On-site battery health assessment method based on multi-impedance measurement

Abstract

The invention discloses an on-site battery health assessment method based on multi-impedance measurement. The method comprises the following steps: measuring real part impedance values and imaginary part impedance values corresponding to a plurality of preset frequencies of an on-site battery under any on-site working condition; predicting real part impedance values and imaginary part impedance values corresponding to a plurality of same preset frequencies under the laboratory working condition; predicting an electrochemical impedance spectrum under a laboratory working condition; predicting a charge / discharge curve and a relaxation voltage curve under the laboratory working condition; and carrying out feature extraction and health assessment to obtain a health assessment result of the on-site battery. According to the method, field data and laboratory data are bridged through machine learning, and compared with an existing battery health assessment method based on field historical data, the method has the advantages of being high in precision, low in cost, rapid, capable of being used anytime and anywhere, free of privacy data and the like.

Description

Technical Field

[0001] The present invention relates to an online health assessment method in the field of batteries, and in particular to an on-site battery health assessment method based on multi-impedance measurement. Background Art

[0002] Batteries have been commercialized on a large scale in portable electronic devices, transportation vehicles, and stationary energy storage, all of which have an urgent need for longer cycle life and safer batteries. Accurate battery health assessment (including health diagnosis and health prognosis) information can provide a reliable basis for the optimization of predictive maintenance and management strategies for on-site batteries, thereby reducing the occurrence of safety accidents and extending battery life. In recent years, thanks to the rapid development of artificial intelligence technology and the open source of many battery test data sets, data-driven methods combining machine learning and feature engineering have achieved superior performance in battery health assessment and have shown great application potential in the entire battery life cycle, including R&D, production, research, application, and recycling. However, most of the current data-driven methods and their feature engineering rely on laboratory data at specific stages under specific operating conditions, such as charge and discharge data, impedance data, relaxation data, etc. Unlike laboratory data with relatively stable operating environments and operating conditions, battery field data is affected by the coupling of many random uncertain factors such as application scenarios, operating environments, and user habits, and has great randomness and uncertainty. Therefore, most data-driven methods developed based on laboratory data cannot be directly applied to field data.

[0003] Researchers have also conducted some research on battery health assessment methods based on field data. The main idea is to develop data-driven methods based on the massive historical data generated by field battery operation. According to the field battery's needs for operation, maintenance and optimization management, the ideal battery health assessment method should have five main characteristics: (1) High accuracy. The higher the accuracy of the health assessment method, the more accurate the predictive maintenance and management strategy optimization of the field battery can be, which is more beneficial to the safe and long-life operation of the field lithium battery. On the contrary, inaccurate health assessment may have a negative impact on the safe and long-life operation of the field battery. (2) Low cost. The battery health assessment method should have low development cost, low deployment cost, and low use cost, which is more conducive to large-scale commercial promotion. (3) Use anytime and anywhere. The battery health assessment method can carry out field battery health assessment anytime and anywhere according to user needs. (4) Fast. The battery health assessment method can quickly complete the field battery health assessment and will not have a negative impact on the normal operation of the field battery. (5) Does not involve user privacy data. As users' awareness of data privacy continues to increase, the protection of battery operation data will become more and more stringent, and the amount of data available will become less and less. The data on which the battery health assessment method relies should try not to involve the user's privacy data, such as historical operation data, which may include historical travel data, historical charging data, and user usage habit data.

[0004] Compared with the five characteristics that an ideal battery health assessment method should have, the existing data-driven methods developed based on massive field historical data are not ideal in all aspects. In terms of accuracy, under the same data volume, the health assessment accuracy of the field data-based method is not yet as good as that of the laboratory data-based method. In terms of cost, the field data-based method needs to accumulate massive field historical data, and the development cost is high. In terms of use, the field data-based method needs to collect complete or specific stage operation data, which cannot be used anytime and anywhere, and the speed is also slow. In terms of privacy, the field data-based method uses massive historical data, which may involve the user's privacy data. In addition, the existing field data-based health assessment methods have not made full use of the many excellent research results based on laboratory data. Most of the methods are developed from scratch, which is a waste of research resources. According to the first principle idea, based on the excellent results of many laboratory data-based methods, the development of the lowest cost field battery health assessment method should make use of the existing excellent results as much as possible, rather than developing from scratch. Summary of the invention

[0005] In order to solve the problem of on-site battery health assessment, the present invention proposes an on-site battery health assessment method based on multi-impedance measurement. The present invention adopts a new idea different from the existing methods: in actual field applications, a small amount of on-site battery impedance is measured online, combined with machine learning model prediction to obtain high-quality laboratory data of the battery under laboratory conditions. Since there have been a lot of research results and excellent performance achieved in battery health assessment based on high-quality laboratory data, many existing research results can be directly used to perform high-precision battery health assessment.

[0006] The scheme adopted by the present invention is:

[0007] 1. A method for on-site battery health assessment based on multi-impedance measurement

[0008] 1) Online measurement of the real impedance value and imaginary impedance value of the on-site battery corresponding to multiple preset frequencies under any on-site working conditions;

[0009] 2) based on the measured real impedance values ​​and imaginary impedance values ​​corresponding to multiple preset frequencies under field conditions, respectively predict the real impedance values ​​and imaginary impedance values ​​corresponding to multiple identical preset frequencies under laboratory conditions;

[0010] 3) Predicting and obtaining EIS data under laboratory conditions based on the predicted real impedance values ​​and imaginary impedance values ​​corresponding to multiple preset frequencies under laboratory conditions;

[0011] 4) Inputting the EIS data under laboratory conditions into the laboratory conditions charging Q / V data prediction model, the laboratory conditions discharging Q / V data prediction model, and the laboratory conditions relaxation V / t data prediction model, respectively, predicts the charging Q / V data, discharging Q / V data, and relaxation V / t data under laboratory conditions;

[0012] 5) Perform feature extraction and battery health assessment based on the EIS data, charging Q / V data, discharging Q / V data and relaxation V / t data under laboratory conditions to obtain the health assessment results of the on-site battery.

[0013] In the above 2), the prediction process of the real impedance value and the imaginary impedance value corresponding to any preset frequency under the measured field working conditions is specifically as follows:

[0014] The real impedance value corresponding to the current preset frequency under the field working conditions is input into the real impedance value prediction model of the laboratory working conditions of the preset frequency to obtain the real impedance value corresponding to the preset frequency under the laboratory working conditions; the imaginary impedance value corresponding to the current preset frequency under the field working conditions is input into the imaginary impedance value prediction model of the laboratory working conditions of the preset frequency to obtain the imaginary impedance value corresponding to the preset frequency under the laboratory working conditions.

[0015] The specific aspects of 3) are:

[0016] The real impedance values ​​and imaginary impedance values ​​corresponding to multiple preset frequencies under laboratory conditions are input into the laboratory conditions EIS prediction model to predict and obtain the EIS data under laboratory conditions.

[0017] In the above 5), the health assessment result of the on-site battery includes the aging status and / or the remaining service life.

[0018] When the health assessment result of the on-site battery is an aging state, diagnostic features are extracted based on the EIS data, charging Q / V data, discharging Q / V data and relaxation V / t data under laboratory conditions, and the extracted diagnostic features are input into the battery health diagnosis model. The model outputs the current aging state of the on-site battery.

[0019] When the health assessment result of the on-site battery is the remaining service life, the prognostic features are extracted based on the EIS data, charging Q / V data, discharging Q / V data and relaxation V / t data under laboratory conditions, and the extracted prognostic features are input into the battery health prognosis model. The model outputs the current remaining service life of the on-site battery.

[0020] The plurality of preset frequencies are selected and obtained through expert experience method, filtering method, packaging method, embedding method or clustering method.

[0021] 2. An on-site battery health assessment system based on multi-impedance measurement

[0022] An impedance data acquisition unit of a field battery, used to acquire real impedance values ​​and imaginary impedance values ​​of the field battery corresponding to multiple preset frequencies under any field working conditions;

[0023] The impedance data generation unit of the laboratory working condition is used to generate the real impedance values ​​and imaginary impedance values ​​corresponding to the same preset frequencies under the laboratory working condition according to the real impedance values ​​and imaginary impedance values ​​corresponding to the preset frequencies under the field working condition obtained by measurement;

[0024] An EIS data generation unit for laboratory working conditions, used to predict and obtain EIS data under laboratory working conditions according to the predicted real impedance values ​​and imaginary impedance values ​​corresponding to multiple preset frequencies under laboratory working conditions;

[0025] A unit for generating charging Q / V data, discharging Q / V data and relaxing V / t data under laboratory conditions, used for predicting the charging Q / V data, discharging Q / V data and relaxing V / t data under laboratory conditions according to the predicted EIS data under laboratory conditions;

[0026] The health assessment unit is used to extract features based on the predicted EIS data, charging Q / V data, discharging Q / V data and relaxation V / t data under laboratory conditions to obtain the health assessment results of the on-site battery.

[0027] 3. A computer device

[0028] The device comprises a memory and a processor, wherein the memory stores a computer program, and the steps of the method are implemented when the processor executes the computer program.

[0029] 4. A computer-readable storage medium

[0030] The medium has a computer program stored thereon, and when the computer program is executed by a processor, the steps of the method are implemented.

[0031] The beneficial effects of the present invention are:

[0032] 1. The present invention proposes an on-site battery health assessment method based on multi-impedance measurement. Compared with the existing method based on on-site historical data, it has the advantages of high accuracy, low cost, fast speed, can be used anytime and anywhere, and does not involve privacy data;

[0033] 2. The present invention uses machine learning to bridge the field data and laboratory data of the battery, filling the gap between the battery health assessment research results based on laboratory data and field applications, so that all battery health assessment research results based on laboratory data can be directly applied to actual field batteries;

[0034] 3. The present invention uses the battery aging status index and remaining service life index under laboratory conditions to accurately and quantitatively evaluate the current health status of the on-site battery. Compared with the features / indicators extracted from random on-site battery data, the battery health assessment results are more accurate.

[0035] 4. The present invention overturns the existing fixed mode of battery health assessment based on field data, and opens up a new direction for online health assessment of field batteries. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is the overall flow chart of the present invention.

[0037] Figure 2 are two preset frequencies (f 1 =10.0Hz and f 2 =312.5Hz) corresponding to the real impedance value and the imaginary impedance value.

[0038] Figure 3It is a comparison chart between EIS prediction data and actual measurement data under laboratory conditions in an embodiment of the present invention.

[0039] Figure 4 It is a comparison diagram of the relaxation time distribution (DRT) data corresponding to the EIS obtained by prediction in the embodiment of the present invention and the DRT data corresponding to the EIS obtained by actual measurement.

[0040] Figure 5 It is a comparison chart between the predicted values ​​and the actual measured values ​​of the charging Q / V data and the discharging Q / V data under laboratory conditions in an embodiment of the present invention.

[0041] Figure 6 It is a comparison chart of incremental capacity (IC) data corresponding to the charging Q / V data predicted in an embodiment of the present invention and IC data corresponding to the charging Q / V data actually measured.

[0042] Figure 7 It is a comparison diagram of differential voltage (DV) data corresponding to the charging Q / V data predicted in an embodiment of the present invention and DV data corresponding to the charging Q / V data actually measured. DETAILED DESCRIPTION

[0043] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0044] In this embodiment, the selected battery chemistry type is NCM, the manufacturer is LG, the model is INR18650HG2, and the rated capacity is 3000mAh. The number of preset frequencies is set to two. The two preset frequencies are specifically one of the values ​​​​in the range of 50mHz to 10KHz. 1 Named as the first frequency, f 2 Named as the second frequency.

[0045] Two preset frequencies (f 1 and f 2) can be selected through expert experience, filtering, packaging, embedding or clustering. In this embodiment, the K-Means clustering algorithm in the clustering method is used to select two preset frequencies. Specifically, 16 data points in the high frequency part (1Hz~1KHz) of EIS are selected, and each data point contains a frequency value, a real impedance value and an imaginary impedance value. The 16 data points are divided into two clusters, and the K-means clustering algorithm is used to extract two preset frequencies from the 16 frequencies corresponding to the 16 data points. Among them, the number of clusters of the K-means clustering algorithm is set to 2. The 16 data points are input into the K-means clustering algorithm for clustering to obtain the cluster label of each data point. After clustering is completed, the mean of the frequencies corresponding to all data points in each cluster is calculated. Next, in each cluster, find the data point closest to the frequency mean in the cluster, and use the frequency value corresponding to the data point as the preset frequency of the cluster. The two preset frequency values ​​obtained are f and f respectively. 1 =10.0Hz and f 2 =312.5Hz.

[0046] like Figure 1 As shown, the present invention comprises the following steps:

[0047] 1) Online measurement of the real impedance value and imaginary impedance value corresponding to multiple preset frequencies of the field battery under any field working conditions, such as Figure 2 As shown. Any field operating condition includes any value within the battery operating temperature range and any value within the battery SOC range. In this embodiment, the battery operating temperature range is [0°C, 40°C], and the battery SOC range is [0%, 100%].

[0048] 2) Based on the real impedance values ​​and imaginary impedance values ​​corresponding to multiple preset frequencies under the field conditions obtained by measurement, the real impedance values ​​and imaginary impedance values ​​corresponding to multiple identical preset frequencies under the laboratory conditions are predicted. The laboratory conditions include a preset battery operating temperature and a preset battery SOC. In this embodiment, the battery operating temperature under the laboratory conditions is 25°C, and the battery SOC is 90% during the charging process. The first frequency f under the field conditions is 1 The corresponding real impedance value Input it into the real impedance value prediction model of the first frequency laboratory condition to obtain the first frequency f under laboratory conditions 1 The corresponding real impedance value The first frequency f under the on-site working conditions 1 The corresponding imaginary impedance value Input to the first frequency f 1 The first frequency f under laboratory conditions is obtained in the laboratory working condition imaginary impedance value prediction model. 1 The corresponding imaginary impedance value The same method is used to predict the real impedance value and imaginary impedance value corresponding to the second frequency under laboratory conditions. The input and output data range of the laboratory impedance value prediction model used to train each preset frequency should cover the full life cycle of the battery, the entire operating temperature range and the entire SOC range. In this embodiment, the two laboratory impedance value prediction models use a random forest (RF) regression model, the battery operating temperature range is [0°C, 40°C], the battery SOC range is [0%, 100%], and the battery full life cycle is the capacity under laboratory conditions. The decay from the initial 100% to 80%. The multiple preset frequencies described in step 2) are exactly the same as the multiple preset frequencies in step 1).

[0049] 3) The real impedance values ​​and imaginary impedance values ​​corresponding to multiple preset frequencies under laboratory conditions obtained by prediction are input into the laboratory conditions EIS prediction model to predict the EIS data under laboratory conditions, such as Figure 3 The predicted EIS data under laboratory conditions and the measured EIS data are processed for relaxation time distribution and converted into DRT curves. The accuracy of EIS data prediction can be further verified by comparing the error between the two DRT curves, as shown in Figure 4 The input and output data range for training the laboratory working condition EIS prediction model should cover the battery life cycle. In this embodiment, the laboratory working condition EIS prediction model adopts the RF regression model. The input of the laboratory working condition EIS prediction model is two frequencies (f 1 and f 2 )'s real impedance value ( and ) and the imaginary impedance value ( and ), the output is the EIS prediction data under laboratory conditions (1Hz~1KHz).

[0050] 4) Input the EIS data under laboratory conditions into the laboratory conditions charging capacity vs. voltage (Q / V) data prediction model, laboratory conditions discharging Q / V data prediction model, and laboratory conditions relaxation voltage vs. time (V / t) data prediction model, respectively, and predict the charging Q / V data, discharging Q / V data, and relaxation V / t data under laboratory conditions, respectively, as follows: Figure 5 (a) and Figure 5 The predicted charging Q / V data under laboratory conditions and the measured charging Q / V data are processed for capacity derivation and converted into charging incremental capacity (IC) curves. The accuracy of the charging Q / V data prediction can be further verified by comparing the error between the two IC curves, as shown in (b). Figure 6The predicted charging Q / V data under laboratory conditions and the measured charging Q / V data are respectively subjected to voltage derivative processing and converted into charging differential voltage (DV) curves. Comparing the error between the two DV curves can further verify the accuracy of the charging Q / V data prediction, as shown in Figure 7 As shown. The same method can be used to verify the accuracy of the prediction of discharge Q / V data and relaxation V / t data. The inputs of the laboratory condition charging Q / V data prediction model, the laboratory condition discharging Q / V data prediction model, and the laboratory condition relaxation V / t data prediction model are all the predicted EIS data under laboratory conditions. The output of the laboratory condition charging Q / V data prediction model is the constant current charging Q / V data or constant voltage charging capacity vs. current (Q / I) data under laboratory conditions. The output of the laboratory condition discharging Q / V data prediction model is the constant current discharge Q / V data or constant voltage discharge Q / I data under laboratory conditions. The output of the laboratory condition relaxation V / t data prediction model is the relaxation V / t data under laboratory conditions (relaxation V / t data after charging or relaxation V / t data after discharging). The input and output data ranges used to train the laboratory condition charging Q / V data prediction model, the laboratory condition discharging Q / V data prediction model, and the laboratory condition relaxation V / t data prediction model should cover the full life cycle of the battery. In this embodiment, the constant current charging current under laboratory conditions is 1 / 3C, the constant current charging cut-off voltage is 4.2V, the constant voltage charging voltage is 4.2V, and the constant voltage charging cut-off current is 1 / 20C. The constant current discharge current under laboratory conditions is 1 / 3C, the constant current discharge cut-off voltage is 2.5V, the constant voltage discharge voltage is 2.5V, and the constant voltage discharge cut-off current is 1 / 20C. The laboratory working condition charging Q / V data prediction model, the laboratory working condition discharge Q / V data prediction model, and the laboratory working condition relaxation V / t data prediction model all use RF regression models.

[0051] 5) Perform feature extraction and battery health assessment based on the EIS data, charging Q / V data, discharging Q / V data and relaxation V / t data under laboratory conditions to obtain the health assessment results of the on-site battery. The features extracted from the EIS data, charging Q / V data, discharging Q / V data and relaxation V / t data under laboratory conditions include instantaneous features, statistical features and model parameter features, etc. The health assessment results of the on-site battery include aging status and / or remaining service life. When the health assessment result of the on-site battery is an aging state, diagnostic features are extracted based on the EIS data, charging Q / V data, discharging Q / V data and relaxation V / t data under laboratory conditions, and the extracted diagnostic features are input into the battery health diagnosis model. The model outputs the current aging status of the on-site battery, such as the current battery's retained capacity, internal resistance, lithium inventory loss, active material loss, etc. under laboratory conditions. When the health assessment result of the on-site battery is the remaining service life, the prognostic features are extracted based on the EIS data, charging Q / V data, discharging Q / V data and relaxation V / t data under laboratory conditions, and the extracted prognostic features are input into the battery health prognosis model. The model outputs the current remaining service life of the on-site battery, such as the remaining number of cycles of the battery under laboratory conditions, the remaining cumulative charge and discharge capacity, etc. The input and output data range for training the battery health diagnosis model and the battery health prognosis model should cover the full life cycle of the battery. In this embodiment, the features extracted from the EIS data, the charging Q / V data, and the discharging Q / V data are statistical features. The battery health diagnosis model and the battery health prognosis model both use RF regression models. The battery aging status indicator uses the battery's retained capacity under laboratory conditions, and the battery remaining service life indicator uses the remaining number of cycles of the battery under laboratory conditions.

[0052] Finally, it should be noted that the above embodiments and explanations are only used to illustrate the technical solution of the present invention rather than to limit it. Those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope disclosed in the technical solution of the present invention, which should be included in the scope of protection of the claims of the present invention.

Claims

1. A method for on-site battery health assessment based on multi-impedance measurement, characterized in that: The following steps are involved: 1) Online measurement of the real impedance value and imaginary impedance value of the on-site battery corresponding to multiple preset frequencies under any on-site working conditions; 2) based on the measured real impedance values ​​and imaginary impedance values ​​corresponding to multiple preset frequencies under field conditions, respectively predict the real impedance values ​​and imaginary impedance values ​​corresponding to multiple identical preset frequencies under laboratory conditions; 3) Predicting and obtaining EIS data under laboratory conditions based on the predicted real impedance values ​​and imaginary impedance values ​​corresponding to multiple preset frequencies under laboratory conditions; 4) Inputting the EIS data under laboratory conditions into the laboratory conditions charging Q / V data prediction model, the laboratory conditions discharging Q / V data prediction model, and the laboratory conditions relaxation V / t data prediction model, respectively, predicts the charging Q / V data, discharging Q / V data, and relaxation V / t data under laboratory conditions; 5) Perform feature extraction and battery health assessment based on the EIS data, charging Q / V data, discharging Q / V data and relaxation V / t data under laboratory conditions to obtain the health assessment results of the on-site battery.

2. According to claim 1, a method for on-site battery health assessment based on multi-impedance measurement is characterized in that: In the above 2), the prediction process of the real impedance value and the imaginary impedance value corresponding to any preset frequency under the measured field working conditions is specifically as follows: Inputting the real impedance value corresponding to the current preset frequency under the field working condition into the real impedance value prediction model of the laboratory working condition of the preset frequency to obtain the real impedance value corresponding to the preset frequency under the laboratory working condition; The imaginary impedance value corresponding to the current preset frequency under the field working condition is input into the imaginary impedance value prediction model of the laboratory working condition of the preset frequency to obtain the imaginary impedance value corresponding to the preset frequency under the laboratory working condition.

3. The on-site battery health assessment method based on multi-impedance measurement according to claim 1, characterized in that: The specific aspects of 3) are: The real impedance values ​​and imaginary impedance values ​​corresponding to multiple preset frequencies under laboratory conditions are input into the laboratory conditions EIS prediction model to predict and obtain the EIS data under laboratory conditions.

4. The on-site battery health assessment method based on multi-impedance measurement according to claim 1, characterized in that: In the above 5), the health assessment result of the on-site battery includes the aging status and / or the remaining service life.

5. The on-site battery health assessment method based on multi-impedance measurement according to claim 4, characterized in that: When the health assessment result of the on-site battery is an aging state, diagnostic features are extracted based on the EIS data, charging Q / V data, discharging Q / V data and relaxation V / t data under laboratory conditions, and the extracted diagnostic features are input into the battery health diagnosis model. The model outputs the current aging state of the on-site battery.

6. The on-site battery health assessment method based on dual impedance measurement according to claim 4, characterized in that: When the health assessment result of the on-site battery is the remaining service life, the prognostic features are extracted based on the EIS data, charging Q / V data, discharging Q / V data and relaxation V / t data under laboratory conditions, and the extracted prognostic features are input into the battery health prognosis model. The model outputs the current remaining service life of the on-site battery.

7. The on-site battery health assessment method based on multi-impedance measurement according to claim 1, characterized in that: The plurality of preset frequencies are selected and obtained through expert experience method, filtering method, packaging method, embedding method or clustering method.

8. An on-site battery health assessment system based on multi-impedance measurement, characterized in that: include: An impedance data acquisition unit of a field battery, used to acquire real impedance values ​​and imaginary impedance values ​​of the field battery corresponding to multiple preset frequencies under any field working conditions; The impedance data generation unit of the laboratory working condition is used to generate the real impedance values ​​and imaginary impedance values ​​corresponding to the same preset frequencies under the laboratory working condition according to the real impedance values ​​and imaginary impedance values ​​corresponding to the preset frequencies under the field working condition obtained by measurement; An EIS data generation unit for laboratory working conditions, used to predict and obtain EIS data under laboratory working conditions according to the predicted real impedance values ​​and imaginary impedance values ​​corresponding to multiple preset frequencies under laboratory working conditions; A unit for generating charging Q / V data, discharging Q / V data and relaxing V / t data under laboratory conditions, used for predicting the charging Q / V data, discharging Q / V data and relaxing V / t data under laboratory conditions according to the predicted EIS data under laboratory conditions; The health assessment unit is used to extract features based on the predicted EIS data, charging Q / V data, discharging Q / V data and relaxation V / t data under laboratory conditions to obtain the health assessment results of the on-site battery.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images