A wireless charging coil parameter optimization method, system and storage medium
By transforming the wireless charging coil parameters into nonlinear equations and optimizing them using deep learning, the problems of transmission loss and device reliability in the wireless charging process are solved, achieving the optimization of an efficient and low-cost wireless charging system for electric vehicles.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- 深圳市旋依科技有限公司
- Filing Date
- 2022-12-26
- Publication Date
- 2026-05-29
Smart Images
Figure CN115983118B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of wireless charging, and more specifically to a method, system, and storage medium for optimizing wireless charging coil parameters. Background Technology
[0002] Electric vehicles (EVs) offer advantages over traditional gasoline-powered vehicles, including lower noise, zero emissions, and energy efficiency. In recent years, the development of EVs has received strong support from governments of major automobile-producing countries. Currently, the primary charging method for EVs is traditional wired charging. This method inevitably incurs transmission losses during the energy transfer process, and factors such as aging wiring and point discharge significantly reduce the reliability and safety of the power supply.
[0003] Wireless charging technology boasts advantages such as high transmission power, high transmission efficiency, long transmission distance, low directionality requirements, and minimal impact on living organisms. Wireless charging technology for electric vehicles is a power battery charging technology that achieves wireless power transmission through electromagnetic coupling between transmitting and receiving coils. It mainly includes two types: static and dynamic wireless charging. Compared to plug-in charging, wireless charging enables long-distance, contactless power transmission, eliminates the need for manual plugging and unplugging, and offers a high degree of automation.
[0004] Therefore, how to achieve efficient wireless charging and optimize wireless charging parameters is a research topic that urgently needs to be studied by those skilled in the art. Summary of the Invention
[0005] In view of this, the present invention provides a method, system, and storage medium for optimizing wireless charging coil parameters. The method equates coil parameter optimization to solving a nonlinear equation, which is then transformed into a deep learning nonlinear fitting problem. By training the model offline and importing the trained model into a microcontroller, simultaneous online identification of load and mutual inductance can be achieved. This results in fast identification speed, high accuracy, and facilitates real-time system control. Furthermore, the method is low-cost and easy to implement.
[0006] To achieve the above objectives, the present invention adopts the following technical solution:
[0007] One method for optimizing wireless charging coil parameters is disclosed, including the following steps:
[0008] The design parameters of the wireless charging system are obtained based on the application scenarios of wireless charging, and the design parameters of the wireless charging system are divided into test set and training set;
[0009] Input the training set into the TensorFlow network framework and set the training parameters for the TensorFlow network framework;
[0010] The Tensorflow network framework outputs the first weight and the first bias value;
[0011] Compare the first weight and the first bias value with the preset weight and the preset bias value;
[0012] If the preset weights and preset biases are not reached, the Tensorflow network framework will continue to be trained iteratively.
[0013] When the preset weights and preset bias values are reached, the training of the Tensorflow network framework is completed, and the wireless charging coil parameter optimization model is obtained.
[0014] The design parameters of the wireless charging system under test are input into the wireless charging coil parameter optimization model to obtain the optimized parameters.
[0015] Optionally, training parameters include the number of training iterations and the learning rate, where the learning rate is calculated using the following formula:
[0016]
[0017] Where α0 represents the initial learning rate, α t Let t represent the learning rate at the t-th training session, β represent the learning rate decay coefficient, and T represent the number of training sessions required to complete one learning rate decay.
[0018] Optionally, it also includes preprocessing the design parameters of the wireless charging system, specifically: first, using a Gaussian function to remove noise points, and then using a Sigmoid function to standardize the design parameters.
[0019] Optionally, the Tensorflow network framework includes an input layer, a hidden layer, and an output layer, which are connected sequentially. The input layer includes multiple input ports for inputting design parameters; the hidden layer is used to learn and process the design parameters; and the output layer is used to output weights and biases.
[0020] Optionally, the design parameters of the wireless charging system include: the coil structure type of the transmitting coil, the parameters of the transmitting coil structure, the structure type of the receiving coil, the coil structure parameters of the receiving coil, the load parameters, the transmission distance, and the offset distance.
[0021] Optionally, it also includes testing the wireless charging coil parameter optimization model using a test set, and selecting a compensation function to compensate the wireless charging coil parameter optimization model based on the test results.
[0022] On the other hand, a wireless charging coil parameter optimization system is also disclosed, including:
[0023] Obtaining Design Parameters: This is used to obtain the design parameters of the wireless charging system based on the application scenario of wireless charging, and the design parameters of the wireless charging system are divided into a test set and a training set.
[0024] Training parameter setting module: Used to input the training set into the TensorFlow network framework and set the training parameters for the TensorFlow network framework;
[0025] The Tensorflow network framework training module is used to output the first weight and the first bias value based on the Tensorflow network framework; compare the first weight and the first bias value with the preset weight and the preset bias value; if the preset weight and the preset bias value are not reached, the Tensorflow network framework continues to be trained iteratively.
[0026] Wireless charging coil parameter optimization model acquisition module: When the preset weights and preset bias values are reached, the training of the Tensorflow network framework is completed, and the wireless charging coil parameter optimization model is obtained.
[0027] The design parameter optimization module is used to input the design parameters of the wireless charging system under test into the wireless charging coil parameter optimization model to obtain the optimized parameters.
[0028] Finally, a computer storage medium is disclosed, on which a computer program is stored, which, when executed by a processor, implements the steps of any one of the wireless charging coil parameter optimization methods.
[0029] As can be seen from the above technical solutions, compared with the prior art, the present invention discloses a method, system and storage medium for optimizing wireless charging coil parameters, which greatly improves the accuracy and speed of charging, facilitates real-time control of the system, and has low cost, is easy to implement, and is conducive to engineering promotion and application. Attached Figure Description
[0030] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.
[0031] Figure 1 This is a schematic diagram of the process of the present invention;
[0032] Figure 2 This is a schematic diagram of the structure of the present invention. Detailed Implementation
[0033] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0034] This invention discloses a method for optimizing wireless charging coil parameters, comprising the following steps:
[0035] S1: Obtain the design parameters of the wireless charging system based on the application scenarios of wireless charging, and divide the design parameters of the wireless charging system into a test set and a training set;
[0036] S2: Input the training set into the Tensorflow network framework and set the training parameters for the Tensorflow network framework;
[0037] S3: The Tensorflow network framework outputs the first weight and the first bias value;
[0038] S4: Compare the first weight and the first bias value with the preset weight and the preset bias value;
[0039] S5: If the preset weights and preset biases are not reached, continue iterative training of the Tensorflow network framework;
[0040] S6: When the preset weights and preset bias values are reached, the training of the Tensorflow network framework is completed, and the wireless charging coil parameter optimization model is obtained.
[0041] S7: Input the design parameters of the wireless charging system under test into the wireless charging coil parameter optimization model to obtain the optimized parameters.
[0042] In the gradient descent optimization algorithm, the learning rate needs to be specified before model training. A learning rate that is too small will slow down the training speed of the neural network model and increase training time; while a learning rate that is too large may cause the model to fail to converge. Therefore, setting an appropriate learning rate is crucial for model training. There are two main methods for setting the learning rate: constant learning rate and learning rate decay. In S2, the training parameters include the number of training iterations and the learning rate. This embodiment uses the learning rate decay method, where the formula for calculating the learning rate is as follows:
[0043]
[0044] Where α0 represents the initial learning rate, α tLet t represent the learning rate at the t-th training session, β represent the learning rate decay coefficient, and T represent the number of training sessions required to complete one learning rate decay.
[0045] This embodiment also includes preprocessing the design parameters of the wireless charging system, specifically: first, using a Gaussian function to remove noise points, and then using a Sigmoid function to standardize the design parameters.
[0046] In S2 or S3, the Tensorflow network framework includes an input layer, a hidden layer, and an output layer, which are connected sequentially. The input layer includes multiple input ports for inputting design parameters; the hidden layer is used to learn and process the design parameters; and the output layer is used to output weights and biases.
[0047] The training process of a neural network model is essentially the process of automatically adjusting the parameters within the network. Therefore, the initial values of these parameters determine the direction of the model's training. The weights are the most crucial parameters in a neural network model, and their initialization method significantly impacts both the training effectiveness and speed. Choosing an appropriate weight initialization method not only helps resolve gradient vanishing or exploding problems but also improves training speed and performance, and in some network structures, can even enhance model accuracy. Therefore, in this embodiment, the Sigmoid activation function from the TensorFlow framework is used for the nonlinear activation function of the hidden layers in the nonlinear model, and the Xavier initialization method is adopted for the model weights.
[0048] In S1, the design parameters of the wireless charging system include: the coil structure type of the transmitting coil, the parameters of the transmitting coil structure, the structure type of the receiving coil, the parameters of the receiving coil structure, the load parameters, the transmission distance, and the offset distance. The research on the coupling coil mainly focuses on the selection and optimization of the coil and ferrite materials, shapes, and dimensions.
[0049] This embodiment also includes testing the wireless charging coil parameter optimization model using a test set, and selecting a compensation function to compensate the wireless charging coil parameter optimization model based on the test results.
[0050] and Figure 1 Corresponding to the method shown, the present invention also discloses a wireless charging coil parameter optimization system for optimizing wireless charging coil parameters. Figure 1 The implementation of the method, specifically its structure, is as follows: Figure 2 As shown, it includes:
[0051] Obtaining Design Parameters: This is used to obtain the design parameters of the wireless charging system based on the application scenario of wireless charging, and the design parameters of the wireless charging system are divided into a test set and a training set.
[0052] Training parameter setting module: Used to input the training set into the TensorFlow network framework and set the training parameters for the TensorFlow network framework;
[0053] The Tensorflow network framework training module is used to output the first weight and the first bias value based on the Tensorflow network framework; compare the first weight and the first bias value with the preset weight and the preset bias value; if the preset weight and the preset bias value are not reached, the Tensorflow network framework continues to be trained iteratively.
[0054] Wireless charging coil parameter optimization model acquisition module: When the preset weights and preset bias values are reached, the training of the Tensorflow network framework is completed, and the wireless charging coil parameter optimization model is obtained.
[0055] The design parameter optimization module is used to input the design parameters of the wireless charging system under test into the wireless charging coil parameter optimization model to obtain the optimized parameters.
[0056] Finally, a computer storage medium is disclosed, on which a computer program is stored, and when the computer program is executed by a processor, the steps of a wireless charging coil parameter optimization method are implemented, which can be any one of the steps.
[0057] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.
[0058] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A method for optimizing wireless charging coil parameters, characterized in that, Includes the following steps: The design parameters of the wireless charging system are obtained based on the application scenarios of wireless charging, and the design parameters of the wireless charging system are divided into test set and training set; Input the training set into the TensorFlow network framework and set the training parameters for the TensorFlow network framework; The Tensorflow network framework outputs the first weight and the first bias value; Compare the first weight and the first bias value with the preset weight and the preset bias value; If the preset weights and preset biases are not reached, the Tensorflow network framework will continue to be trained iteratively. When the preset weights and preset bias values are reached, the training of the Tensorflow network framework is completed, and the wireless charging coil parameter optimization model is obtained. The design parameters of the wireless charging system under test are input into the wireless charging coil parameter optimization model to obtain the optimized parameters. The design parameters of the wireless charging system include: the coil structure type of the transmitting coil, the parameters of the transmitting coil structure, the structure type of the receiving coil, the coil structure parameters of the receiving coil, the load parameters, the transmission distance, and the offset distance. Training parameters include the number of training iterations and the learning rate, where the learning rate is calculated using the following formula: ； in, This represents the initial learning rate. This represents the learning rate during the t-th training iteration. This represents the learning rate decay coefficient, where T represents the number of training iterations required to complete one learning rate decay.
2. The method for optimizing wireless charging coil parameters according to claim 1, characterized in that, It also includes preprocessing the design parameters of the wireless charging system, specifically: first, using the Gaussian function to remove noise points, and then using the Sigmoid function to standardize the design parameters.
3. The method for optimizing wireless charging coil parameters according to claim 1, characterized in that, The TensorFlow network framework consists of an input layer, hidden layers, and an output layer, which are connected sequentially. The input layer includes multiple input ports for inputting design parameters; the hidden layers are used to learn and process the design parameters; and the output layer is used to output weights and biases.
4. The method for optimizing wireless charging coil parameters according to claim 1, characterized in that, It also includes testing the wireless charging coil parameter optimization model using a test set, and selecting a compensation function to compensate the wireless charging coil parameter optimization model based on the test results.
5. A wireless charging coil parameter optimization system, characterized in that, include: Obtaining Design Parameters: This is used to obtain the design parameters of the wireless charging system based on the application scenario of wireless charging, and the design parameters of the wireless charging system are divided into a test set and a training set. Training parameter setting module: Used to input the training set into the TensorFlow network framework and set the training parameters for the TensorFlow network framework; Tensorflow network framework training module: used to output the first weight and the first bias value based on the Tensorflow network framework; compare the first weight and the first bias value with the preset weight and the preset bias value; If the preset weights and preset biases are not reached, the Tensorflow network framework will continue to be trained iteratively. Wireless charging coil parameter optimization model acquisition module: When the preset weights and preset bias values are reached, the training of the Tensorflow network framework is completed, and the wireless charging coil parameter optimization model is obtained. The design parameter optimization module is used to input the design parameters of the wireless charging system under test into the wireless charging coil parameter optimization model to obtain the optimized parameters. The design parameters of the wireless charging system include: the coil structure type of the transmitting coil, the parameters of the transmitting coil structure, the structure type of the receiving coil, the coil structure parameters of the receiving coil, the load parameters, the transmission distance, and the offset distance. Training parameters include the number of training iterations and the learning rate, where the learning rate is calculated using the following formula: ； in, This represents the initial learning rate. This represents the learning rate during the t-th training iteration. This represents the learning rate decay coefficient, where T represents the number of training iterations required to complete one learning rate decay.
6. A computer storage medium, characterized in that, The computer storage medium stores a computer program, which, when executed by a processor, implements the steps of a wireless charging coil parameter optimization method as described in any one of claims 1-4.