A method for end-cloud collaborative estimation of battery health status based on constant voltage charging segments

Abstract

A method for end-cloud collaborative estimation of battery health status based on constant voltage charging segments belongs to the field of battery technology. The present invention includes the following steps: Step 1, obtaining battery constant current and constant voltage charging test data, and intercepting a local segment of the constant voltage charging curve; selectively extracting HI, CV charging time and CV charging capacity features based on different sampling density data of the segment; Step 2, using the Pearson correlation coefficient to analyze the linear correlation between HI, CV charging time and CV charging capacity characteristics and battery capacity, and constructing a feature mapping library; Step 3, obtaining vehicle charging data, and performing end-cloud collaborative estimation of battery SOH based on the feature mapping library. The present invention proposes a cloud-based SOH estimation method, which can obtain satisfactory estimation results even in the case of sparse data in the cloud, which makes it possible to achieve high-precision and reliable SOH estimation in scenarios such as "partial loss of sampling data, low-cost BMS computing power and insufficient storage".

Description

Technical Field

[0001] The present invention belongs to the field of battery technology, and specifically relates to a method for end-cloud collaborative estimation of battery health status based on constant voltage charging segments. Background Art

[0002] As environmental pollution and energy security issues become increasingly prominent, vigorously developing clean energy has become a major strategy for almost all countries in the world, especially new energy vehicles represented by electric vehicles. As the most widely used energy storage system in electric vehicles, lithium-ion batteries have the advantages of high energy density, long service life and environmental protection. However, as an electrochemical system, lithium-ion batteries will degrade during use, resulting in reduced safety and power, and even uncontrollable failures and more serious accidents. Therefore, it is necessary to monitor the battery's state of health (SOH).

[0003] Existing advanced SOH estimation methods, such as artificial neural networks, support vector machines, Gaussian process regression, etc., can achieve good SOH estimation accuracy, but these methods require high-quality sampling data (such as constant current / constant voltage charging or discharging data is sufficiently complete, the sampling frequency is high enough, etc.), which is not easy to meet in practical applications; on the other hand, they require sufficient online computing and storage capabilities, which are difficult to achieve on some low-cost BMS. Regarding data quality issues, due to the stability of charging data, more and more research has begun to focus on extracting features based on charging data and then estimating SOH in recent years, but the conditions for feature extraction usually require a complete constant current-constant voltage (CCCV) or constant voltage (CV) charging process; For computing power and storage issues, cloud-based data monitoring and management is an effective method, but the data recording cycle of automobile or battery companies in the cloud is usually 10 to 30 seconds. Currently, there is little research on battery SOH estimation based on sparse cloud data;

[0004] Therefore, providing a battery health status estimation method that is easy to obtain, accurate, reliable and applicable to sparse data in the cloud is one of the urgent problems to be solved by technical personnel in this field. Summary of the invention

[0005] The present invention mainly solves the technical problems existing in the above-mentioned prior art and provides a method for end-cloud collaborative estimation of battery health status based on constant voltage charging segments; for typical scenarios where the CV charging process is incomplete, a health factor (Healthindictor, HI) suitable for the constant voltage charging process is extracted from the first-order equivalent circuit model according to the current data of the CV charging segment, and the CV charging time feature and the CV charging capacity feature are further extracted; under a public battery data set, analysis shows that when the sampling data is dense, the above three features are strongly correlated with the battery capacity; when the sampling data is sparse, HI still has a high correlation with the capacity; using the above multiple features or single features as input, simple linear regression models can be established to predict the battery SOH under dense data or sparse data.

[0006] The above technical problem of the present invention is mainly solved by the following technical solution: a method for end-cloud collaborative estimation of battery health status based on constant voltage charging segment, comprising the following steps:

[0007] Step 1, obtaining constant current and constant voltage charging test data of the battery, and intercepting a local segment of the constant voltage charging curve; selectively extracting HI, CV charging time and CV charging capacity features based on different sampling density data of the segment;

[0008] Step 2, using the Pearson correlation coefficient to analyze the linear correlation between HI, CV charging time and CV charging capacity characteristics and battery capacity, and constructing a feature mapping library;

[0009] Step 3: Obtain vehicle charging data and perform end-cloud collaborative estimation of battery SOH based on the feature mapping library.

[0010] Preferably, step 1 comprises the following steps:

[0011] Step 1.1, obtaining constant current and constant voltage charging test data of the battery, and intercepting a constant voltage charging segment; wherein the constant voltage charging segment includes a sampling time and current;

[0012] Step 1.2, extracting HI, CV charging time and CV charging capacity characteristics in sequence based on the constant voltage charging segment in step 1.1;

[0013] Step 1.2.1, the extraction of HI is based on the external electrical behavior of the equivalent circuit model in the CV stage, using the Thevenin model as the model to describe the external characteristics of the battery:

[0014]

[0015] Among them, R int is the series internal resistance, C p is the polarization capacitance, R p is the polarization internal resistance, Vp is the voltage across the first-order RC network, V OC is the open circuit voltage of the battery, I l is the load current;

[0016] Performing Laplace transform and Z transform on equation (1) yields:

[0017]

[0018] Among them, T s is the sampling period, k and k-1 are the current sampling time and the sampling time of the previous period;

[0019] Assume that CV segment V oc It is a linear function of SOC, with a coefficient of m. The value of m varies with different battery aging levels, and we get:

[0020]

[0021] Since the CV segment voltage remains unchanged, equation (2) can be simplified to

[0022] I k+1 =θ s I k (4)

[0023] in,

[0024]

[0025] Define a new parameter h to describe new and old batteries. The value of a new battery is 1, and the more severe the aging, the closer the value is to 0. The expression is as follows:

[0026]

[0027] Let the constant voltage charging current data in step 1.1 be I = (I1, I2, ..., I n ), (I1, I2, …, I n-1 ) and (I2, I3, …, I n ) are used as the input and output of the least squares method, respectively, to identify the HI in each cycle;

[0028] Step 1.2.2: The CV charging time and CV charging capacity characteristics are the sampled consumption time value and capacity growth of the constant voltage charging segment in step 1.1:

[0029]

[0030] T start and T end are the initial and cutoff sampling times, tim addand cap add That is, CV charging time and CV charging capacity;

[0031] Step 1.2.3: Downsample the constant voltage charging segment in step 1.1 to obtain a sparse data segment, and extract HI based on the segment using step 1.2.1.

[0032] Preferably, step 2 comprises the following steps:

[0033] Step 2.1: The Pearson correlation coefficient is used to analyze the linear correlation between HI, CV charging time and CV charging capacity characteristics and battery capacity respectively. The calculation formula is as follows:

[0034]

[0035] Where p is the characteristic sequence, Q is the capacity sequence, σ p and σ Q are the average values ​​of the characteristic sequence and capacity sequence respectively, and ρ is the correlation coefficient;

[0036] Step 2.2: Establish a prediction model for battery capacity with respect to HI, CV charging time and CV charging capacity characteristics, use the HI sequence in step 1.2.1, the CV charging time feature sequence and the CV charging capacity feature sequence in step 1.2.2, and the battery capacity sequence as training sets, and obtain the trained model parameter group as a dense data mapping library; similarly, establish a prediction model for battery capacity with respect to HI, use the HI sequence and the battery capacity sequence in step 1.2.3 as training sets, and obtain the trained model parameter group as a sparse data mapping library.

[0037] Preferably, step 3 comprises the following steps:

[0038] Step 3.1: Obtain vehicle charging data and choose whether end-cloud collaboration is required based on data quality, computing power and storage conditions, and the enterprise's cloud monitoring and management requirements; if end-cloud collaboration is not required, proceed to step 3.2, otherwise proceed to step 3.3;

[0039] Step 3.2: Determine whether the feature extraction conditions are met based on the charging data collected and stored by the BMS. If so, obtain the characteristics of the charging process according to the HI extraction method in step 2.1 and the CV charging time and CV charging capacity feature extraction method in step 2.2, and estimate the SOH based on the dense data mapping library in step 2; otherwise, do not process and wait for the next charge;

[0040] Step 3.3: Based on the sparse data of this charging uploaded by the vehicle-side Tbox and received and stored by the cloud platform, determine whether the feature extraction conditions are met. If so, obtain the characteristics of this charging process according to the HI extraction method in step 2.3, and estimate SOH based on the dense data mapping library in step 2; otherwise, do not process it and wait for the next charging.

[0041] Preferably, the constant voltage charging segment in step 1.1 is based on the current during constant current charging, and a 0.2C-0.7C window is intercepted to obtain a curve segment, with the reference being 1C.

[0042] Preferably, the sparse data segment in step 1.2 is a current data sequence obtained by sampling the constant voltage charging segment in step 1 at any value between 10s and 30s.

[0043] Preferably, the prediction model in step 2.2 is a linear regression model.

[0044] Preferably, the vehicle charging data in step 3.1 is sequence data of the current changing with time during the constant current and constant voltage charging stage of the battery.

[0045] Preferably, the feature extraction condition is based on the current during constant current charging, and the vehicle charging data undergoes a process in which the constant voltage charging current decreases from 0.7C to 0.2C, with the benchmark being 1C.

[0046] The present invention has the beneficial effects:

[0047] 1. The present invention proposes a feature extraction method based on constant voltage charging segments, which only requires a small amount of local constant voltage charging data to perform SOH estimation, does not require the battery to be fully charged and is applicable to any discharge condition.

[0048] 2. The present invention proposes three features that are strongly correlated with battery capacity and has been verified on multiple batteries, which can achieve high-precision and high-robustness SOH estimation.

[0049] 3. The present invention proposes a cloud-based SOH estimation method, which can obtain satisfactory estimation results even in the case of sparse data in the cloud, making it possible to achieve high-precision and reliable SOH estimation in scenarios such as "partial loss of sampling data, low-cost BMS computing power and insufficient storage".

[0050] 4. The present invention provides a complete, clear, high-performance, and easy-to-implement SOH end-cloud collaborative estimation method, which helps enterprises better monitor and manage vehicle batteries throughout their life cycle. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a schematic diagram of a process of the present invention;

[0052] Figure 2 is a schematic diagram of a constant voltage charging segment in an embodiment of the present invention;

[0053] Figure 3 is a schematic diagram of a CV charging current prediction error based on HI in an embodiment of the present invention;

[0054] Figure 4 It is a schematic diagram of the correlation results of HI characteristics, time characteristics, capacity characteristics and battery capacity in an embodiment of the present invention;

[0055] Figure 5 is a schematic diagram of the correlation results between HI and capacitance under different sparsity in an embodiment of the present invention;

[0056] Figure 6 It is a schematic diagram of the SOH end-cloud collaborative estimation result in an embodiment of the present invention. DETAILED DESCRIPTION

[0057] The technical solution of the present invention is further specifically described below through embodiments and in conjunction with the accompanying drawings.

[0058] Embodiment: A method for end-cloud collaborative estimation of battery health status based on constant voltage charging segments, including two stages of offline construction of a charging feature mapping library and end-cloud collaborative estimation of SOH, specifically including the following steps:

[0059] Step 1, obtaining constant current and constant voltage charging test data of the battery, and intercepting a local segment of the constant voltage charging curve; selectively extracting HI, CV charging time and CV charging capacity features based on different sampling density data of the segment;

[0060] Step 2, using the Pearson correlation coefficient to analyze the linear correlation between HI, CV charging time and CV charging capacity characteristics and battery capacity, and constructing a feature mapping library;

[0061] Step 3: Obtain vehicle charging data and perform end-cloud collaborative estimation of battery SOH based on the feature mapping library.

[0062] Among them, step 1 and step 2 are the stages of offline construction of the charging feature mapping library, and step 3 is the SOH end-cloud collaborative estimation stage.

[0063] Specifically, step 1 includes the following steps:

[0064] Step 1.1, obtain the battery constant current and constant voltage charging test data, take the current during constant current charging as the benchmark (recorded as "1C"), extract the curve segment within the 0.2C to 0.7C window, that is, the intercepted constant voltage charging segment, such as Figure 2 As shown; wherein the constant voltage charging segment includes sampling time and current;

[0065] Step 1.2: Extract the charging time of HI and CV and the charging capacity characteristics of CV in sequence based on the constant voltage charging segment in step 1.1;

[0066] Step 1.2.1, the extraction of HI is based on the external electrical behavior of the equivalent circuit model in the CV stage, using the Thevenin model as the model to describe the external characteristics of the battery:

[0067]

[0068] Among them, R int is the series internal resistance, C p is the polarization capacitance, R p is the polarization internal resistance, V p is the voltage across the first-order RC network, V OC is the open circuit voltage of the battery, I l is the load current;

[0069] Performing Laplace transform and Z transform on equation (1) yields:

[0070]

[0071] Among them, T s is the sampling period, k and k-1 are the current sampling time and the sampling time of the previous period;

[0072] Assume that CV segment V oc It is a linear function of SOC, with a coefficient of m. The value of m varies with different battery aging levels, and we get:

[0073]

[0074] Since the CV segment voltage remains unchanged, equation (2) can be simplified to

[0075] I k+1 =θ s I k (4)

[0076] in,

[0077]

[0078] Let the constant voltage charging current data in step 1.1 be I = (I1, I2, ..., I n ), (I1, I2, …, I n-1 ) and (I2, I3, …, I n ) are used as the input and output of the least squares method, and the θ of each cycle in formula (4) can be identified. s ;

[0079] Define a new parameter h (parameter h is HI) to describe new batteries and old batteries. The value of a new battery is 1, and the more severe the aging, the closer the value is to 0. The expression is as follows:

[0080]

[0081] The calculated parameter h is the HI to be extracted in this embodiment;

[0082] like Figure 3 As shown, in this embodiment, taking the first cycle and the last cycle as examples, the correlation coefficient R between the predicted current based on HI and the original current data is 2 Both are greater than 0.99, and the prediction relative errors are within 0.3% and 3%, respectively.

[0083]

[0084] Among them, n is the data length, y k and estimation,k are experimental data and predicted data respectively;

[0085] Step 1.2.2: The CV charging time and CV charging capacity characteristics are the sampled consumption time value and capacity growth of the constant voltage charging segment in step 1.1. The extraction method can be calculated by formula (7):

[0086]

[0087] Among them, T start and T end are the initial and cutoff sampling times, tim add and cap add That is, CV charging time and CV charging capacity;

[0088] Step 1.2.3: Downsample the constant voltage charging segment in step 1.1, and take the sampling interval as 10s to obtain a sparse data segment. In this embodiment, the current data length is reduced from the original 2339 to 234; extract HI based on the segment using step 1.2.1;

[0089] Specifically, step 2 includes the following steps:

[0090] Step 2.1: Use the Pearson correlation coefficient to analyze the linear correlation between HI, CV charging time and CV charging capacity characteristics and battery capacity respectively. The calculation formula is as follows:

[0091]

[0092] Where p is the characteristic sequence, Q is the capacity sequence, σ p and σ Qare the average values ​​of the characteristic sequence and capacity sequence respectively, and ρ is the correlation coefficient; Figure 4 As shown in Figure 2, in this implementation, each feature has a high correlation with capacity (ρ>0.97); at the same time, Figure 5 As shown, the correlation between HI features and capacity under sparse data is not significantly reduced (ρ>0.96);

[0093] Step 2.2: Establish a prediction model for battery capacity with respect to HI, CV charging time and CV charging capacity characteristics, use the HI sequence in step 1.2.1, the CV charging time feature sequence and the CV charging capacity feature sequence in step 1.2.2 and the battery capacity sequence as training sets, and obtain the trained model parameter group as a dense data mapping library; similarly, establish a prediction model for battery capacity with respect to HI, use the HI sequence and the battery capacity sequence in step 1.2.3 as training sets, and obtain the trained model parameter group as a sparse data mapping library; preferably, the prediction model adopts a linear regression model.

[0094] Specifically, step 3 includes the following steps:

[0095] Step 3.1: Obtain vehicle charging data and choose whether end-cloud collaboration is required based on data quality, computing power and storage conditions, and the enterprise's cloud monitoring and management requirements; if end-cloud collaboration is not required, proceed to step 3.2, otherwise proceed to step 3.3;

[0096] Step 3.2: Determine whether the feature extraction conditions are met based on the charging data collected and stored by the BMS. If so, obtain the characteristics of the charging process according to the HI extraction method in step 2.1 and the CV charging time and CV charging capacity feature extraction method in step 2.2, and estimate SOH based on the dense data mapping library in step 2 to obtain Figure 6 The vehicle-side estimation result is shown; otherwise, no processing is performed and the battery waits for the next charging.

[0097] Step 3.3: According to the sparse data of this charging uploaded by the vehicle-side Tbox received and stored by the cloud platform, determine whether the feature extraction conditions are met. If so, obtain the characteristics of this charging process according to the HI extraction method in step 2.3, and estimate SOH based on the dense data mapping library in step 2 to obtain Figure 6 The cloud-based estimation result is shown; otherwise, no processing is performed and the battery waits for the next charging.

[0098] Among them, the vehicle charging data in step 3.1 is the sequence data of the current changing with time during the constant current and constant voltage charging stage of the battery; the feature extraction condition is based on the current during constant current charging (denoted as "1C"), and the vehicle charging data is the process of experiencing the constant voltage charging current decreasing from 0.7C to 0.2C.

[0099] Finally, it should be pointed out that the above embodiments are only representative examples of the present invention. Obviously, the present invention is not limited to the above embodiments, and there are many variations. Any simple modification, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention should be considered to belong to the protection scope of the present invention.

Claims

1. A method for end-cloud collaborative estimation of battery health status based on constant voltage charging segments, characterized in that: The following steps are involved: Step 1, obtaining constant current and constant voltage charging test data of the battery, and intercepting a local segment of the constant voltage charging curve; selectively extracting HI, CV charging time and CV charging capacity features based on different sampling density data of the segment; Step 2, using the Pearson correlation coefficient to analyze the linear correlation between HI, CV charging time and CV charging capacity characteristics and battery capacity, and constructing a feature mapping library; Step 3: Obtain vehicle charging data and perform end-cloud collaborative estimation of battery SOH based on the feature mapping library; Step 1 includes the following steps: Step 1.1, obtaining constant current and constant voltage charging test data of the battery, and intercepting a constant voltage charging segment; wherein the constant voltage charging segment includes a sampling time and current; Step 1.2, extracting HI, CV charging time and CV charging capacity characteristics in sequence based on the constant voltage charging segment in step 1.1; Step 1.2.1, the extraction of HI is based on the external electrical behavior of the equivalent circuit model in the CV stage, using the Thevenin model as the model to describe the external characteristics of the battery: （1） in, is the polarized capacitor, is the polarization internal resistance, is the voltage across the first-order RC network, is the open circuit voltage of the battery, is the load current, is the series resistance; Performing Laplace transform and Z transform on equation (1) yields: （2） in, is the sampling period, and is the current sampling time and the sampling time of the previous cycle; Assuming CV segment About A linear function with coefficients , under different battery aging levels The values ​​are different, and we get: （3） Since the CV segment voltage remains unchanged, equation (2) can be simplified to （4） in, Define a new parameter To describe new batteries and old batteries, the value of a new battery is 1, and the more severe the aging, the closer the value is to 0. The expression is as follows: （5） Let the current data of the constant voltage charging segment in step 1.1 be ,Will and As the input and output of the least square method, the HI in each cycle can be identified; Step 1.2.2: The CV charging time and CV charging capacity characteristics are the sampled consumption time value and capacity growth of the constant voltage charging segment in step 1.1: （6） and are the initial and cutoff sampling times, and That is, CV charging time and CV charging capacity; Step 1.2.3: Downsample the constant voltage charging segment in step 1.1 to obtain a sparse data segment, and extract HI based on the segment using step 1.2.

1.

2. According to claim 1, a method for end-cloud collaborative estimation of battery health status based on constant voltage charging segments is characterized in that: Step 2 includes the following steps: Step 2.1: The Pearson correlation coefficient is used to analyze the linear correlation between HI, CV charging time and CV charging capacity characteristics and battery capacity respectively. The calculation formula is as follows: （7） in is the feature sequence, is the capacity sequence, and are the average values ​​of the characteristic sequence and capacity sequence, respectively. is the correlation coefficient; Step 2.2: Establish a prediction model for battery capacity with respect to HI, CV charging time and CV charging capacity characteristics, use the HI sequence in step 1.2.1, the CV charging time feature sequence and the CV charging capacity feature sequence in step 1.2.2, and the battery capacity sequence as training sets, and obtain the trained model parameter group as a dense data mapping library; similarly, establish a prediction model for battery capacity with respect to HI, use the HI sequence and the battery capacity sequence in step 1.2.3 as training sets, and obtain the trained model parameter group as a sparse data mapping library.

3. According to claim 1, a method for end-cloud collaborative estimation of battery health status based on constant voltage charging segments is characterized in that: Step 3 includes the following steps: Step 3.1: Obtain vehicle charging data and choose whether end-cloud collaboration is required based on data quality, computing power and storage conditions, and the enterprise's cloud monitoring and management requirements; if end-cloud collaboration is not required, proceed to step 3.2, otherwise proceed to step 3.3; Step 3.2: Determine whether the feature extraction conditions are met based on the charging data collected and stored by the BMS. If so, obtain the characteristics of the charging process according to the HI extraction method in step 1.2.1 and the CV charging time and CV charging capacity feature extraction method in step 1.2.2, and estimate the SOH based on the dense data mapping library in step 2; otherwise, do not process and wait for the next charge; Step 3.3: Based on the sparse data of this charging uploaded by the vehicle-side Tbox and received and stored by the cloud platform, determine whether the feature extraction conditions are met. If so, obtain the characteristics of this charging process according to the HI extraction method in step 1.2.3, and estimate SOH based on the sparse data mapping library in step 2; otherwise, do not process it and wait for the next charging.

4. According to claim 1, a method for end-cloud collaborative estimation of battery health status based on constant voltage charging segments, characterized in that: The constant voltage charging segment in step 1.1 is based on the current during constant current charging, and the 0.2C~0.7C window is intercepted to obtain the curve segment, with the benchmark being 1C.

5. According to claim 1, a method for end-cloud collaborative estimation of battery health status based on constant voltage charging segments, characterized in that: The sparse data segment in step 1.2 is a current data sequence obtained by sampling the constant voltage charging segment in step 1 at any value between 10s and 30s.

6. According to claim 2, a method for end-cloud collaborative estimation of battery health status based on constant voltage charging segments, characterized in that: The prediction model in step 2.2 is a linear regression model.

7. According to claim 3, a method for end-cloud collaborative estimation of battery health status based on constant voltage charging segments, characterized in that: The vehicle charging data in step 3.1 is the sequence data of the current changing with time during the constant current and constant voltage charging stage of the battery.

8. According to claim 3, a method for end-cloud collaborative estimation of battery health status based on constant voltage charging segments is characterized in that: The feature extraction condition is based on the current during constant current charging. The vehicle charging data undergoes a process where the constant voltage charging current decreases from 0.7C to 0.2C, with the benchmark being 1C.

Images