Electric heavy truck lithium battery SOC calibration method and device based on neural network, medium and product
Through the lithium battery calibration model based on neural network, the problem of insufficient SOC estimation accuracy of lithium batteries in electric heavy trucks is solved, and more accurate SOC calibration is achieved, reducing mileage errors and improving user experience.
Patent Information
- Application Number
- CN202510345182.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-03-24
- Publication Date
- 2025-07-08
AI Technical Summary
In the prior art, the SOC estimation accuracy of the lithium battery of electric heavy trucks is poor, and cannot accurately reflect the actual mileage, affecting the vehicle's endurance and user experience.
By obtaining the battery state parameters before charging of the lithium battery, using the neural network to build a lithium battery calibration model, dividing the training set and test set, using the neural network to train the model, select the optimal parameters, and realize the calibration of the lithium battery.
It improves the accuracy of lithium battery SOC calibration, reduces the mileage error of electric heavy trucks, and improves user experience.
Smart Images
Figure CN120275845A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of state of charge (SOC) of lithium batteries, and particularly to a method, device, medium and product for calibrating the SOC of lithium batteries of electric heavy trucks based on a neural network. Background Art
[0002] With the rapid development of electric heavy trucks, lithium batteries, as their core power source, the accurate estimation of the state of charge (SOC) is crucial for vehicle driving range prediction, energy management, and safety. Due to the huge load borne by electric heavy trucks during use and the influence of long-term complex charging and discharging working conditions, the battery parameters of their lithium batteries change greatly. Therefore, during the repeated charging and discharging process of electric heavy trucks, the SOC value needs to be calibrated frequently.
[0003] Currently, the lithium battery SOC estimation technology still faces many challenges in practical applications. Traditional SOC estimation methods have poor estimation accuracy for the lithium batteries of electric heavy trucks, cannot accurately reflect the actual driving mileage, and are difficult to meet the requirements of high-precision SOC estimation for electric heavy trucks, thus affecting the driving mileage error and user experience of electric heavy trucks. Summary of the Invention
[0004] In view of this, the present invention provides a method, device, medium and product for calibrating the SOC of lithium batteries of electric heavy trucks based on a neural network. The present invention can improve the accuracy of calibrating the SOC of lithium batteries of electric heavy trucks and reduce the driving mileage error of electric heavy trucks.
[0005] To achieve the above object, the present invention provides the following solutions:
[0006] First aspect, the present invention provides a method for calibrating the state of charge (SOC) of lithium batteries in electric heavy trucks based on a neural network. The method for calibrating the SOC of lithium batteries in electric heavy trucks based on a neural network includes: obtaining the state-of-charge parameters of multiple lithium batteries before charging; the state-of-charge parameters include the SOC value to be calibrated and the cut-off voltage; the lithium batteries are lithium batteries for electric heavy trucks; screening out multiple lithium batteries to be calibrated through SOC calibration based on the state-of-charge parameters before charging, obtaining the SOC error value of each lithium battery to be calibrated and the calibration value of the lithium battery to be calibrated; dividing the SOC error values of all lithium batteries to be calibrated and the calibration values of the lithium batteries to be calibrated into a training set and a test set; constructing a lithium battery calibration model based on a neural network; inputting the training set into the lithium battery calibration model to train the lithium battery calibration model; using a validation set to select the optimal parameters of the lithium battery calibration model, and taking the model corresponding to the minimum value of the loss function as the trained lightweight model; inputting the SOC error value of the target lithium battery into the trained lithium battery calibration model to obtain the calibration value of the target lithium battery; performing SOC calibration on the target lithium battery based on the calibration value of the target lithium battery.
[0007] Second aspect, the present invention provides a computer device, including: a memory, a processor, and a computer program stored on the memory and executable on the processor. The processor executes the computer program to implement the above-mentioned method for calibrating the SOC of lithium batteries in electric heavy trucks based on a neural network.
[0008] Third aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the above-mentioned method for calibrating the SOC of lithium batteries in electric heavy trucks based on a neural network.
[0009] Fourth aspect, the present invention provides a computer program product, including a computer program. When the computer program is executed by a processor, it implements the above-mentioned method for calibrating the SOC of lithium batteries in electric heavy trucks based on a neural network.
[0010] According to the specific embodiments provided by the present invention, the following technical effects are disclosed by the present invention:
[0011] The present invention constructs a lithium battery calibration model by using a neural network, and uses the SOC error values and calibration values of multiple lithium batteries as a training set to train the lithium battery calibration model. The trained lithium battery calibration model is used to complete the calibration of the target lithium battery, improving the accuracy of the SOC calibration of lithium batteries in electric heavy trucks and reducing the driving range error of electric heavy trucks. Description of the Drawings
[0012] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.
[0013] Figure 1 It is a schematic flowchart of a method for calibrating the SOC of a lithium battery of an electric heavy truck based on a neural network provided by an embodiment of the present invention.
[0014] Figure 2 It is a schematic structural diagram of a computer device provided by an embodiment of the present invention. Specific embodiments
[0015] In order to enable those skilled in the art to better understand the solutions of the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.
[0016] Embodiment 1, as Figure 1 shown, this embodiment provides a method for calibrating the SOC of a lithium battery of an electric heavy truck based on a neural network. The method for calibrating the SOC of a lithium battery of an electric heavy truck based on a neural network includes:
[0017] S1. Obtain the power state parameters of multiple lithium batteries before charging; the power state parameters include the SOC value to be calibrated and the cut-off voltage; the lithium batteries are lithium batteries of electric heavy trucks.
[0018] S2. Based on the power state parameters before charging, screen out multiple lithium batteries to be calibrated through SOC calibration, and obtain the SOC error value of each lithium battery to be calibrated and the calibration value of the lithium battery to be calibrated.
[0019] Step S2 specifically includes:
[0020] Perform the following operations on each lithium battery:
[0021] S21. Charge the lithium battery with a preset amount of power to obtain the SOC value to be calibrated and the cut-off voltage after charging.
[0022] S22. Calculate the first proportionality coefficient; the first proportionality coefficient is the ratio of the difference between the SOC value to be calibrated after charging and the SOC value to be calibrated before charging, to the difference between the cut-off voltage to be calibrated after charging and the cut-off voltage to be calibrated before charging.
[0023] In some embodiments, the calculation formula of the first proportionality coefficient is as follows:
[0024] R = ΔQ - ΔV.
[0025] In the formula, R is the first proportionality coefficient, ΔQ is the difference between the calibrated SOC value after charging and the calibrated SOC value before charging, and ΔV is the difference between the calibrated cut-off voltage after charging and the calibrated cut-off voltage before charging.
[0026] S23. Repeatedly obtain multiple sets of proportionality coefficients.
[0027] S24. Fit the multiple sets of proportionality coefficients to obtain a proportionality coefficient curve.
[0028] In some embodiments, the fitting methods of the proportionality coefficient curve include at least any one of polynomial fitting and least squares fitting.
[0029] S25. Select the proportionality coefficient with the smallest difference from the median of the proportionality coefficient curve as the target proportionality coefficient.
[0030] In some embodiments, select the minimum value of the proportionality coefficient curve as the target proportionality coefficient.
[0031] In some embodiments, among multiple sets of proportionality coefficients, if the current proportionality coefficient is less than the previous proportionality coefficient, then take the current proportionality coefficient as the target proportionality coefficient.
[0032] S26. When the second proportionality coefficient is greater than the threshold, regard the current lithium battery as the lithium battery to be calibrated, calculate the SOC error value of the lithium battery to be calibrated, and take the target proportionality coefficient as the calibration value of the lithium battery to be calibrated; the SOC error value is the difference between the calibrated SOC value after charging corresponding to the target proportionality coefficient and the actual SOC value after charging calculated by the battery management controller; the second proportionality coefficient is the ratio of the SOC error value to the maximum capacity of the lithium battery to be calibrated.
[0033] In some embodiments, the calculation formula of the second proportionality coefficient is as follows:
[0034] S = ΔW / C.
[0035] In the formula, S is the second proportionality coefficient, ΔW is the SOC error value of the lithium battery to be calibrated, and C is the maximum capacity of the lithium battery to be calibrated.
[0036] S3. Divide the SOC error values of all lithium batteries to be calibrated and the calibration values of the lithium batteries to be calibrated into a training set and a test set.
[0037] S4. Construct a lithium battery calibration model based on a neural network.
[0038] In some embodiments, the neural network is a convolutional neural network.
[0039] S5. Input the training set into the lithium battery calibration model and train the lithium battery calibration model.
[0040] S6. Use the validation set to select the optimal parameters of the lithium battery calibration model, and take the model corresponding to the minimum value of the loss function as the trained lightweight model.
[0041] In some embodiments, the parameters of the lithium battery calibration model are adjusted by the honey badger algorithm.
[0042] S7. Input the SOC error value of the target lithium battery into the trained lithium battery calibration model to obtain the calibration value of the target lithium battery.
[0043] S8. Perform SOC calibration on the target lithium battery based on the calibration value of the target lithium battery.
[0044] Step S8 specifically includes: charging the target lithium battery with the power corresponding to the calibration value of the target lithium battery, and simultaneously calibrating and updating the battery management controller of the target lithium battery.
[0045] The technical effects of the present invention are as follows:
[0046] The present invention determines the target proportional coefficient of the lithium battery of the electric heavy truck by means of linear fitting and taking the median, calculates the calibration value based on this, then constructs a lithium battery calibration model using a neural network, and uses the SOC error values and calibration values of multiple lithium batteries as the training set to train the lithium battery calibration model, and completes the calibration of the target lithium battery with the trained lithium battery calibration model, improving the accuracy of the SOC calibration of the lithium battery of the electric heavy truck, reducing the driving range error of the electric heavy truck, and improving the user experience of the electric heavy truck user.
[0047] Example 2, as Figure 2As shown, the present invention also provides a computer device, which can be a server or a terminal. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O), and a communication interface. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store processed data. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals through a network connection. The computer program, when executed by the processor, implements the above-mentioned method.
[0048] Those skilled in the art can understand that Figure 2 the structure shown in is only a block diagram of some structures related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.
[0049] In Embodiment 3, the present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-mentioned various methods.
[0050] In Embodiment 4, the present invention also provides a computer program product including a computer program, which, when executed by a processor, implements the above-mentioned various methods.
[0051] In several embodiments provided by the present invention, it can be understood that each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, and the module, the program segment, or the part of code includes one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved.
[0052] In several embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection to each other can be through some interfaces. The indirect coupling or communication connection of devices or units can be in electrical, mechanical or other forms.
[0053] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than limiting them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.
Claims
1. A method for calibrating the SOC of a lithium battery of an electric heavy-duty truck based on a neural network, characterized in that, The method for calibrating the SOC of the lithium battery of an electric heavy truck based on a neural network includes: Obtaining the state-of-charge parameters of multiple lithium batteries before charging; the state-of-charge parameters include the SOC value to be calibrated and the cut-off voltage; the lithium battery is a lithium battery of an electric heavy truck; Based on the state-of-charge parameters before charging, screening out multiple lithium batteries to be calibrated through SOC calibration, and obtaining the SOC error value of each lithium battery to be calibrated and the calibration value of the lithium battery to be calibrated; Dividing the SOC error values of all lithium batteries to be calibrated and the calibration values of the lithium batteries to be calibrated into a training set and a test set; Constructing a lithium battery calibration model based on a neural network; Inputting the training set into the lithium battery calibration model to train the lithium battery calibration model; Using the validation set to select the optimal parameters of the lithium battery calibration model, and taking the model corresponding to the minimum value of the loss function as the trained lightweight model; Inputting the SOC error value of the target lithium battery into the trained lithium battery calibration model to obtain the calibration value of the target lithium battery; Performing SOC calibration on the target lithium battery based on the calibration value of the target lithium battery.
2. The method for calibrating the SOC of the lithium battery of the electric heavy truck based on the neural network according to claim 1, wherein Based on the state-of-charge parameters before charging, screening out multiple lithium batteries to be calibrated through SOC calibration, and obtaining the SOC error value of each lithium battery to be calibrated and the calibration value of the lithium battery to be calibrated, specifically including: Performing the following operations on each lithium battery: Charging the lithium battery with a preset amount of electricity to obtain the SOC value to be calibrated and the cut-off voltage after charging; Calculating the first proportionality coefficient; the first proportionality coefficient is the ratio of the difference between the SOC value to be calibrated after charging and the SOC value to be calibrated before charging, to the difference between the cut-off voltage to be calibrated after charging and the cut-off voltage to be calibrated before charging; Repeatedly obtaining multiple sets of proportionality coefficients; Fitting the multiple sets of proportionality coefficients to obtain a proportionality coefficient curve; Selecting the proportionality coefficient with the smallest difference from the median of the proportionality coefficient curve as the target proportionality coefficient; When the second proportionality coefficient is greater than the threshold, taking the current lithium battery as the lithium battery to be calibrated, calculating the SOC error value of the lithium battery to be calibrated, and taking the target proportionality coefficient as the calibration value of the lithium battery to be calibrated; the SOC error value is the difference between the SOC value to be calibrated after charging corresponding to the target proportionality coefficient and the actual SOC value after charging calculated by the battery management controller; the second proportionality coefficient is the ratio of the SOC error value to the maximum capacity of the lithium battery to be calibrated.
3. The method for calibrating the SOC of the lithium battery of an electric heavy truck based on a neural network according to claim 2, wherein, The calculation formula of the first proportionality coefficient is as follows: R = ΔQ - ΔV; In the formula, R is the first proportionality coefficient, ΔQ is the difference between the SOC value to be calibrated after charging and the SOC value to be calibrated before charging, and ΔV is the difference between the cut-off voltage to be calibrated after charging and the cut-off voltage to be calibrated before charging.
4. The method for calibrating the SOC of the lithium battery of the electric heavy truck based on the neural network according to claim 2, characterized in that, The calculation formula of the second proportionality coefficient is as follows: S = ΔW / C; In the formula, S is the second proportionality coefficient, ΔW is the SOC error value of the lithium battery to be calibrated, and C is the maximum capacity of the lithium battery to be calibrated.
5. The method for calibrating the SOC of the lithium battery of an electric heavy truck based on a neural network according to claim 2, wherein The fitting method of the proportionality coefficient curve includes at least any one of polynomial fitting and least squares fitting.
6. The method for calibrating the SOC of the lithium battery of an electric heavy truck based on a neural network according to claim 1, characterized in that, The neural network is a convolutional neural network.
7. The method for calibrating the SOC of the lithium battery of the electric heavy truck based on the neural network according to claim 1, wherein, Performing SOC calibration on the target lithium battery based on the calibration value of the target lithium battery, specifically including: Charge the target lithium battery with the power corresponding to the calibration value of the target lithium battery, and at the same time calibrate and update the battery management controller of the target lithium battery.
8. A computer device, comprising: A memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor executes the computer program to implement the method for calibrating the SOC of a lithium battery of an electric heavy truck based on a neural network according to any one of claims 1-7.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the method for calibrating the SOC of a lithium battery of an electric heavy truck based on a neural network according to any one of claims 1-7.
10. A computer program product comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the method for calibrating the SOC of a lithium battery of an electric heavy truck based on a neural network according to any one of claims 1-7.