CN115308609B - Lithium ion battery thickness prediction method and device and lithium ion battery - Google Patents
Lithium ion battery thickness prediction method and device and lithium ion battery Download PDFInfo
- Publication number
- CN115308609B CN115308609B CN202210923857.5A CN202210923857A CN115308609B CN 115308609 B CN115308609 B CN 115308609B CN 202210923857 A CN202210923857 A CN 202210923857A CN 115308609 B CN115308609 B CN 115308609B
- Authority
- CN
- China
- Prior art keywords
- lithium ion
- discharge
- ion battery
- charge
- thickness
- 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.)
- Active
Links
- HBBGRARXTFLTSG-UHFFFAOYSA-N Lithium ion Chemical compound [Li+] HBBGRARXTFLTSG-UHFFFAOYSA-N 0.000 title claims abstract description 75
- 229910001416 lithium ion Inorganic materials 0.000 title claims abstract description 75
- 238000000034 method Methods 0.000 title claims abstract description 56
- 238000003062 neural network model Methods 0.000 claims abstract description 23
- 238000012545 processing Methods 0.000 claims abstract description 9
- 238000000605 extraction Methods 0.000 claims abstract description 5
- 208000028659 discharge Diseases 0.000 claims description 117
- 238000007600 charging Methods 0.000 claims description 31
- 238000007599 discharging Methods 0.000 claims description 17
- 238000013461 design Methods 0.000 claims description 12
- 238000013528 artificial neural network Methods 0.000 claims description 9
- 230000001537 neural effect Effects 0.000 claims description 3
- 238000001514 detection method Methods 0.000 abstract description 11
- 238000012360 testing method Methods 0.000 description 15
- 238000012549 training Methods 0.000 description 9
- 238000012544 monitoring process Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000007726 management method Methods 0.000 description 3
- WHXSMMKQMYFTQS-UHFFFAOYSA-N Lithium Chemical compound [Li] WHXSMMKQMYFTQS-UHFFFAOYSA-N 0.000 description 2
- 238000004140 cleaning Methods 0.000 description 2
- 238000001125 extrusion Methods 0.000 description 2
- 229910052744 lithium Inorganic materials 0.000 description 2
- 238000001556 precipitation Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010280 constant potential charging Methods 0.000 description 1
- 238000010277 constant-current charging Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000003792 electrolyte Substances 0.000 description 1
- 238000004146 energy storage Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 229910002804 graphite Inorganic materials 0.000 description 1
- 239000010439 graphite Substances 0.000 description 1
- 231100001261 hazardous Toxicity 0.000 description 1
- 238000009830 intercalation Methods 0.000 description 1
- 230000002687 intercalation Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005215 recombination Methods 0.000 description 1
- 230000006798 recombination Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000007086 side reaction Methods 0.000 description 1
- 239000007784 solid electrolyte Substances 0.000 description 1
- 230000008719 thickening Effects 0.000 description 1
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]
- G01R31/378—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC] specially adapted for the type of battery or accumulator
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Secondary Cells (AREA)
Abstract
The invention belongs to the technical field of lithium ion batteries, and particularly relates to a lithium ion battery thickness prediction method, a lithium ion battery thickness prediction device and a lithium ion battery. The prediction method comprises the following steps: acquiring charge and discharge data of the lithium ion battery to be tested in the charge and discharge cycle process under the current cycle number; after the current cycle number is cut off, extracting characteristic values of the charge and discharge data to obtain a plurality of characteristic values for prediction; and processing the plurality of predicted characteristic values by using a pre-constructed neural network model to obtain the predicted thickness value of the lithium ion battery to be detected. The prediction device includes: the device comprises an acquisition module, an extraction module and an obtaining module. The lithium ion battery comprises a battery management system with a prediction means of the lithium ion battery. Through the technical scheme, the thickness change of the battery can be detected on line in real time, the artificial error of off-line detection is avoided, and the advance prediction with higher precision can be provided for the judgment of the battery thickness failure.
Description
Technical Field
The invention belongs to the technical field of lithium ion batteries, and particularly relates to a lithium ion battery thickness prediction method, a lithium ion battery thickness prediction device and a lithium ion battery.
Background
At present, with the rapid development of 3C portable electronic consumer products, power automobiles and terminal energy storage devices, the demand for lithium ion batteries is increasing. The lithium ion battery has the advantages of high energy, long service life, recoverability and the like, but certain potential safety hazards such as thermal failure caused by short circuit in the battery, hazardous gas generated by battery inflation, precipitation risk of the battery and the like exist in the use process of the lithium ion battery. Monitoring the safety performance or failure index of the batteries is an important guarantee for realizing the safe work of the lithium ion batteries.
The thickness change of the battery in the cycle process is always an important index concerned by manufacturers and terminal customers, and mainly comes from physical expansion caused by lithium intercalation at the negative electrode, SEI (Solid Electrolyte Interface) thickening caused by continuous fracture and recombination of the SEI in the cycle process, and the battery at the later cycle period is often accompanied with the generation of lithium precipitation and severe generation of a large amount of harmful gas due to side reaction of the Electrolyte under severe conditions. Thus, when the thickness of the battery reaches a certain value at a certain time node, the battery is considered to have failed, with a certain safety risk. Therefore, monitoring the thickness of the battery is of great practical value.
The method for monitoring the thickness of the battery mainly adopts an on-line detection method and an off-line detection method, wherein the off-line detection method mainly comprises the following steps: after the battery is circulated for 50 weeks or 100 weeks, taking the battery out of the cabinet, and measuring the thickness of the battery by adopting an infrared thickness meter; another online monitoring method mainly comprises the following steps: and calculating the thickness of the battery under each cycle by using the corresponding relation between the expansion stress generated by battery expansion and the thickness of the battery core.
The current market and most researchers have limitations in the study of cell thickness. The offline detection method can save cost, but real-time monitoring cannot be performed, and the problems of artificial errors of the cell test, test consistency and the like can be caused by repeated loading and unloading of the battery. The online detection method can really monitor the thickness of the battery in real time, but has little effect in the application field, and mainly has the advantages that due to the high cost, hundreds of thousands of equipment expenses need to be consumed for testing single electrical property, and the difficulty in testing a plurality of batteries is high; and secondly, the practical application value of the continuity test is not obvious. Finally, no matter the thickness of the battery is detected on line or off line, the thickness detection method cannot be applied to terminal products. Thus the disadvantages are obvious.
Disclosure of Invention
Aiming at the defects of the prior art, the invention aims to provide a method for predicting the thickness of a lithium ion battery, which can predict the thickness of the battery on line in real time based on a big data model constructed in advance, provides accurate early warning for the failure of the battery and is suitable for terminal products.
In order to solve the technical problems, the invention provides the following technical scheme:
in one aspect, a method for predicting a thickness of a lithium ion battery is provided, where the method includes: acquiring charge and discharge data of the lithium ion battery to be tested in the charge and discharge cycle process under the current cycle number; after the current cycle number is cut off, extracting characteristic values of the charge and discharge data to obtain a plurality of characteristic values for prediction; and processing the plurality of predicted characteristic values by using a pre-constructed neural network model to obtain the predicted thickness value of the lithium ion battery to be detected.
In the prediction method as described above, optionally, the charge and discharge data includes: voltage, capacity, design thickness and discharge time of the lithium ion battery to be tested corresponding to the current cycle number.
In the prediction method as described above, optionally, the plurality of feature values for prediction includes: the current cycle number n and the design thickness d of the lithium ion battery to be tested corresponding to the current cycle number 0 Discharge capacity Q in the nth charge-discharge cycle n And initial discharge capacity Q 0 Ratio eta of, charging time t in the process of the nth charge-discharge cycle n And the voltage difference delta V during the standing period after charging in the nth charging and discharging circulation process n (ii) a And an average mean value mean of dV/dQ ((dV/dQ)) in a discharge phase of the nth charge-discharge cycle n ) Variance var ((dV/dQ) n ) Range ((dV/dQ) n ) (ii) a Wherein, in the interval of 5% -95% of discharge DV curve SOC in the discharge stage of the nth charge-discharge cycle, values are calculated every 0.1% of SOC, the values of dV/dQ are respectively calculated, and then the average value of all the obtained dV/dQ values is mean ((dV/dQ) in the nth charge-discharge cycle) n ) The variance of all the dV/dQ values is calculated as var ((dV/dQ)) in the nth charge-discharge cycle n ) And the range ((dV/dQ) in the nth charge-discharge cycle) is obtained by calculating the range of all the dV/dQ values n )。
In the above prediction method, optionally, before the collecting charge and discharge data of the lithium ion battery to be measured in the charge and discharge cycle process under the current cycle number, the prediction method further includes: constructing a neural network model; wherein, a Neural Net Fitting tool in matlab software is adopted for modeling, and the algorithm is an artificial Neural network.
In the prediction method as described above, optionally, the neural network model has: an input layer, a hidden layer and an output layer; the number of units of the hidden layer is 15, and the algorithm is levenberg-Marquardt.
In the prediction method as described above, optionally, the lithium ion battery is a soft-package lithium ion battery.
In another aspect, an apparatus for predicting a thickness of a lithium ion battery is provided, which includes: the acquisition module is used for acquiring charge and discharge data of the lithium ion battery to be detected in the charge and discharge cycle process under the current cycle number; the extraction module is used for extracting characteristic values of the charge and discharge data after the current cycle number is cut off to obtain a plurality of characteristic values for prediction; and the obtaining module is used for processing the plurality of predicted characteristic values by using a pre-constructed neural network model to obtain the predicted thickness value of the lithium ion battery to be detected.
In the prediction apparatus as described above, optionally, the charge and discharge data includes: voltage, capacity, design thickness and charging time of the lithium ion battery to be tested corresponding to the current cycle number.
In another aspect, a lithium ion battery is provided, which includes a battery management system having the above prediction device for lithium ion battery.
The technical scheme of the invention has the following beneficial effects:
1. the thickness change of the battery can be detected in real time on line, the artificial error of off-line detection is avoided, and the battery thickness failure judgment can be predicted in advance with relatively high precision.
2. The method is not only suitable for battery manufacturing, but also suitable for terminal products (such as mobile phones and computers), the change situation of the battery thickness in the terminal products is monitored in real time, early warning is carried out after the battery thickness reaches a certain degree, and safety accidents caused by extrusion of the battery are avoided.
Drawings
Fig. 1 is a schematic flowchart of a method for predicting a thickness of a lithium ion battery according to an embodiment of the present invention;
FIG. 2 is a diagram illustrating a training result of a neural network model using a training set according to an embodiment of the present invention;
fig. 3 is a schematic diagram of a test result of a neural network model using a test set according to an embodiment of the present invention.
Fig. 4 is a schematic structural diagram of a device for predicting a thickness of a lithium ion battery according to an embodiment of the present invention.
Detailed Description
The invention provides a method for predicting the thickness of a lithium ion battery based on a big data artificial intelligence algorithm, which can realize the nondestructive online detection of the thickness of the battery. To make the objects, technical solutions and advantages of the present disclosure more apparent, embodiments of the present disclosure will be described in detail with reference to the accompanying drawings.
Referring to fig. 1, an embodiment of the present invention provides a method for predicting a thickness of a lithium ion battery, including the following steps:
The charge and discharge data of the lithium ion battery to be tested in the charge and discharge cycle process are collected and recorded, and generally, one lithium ion battery can be charged and discharged for many times, such as 600 times, 800 times, 1000 times and the like. For each charge-discharge cycle, called cycle week, i.e. the charge-discharge data during the charge-discharge cycle under the corresponding cycle number is collected, the frequency of collection may be 20-40 ms/time, preferably 30 ms/time. The battery can be charged with constant current and constant voltage by adopting any charging rate of 1C-6C, when discharging, the battery can be discharged with constant current by adopting 1C discharging rate, and the battery can be kept still for a preset time (or called standing period) after the charging is ended before discharging and using, wherein the preset time can be 5min or 4.5min, and the embodiment does not limit the length of the preset time. The charge and discharge data include: voltage, capacity, design thickness, and charging time. The thickness of the lithium ion battery to be tested can change along with multiple charging and discharging cycles of the lithium ion battery, and the designed thickness is the thickness of the lithium ion battery corresponding to the current cycle number and can be obtained from a battery manufacturer. The voltage may be obtained by a voltage sampling circuit. The capacity can be calculated from the discharge time and the discharge current.
In order to improve the prediction accuracy, after the acquired charge and discharge data, the charge and discharge data are cleaned, namely, invalid data, data with errors and data which are seriously deviated from the actual condition (such as exceeding 3 sigma) are eliminated.
And step 102, after the current cycle number is cut off, extracting characteristic values of the charge and discharge data to obtain a plurality of characteristic values for prediction.
The extracted characteristic values for prediction include: the current cycle number n and the design thickness d of the lithium ion battery to be tested corresponding to the current cycle number n 0 And the discharge capacity Q of the lithium ion battery to be tested in the nth charge-discharge cycle n And initial discharge capacity Q 0 A ratio eta of the charging time t in the nth charge-discharge cycle n Voltage difference Δ V during standing after charging in nth charge-discharge cycle n 。
It should be noted that: and the initial discharge capacity is the discharge capacity of the lithium ion battery to be tested in the 1 st charge-discharge cycle. The charging time is a charging time from the start of charging to the end of charging in the present charge-discharge cycle. Voltage difference Δ V n The difference between the charging ending voltage in the nth charging and discharging cycle and the ending voltage after the preset time is set aside for the charging ending.
The extracted characteristic values for prediction further include: mean value mean of dV/dQ ((dV/dQ)) in discharge phase of nth charge-discharge cycle n ) Variance var ((dV/dQ) n ) Range ((dV/dQ) n ) (ii) a Wherein, in the interval of 5% -95% of discharging DV curve SOC in the discharging stage of the nth charge-discharge cycle, values are calculated every 0.1% SOC, the numerical values of dV/dQ are respectively calculated, and then the average value of all the calculated dV/dQ numerical values is mean ((dV/dQ) in the nth charge-discharge cycle n ) The variance of all the dV/dQ values is calculated as var ((dV/dQ) in the nth charge-discharge cycle) n ) And the range ((dV/dQ) in the nth charge-discharge cycle) is obtained by calculating the range of all the dV/dQ values n ). The characteristic value can be extracted by utilizing matlab for data processing.
And 103, processing the plurality of predicted characteristic values by using a pre-constructed neural network model to obtain the predicted thickness value of the lithium ion battery to be detected.
The Neural Network model is modeled by adopting a Neural Network Fitting tool in matlab software, and the algorithm is an Artificial Neural Network (ANN). The neural network model after training comprises the following steps: one input layer, one hidden layer and one output layer. The number of cells of the hidden layer is 15 and the algorithm is levenberg-Marquardt (levenberg-Marquardt).
The following describes the construction process of the neural network model in detail:
1) Data collection: the sample battery adopts a soft package battery of an LPF/graphite system, data of the sample battery in the charge-discharge cycle process are collected and processed, and the method adopts a big data method, so that a large amount of data is needed to be used as support, the sample battery is subjected to 800-week charge-discharge cycle, and the data are collected to construct a model.
The battery charging adopts 1C-6C different multiplying power to carry out constant current and constant voltage charging, the discharging multiplying power selects 1C to carry out constant current discharging, the charging and discharging are shelved for 5min after being stopped, wherein after each 50 times of circulation, the thickness of the battery is subjected to one off-line detection, the battery is circulated for 800 weeks at the normal temperature of 25 ℃,3 parallel samples are arranged in the same group to carry out parallel experiments, the voltage, the current and the capacity in the charging and discharging process are recorded, and the recording frequency is 30 ms/time.
2) Data cleaning: the method comprises the steps of 1) collecting original data, then cleaning the original data, eliminating invalid data and data with errors, and improving the accuracy of a model. Invalid data and data with errors can be data which is missed, misdetected and seriously deviated from the actual condition (exceeding 3 sigma), and the like.
3) Extracting a characteristic value: and processing the cleaned data, and extracting a series of characteristic values to be used as input data of the neural network model.
The following feature values were extracted as input data: number of cycles n of charge and discharge, design thickness d of battery 0 And the discharge capacity Q of the battery in the n-th charge-discharge cycle n Initial discharge capacity Q of battery 0 A ratio η ofTime t n Voltage difference Δ V during standing after charging n And mean ((dV/dQ) n )、var((dV/dQ) n ) And range ((dV/dQ) n ) In the section where the discharge DV curve SOC at the discharge stage of the nth charge-discharge cycle is 5% -95%, values are taken every 0.1% SOC, the values of dV/dQ are calculated, and then the average value of all the dV/dQ values is calculated as mean ((dV/dQ) in the nth charge-discharge cycle) n ) The variance of all the dV/dQ values is calculated as var ((dV/dQ) in the nth charge-discharge cycle) n ) The range ((dV/dQ) in the nth charge-discharge cycle) is determined by calculating the range of all the dV/dQ values n )。
4) Modeling big data: the Neural Network model is modeled by using a Neural Network Fitting tool in matlab software, and the algorithm is Artificial Neural Network (ANN). The neural network model has an input layer, a hidden layer, and an output layer. The number of the units of the hidden layer is multiple, the algorithm is levenberg-Marquardt (Levenberg-Marquardt), the number of the units of the hidden layer is determined through training and testing, and after the training and testing are completed, the number of the units of the hidden layer is determined to be 15.
According to the characteristic value extracted in the step 3), the extracted characteristic value is used as input data, and the thickness of the battery corresponding to the node of the current characteristic value is used as output data. Since the thickness of the battery is usually detected off-line every preset number of weeks, the thickness of the battery corresponding to the node of the current characteristic value is the thickness of the monitoring point closest to the node of the current characteristic value. When training the model, 70% of the raw data will be used as the training set and 30% as the test set. The training result of the training set is shown in fig. 2, in the figure, the abscissa target is the true thickness expansion rate, the ordinate Output is the predicted thickness expansion rate, Y represents Output, T represents target, fit is the linear Fit between Output and target, and the more the slope k value is close to 1, the more accurate the prediction effect is. Test set R 2 And (5) storing the product. The thickness expansion ratio and the thickness are variables of the same property, and since the two pearson coefficients r =1, the thickness is converted into the thickness expansion ratio. Thickness expansion ratio = (whenFront node full electrical thickness/first perimeter full electrical thickness-1) × 100%. The current node full electrical thickness corresponds to the current cycle number full electrical thickness. The first full electric thickness of the cycle is the full electric thickness of the first charge-discharge cycle.
5) And (3) model verification: inputting the test set into the model obtained in the step 4) for verification, and evaluating the quality of the model.
The accuracy of the model is optimized by adjusting the number of the units of the hidden layer in the neural network model, and the number of the adjusted units of the hidden layer is 15. The test result of the test set is shown in fig. 3, in the graph, the abscissa target is the true thickness expansion rate, the ordinate Output is the predicted thickness expansion rate, Y represents Output, T represents target, fit is linear fitting between Output and target, and the more the slope k value is close to 1, the more accurate the prediction effect is. Test set R 2 =0.91, stored and used for on-line battery thickness prediction.
The method comprises the steps of obtaining charge and discharge data of the lithium ion battery to be detected in the charge and discharge cycle process under the current cycle number, extracting characteristic values of the charge and discharge data after the current cycle number is ended to obtain a plurality of characteristic values for prediction, and processing the plurality of predicted characteristic values by using a pre-constructed neural network model to obtain the predicted thickness value of the lithium ion battery to be detected, so that the thickness change of the battery can be detected on line in real time, the artificial error of off-line detection is avoided, and the more accurate advanced prediction can be provided for the judgment of the battery thickness failure.
Based on the content of the foregoing embodiment, referring to fig. 4, another embodiment of the present invention provides a device for predicting a thickness of a lithium ion battery, including: an acquisition module 201, an extraction module 202 and an acquisition module 203.
The obtaining module 201 is configured to obtain charge and discharge data of the lithium ion battery to be tested in the charge and discharge cycle process under the current cycle number. The extraction module 202 is configured to extract a feature value of the charge and discharge data after the current cycle number expires, so as to obtain a plurality of feature values for prediction. The obtaining module 203 is configured to process the plurality of predicted characteristic values by using a pre-established neural network model to obtain a predicted thickness value of the lithium ion battery to be measured.
Optionally, the charge and discharge data includes: voltage, capacity, design thickness and discharge time of the lithium ion battery to be tested corresponding to the current cycle number.
Optionally, the plurality of characteristic values for prediction include:
the current cycle number n and the design thickness d of the lithium ion battery to be tested corresponding to the current cycle number 0 Discharge capacity Q in the nth charge-discharge cycle n And initial discharge capacity Q 0 Ratio eta of, charging time t in the process of the nth charge-discharge cycle n And the voltage difference delta V during the standing period after charging in the nth charging and discharging cycle process n (ii) a And an average mean value mean of dV/dQ ((dV/dQ)) in a discharge phase of the nth charge-discharge cycle n ) Variance var ((dV/dQ) n ) Range ((dV/dQ) n ) (ii) a Wherein, in the interval of 5% -95% of discharge DV curve SOC in the discharge stage of the nth charge-discharge cycle, values are calculated every 0.1% of SOC, the values of dV/dQ are respectively calculated, and then the average value of all the obtained dV/dQ values is mean ((dV/dQ) in the nth charge-discharge cycle) n ) The variance of all the dV/dQ values is calculated as var ((dV/dQ) in the nth charge-discharge cycle) n ) And the range ((dV/dQ) in the nth charge-discharge cycle) is obtained by calculating the range of all the dV/dQ values n )。
Optionally, the Neural network model is modeled by a Neural Net Fitting tool in matlab software, and the algorithm is an artificial Neural network.
Optionally, the neural network model has: an input layer, a hidden layer and an output layer; the number of units of the hidden layer is 15, and the algorithm is levenberg-Marquardt.
Optionally, the lithium ion battery is a soft package lithium ion battery.
Based on the content of the above embodiments, another embodiment of the present invention provides a lithium ion Battery, which includes a Battery Management System (BMS). The battery management system is integrated with the lithium ion battery thickness prediction device, so that the change condition of the battery thickness in a terminal product can be monitored in real time, early warning is carried out after the battery thickness is detected to a certain degree, and safety accidents caused by extrusion of the battery are avoided.
It will be appreciated by those skilled in the art that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The embodiments disclosed above are therefore to be considered in all respects as illustrative and not restrictive. All changes which come within the scope of or equivalence to the invention are intended to be embraced therein.
Claims (8)
1. A method for predicting the thickness of a lithium ion battery is characterized by comprising the following steps:
acquiring charge and discharge data of the lithium ion battery to be tested in the charge and discharge cycle process under the current cycle number;
after the current cycle number is cut off, extracting characteristic values of the charge and discharge data to obtain a plurality of characteristic values for prediction;
processing the plurality of predicted characteristic values by using a pre-constructed neural network model to obtain a predicted thickness value of the lithium ion battery to be detected;
the plurality of characteristic values for prediction include:
the current cycle number n and the design thickness d of the lithium ion battery to be tested corresponding to the current cycle number 0 And discharge capacity Q in the nth charge-discharge cycle n And initial discharge capacity Q 0 Ratio eta of, charging time t in the process of the nth charge-discharge cycle n And the voltage difference delta V during the standing period after charging in the nth charging and discharging circulation process n (ii) a And
mean value mean of dV/dQ ((dV/dQ)) in discharge phase of nth charge-discharge cycle n ) Variance var ((dV/dQ) n ) Range ((dV/dQ) n ) (ii) a Wherein, in the interval of 5% -95% of discharging DV curve SOC in the discharging stage of the nth charge-discharge cycle, values are calculated every 0.1% SOC, the numerical values of dV/dQ are respectively calculated, and then the average value of all the calculated dV/dQ numerical values is mean ((dV/dQ) in the nth charge-discharge cycle n )，The variance of all the dV/dQ values is calculated as var in the nth charge-discharge cycle (dV/dQ) n ) And the range ((dV/dQ) in the nth charge-discharge cycle) is obtained by calculating the range of all the dV/dQ values n )。
2. The prediction method according to claim 1, wherein the charge and discharge data includes: voltage, capacity, design thickness and discharge time of the lithium ion battery to be tested corresponding to the current cycle number.
3. The prediction method according to claim 1, wherein before the obtaining of the charge and discharge data of the lithium ion battery to be tested in the charge and discharge cycle process under the current cycle number, the prediction method further comprises: constructing a neural network model;
wherein, a Neural Net Fitting tool in matlab software is adopted for modeling, and the algorithm is an artificial Neural network.
4. The prediction method according to claim 3, wherein the neural network model has: an input layer, a hidden layer and an output layer;
the number of units of the hidden layer is 15, and the algorithm is levenberg-Marquardt.
5. The prediction method according to claim 1, wherein the lithium ion battery is a soft pack lithium ion battery.
6. An apparatus for predicting a thickness of a lithium ion battery, the apparatus comprising:
the acquisition module is used for acquiring charge and discharge data of the lithium ion battery to be detected in the charge and discharge cycle process under the current cycle number;
the extraction module is used for extracting characteristic values of the charge and discharge data after the current cycle number is cut off to obtain a plurality of characteristic values for prediction;
the obtaining module is used for processing the plurality of predicted characteristic values by using a pre-constructed neural network model to obtain predicted thickness values of the lithium ion battery to be detected;
the plurality of characteristic values for prediction include:
the current cycle number n and the design thickness d of the lithium ion battery to be tested corresponding to the current cycle number 0 And discharge capacity Q in the nth charge-discharge cycle n And initial discharge capacity Q 0 Ratio eta of, charging time t in the process of the nth charge-discharge cycle n And the voltage difference delta V during the standing period after charging in the nth charging and discharging cycle process n (ii) a And
mean value mean of dV/dQ ((dV/dQ)) in discharge phase of nth charge-discharge cycle n ) Variance var ((dV/dQ) n ) Range ((dV/dQ) n ) (ii) a Wherein, in the interval of 5% -95% of discharge DV curve SOC in the discharge stage of the nth charge-discharge cycle, values are calculated every 0.1% of SOC, the values of dV/dQ are respectively calculated, and then the average value of all the obtained dV/dQ values is mean ((dV/dQ) in the nth charge-discharge cycle) n ) The variance of all the dV/dQ values is calculated as var ((dV/dQ) in the nth charge-discharge cycle) n ) And the range ((dV/dQ) in the nth charge-discharge cycle) is obtained by calculating the range of all the dV/dQ values n )。
7. The prediction apparatus according to claim 6, wherein the charge/discharge data includes: voltage, capacity, design thickness and charging time of the lithium ion battery to be tested corresponding to the current cycle number.
8. A lithium ion battery comprising a battery management system, wherein the battery management system has the prediction device of the lithium ion battery of any one of claims 6 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210923857.5A CN115308609B (en) | 2022-08-02 | 2022-08-02 | Lithium ion battery thickness prediction method and device and lithium ion battery |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210923857.5A CN115308609B (en) | 2022-08-02 | 2022-08-02 | Lithium ion battery thickness prediction method and device and lithium ion battery |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115308609A CN115308609A (en) | 2022-11-08 |
| CN115308609B true CN115308609B (en) | 2023-04-14 |
Family
ID=83859574
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210923857.5A Active CN115308609B (en) | 2022-08-02 | 2022-08-02 | Lithium ion battery thickness prediction method and device and lithium ion battery |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115308609B (en) |
Family Cites Families (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR102108279B1 (en) * | 2013-05-22 | 2020-05-07 | 삼성에스디아이 주식회사 | System for predicting thickness of battery and method for predicting thickness of battery |
| CN108761344A (en) * | 2018-06-06 | 2018-11-06 | 江苏塔菲尔新能源科技股份有限公司 | A kind of detection method and system of lithium ion battery analysis lithium |
| CN112363069B (en) * | 2020-09-18 | 2023-07-07 | 万向一二三股份公司 | Method for detecting breakage of tab of lithium ion battery |
| CN112464465B (en) * | 2020-11-23 | 2022-12-09 | 西安交通大学 | Battery equivalent mechanical model for soft package lithium ion battery thickness estimation and modeling method thereof |
| CN113189498A (en) * | 2021-05-16 | 2021-07-30 | 广州菲利斯太阳能科技有限公司 | Nondestructive lithium analysis detection method |
| CN113820614B (en) * | 2021-08-18 | 2023-10-24 | 浙江南都电源动力股份有限公司 | Method for predicting cycle life of lithium ion battery |
| CN114114049B (en) * | 2021-09-30 | 2022-10-18 | 北京交通大学 | Lithium ion battery life prediction method based on sample migration |
| CN113917336A (en) * | 2021-10-13 | 2022-01-11 | 国网福建省电力有限公司 | Lithium ion battery health state prediction method based on segment charging time and GRU |
| CN113759269B (en) * | 2021-11-10 | 2022-02-25 | 武汉理工大学 | Method and system for monitoring health state of battery of electric vehicle |
| CN114137430A (en) * | 2021-11-22 | 2022-03-04 | 北京胜能能源科技有限公司 | Estimation method for residual life of retired lithium battery |
-
2022
- 2022-08-02 CN CN202210923857.5A patent/CN115308609B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN115308609A (en) | 2022-11-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107957562B (en) | Online prediction method for residual life of lithium ion battery | |
| CN108254696B (en) | Battery health state evaluation method and system | |
| Yang et al. | On-board diagnosis of soft short circuit fault in lithium-ion battery packs for electric vehicles using an extended Kalman filter | |
| CN103412264B (en) | The conforming evaluation method of cell in battery pack | |
| CN111458649A (en) | Rapid detection method for health degree of battery module | |
| CN116401585B (en) | Energy storage battery failure risk assessment method based on big data | |
| CN112526352B (en) | SOH estimation method for retired lithium ion battery | |
| CN115184809B (en) | Multi-dimensional evaluation method for energy storage battery system based on temperature angle | |
| Zheng et al. | Quantitative short circuit identification for single lithium-ion cell applications based on charge and discharge capacity estimation | |
| CN115267555A (en) | Battery SOH (State of health) evaluation system of energy storage system based on battery multipoint temperature measurement | |
| Li et al. | The open-circuit voltage characteristic and state of charge estimation for lithium-ion batteries based on an improved estimation algorithm | |
| Kamali et al. | Novel SOH estimation of lithium-ion batteries for real-time embedded applications | |
| CN113687251B (en) | Double-model-based lithium ion battery pack voltage abnormality fault diagnosis method | |
| CN113484784B (en) | Lithium battery online aging diagnosis method based on two-point impedance aging characteristics | |
| Ma et al. | State of health estimation of retired battery for echelon utilization based on charging curve | |
| CN115308609B (en) | Lithium ion battery thickness prediction method and device and lithium ion battery | |
| CN117007975A (en) | Method for performing reinforcement learning on battery capacity attenuation assessment by collecting multi-point temperatures of battery cells of energy storage battery | |
| CN113466700B (en) | Lithium battery online life prediction method based on two-point impedance life characteristics | |
| CN110927585A (en) | Lithium battery SOH estimation system and method based on self-circulation correction | |
| CN114594380A (en) | Battery SOC prediction method based on BP neural network | |
| Zhao et al. | Micro-short circuit fault diagnosis of the parallel battery module based on increment capacity curve | |
| CN113866644A (en) | Method and device for predicting usable time and capacity of battery | |
| CN117805639A (en) | Method and system for judging micro-short circuit fault of lithium battery of electrochemical energy storage power station | |
| CN114002601B (en) | Method and device for calculating capacity of battery core of retired lithium ion battery | |
| Ye et al. | Lithium battery SOH estimation through FFNN |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP02 | Change in the address of a patent holder |
Address after: No. 1, Chuneng Avenue, Lingang Economic Zone, Xiaogan City, Hubei Province, 430051 Patentee after: CORNEX NEW ENERGY Co.,Ltd. Address before: 430051 No. 1, Longyang Third Road, Hanyang District, Wuhan City, Hubei Province Patentee before: CORNEX NEW ENERGY Co.,Ltd. |
|
| CP02 | Change in the address of a patent holder |