characteristic impedance degradation track data, online estimation is carried out on the lithium analysis probability of the lithium ion battery to be diagnosed through the probability machine learning model, and the online lithium analysis diagnosis result of the lithium ion battery to be diagnosed is obtained.
2. The lithium ion battery lithium analysis diagnosis method according to claim 1, wherein the online monitoring of the acquisition process of the capacity degradation trajectory data comprises:
acquiring battery voltage data and current data for online monitoring of the lithium ion battery to be diagnosed, and performing data preprocessing on the battery voltage data and the current data;
and performing online estimation on the battery capacity of the lithium ion battery to be diagnosed in a monitoring period based on an online capacity estimation method according to the preprocessed battery voltage data and the preprocessed current data to obtain online monitoring capacity degradation track data.
3. The lithium ion battery lithium analysis diagnosis method according to claim 1, wherein the process of online monitoring of the acquisition of characteristic impedance degradation trajectory data comprises:
acquiring a characteristic impedance value and battery temperature data measured on line on the lithium ion battery to be diagnosed under a set frequency band;
performing impedance conversion according to the characteristic impedance value and the battery temperature data to obtain a characteristic impedance value at normal temperature;
and acquiring the characteristic impedance value at normal temperature in a monitoring period to obtain the online monitoring characteristic impedance degradation track data.
4. The lithium ion battery lithium analysis diagnosis method according to claim 1, wherein the capacity degradation trajectory data in the lithium ion battery aging process is obtained by performing a battery capacity test in an offline cyclic aging test process performed on the lithium ion battery under different temperature stress levels and charging current stress levels.
5. The lithium ion battery lithium analysis diagnosis method according to claim 1, wherein the ac impedance degradation trajectory data in the lithium ion battery aging process is obtained by performing an ac impedance spectrum test in an off-line cyclic aging test process of the lithium ion battery under different temperature stress levels and charging current stress levels.
6. The lithium ion battery lithium analysis diagnosis method according to claim 5, wherein the alternating current impedance degeneration trajectory data is impedance data obtained by performing an alternating current impedance spectrum test within a frequency range of 1kHz-1 Hz.
7. A lithium ion battery lithium deposition diagnostic device is characterized by comprising:
the learning calling module is used for calling the trained probability machine learning model; the probability machine learning model is obtained by training on the basis of off-line battery aging data obtained by off-line testing, wherein the off-line battery aging data comprises capacity degradation track data, alternating current impedance degradation track data and label data of battery lithium analysis in the lithium ion battery aging process;
the online data module is used for acquiring online monitoring capacity degradation track data and online monitoring characteristic impedance degradation track data of the lithium ion battery to be diagnosed;
and the online diagnosis module is used for online estimating the lithium analysis probability of the lithium ion battery to be diagnosed through the probability machine learning model according to the online monitoring capacity degradation track data and the online monitoring characteristic impedance degradation track data to obtain the lithium analysis online diagnosis result of the lithium ion battery to be diagnosed.
8. The lithium ion battery lithium analysis diagnostic device of claim 7, wherein the online data module comprises:
the preprocessing submodule is used for acquiring battery voltage data and current data of the lithium ion battery to be diagnosed, and performing data preprocessing on the battery voltage data and the current data;
and the estimation submodule is used for carrying out online estimation on the battery capacity of the lithium ion battery to be diagnosed in a monitoring period based on an online capacity estimation method according to the preprocessed battery voltage data and the preprocessed current data to obtain online monitoring capacity degradation track data.
9. A battery diagnosis apparatus comprising a memory and a processor, the memory storing a computer program, wherein the processor implements the steps of the lithium ion battery lithium analysis diagnosis method according to any one of claims 1 to 6 when executing the computer program.
10. A computer-readable storage medium, on which a computer program is stored, wherein the computer program, when being executed by a processor, implements the steps of the lithium ion battery lithium analysis diagnostic method according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210044751.8A CN114879043A (en) | 2022-01-14 | 2022-01-14 | Lithium ion battery lithium analysis diagnosis method, device, equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210044751.8A CN114879043A (en) | 2022-01-14 | 2022-01-14 | Lithium ion battery lithium analysis diagnosis method, device, equipment and medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114879043A true CN114879043A (en) | 2022-08-09 |
Family
ID=82666661
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210044751.8A Pending CN114879043A (en) | 2022-01-14 | 2022-01-14 | Lithium ion battery lithium analysis diagnosis method, device, equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114879043A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117521857A (en) * | 2024-01-05 | 2024-02-06 | 宁德时代新能源科技股份有限公司 | Battery cell lithium analysis method and device, readable storage medium and electronic equipment |
-
2022
- 2022-01-14 CN CN202210044751.8A patent/CN114879043A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117521857A (en) * | 2024-01-05 | 2024-02-06 | 宁德时代新能源科技股份有限公司 | Battery cell lithium analysis method and device, readable storage medium and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110794305B (en) | Power battery fault diagnosis method and system | |
| CN110161414B (en) | Power battery thermal runaway online prediction method and system | |
| Gan et al. | Data-driven fault diagnosis of lithium-ion battery overdischarge in electric vehicles | |
| EP3721499B1 (en) | Method and system for online assessing state of health of a battery | |
| Berecibar et al. | Critical review of state of health estimation methods of Li-ion batteries for real applications | |
| CN112834945B (en) | Evaluation model establishment method, battery health state evaluation method and related products | |
| CN111257755B (en) | Method for preventive detection and diagnosis of battery pack | |
| CN112098850B (en) | Lithium ion battery voltage fault diagnosis method and system based on SDO algorithm | |
| CN113447828B (en) | Lithium battery temperature estimation method and system based on Bayesian neural network | |
| CN110988695A (en) | Power battery health state evaluation method and device, storage medium and electronic equipment | |
| CN111983474A (en) | Lithium ion battery life prediction method and system based on capacity decline model | |
| CN114280479A (en) | Electrochemical impedance spectrum-based rapid sorting method for retired batteries | |
| CN114580496A (en) | Self-detection method for single fault of lithium ion battery | |
| CN114879043A (en) | Lithium ion battery lithium analysis diagnosis method, device, equipment and medium | |
| CN117330987B (en) | Method, system, medium and apparatus for time-based battery state of health assessment | |
| CN113391214A (en) | Battery micro-fault diagnosis method based on battery charging voltage ranking change | |
| CN113687251A (en) | Dual-model-based lithium ion battery pack voltage abnormity fault diagnosis method | |
| CN115792636A (en) | Battery pack tiny fault quantitative diagnosis method and system based on maximum voltage value | |
| CN115128468A (en) | Chemical energy storage battery PHM undervoltage fault prediction method | |
| CN111190112B (en) | Battery charging and discharging prediction method and system based on big data analysis | |
| Zhou et al. | A review of state-of-health estimation of lithiumion batteries: Experiments and data | |
| Bose et al. | Sensitivity Analysis of Battery Digital Twin Design Variables Using Genetic Programming | |
| CN117970128B (en) | Battery comprehensive experiment debugging method and system based on real-time feedback control | |
| Zhang et al. | Support vector machine based lithium-ion battery electrolyte leakage fault diagnosis method | |
| Mao et al. | A Multi-Fault Diagnosis Method for Battery Packs Based on Low-Redundancy Representation Learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |