New energy vehicle driving range estimation method based on data driving

Abstract

The invention relates to a new energy automobile driving range estimation method based on data driving, and belongs to the technical field of batteries. The method comprises the following steps: collecting operation data of a new energy vehicle of a corresponding vehicle type, and establishing a new energy vehicle operation database; establishing an estimation model based on the charging data and a data enhancement technology, and calculating the current available total capacity of the vehicle according to an estimation result and an ampere integral method; based on a Pearson correlation coefficient index, extracting a charging feature set and a driving feature set which have correlation with the driving mileage from the driving mileage estimation feature set; and constructing a driving range estimation model based on XGBoost according to the selected charging feature set and the driving feature set, and estimating the driving range through the driving range estimation model. The current available total energy of the battery is calculated by using the historical operation data of the vehicle, the influence of battery aging on the driving range is effectively considered, the characteristics are extracted from the battery state and the driving behavior, and the actual condition of vehicle operation is comprehensively reflected.

Description

Technical Field

[0001] The invention belongs to the technical field of batteries and relates to a method for estimating the driving range of a new energy vehicle based on data-driven. Background Art

[0002] The methods for estimating the driving range of electric vehicles can generally be divided into three categories: based on energy calculation, based on vehicle parameters and operating environment, and based on data-driven. The energy calculation method calculates the remaining available energy of the battery at a certain moment, and estimates the energy consumption value of the electric vehicle at a future moment based on the historical data of the electric vehicle's past operation, and divides the two values ​​to obtain the remaining driving range of the vehicle. The method based on vehicle parameters and operating environment takes into account the battery power, real-time traffic environment and driving status, and estimates the remaining driving range through some parameters (such as driving speed, vehicle mass, driving resistance, etc.). The data-driven method has received widespread attention in recent years because it relies on sufficient data to achieve high-precision estimation. The basic process of using the data-driven method to estimate the remaining driving range of a vehicle is: vehicle operation data extraction, data preprocessing and model training and tuning, to obtain an empirical model, and to predict the remaining driving range of the vehicle through this empirical model. Since the data-driven method is a prediction model learned from the historical data of vehicle operation, the obtained model has good versatility and high prediction accuracy. However, these methods rely on the vehicle's current SOC or SOE to estimate the available energy of electric vehicles and ignore the impact of battery aging, which makes the model unusable for long-term mileage estimation of electric vehicles.

[0003] In response to the above problems, there is currently no effective estimation method based on machine learning and historical operating data that takes battery aging into consideration to efficiently and accurately estimate the driving range of new energy vehicles. Summary of the invention

[0004] In view of this, an object of the present invention is to provide a data-driven method for estimating the driving range of new energy vehicles.

[0005] In order to achieve the above object, the present invention provides the following technical solutions:

[0006] A method for estimating the driving range of a new energy vehicle based on data drive, the method comprising the following steps:

[0007] S1. Collect the operation data of new energy vehicles of corresponding models and establish a new energy vehicle operation database;

[0008] S2. Establish an estimation model based on charging data and data enhancement technology, and calculate the current total available capacity of the vehicle according to the estimation results and the ampere integral method;

[0009] S3, extracting a charging feature set and a driving feature set that are correlated with the driving range from the driving range estimation feature set based on the Pearson correlation coefficient index;

[0010] S4. According to the selected charging feature set and driving feature set, a driving range estimation model is constructed based on XGBoost, and the driving range estimation model is used to estimate the driving range.

[0011] Further, in step S1, the collected operating data of the new energy vehicle include: charging and discharging current, maximum battery cell voltage, minimum battery cell voltage, maximum battery cell temperature, minimum battery cell temperature, time, charging status, vehicle speed, mileage and SOC; then, based on the collected vehicle operating data, a battery pack operating database corresponding to the new energy vehicle is established.

[0012] Further, in step S2, the following steps are specifically included:

[0013] S21, analyzing the charging data of new energy vehicles and extracting stable charging segments therefrom;

[0014] S22, calculating the charging capacity of the sudden charge SOC interval by the ampere-hour integration method, and using it as a label;

[0015] S23, arbitrarily intercepting charging data with an SOC span of 20% from a charging segment in which the charging cut-off SOC is greater than 99% and the starting SOC is less than 80% to extract features related to the charging capacity in the sudden charging SOC interval as a capacity estimation feature set;

[0016] S24. According to the capacity estimation feature set, a sudden charge SOC interval capacity estimation model is constructed based on GPR, and the training set is expanded by data enhancement technology;

[0017] S25. Calculate the charging capacity of the stable charging SOC interval using the ampere-hour integration method, and then estimate the charging capacity of the sudden charging SOC interval using the sudden charging SOC interval capacity estimation model to obtain the current total available capacity of the vehicle.

[0018] Further, in step S21, the stable charging segment refers to a charging segment in which the charging cut-off SOC is greater than 99% and the starting SOC is less than 80%;

[0019] In step S22, the calculation method of the data-driven sudden charge SOC interval capacity estimation tag is:

[0020]

[0021] Where I(t) is the total current when the vehicle is charging, and [90,100] is the sudden charge SOC interval;

[0022] In step S23, the charging data of the charging segment with a charging SOC span of 20% includes the current mileage, average current, average temperature, the highest cell voltage at the end of charging, the difference between the highest cell voltage and the lowest cell voltage at the end of charging, the maximum cell voltage change in the 20% SOC interval, the minimum cell voltage change in the 20% SOC interval, the charging cut-off SOC, and the characteristics of the charged power;

[0023] In step S24, the sudden charge SOC interval capacity estimation model constructed based on GPR includes: processing the input features into input feature vectors, constructing a mapping relationship between the input features and the target output based on the kernel function, and optimizing the parameters in the kernel function through learning of training samples, so that the model can accurately capture the complex relationship between the input features and the target output;

[0024] Data enhancement technology refers to dividing the stable charging segment into multiple charging sub-segments with equal span SOC, extracting a sample from each charging sub-segment; extracting mileage, average charging current, average temperature during charging, the difference between the highest cell voltage at the beginning and the end, the difference between the lowest cell voltage at the beginning and the end, and the difference between the highest cell voltage and the lowest cell voltage at the end of charging from all the expanded samples to construct a capacity estimation feature set;

[0025] In step S25, the current total available capacity of the new energy vehicle is calculated as:

[0026]

[0027] Where Q is the current total available capacity; Q 90_100 ' is the capacity of the 90%-100% SOC interval estimated by the sudden charge SOC interval capacity estimation model; x is the starting SOC value of the charging segment; y is the ending SOC value of the charging segment.

[0028] Further, in step S3, the following steps are included:

[0029] S31, analyzing the vehicle operation data, extracting complete charging data and partial data of its corresponding discharging process;

[0030] S32, calculating statistical characteristics of the charging data and the driving data, including average values ​​and standard deviation values, based on the selected charging data and driving data, and extracting a driving range estimation feature set;

[0031] S33. Screen out a driving range estimation feature set that is correlated with the driving range through a correlation analysis based on the Pearson correlation coefficient, wherein the driving range estimation feature set includes a charging feature set and a driving feature set.

[0032] Further, in step S33, the Pearson correlation coefficient PCC can provide the direction and strength information of the linear correlation between the features in the driving range estimation feature set and the driving range, and its calculation method is:

[0033]

[0034] In the formula, x i represents the i-th feature in the range estimation feature set, y represents the actual range, and Represent the average values ​​of the corresponding feature sequence and mileage sequence respectively.

[0035] Furthermore, the charging feature set screened according to the Pearson correlation coefficient includes average current, maximum battery cell voltage, minimum battery cell voltage, average temperature of the charging process, total available energy, and charging starting SOC; the driving feature set includes energy consumption of part of the driving process, average temperature of the driving process, depth of discharge, average vehicle speed, average current of the discharge process, and current variance of the discharge process.

[0036] Further, in step S4, the following steps are included:

[0037] S41. Using the selected range estimation feature set as input, a vehicle range estimation model is established based on XGBoost, and the model is iteratively trained through a gradient boosting framework; the first round of iteration is based on a training sample set containing feature vectors and real mileage, and the first decision tree is constructed: a greedy algorithm is used to divide the feature space, and the optimal split point is determined by maximizing information gain; subsequent iterations generate new trees to optimize feature combinations and split points;

[0038] S42, inputting the test data after the feature set is extracted into the trained vehicle range estimation model to perform vehicle range estimation.

[0039] The beneficial effects of the present invention are:

[0040] The present invention uses the historical operation data of the vehicle to calculate the total energy currently available in the battery, effectively considering the impact of battery aging on the driving range. The present invention requires less data, reduces the high data collection cost, accurately estimates the vehicle's driving range, and helps users overcome range anxiety. The present invention extracts features from both battery status and driving behavior, and can more comprehensively reflect the actual operating conditions of the vehicle.

[0041] Other advantages, objectives and features of the present invention will be described in the following description to some extent, and to some extent, will be obvious to those skilled in the art based on the following examination and study, or can be taught from the practice of the present invention. The objectives and other advantages of the present invention can be realized and obtained through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be described in detail below in conjunction with the accompanying drawings, wherein:

[0043] Figure 1 The overall flow chart of the data-driven new energy vehicle range estimation method of the present invention;

[0044] Figure 2 The overall architecture diagram of the data-driven new energy vehicle range estimation method of the present invention;

[0045] Figure 3 It is a schematic diagram of the analysis results of the correlation between the health indicator set and the battery health status under the embodiment. DETAILED DESCRIPTION

[0046] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments only illustrate the basic concept of the present invention in a schematic manner, and the following embodiments and features in the embodiments can be combined with each other without conflict.

[0047] Among them, the drawings are only used for illustrative explanations, and they only represent schematic diagrams rather than actual pictures, and should not be understood as limitations on the present invention. In order to better illustrate the embodiments of the present invention, some parts of the drawings may be omitted, enlarged or reduced, and do not represent the size of actual products. For those skilled in the art, it is understandable that some well-known structures and their descriptions in the drawings may be omitted.

[0048] The same or similar numbers in the drawings of the embodiments of the present invention correspond to the same or similar parts; in the description of the present invention, it should be understood that if the terms "upper", "lower", "left", "right", "front", "rear", etc. indicate the orientation or position relationship, they are based on the orientation or position relationship shown in the drawings, which is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation. Therefore, the terms describing the position relationship in the drawings are only used for illustrative purposes and cannot be understood as limiting the present invention. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to specific circumstances.

[0049] See also Figure 1 to Figure 3 , which is a data-driven method for estimating the driving range of new energy vehicles.

[0050] Example

[0051] Figure 1 The overall flow chart of a data-driven new energy vehicle mileage estimation method of the present invention is shown, which can be specifically divided into the following steps:

[0052] Step S1: Collecting the operating data of new energy vehicles of the same model, including charging and discharging data, and establishing a new energy vehicle operating database;

[0053] Step S2: Based on the charging data and data enhancement technology, an estimation model is established, and the total available capacity of the vehicle is calculated using the estimation results and the ampere integral method;

[0054] Step S3: extracting a charging feature set and a driving feature set that are correlated with the driving range based on a Pearson correlation coefficient (PCC) indicator;

[0055] Step S4: constructing a driving range estimation model based on XGBoost according to the selected charging feature set and driving feature set;

[0056] As an optional embodiment, the complete technical roadmap of this solution is as follows: Figure 2 shown.

[0057] As an optional embodiment, the above step S1 specifically includes steps S11-S12:

[0058] Step S11: collecting operation data of a certain electric vehicle, including charging and discharging current, maximum value of battery cell voltage, minimum value of battery cell voltage, maximum value of battery cell temperature, minimum value of battery cell temperature, time, charging state, vehicle speed, mileage and SOC;

[0059] Step S12: Establish a battery pack operation database for a certain car based on the collected vehicle operation data.

[0060] As an optional embodiment, the above step S2 specifically includes steps S21-S23:

[0061] Step S21: Analyze the charging data of the electric vehicle. As the battery ages, when the battery is fully charged, the vehicle SOC will suddenly change from 90 to 100 in the range of 90-100 to 100. In order to more accurately calculate the capacity corresponding to this part, select the charging segment with a cut-off SOC greater than 99 and a starting SOC less than 80;

[0062] Step S22: Calculate the charging capacity in the 90%-100% SOC range by the ampere-hour integration method as a label;

[0063] Step S23: extracting features related to the charging capacity in the SOC interval of 90%-100% according to the charging data with an SOC span of 20%, as a capacity estimation feature set;

[0064] Step S24: constructing a sudden charge SOC interval capacity estimation model based on the GPR according to the capacity estimation feature set, and expanding the training set using data enhancement technology;

[0065] Step S25: Calculate the capacity below 90% SOC using the ampere-hour integration method, and then use the trained mutation charge SOC interval capacity estimation model to estimate the capacity in the 90%-100% SOC interval. Add the two together to get the current total available capacity, which reflects the current health status of the battery. As the vehicle is used, its health status will decrease, the total available capacity of the battery will decrease, and the driving range will decrease; the current total available capacity is used as a feature and input into the driving range prediction model.

[0066] As an optional embodiment, an estimation model is established based on charging data and data enhancement technology, and the total available capacity of the vehicle is calculated using the estimation results and the ampere integral method, specifically including:

[0067] Screening charging segments that meet the SOC usage range of 90%-100% and a span greater than 20%;

[0068] The charging capacity in the 90%-100% SOC interval is calculated by the ampere-hour integration method as a data-driven mutation charging SOC interval capacity estimation label. The formula is as follows:

[0069]

[0070] Extract the current mileage, average current, average temperature, the highest cell voltage at the end of charging, the difference between the highest cell voltage and the lowest cell voltage at the end of charging, the change of the highest cell voltage in the 20% SOC interval, the change of the lowest cell voltage in the 20% SOC interval, the charging cut-off SOC, the charged power, etc. from the charging data with an SOC span of 20% as features for capacity estimation in the 90%-100% SOC interval;

[0071] The training set was expanded using data enhancement technology. In the charging segments with an SOC interval span greater than 20, multiple charging segments with an SOC span equal to 20 were divided and a sample was extracted from each of them. In all the expanded samples, the mileage, average charging current, average temperature during the charging process, the difference between the highest cell voltage at the beginning and the end, the difference between the lowest cell voltage at the beginning and the end, and the difference between the highest cell voltage and the lowest cell voltage at the end of charging were extracted as features. A sudden charge SOC interval capacity estimation model was constructed based on GPR to estimate Q 90_100 ′.

[0072] The capacity below 90% SOC is calculated using the ampere-hour integration method, and then the capacity in the 90%-100% SOC interval is estimated using the trained mutation charge SOC interval capacity estimation model. The two are added together to obtain the current total available capacity, as shown in the following formula:

[0073]

[0074] Where Q is the current total available capacity; Q 90_100 ' is the capacity of the 90%-100% SOC interval estimated by the sudden charge SOC interval capacity estimation model; x is the starting SOC value of the charging segment (without the percentage sign); y is the cutoff SOC value of the charging segment, and the maximum value of y is 90;

[0075] As an optional embodiment, the above step S3 specifically includes steps S31-S33:

[0076] Step S31: Analyze the vehicle operation data to extract complete charging data and partial data of the corresponding discharging process;

[0077] Step S32: Calculate the statistical characteristics of the charging data and the driving data, including the average value and the standard deviation value, based on the selected charging data and the driving data, and extract the driving range estimation feature set;

[0078] Step S33: Filter out a driving range estimation feature set that has a high correlation with the driving range through correlation analysis based on the Pearson correlation coefficient (PCC).

[0079] As an optional embodiment, the driving range estimation feature set includes a charging feature set and a driving feature set. The charging feature set includes average current, maximum battery cell voltage, minimum battery cell voltage, average temperature during charging, total available energy, and charging start SOC. The driving feature set includes energy consumption during part of the driving process, average temperature during the driving process, depth of discharge, average vehicle speed, average current during the discharge process, and current variance during the discharge process.

[0080] In order to describe the factors affecting the driving range from multiple aspects, the charging feature set and the driving feature set are simultaneously used to establish a vehicle driving range estimation model to achieve the estimation of the vehicle driving range.

[0081] As an optional embodiment, the Pearson correlation coefficient (PCC) indicator is used to evaluate the correlation between the driving range estimation feature set and the driving range, such as Figure 3 As shown, it is a schematic diagram of the analysis of the correlation results between the driving range estimation feature set and the driving range of the embodiment; the Pearson correlation coefficient (PCC) provides information on the direction and strength of the linear correlation, and the calculation formula is shown as follows:

[0082]

[0083] In the formula, x i represents the i-th feature in the range estimation feature set, y represents the actual range, and Represent the average values ​​of the corresponding feature sequence and mileage sequence respectively.

[0084] As an optional embodiment, in order to consider the impact on the driving range from multiple aspects, the relevant threshold is set to 0.1.

[0085] As an optional embodiment, the above step S4 specifically includes steps S41-S42:

[0086] Step S41: Take the selected range estimation feature set as input, establish a vehicle range estimation model based on XGBoost, and use the training set to train and tune the model; the model is iteratively trained through the gradient boosting framework. The first round of iteration is based on the training sample set containing feature vectors and real mileage, and the first decision tree is constructed: the feature space (such as speed segmentation) is divided by a greedy algorithm, and the optimal split point is determined by maximizing information gain. Subsequent iterations generate new trees, optimize feature combinations and split points to reduce residuals and improve model accuracy.

[0087] Step S42: Input the test data after the feature set is extracted into the trained vehicle range estimation model to estimate the vehicle range.

[0088] As an optional embodiment, the mean absolute error and the root mean square error are used to evaluate the driving range estimation effect.

[0089] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, 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 purpose and scope of the technical solution, which should be included in the scope of the claims of the present invention.

Claims

1. A data-driven method for estimating the driving range of new energy vehicles, characterized in that: The method comprises the following steps: S1. Collect the operation data of new energy vehicles of corresponding models and establish a new energy vehicle operation database; S2. Establish an estimation model based on charging data and data enhancement technology, and calculate the current total available capacity of the vehicle according to the estimation results and the ampere integral method; S3, extracting a charging feature set and a driving feature set that are correlated with the driving range from the driving range estimation feature set based on the Pearson correlation coefficient index; S4. According to the selected charging feature set and driving feature set, a driving range estimation model is constructed based on XGBoost, and the driving range estimation model is used to estimate the driving range.

2. The method for estimating the driving range of new energy vehicles based on data drive according to claim 1, characterized in that: In step S1, the collected operating data of the new energy vehicle include: charging and discharging current, maximum battery cell voltage, minimum battery cell voltage, maximum battery cell temperature, minimum battery cell temperature, time, charging status, vehicle speed, mileage and SOC; then, based on the collected vehicle operating data, a battery pack operating database corresponding to the new energy vehicle is established.

3. The method for estimating the driving range of new energy vehicles based on data drive according to claim 2, characterized in that: In step S2, the following steps are specifically included: S21, analyzing the charging data of new energy vehicles and extracting stable charging segments therefrom; S22, calculating the charging capacity of the sudden charge SOC interval by the ampere-hour integration method, and using it as a label; S23, arbitrarily intercepting charging data with an SOC span of 20% from a charging segment in which the charging cut-off SOC is greater than 99% and the starting SOC is less than 80% to extract features related to the charging capacity in the sudden charging SOC interval as a capacity estimation feature set; S24. According to the capacity estimation feature set, a sudden charge SOC interval capacity estimation model is constructed based on GPR, and the training set is expanded by data enhancement technology; S25. Calculate the charging capacity of the stable charging SOC interval using the ampere-hour integration method, and then estimate the charging capacity of the sudden charging SOC interval using the sudden charging SOC interval capacity estimation model to obtain the current total available capacity of the vehicle.

4. The method for estimating the driving range of new energy vehicles based on data drive according to claim 3 is characterized in that: In step S21, the stable charging segment refers to a charging segment in which the charging cut-off SOC is greater than 99% and the starting SOC is less than 80%; In step S22, the calculation method of the data-driven sudden charge SOC interval capacity estimation tag is: Where I(t) is the total current when the vehicle is charging, and [90,100] is the sudden charge SOC interval; In step S23, the charging data of the charging segment with a charging SOC span of 20% includes the current mileage, average current, average temperature, the highest cell voltage at the end of charging, the difference between the highest cell voltage and the lowest cell voltage at the end of charging, the maximum cell voltage change in the 20% SOC interval, the minimum cell voltage change in the 20% SOC interval, the charging cut-off SOC, and the characteristics of the charged power; In step S24, the sudden charge SOC interval capacity estimation model constructed based on GPR includes: processing the input features into input feature vectors, constructing a mapping relationship between the input features and the target output based on the kernel function, and optimizing the parameters in the kernel function through learning of training samples, so that the model can accurately capture the complex relationship between the input features and the target output; Data enhancement technology refers to dividing the stable charging segment into multiple charging sub-segments with equal span SOC, extracting a sample from each charging sub-segment; extracting mileage, average charging current, average temperature during charging, the difference between the highest cell voltage at the beginning and the end, the difference between the lowest cell voltage at the beginning and the end, and the difference between the highest cell voltage and the lowest cell voltage at the end of charging from all the expanded samples to construct a capacity estimation feature set; In step S25, the current total available capacity of the new energy vehicle is calculated as: Where Q is the current total available capacity; Q 90_100 ' is the capacity of the 90%-100% SOC interval estimated by the sudden charge SOC interval capacity estimation model; x is the starting SOC value of the charging segment; y is the ending SOC value of the charging segment.

5. The method for estimating the driving range of new energy vehicles based on data drive according to claim 3, characterized in that: In step S3, the following steps are included: S31, analyzing the vehicle operation data, extracting complete charging data and partial data of its corresponding discharging process; S32, calculating statistical characteristics of the charging data and the driving data, including average values ​​and standard deviation values, based on the selected charging data and driving data, and extracting a driving range estimation feature set; S33. Screen out a driving range estimation feature set that is correlated with the driving range through a correlation analysis based on the Pearson correlation coefficient, wherein the driving range estimation feature set includes a charging feature set and a driving feature set.

6. The method for estimating the driving range of new energy vehicles based on data drive according to claim 5, characterized in that: In step S33, the Pearson correlation coefficient PCC can provide the direction and strength information of the linear correlation between the features in the driving range estimation feature set and the driving range, and is calculated as follows: In the formula, x i represents the i-th feature in the range estimation feature set, y represents the actual range, and Represent the average values ​​of the corresponding feature sequence and mileage sequence respectively.

7. The method for estimating the driving range of new energy vehicles based on data drive according to claim 6 is characterized by: The charging feature set selected based on the Pearson correlation coefficient includes average current, maximum battery cell voltage, minimum battery cell voltage, average temperature during charging, total available energy, and charging start SOC; The driving feature set includes the energy consumption of part of the driving process, the average temperature of the driving process, the discharge depth, the average vehicle speed, the average current of the discharge process, and the current variance of the discharge process.

8. The method for estimating the driving range of new energy vehicles based on data drive according to claim 5, characterized in that: In step S4, the following steps are included: S41. Using the selected range estimation feature set as input, a vehicle range estimation model is established based on XGBoost, and the model is iteratively trained through a gradient boosting framework; the first round of iteration is based on a training sample set containing feature vectors and real mileage, and the first decision tree is constructed: a greedy algorithm is used to divide the feature space, and the optimal split point is determined by maximizing information gain; subsequent iterations generate new trees to optimize feature combinations and split points; S42, inputting the test data after the feature set is extracted into the trained vehicle range estimation model to perform vehicle range estimation.

Images