New energy automobile power battery health state detection method based on migration from single body to battery pack

Abstract

The invention relates to a new energy automobile power battery SOH detection method based on migration from a single battery to a battery pack, and the method comprises the steps: collecting the voltage data of a single battery in a cycle charging stage, and constructing a pre-trained single battery SOH evaluation model; extracting a charging stage voltage curve of the battery pack; performing deep feature migration on the charging stage voltage curve of the battery pack to obtain target battery pack data; based on the target battery pack data, retraining the single battery SOH evaluation model to obtain a battery pack SOH evaluation model; and performing new energy automobile power battery health state detection by using the battery pack evaluation model. According to the invention, the SOH evaluation method for migration from the single battery to the vehicle battery pack is designed, an evaluation model is optimized by using small-scale sample data, and rapid evaluation of the SOH of the battery pack through a small-scale sample data set is realized by selecting a proper model main network and an innovative deep neuron feature optimization method.

Description

Technical Field

[0001] The present invention relates to the technical field of electric vehicle detection, and particularly to a method for detecting the health state of a power battery of a new energy vehicle based on the migration from a single cell to a battery pack. Background Art

[0002] New energy vehicles have become an important direction for the development of the automotive industry due to their advantages of environmental friendliness and high energy efficiency. As the core component of new energy vehicles, the health state (State of Health, SOH) of the battery pack directly affects the vehicle's driving range, charging speed, and safety. In actual use, the health state of the battery pack is affected by various factors, including temperature, charge-discharge cycle times, depth of discharge, and environmental conditions. These factors can cause battery capacity attenuation and internal resistance increase, thereby affecting the overall vehicle performance and posing safety hazards. Accurate assessment of the battery health state can not only predict the trend of battery performance decline, provide maintenance suggestions for vehicle owners, but also improve the safety and reliability of the overall vehicle operation.

[0003] However, at present, there is a lack of direct detection means for the health state of vehicle battery packs, especially without unpacking. The main reasons are as follows: First, lithium-ion batteries convert chemical energy into electrical energy through a series of electrochemical coupling reactions. Without unpacking, the battery pack is equivalent to a "black box" to the outside world, and it is difficult to detect the microchemical reactions related to internal aging. During actual detection, only physical quantities such as voltage and current during the battery charge-discharge process can be used for indirect estimation, which brings challenges to detection; Second, vehicle power-level battery packs have characteristics such as high voltage, large capacity, and large volume. Therefore, it is difficult and costly to collect data from vehicle-grade battery packs. In the new development situation, a more rapid, accurate, and intelligent method for assessing the health state of vehicle battery packs is needed.

[0004] The continuous progress of big data science and the extensive application of deep learning technology in the engineering field have provided an advanced approach driven by operation data for the SOH assessment of vehicle battery packs, meeting the new requirements for assessment methods.

[0005] The latest existing methods obtain characteristic data during the battery charging process, such as current, voltage, temperature, etc. By analyzing and modeling these data, the health state of the battery is evaluated. This type of method is called the data-driven method. The general process is divided into: 1. Data collection. Obtain data such as current, voltage, temperature, etc. when the battery is charging from the battery system; 2. Data preprocessing. Clean, correct, and preprocess the collected data to ensure data consistency and accuracy; 3. Feature extraction. Extract features related to the battery health state from the processed data based on deep learning or machine learning; 4. Modeling. Use the extracted features to establish a model, usually a mathematical model that correlates the battery SOH and the features. 5. Calculate the result. Calculate the battery SOH through the model. Figure 1 It is a schematic diagram of the data-driven method process.

[0006] The above method has been successfully applied to the health state detection of small single cells such as 18650 model batteries at the current stage and achieved ideal evaluation results. However, when the evaluation object is changed to a larger-capacity vehicle-grade battery pack, due to the differences in data distribution between single cells and battery packs, the evaluation accuracy is reduced or the model fails. Especially at present, there are many types of battery packs on the market and the chemical structures are complex. Therefore, the operation data of the battery pack to be tested must be collected to retrain the model in order to fully utilize the performance of the evaluation model. Also, due to the characteristics of high voltage and large capacity of the battery pack, obtaining the operation data required for retraining takes a huge amount of time and labor costs, seriously hindering the smooth application of this method.

[0007] How to improve the cross-object evaluation ability of the model and shorten the time and work for model retraining is the key issue for the actual application of the data-driven method in the health state evaluation of vehicle battery packs. Summary of the Invention

[0008] The purpose of the present invention is to propose a method for detecting the health state of power batteries for new energy vehicles based on the migration from single cells to battery packs to solve the problems existing in the above-mentioned prior art. The present invention designs an SOH evaluation method for migrating from battery single cells to vehicle battery packs, successfully extending the method that could only be applied to the SOH detection of battery single cells to the vehicle detection field. The evaluation model is optimized using small-scale sample data. By selecting a suitable model main network and an innovative deep neuron feature optimization method, finally, based on the battery single cell SOH evaluation model, the rapid evaluation of the battery pack SOH through a small-scale sample data set is realized.

[0009] To achieve the above object, the present invention provides the following solution:

[0010] A method for detecting the health state of power batteries for new energy vehicles based on the migration from single cells to battery packs, including:

[0011] Collect the voltage data of a single battery during the cyclic charging stage, and construct a pre-trained SOH evaluation model for the single battery; wherein, the chemical composition of the single battery is the same as that of the battery pack;

[0012] Extract the voltage curve of the battery pack during the charging stage;

[0013] Perform deep feature migration on the voltage curve of the battery pack during the charging stage to obtain the target battery pack data;

[0014] Based on the target battery pack data, retrain the SOH evaluation model for the single battery to obtain a SOH evaluation model for the battery pack;

[0015] Use the SOH evaluation model for the battery pack to detect the health state of the power battery of a new energy vehicle.

[0016] Optionally, collecting the voltage data of a single battery during the cyclic charging stage includes:

[0017] Prepare a battery cell with the same chemical composition as the battery pack as an experimental battery;

[0018] Perform charge and discharge experiments on the experimental battery;

[0019] Detect the maximum discharge capacity of the battery after charge and discharge. If the ratio of the maximum discharge capacity to the rated capacity is higher than the preset threshold, repeat the charge and discharge experiment; when the ratio of the maximum discharge capacity to the rated capacity is less than the preset threshold, extract all the voltage data curves during the charging stage.

[0020] Optionally, performing the charge and discharge experiment on the experimental battery includes:

[0021] Charge the battery cell at a constant current with a preset rate until the maximum allowable charging voltage, and then discharge it at the constant current with the preset rate until the maximum allowable discharge voltage.

[0022] Optionally, establishing a pre-trained SOH evaluation model for a single battery includes:

[0023] Based on the long short-term memory neural network model and the three-layer fully connected neural network model, construct an initial network model;

[0024] Normalize the voltage data of the single battery during the cyclic charging stage;

[0025] Input the normalized voltage data into the initial network model, and train the initial network model to obtain a SOH evaluation model for the single battery.

[0026] Optionally, the expression for normalizing the voltage data of the single battery during the cyclic charging stage is:

[0027] x' = (x - μ) / σ

[0028] Among them, x' is the voltage data after standardization, x is the original data, μ is the mean of the feature, and σ is the standard deviation of the feature.

[0029] Optionally, inputting the voltage data after standardization into the initial network model includes:

[0030] Using a long short-term memory neural network to extract the temporal features in the data;

[0031] Using a three-layer fully connected neural network to extract the deep neuron features in the data.

[0032] Optionally, using a long short-term memory neural network to extract the temporal features in the data includes:

[0033] Organize the voltage data after standardization into a three-dimensional tensor, namely the number of samples, the number of time steps, and the number of features, and cut the organized data in the form of a sliding window;

[0034] Input the cut data into the long short-term memory network in chronological order, and calculate the hidden state of the current time step from the hidden state of the previous time step and the input of the current time step, so as to extract the temporal features.

[0035] Optionally, the method for performing deep feature migration on the voltage curve in the charging stage is:

[0036] U' = U × U 单max / U 包max

[0037] Among them, U' is the migrated feature, U is the feature of the battery pack before migration, U 单max is the maximum charging voltage of a single cell, and U 包max is the maximum charging voltage of the battery pack.

[0038] Optionally, retraining the SOH evaluation model of the single cell battery includes:

[0039] Preprocess the SOH evaluation model of the single cell battery;

[0040] Use the preprocessed model to retrain the SOH evaluation model of the single cell battery;

[0041] Among them, the preprocessing includes: freezing the neurons of the long short-term memory network model and only keeping the neurons of the fully neural network model activated.

[0042] The beneficial effects of the present invention are:

[0043] Compared with the existing methods, the advantages of the present invention are mainly reflected in the following three aspects: First, the present invention constructs an SOH pre-training model based on the cyclic aging data of small battery cells, and then migrates the battery pack data to make it have the same distribution as the cell data, so that the model can effectively detect the SOH of the battery pack. Compared with the method of directly using battery pack data to construct a model, the labor and time costs of data collection are effectively reduced. Second, by retraining the pre-training model with a small amount of battery pack data, the accuracy of the model for evaluating the SOH of the battery pack can be effectively improved, and an SOH evaluation model for the battery pack can be established quickly and at low cost. Third, a battery SOH detection model is constructed based on an artificial intelligence large model, and the conversion relationship between battery data and battery SOH is extracted, obtaining a higher accuracy rate compared with traditional models. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings based on these drawings without creative efforts.

[0045] Figure 1 It is a schematic diagram of the data-driven method;

[0046] Figure 2 It is a schematic flow chart of a method for detecting the health state of a power battery for a new energy vehicle based on migration from a single cell to a battery pack according to an embodiment of the present invention;

[0047] Figure 3 It is a schematic flow chart of a method for detecting the health state of a power battery for a new energy vehicle applied to an EV power battery testing station according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0048] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, not all of them. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0049] To make the above objects, features, and advantages of the present invention more obvious and understandable, the present invention will be further described in detail below with reference to the drawings and specific embodiments.

[0050] As Figure 2 shown, this embodiment proposes a method for detecting the health state of a power battery for a new energy vehicle based on migration from a single cell to a battery pack, including:

[0051] Collect the voltage data of a single battery during the cyclic charging stage and construct a pre-trained SOH evaluation model for the single battery; among them, the chemical composition of the single battery is the same as that of the battery pack;

[0052] Extract the voltage curve of the battery pack during the charging stage;

[0053] Perform deep feature migration on the voltage curve of the battery pack during the charging stage to obtain the target battery pack data;

[0054] Retrain the SOH evaluation model for the single battery based on the target battery pack data to obtain the SOH evaluation model for the battery pack;

[0055] Use the SOH evaluation model of the battery pack to detect the health status of the power battery of new energy vehicles.

[0056] Specifically, in this embodiment, a pre-trained health degree evaluation model is first established based on a single battery, and then the deep feature migration method is used to transfer the battery pack features and retrain the pre-trained model with a small amount of battery pack data. Finally, a transfer learning model that can effectively evaluate the SOH of the battery pack is obtained. Specifically, in the first stage, a battery cell with the same chemical composition as the battery pack to be tested is prepared as an experimental battery. Since the battery cell has a small capacity and low voltage, it can be quickly aged at a low artificial and time cost, and its aging data is collected. The battery cell is charged at a constant current of C1 rate until the maximum allowable charging voltage, and then discharged at a constant current of C1 rate until the maximum allowable discharge voltage. The maximum discharge capacity of the battery is detected at this time. If the ratio (SOH) of the maximum discharge capacity to the rated capacity is higher than 80%, the above charge and discharge experiments are repeated. When the SOH is less than 80%, all voltage data curves during the charging stage are extracted as the training database. A pre-trained SOH evaluation model for the single battery is established based on the training database. In the second stage, first communicate with the vehicle to obtain the rated capacity Qr and the maximum charging voltage Umax of the battery pack. Calculate the charging current I according to I = Qr * C1. Then, charge the battery pack at a constant current of I to the maximum charging voltage and extract the voltage curve during the charging stage. Perform deep feature migration on the extracted voltage curve of the battery pack to make it have the same distribution as the battery cell data. The specific method is U’ = U × U 单 max / U 包 max, where U’ is the migrated feature, U is the feature of the battery pack before migration, and U 单 max is the maximum charging voltage of the single battery, and U 包max is the maximum charging voltage of the battery pack. After migrating the data, the model is retrained based on a small amount of battery pack data. The retraining method is to freeze some neurons in the part of the model for extracting time series features, keep the deep neurons activated, and after importing a small part of the battery pack features, obtain the battery pack SOH evaluation model.

[0057] Further, the voltage data collected during the cyclic charging stage of the single battery includes:

[0058] Prepare battery cells with the same chemical composition as the battery pack as experimental batteries;

[0059] Conduct charge and discharge experiments on the experimental batteries;

[0060] Detect the maximum discharge capacity of the battery after charge and discharge. If the ratio of the maximum discharge capacity to the rated capacity is higher than the preset threshold, repeat the charge and discharge experiment; when the ratio of the maximum discharge capacity to the rated capacity is less than the preset threshold, extract all the voltage data curves during the charging stage.

[0061] Further, conducting charge and discharge experiments on the experimental batteries includes:

[0062] Charge the single battery at a constant current with a preset rate until the maximum allowable charging voltage, and then discharge it at a constant current with the preset rate until the maximum allowable discharge voltage.

[0063] Specifically, apply this embodiment to the EV power battery testing station; its technical process is as Figure 3 shown; Data acquisition stage: Prepare battery single cells with the same chemical materials as the vehicle to be tested, and perform constant current cyclic charge and discharge on them at a current rate of 0.5C until its maximum available capacity is less than 80% of the rated available capacity. Collect all the voltage data during the charging stage to form a training data set. Use the EV battery pack data reading device to read the battery pack information of the vehicle to be tested, and obtain its rated capacity and maximum charging voltage. Perform a single charge and discharge operation on the vehicle to be tested at a current rate of 0.5C as well, and collect the voltage data during its charging stage to form test data.

[0064] Further, establishing a pre-trained single battery SOH evaluation model includes:

[0065] Based on the long short-term memory neural network model and the three-layer fully connected neural network model, construct an initial network model;

[0066] Standardize the voltage data during the cyclic charging stage of the single battery;

[0067] Input the standardized voltage data into the initial network model, and train the initial network model to obtain the single battery SOH evaluation model.

[0068] Further, inputting the standardized voltage data into the initial network model includes:

[0069] Using a long short-term memory neural network to extract the temporal features in the data;

[0070] Using a three-layer fully connected neural network to extract the deep neuron features in the data.

[0071] Further, using a long short-term memory neural network to extract the temporal features in the data includes:

[0072] Organize the standardized voltage data into a three-dimensional tensor, namely the number of samples, the time step, and the number of features, and cut the organized data in the form of a sliding window;

[0073] Input the cut data into the long short-term memory network in chronological order, and calculate the hidden state of the current time step from the hidden state of the previous time step and the input of the current time step, so as to extract the temporal features.

[0074] Further, retraining the single-cell battery SOH evaluation model includes:

[0075] Preprocess the single-cell battery SOH evaluation model;

[0076] Use the preprocessed model to retrain the single-cell battery SOH evaluation model;

[0077] Among them, the preprocessing includes: freezing the neurons of the long short-term memory network model and only keeping the neurons of the fully neural network model activated.

[0078] Specifically, in this embodiment, feature migration: Based on the formula U’ = U × U 单max / U 包max , perform feature migration on the data of the battery pack to make it have the same distribution as the cell data. Where U’ is the migrated feature, U is the feature of the battery pack before migration, U 单max is the maximum charging voltage of the single cell, and U 包max is the maximum charging voltage of the battery pack.

[0079] Specifically, in this embodiment, standardization: Standardize the extracted features based on the formula x’ = (x - μ) / σ, where x is the original data, μ is the mean of the features, and σ is the standard deviation of the features.

[0080] Specifically, in this embodiment, for the extraction of temporal features: temporal features are extracted from the standardized features. First, the original data is organized into a three-dimensional tensor, namely the number of samples, the time step (= 600), and the number of features (= 1). The entire data segment is cut in the form of a sliding window with a window size of 10 and a step size of 10. The segmented data is sequentially input into the long short-term memory network according to the time order. The hidden state of the current time step is calculated from the hidden state of the previous time step and the input of the current time step, thereby extracting temporal features.

[0081] Specifically, in this embodiment, for the extraction and transfer of deep neuron features: After the long short-term memory network, a three-layer fully-connected neural network is constructed to extract deep neuron features. The number of neurons in each layer is 100, 100, and 50 respectively. When constructing the pre-training model with the data of the power cells, all neurons in the long short-term memory network and the fully-connected neural network are kept activated. When re-training the model with the data of the battery pack, the neurons in the long short-term memory network are frozen, and only the neurons in the fully-connected neural network are kept activated.

[0082] Specifically, in this embodiment, for the construction of the SOH evaluation model: All the data is input into the constructed model, and the SOH evaluation result is output.

[0083] Applying this embodiment to the EV battery pack detection station, through this method, the SOH of the vehicle battery pack can be quickly and effectively evaluated.

[0084] Next, taking the application in the vehicle annual inspection station as an example, the detailed implementation process of the technical solution of this embodiment will be described.

[0085] The fixed vehicle detection equipment is internally equipped with a high-voltage charging gun, a current sensor, a voltage sensor, a memory, and a processor.

[0086] Data acquisition: At the vehicle annual inspection station, the battery health detection equipment is connected to the battery system of the vehicle to be inspected. The vehicle is charged at a constant current with a high-voltage charging gun at a high charging rate (80 A) to ensure that the battery quickly reaches the charging stable state. The data of the battery pack during the charging stage, including current, voltage, and temperature information, is collected and saved as the original data for input into the model.

[0087] Model construction: The health state detection model of the electric vehicle power battery is constructed using the long short-term memory neural network + three-layer fully-connected neural network architecture. The original data is standardized to ensure that the data is suitable for model input. Then, the preprocessed data is input into the network. The long short-term memory neural network extracts the temporal features in the data, and the fully-connected neural network extracts the deep neuron features in the data, and finally the SOH is output.

[0088] Parameter determination: By using battery cells of the same chemical type prepared in advance, divide 80A by the rated capacity of the battery pack to be tested to obtain the charging rate. Charge the battery with a constant current at this charging rate to obtain charging voltage data. Input these data into the model to obtain the predicted SOH value. Use the mean square error between the predicted SOH value and the true SOH value of the battery cell as the training loss function of the model, and adjust the model parameters based on the backpropagation algorithm to minimize the difference between the model output and the actual SOH. The model after parameter determination is deployed in the device memory before the detection device leaves the factory.

[0089] Battery SOH detection: Use a battery health detection device to apply a constant current at the same C1 rate to the vehicle to be tested, and obtain the battery pack data at this time. Based on the formula U’ = U × U_single_max / U_pack_max, perform feature migration on the battery pack data to make it have the same distribution as the single-cell data. Where U’ is the migrated feature, U is the feature of the battery pack before migration, U_single_max is the maximum charging voltage of the single cell, and U_pack_max is the maximum charging voltage of the battery pack. Input the migrated feature into the model after parameter determination to obtain an estimate of the health status (SOH) of the vehicle to be tested.

[0090] Application: At the vehicle annual inspection station, through this technology, it is possible to quickly and accurately evaluate the battery health status, ensuring that the battery system of electric vehicles meets the annual inspection standards. This helps improve the efficiency of the annual inspection station and ensure road traffic safety. At the same time, it provides a convenient way for vehicle owners to timely understand the health status of the vehicle battery so as to take necessary maintenance measures.

[0091] Compared with the existing methods, the advantages of this embodiment are mainly reflected in the following three aspects: First, this method constructs an SOH pre-training model based on the cyclic aging data of small battery cells, and then migrates the battery pack data to make it have the same distribution as the single-cell data, so that the model can effectively detect the SOH of the battery pack. Compared with the method of directly using battery pack data to construct a model, the labor and time costs of data collection are effectively reduced. Second, by retraining the pre-training model with a small amount of battery pack data, the accuracy of the model for evaluating the SOH of the battery pack can be effectively improved, and an SOH evaluation model for the battery pack can be established quickly and at low cost. Third, based on the artificial intelligence large model to construct a battery SOH detection model, extract the conversion relationship between battery data and battery SOH, and obtain a higher accuracy compared with the traditional model.

[0092] A novel method for detecting the SOH of a vehicle battery pack proposed in this embodiment effectively detects the SOH of the internal battery pack of the vehicle without unpacking by constructing an SOH evaluation model through battery cells and migrating data from battery cells to the battery pack.

[0093] Retraining and optimizing the model with a small amount of battery pack sample data can effectively improve the evaluation performance of the SOH evaluation model for battery pack data, and expand the application of data-driven methods in the field of vehicle detection.

[0094] The embodiments described above are only descriptions of the preferred embodiments of the present invention, and do not limit the scope of the present invention. Without departing from the design spirit of the present invention, various deformations and improvements made by those of ordinary skill in the art to the technical solutions of the present invention shall fall within the protection scope determined by the claims of the present invention.

Claims

1. A method for detecting the health status of a power battery of a new energy vehicle based on the migration from a single cell to a battery pack, characterized in that: include: Collect voltage data of a single cell during a cycle charging phase and construct a pre-trained single cell SOH evaluation model; wherein the chemical composition of the single cell is the same as that of the battery pack; Extract the voltage curve of the battery pack during the charging phase; Performing deep feature migration on the voltage curve of the battery pack during the charging phase to obtain target battery pack data; Based on the target battery pack data, retraining the single cell SOH evaluation model to obtain a battery pack SOH evaluation model; The battery pack SOH evaluation model is used to detect the health status of the power battery of a new energy vehicle.

2. The method for detecting the health status of a power battery of a new energy vehicle based on the migration from a single cell to a battery pack according to claim 1 is characterized in that: The voltage data collected during the cycle charging phase of a single battery includes: Prepare battery cells with the same chemical composition as the battery pack as experimental batteries; Carrying out a charge and discharge experiment on the experimental battery; The maximum dischargeable capacity of the battery after charge and discharge is detected. If the ratio of the maximum discharge capacity to the rated capacity is higher than a preset threshold, the charge and discharge experiment is repeated; when the ratio of the maximum discharge capacity to the rated capacity is less than the preset threshold, all voltage data curves in the charging stage are extracted.

3. The method for detecting the health status of a power battery of a new energy vehicle based on the migration from a single cell to a battery pack according to claim 2 is characterized in that: The charging and discharging experiment of the experimental battery includes: The battery cell is charged at a preset rate constant current until the maximum allowed charging voltage is reached, and then discharged at the preset rate constant current until the maximum allowed discharging voltage is reached.

4. The method for detecting the health status of a power battery of a new energy vehicle based on the migration from a single cell to a battery pack according to claim 1, characterized in that: Establishing a pre-trained single cell battery SOH evaluation model includes: Based on the long short-term memory neural network model and the three-layer fully connected neural network model, the initial network model is constructed; Standardize the voltage data of the single battery during the cycle charging stage; The standardized voltage data is input into the initial network model, the initial network model is trained, and a single cell SOH evaluation model is obtained.

5. The method for detecting the health status of a power battery of a new energy vehicle based on the migration from a single cell to a battery pack according to claim 4 is characterized in that: The expression for standardizing the voltage data of a single cell during the cycle charging phase is: x' = (x - μ) / σ Among them, x' is the standardized voltage data, x is the original data, μ is the mean of the feature, and σ is the standard deviation of the feature.

6. The method for detecting the health status of a power battery of a new energy vehicle based on the migration from a single cell to a battery pack according to claim 4 is characterized in that: Inputting the standardized voltage data into the initial network model comprises: Use long short-term memory neural network to extract temporal features from data; A three-layer fully connected neural network is used to extract deep neuron features in the data.

7. The method for detecting the health status of a power battery of a new energy vehicle based on the migration from a single cell to a battery pack according to claim 6 is characterized in that: The time series features extracted from data using long short-term memory neural networks include: The standardized voltage data is organized into three-dimensional tensors, which are the number of samples, time steps, and number of features. The organized data is cut in the form of a sliding window; The cut data are put into the long short-term memory network in chronological order. The hidden state of the current time step is calculated based on the hidden state of the previous time step and the input of the current time step, thereby extracting the time series features.

8. The method for detecting the health status of a power battery of a new energy vehicle based on the migration from a single cell to a battery pack according to claim 1, characterized in that: The method for performing deep feature migration on the voltage curve in the charging stage is: U'=U×U 单max / IN 包max Among them, U' is the characteristic after migration, U is the characteristic of the battery pack before migration, and U 单max is the maximum charging voltage of the single cell, U 包max It is the maximum charging voltage of the battery pack.

9. The method for detecting the health status of a power battery of a new energy vehicle based on the migration from a single cell to a battery pack according to claim 4, characterized in that: Retraining the single cell SOH evaluation model includes: Preprocessing the single cell SOH evaluation model; Retraining the single cell SOH evaluation model using the preprocessed model; Among them, the preprocessing includes: freezing the neurons of the long short-term memory network model and only keeping the neuron activation of the full neural network model.

Images