Method and device for optimizing metering precision of off-board charger of electric vehicle

Through the combination of multimodal iterative fusion and radial basis function neural network, a multi-factor compensation model is constructed, which solves the problem that the measurement accuracy of non-vehicle chargers of electric vehicles is coupled by multiple influencing factors, and realizes the temperature, humidity and electromagnetic field The precise measurement of various factors such as strength improves measurement accuracy and fairness in charging trade.

CN119004988APending Publication Date: 2024-11-22STATE GRID SHANXI MARKETING SERVICE CENT
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411132134.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-19
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

The metering unit of the non-vehicle charger of electric vehicles is coupled by a variety of influencing factors in a complex field environment, which makes it difficult to ensure the measurement accuracy. The existing technology mainly compensates for temperature as the influence of the quantity, and fails to effectively consider other factors such as humidity and electromagnetic field strength.

Method used

The multimodal iterative fusion method is used to fuse various influences such as temperature, humidity and electromagnetic field strength, build a multi-factor compensation model, and build a current and voltage error evaluation model through a radial basis function neural network, and calculate the error compensation coefficient to optimize Measuring accuracy.

Benefits of technology

The measurement accuracy of electric vehicle non-car chargers has been significantly improved. Through the combination of a multi-factor compensation model and an error evaluation model, accurate measurement under various influencing factors has been achieved, and the fairness of charging trade settlement has been improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119004988A_ABST
    Figure CN119004988A_ABST
Patent Text Reader

Abstract

The invention relates to a metering precision optimization method and device for an off-board charger of an electric vehicle. The method comprises the following steps: acquiring historical data of a plurality of influence quantities in an operation condition of the off-board charger of the electric vehicle; fusing the plurality of influence quantities through a multi-modal iterative fusion method to obtain fused influence factor historical data; constructing a metering unit multi-factor compensation model of the non-vehicle-mounted charger of the electric vehicle; constructing a direct current and voltage error evaluation model through a radial basis function neural network method; and acquiring a plurality of real-time influence quantities of the non-vehicle-mounted charger of the electric vehicle, and substituting the influence quantities into the metering unit multi-factor compensation model and the metering unit multi-factor compensation model to obtain a current error compensation coefficient and a voltage error compensation coefficient. According to the method, the multi-factor features are fused through the multi-modal iterative fusion method, and the multi-factor features and the radial basis function neural network method are used for learning the representation of errors and multiple factors, so that the metering precision is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of vehicle detection, and specifically relates to a method and device for optimizing the metering accuracy of an off-board charger for an electric vehicle, and more particularly to a method and device for optimizing the metering accuracy of an off-board charger for an electric vehicle under multiple influencing factors. Background Art In recent years, with the rapid development of electric vehicles and their need for fast charging, off-board chargers for electric vehicles have been increasingly used. As a device that converts AC power into DC power during the charging process of electric vehicles, off-board chargers also have the function of measuring the charging power of electric vehicles. The accuracy of their measurement is related to the fairness of the settlement of electric vehicle charging trade. In 2019, the State Administration for Market Regulation issued a document to include off-board chargers for electric vehicles in the catalog of national mandatory management of measuring instruments. The off-board charger metering unit is mainly composed of a DC shunt and a DC energy meter, which contains a large number of electronic components. The performance of each component directly affects the metering characteristics of the DC charger, and the performance of the components is easily affected by the environment.

[0002] In the prior art, the working principle of DC charger electric energy metering is combined, temperature is selected as the influencing quantity, and the temperature influence simulation test platform is built to measure the metering accuracy of the DC charger metering unit at different temperatures. The temperature influence law of the DC charger is analyzed, and a temperature compensation model of the DC charger metering unit based on the piecewise linear interpolation method is proposed. The experimental results show that the accuracy of the DC charger electric energy metering is significantly improved after temperature compensation, which verifies the accuracy and applicability of the piecewise linear interpolation compensation method. Summary of the invention

[0003] In order to optimize and improve the metering accuracy of a multi-factor off-board charger for electric vehicles, a first aspect of the present invention provides a method for optimizing the metering accuracy of a non-on-board charger for electric vehicles, comprising: obtaining historical data of multiple influencing quantities in the operating conditions of the non-on-board charger for electric vehicles, wherein the multiple influencing quantities include temperature, humidity and electromagnetic field strength; fusing the multiple influencing quantities through a multimodal iterative fusion method to obtain the fused historical data of influencing factors; constructing a multi-factor compensation model of the metering unit of the non-on-board charger for electric vehicles; constructing a DC current and voltage error evaluation model through a radial basis function neural network method based on the historical data of influencing factors, DC current and voltage errors; obtaining multiple real-time influencing quantities of the non-on-board charger for electric vehicles, substituting the multiple influencing quantities into the multi-factor compensation model of the metering unit and calculating to obtain a current error compensation coefficient and a voltage error compensation coefficient.

[0004] In some embodiments of the present invention, the multiple influence quantities are fused through a multimodal iterative fusion method to obtain the fused historical data of influencing factors, including: extracting the features of each influence quantity through a multi-layer neural network, fusing the features of multiple influence quantities through feature splicing and gating functions; calculating the similarity between the fused features and the samples through an exponential kernel function; iterating the multi-layer neural network according to a similarity-based objective function until the similarity is maximized, and using the fused features corresponding to the maximum similarity as the fused historical data of influencing factors.

[0005] Furthermore, the similarity calculation method is expressed as: , in, represents the i-th influence quantity, represents the features after the fusion of the j-th layer of neural network; D(.) represents the distance; Represents the width of the exponential kernel function; It represents the similarity between the features of the i-th influence quantity and the features after fusion of the j-th layer of neural network.

[0006] Preferably, the objective function is expressed as: , in, L Represents the objective function.

[0007] In some embodiments of the present invention, the multi-factor compensation model of the metering unit of the off-board charger of the electric vehicle is constructed, including: determining the error and error ratio of the DC current according to the actual value and the measured value of the current, and determining the error and error ratio of the DC voltage according to the actual value and the measured value of the voltage; defining a DC current compensation coefficient based on the DC current error, the DC current error and the error ratio under the influence of multiple factors; defining a DC voltage compensation coefficient based on the DC voltage error, the DC voltage error and the error ratio under the influence of multiple factors.

[0008] In some embodiments of the present invention, the construction of a DC current and voltage error evaluation model based on historical data of influencing factors, DC current and voltage errors by a radial basis function neural network method includes: evaluating the DC current error by a radial basis function neural network method based on historical data of influencing factors; evaluating the DC voltage error by a radial basis function neural network method based on historical data of influencing factors.

[0009] According to a second aspect of the present invention, a device for optimizing metering accuracy of an off-board charger for electric vehicles is provided, including: an acquisition module for acquiring historical data of multiple influencing quantities in the operating conditions of the off-board charger for electric vehicles, wherein the multiple influencing quantities include temperature, humidity and electromagnetic field strength; a fusion module for fusing the multiple influencing quantities through a multimodal iterative fusion method to obtain the fused historical data of influencing factors; constructing a multi-factor compensation model of a metering unit of an off-board charger for electric vehicles; a construction module for constructing a DC current and voltage error evaluation model through a radial basis function neural network method based on the historical data of influencing factors, DC current and voltage errors; and a calculation module for acquiring multiple real-time influencing quantities of the off-board charger for electric vehicles, substituting the multiple influencing quantities into the multi-factor compensation model of the metering unit and calculating to obtain a current error compensation coefficient and a voltage error compensation coefficient.

[0010] According to a third aspect of the present invention, an electronic device is provided, comprising: one or more processors; a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method for optimizing the metering accuracy of an electric vehicle off-board charger provided in the first aspect of the present invention.

[0011] A fourth aspect of the present invention provides a computer-readable medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method for optimizing the metering accuracy of an off-board charger for an electric vehicle provided in the first aspect of the present invention.

[0012] The beneficial effects of the present invention are: Aiming at the problem of the complex on-site environment of the metering unit of the non-on-board charger of electric vehicles and the coupling of various influencing factors, the present invention proposes a method for optimizing the metering accuracy of the non-on-board charger of electric vehicles under multiple influencing factors, thereby improving the metering accuracy of the non-on-board charger of electric vehicles. The multi-modal iterative fusion algorithm is used to fuse the various influencing quantities of the non-on-board charger of electric vehicles; the fused features are compared by using the similarity calculation formula. With original features The similarity between them is used to select the optimal fusion features. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 A schematic diagram of a basic flow chart of a method for optimizing metering accuracy of an off-board charger for an electric vehicle in some embodiments of the present invention; Figure 2 A schematic diagram of a specific process of a method for optimizing the metering accuracy of an off-board charger for an electric vehicle in some embodiments of the present invention; Figure 3 is a schematic diagram of the principle of a multi-modal iterative fusion method in some embodiments of the present invention; Figure 4 A schematic diagram of a flow chart of a radial basis function neural network error evaluation process in some embodiments of the present invention; Figure 5 It is a structural schematic diagram of a device for optimizing the metering accuracy of an off-board charger for an electric vehicle in some embodiments of the present invention; Figure 6 It is a schematic diagram of the structure of an electronic device in some embodiments of the present invention. DETAILED DESCRIPTION

[0014] The principles and features of the present invention are described below in conjunction with the accompanying drawings. The examples given are only used to explain the present invention and are not used to limit the scope of the present invention.

[0015] refer to Figure 1 and Figure 2 In a first aspect of the present invention, a method for optimizing metering accuracy of an off-board charger for electric vehicles is provided, comprising: S100. Acquiring historical data of multiple influencing quantities in the operating conditions of the off-board charger for electric vehicles, wherein the multiple influencing quantities include temperature, humidity and electromagnetic field strength; S200. Fusing the multiple influencing quantities by a multi-modal iterative fusion method to obtain fused historical data of influencing factors; constructing a multi-factor compensation model of a metering unit of an off-board charger for electric vehicles; S300. Based on the historical data of influencing factors, direct current and voltage errors, constructing a direct current and voltage error evaluation model by a radial basis function neural network method; S400. Acquiring multiple real-time influencing quantities of the off-board charger for electric vehicles, substituting the multiple influencing quantities into the multi-factor compensation model of the metering unit and calculating to obtain a current error compensation coefficient and a voltage error compensation coefficient.

[0016] In step S100 of some embodiments of the present invention, historical data of multiple influencing quantities in the operating conditions of the off-board charger of the electric vehicle are obtained, wherein the multiple influencing quantities include temperature, humidity and electromagnetic field strength; specifically, a simulation test platform for multiple influencing factors of the off-board charger of the electric vehicle is established to obtain historical data of temperature, humidity and electromagnetic field strength in the operating conditions of the off-board charger of the electric vehicle.

[0017] refer to Figure 3 In step S200 of some embodiments of the present invention, the fusion of the multiple influencing factors by the multi-modal iterative fusion method to obtain the fused influencing factor historical data includes: S201. Extract the features of each influencing quantity through a multi-layer neural network, and fuse the features of multiple influencing quantities through feature splicing and gating functions; S202. Calculate the similarity between the fused features and the samples through an exponential kernel function; S203. Iterate the multi-layer neural network according to a similarity-based objective function until the similarity is maximized, and use the fused features corresponding to the maximum similarity as the fused historical data of the influencing factors.

[0018] Specifically, the temperature, humidity, and electromagnetic field strength historical data are used as the first, second, and third influencing factors. , , , and adopts multi-modal iterative fusion to integrate the first, second and third influencing factors , , Fusion, where i=1, 2, 3, , represents the output of the jth hidden layer of the ith influence quantity, represents the i-th influence quantity, represents the parameters of the neural network that processes the i-th influence quantity, represents the depth of the neural network processing the i-th influence quantity.

[0019] (2), (3), (4), (5), (6), in, represents the parameters of the fourth neural network, represents the parameters of the fifth neural network, represents the hidden layer output after j layers are fused, represents the first weighted weight of the jth layer, represents the second weighted weight of the jth layer, represents the first weighted hidden layer output of the jth layer, represents the second weighted hidden layer output of the jth layer, G represents the gating function, represents feature concatenation, represents the Kronecker product.

[0020] Furthermore, the similarity calculation method is expressed as: , in, represents the i-th influence quantity, represents the features after the fusion of the j-th layer of neural network; D(.) represents the distance; Represents the width of the exponential kernel function; It represents the similarity between the features of the i-th influence quantity and the features after fusion of the j-th layer of neural network.

[0021] Specifically, the fused features With original features The similarity between them is calculated as: (7), in, is the distance formula, and the width is defined on D as The exponential kernel function is the sample similarity.

[0022] The objective function of the multimodal iterative fusion algorithm is: (8), According to formula (8), find the optimal , obtain the historical data of influencing factors after fusion .

[0023] In step S200 of some embodiments of the present invention, constructing a multi-factor compensation model for a metering unit of an off-board charger for an electric vehicle includes: S204. Determine the error and error ratio of the DC current based on the actual value and measured value of the current, and determine the error and error ratio of the DC voltage based on the actual value and measured value of the voltage; S205. Define the DC current compensation coefficient based on the DC current error, the DC current error and the error ratio under the influence of multiple factors; S206. Define the DC voltage compensation coefficient based on the DC voltage error, the DC voltage error and the error ratio under the influence of multiple factors.

[0024] Specifically, the mathematical model of DC current and DC voltage measurement error is: (9) in, is the actual current value; To measure the current value; is the actual voltage value; To measure the voltage value; , .

[0025] The metering unit of the off-board charger of electric vehicles will change with the coupling of various influencing factors, thus causing the error to change. The metering accuracy optimization method of the off-board charger of electric vehicles adopts the method of introducing a compensation coefficient before the output of the transformer. ,

[0026] (10), in, , They are respectively the DC current and DC voltage errors under the influence of multiple factors.

[0027] From (10), we can further conclude that the multi-factor compensation model of the metering unit of the off-board charger of electric vehicles is: (11).

[0028] refer to Figure 4 In step S300 of some embodiments of the present invention, the construction of a DC current and voltage error evaluation model based on historical data of influencing factors, DC current and voltage errors by a radial basis function neural network method includes: evaluating the DC current error by a radial basis function neural network method based on historical data of influencing factors; evaluating the DC voltage error by a radial basis function neural network method based on historical data of influencing factors.

[0029] Specifically, based on the DC current and voltage error historical data obtained by the multi-influencing factor simulation test platform of the electric vehicle off-board charger and the fused influencing factor historical data, the RBF (Radial Basis Function Neural Network) algorithm is used to obtain the DC current and voltage error evaluation model. So: (12), (13), in, , Indicates the error data of DC current and voltage. , Represents the first and second implicit functions of RBF.

[0030] In step S400 of some embodiments of the present invention, multiple real-time impact quantities of the off-board charger of the electric vehicle are obtained, and the multiple impact quantities are substituted into the multi-factor compensation model of the metering unit and calculated to obtain the current error compensation coefficient and the voltage error compensation coefficient.

[0031] Specifically, the real-time multi-factor influencing quantities of the off-board charger for electric vehicles are collected and fused and then brought into the evaluation model to obtain the real-time errors of DC current and voltage; the real-time errors of DC current and voltage are then brought into the multi-factor compensation model of the metering unit of the off-board charger for electric vehicles to obtain the current error compensation coefficient and the voltage error compensation coefficient, thereby realizing the optimization of the metering accuracy of the off-board charger for electric vehicles.

[0032] It should be noted that, in order to address the problem of complex on-site environment of the metering unit of the off-board charger for electric vehicles and the coupling of various influencing factors, the present invention proposes a method for optimizing the metering accuracy of the off-board charger for electric vehicles under multiple influencing factors, thereby achieving an improvement in the metering accuracy of the off-board charger for electric vehicles.

[0033] The measurement error before and after compensation was measured using the simulation test platform, and the results are shown in Table 1: Table 1 Changes in measurement accuracy before and after optimization

[0034] It can be seen from Table 1 that after using the compensation coefficient, the metering accuracy of the off-board charger for electric vehicles has been significantly improved, with an average optimization of 1 times.

[0035] Example 2 refer to Figure 5 According to a second aspect of the present invention, a device 1 for optimizing metering accuracy of an off-board charger for electric vehicles is provided, comprising: an acquisition module 11 for acquiring historical data of multiple influencing quantities in the operating conditions of the off-board charger for electric vehicles, wherein the multiple influencing quantities include temperature, humidity and electromagnetic field strength; a fusion module 12 for fusing the multiple influencing quantities through a multimodal iterative fusion method to obtain the fused historical data of influencing factors; and constructing a multi-factor compensation model of a metering unit of an off-board charger for electric vehicles; a construction module 13 for constructing a DC current and voltage error evaluation model through a radial basis function neural network method based on the historical data of influencing factors, DC current and voltage errors; and a calculation module 14 for acquiring multiple real-time influencing quantities of the off-board charger for electric vehicles, substituting the multiple influencing quantities into the multi-factor compensation model of the metering unit and calculating to obtain a current error compensation coefficient and a voltage error compensation coefficient.

[0036] Furthermore, the fusion module 12 includes: a fusion unit, which is used to extract the features of each influencing quantity through a multi-layer neural network, and fuse the features of multiple influencing quantities through feature splicing and gating functions; a calculation unit, which is used to calculate the similarity between the fused features and the samples through an exponential kernel function; an iteration unit, which is used to iterate the multi-layer neural network according to a similarity-based objective function until the similarity is maximized, and use the fused features corresponding to the maximum similarity as the fused historical data of the influencing factors.

[0037] Example 3 refer to Figure 6According to a third aspect of the present invention, an electronic device is provided, comprising: one or more processors; a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method for optimizing the metering accuracy of the non-on-board charger of an electric vehicle according to the first aspect of the present invention.

[0038] The electronic device 500 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 501, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 502 or a program loaded from a storage device 508 into a random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the electronic device 500 are also stored. The processing device 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0039] Typically, the following devices may be connected to the I / O interface 505: an input device 506 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 507 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 508 including, for example, a hard disk, etc.; and a communication device 509. The communication device 509 may allow the electronic device 500 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 6 The electronic device 500 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead. Figure 6 Each block shown in the figure may represent one device, or may represent multiple devices as required.

[0040] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through a communication device 509, or installed from a storage device 508, or installed from a ROM 502. When the computer program is executed by the processing device 501, the above functions defined in the method of the embodiment of the present disclosure are executed. It should be noted that the computer-readable medium described in the embodiment of the present disclosure can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection with one or more conductors, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In an embodiment of the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, an apparatus, or a device. In an embodiment of the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in combination with an instruction execution system, an apparatus, or a device. The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to: wire, optical cable, RF (radio frequency), etc., or any suitable combination of the foregoing.

[0041] The computer-readable medium may be included in the electronic device, or may exist independently without being installed in the electronic device. The computer-readable medium carries one or more computer programs. When the one or more programs are executed by the electronic device, the electronic device: Computer program code for performing the operations of embodiments of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, Python, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on a user's computer, partially on a user's computer, as a separate software package, partially on a user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0042] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0043] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for optimizing the metering accuracy of an off-board charger for an electric vehicle, characterized in that: include: Acquire historical data of multiple influencing quantities in the operating conditions of the off-board charger of the electric vehicle, wherein the multiple influencing quantities include temperature, humidity and electromagnetic field strength; The multiple influencing factors are fused by a multi-modal iterative fusion method to obtain fused historical data of influencing factors; a multi-factor compensation model of a metering unit of an off-board charger for electric vehicles is constructed; Based on the historical data of influencing factors, DC current and voltage errors, a DC current and voltage error evaluation model is constructed using the radial basis function neural network method; A plurality of real-time influencing quantities of the off-board charger of the electric vehicle are obtained, and the plurality of influencing quantities are substituted into the multi-factor compensation model of the metering unit and calculated to obtain a current error compensation coefficient and a voltage error compensation coefficient.

2. The method for optimizing the metering accuracy of an off-board charger for an electric vehicle according to claim 1, characterized in that: The multiple influencing factors are fused by the multi-modal iterative fusion method to obtain the fused influencing factor historical data, including: The features of each influencing quantity are extracted through a multi-layer neural network, and the features of multiple influencing quantities are fused through feature concatenation and gating functions; The similarity between the fusion feature and the sample is calculated using the exponential kernel function; The multi-layer neural network is iterated according to a similarity-based objective function until the similarity is maximized, and the fusion feature corresponding to the maximum similarity is used as the fused influencing factor historical data.

3. The method for optimizing the metering accuracy of an off-board charger for an electric vehicle according to claim 2, characterized in that: The calculation method of the similarity is expressed as: , in, represents the i-th influence quantity, represents the features after the fusion of the j-th layer of neural network; D(.) represents the distance; Represents the width of the exponential kernel function; It represents the similarity between the features of the i-th influence quantity and the features after fusion of the j-th layer of neural network.

4. The method for optimizing the metering accuracy of an off-board charger for an electric vehicle according to claim 3 is characterized in that: The objective function is expressed as: , in, L Represents the objective function.

5. The method for optimizing the metering accuracy of an off-board charger for an electric vehicle according to claim 1, characterized in that: The multi-factor compensation model for the metering unit of the off-board charger of the electric vehicle comprises: Determine the error and error ratio of the DC current based on the actual value and the measured value of the current, and determine the error and error ratio of the DC voltage based on the actual value and the measured value of the voltage; Based on the DC current error under the influence of multiple factors, the DC current error and the error ratio, a DC current compensation coefficient is defined; Based on the DC voltage error under the influence of multiple factors, the DC voltage error and the error ratio, a DC voltage compensation coefficient is defined.

6. The method for optimizing the metering accuracy of an off-board charger for an electric vehicle according to claim 1, characterized in that: The DC current and voltage error evaluation model is constructed by using a radial basis function neural network method based on historical data of influencing factors, DC current and voltage errors, and includes: Based on the historical data of influencing factors, the DC current error is evaluated by radial basis function neural network method; Based on the historical data of influencing factors, the DC voltage error is evaluated by radial basis function neural network method.

7. A device for optimizing the metering accuracy of an off-board charger for an electric vehicle according to claim 1, characterized in that: include: An acquisition module is used to acquire historical data of multiple influencing quantities in the operating conditions of the off-board charger of the electric vehicle, wherein the multiple influencing quantities include temperature, humidity and electromagnetic field strength; A fusion module is used to fuse the multiple influencing factors through a multi-modal iterative fusion method to obtain fused historical data of influencing factors; and to construct a multi-factor compensation model for the metering unit of the off-board charger of the electric vehicle; A building module for building a DC current and voltage error evaluation model through a radial basis function neural network method based on historical data of influencing factors, DC current and voltage errors; The calculation module is used to obtain multiple real-time impact quantities of the off-board charger of the electric vehicle, substitute the multiple impact quantities into the multi-factor compensation model of the metering unit and calculate to obtain the current error compensation coefficient and the voltage error compensation coefficient.

8. The device for optimizing the metering accuracy of an off-board charger for an electric vehicle according to claim 7, characterized in that: The fusion module includes: A fusion unit is used to extract the features of each influencing quantity through a multi-layer neural network, and to fuse the features of multiple influencing quantities through feature concatenation and gating functions; A calculation unit, used to calculate the similarity between the fusion feature and the sample through an exponential kernel function; The iterative unit is used to iterate the multi-layer neural network according to the objective function based on the similarity until the similarity is maximized, and use the fused features corresponding to the maximum similarity as the fused influencing factor historical data.

9. An electronic device, comprising: one or more processors; A storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, enables the one or more processors to implement the method for optimizing the metering accuracy of an off-board charger for an electric vehicle as described in any one of claims 1 to 6.

10. A computer readable medium having a computer program stored thereon, wherein: When the computer program is executed by a processor, the method for optimizing the metering accuracy of an off-board charger for an electric vehicle as claimed in any one of claims 1 to 6 is implemented.

Citation Information

Cited By

  • Multi-modal deep learning network-based overlay mark asymmetry compensation method and apparatus

    WO2026130578A1