Wireless power transmission adaptive frequency tracking method and system based on deep learning

Abstract

The invention provides a wireless power transmission adaptive frequency tracking method and system based on deep learning, and relates to the technical field of wireless transmission, and the method comprises the steps: determining a transmitting end and a plurality of receiving ends; receiving a plurality of pieces of power demand information; performing resonant frequency dynamic matching analysis under the influence of load fluctuation to generate a dynamic emission parameter sequence; establishing a coupling effect network, optimizing the dynamic emission parameter sequence, and generating a target dynamic emission parameter sequence; and performing electric energy transmission, and analyzing the impedance states of the plurality of receiving ends in real time to perform stable tracking of the resonant frequency. According to the invention, the technical problem in the prior art that fine-grained adjustment is difficult to carry out according to real-time load change due to uncertainty, volatility and frequency coupling of a plurality of receiving end loads can be solved, and high-efficiency frequency matching and energy transmission under dynamic load and frequency fluctuation are realized. The technical effect of improving the stability and the high efficiency of wireless power transmission in the multi-device sharing environment is achieved.

Description

Technical Field

[0001] The present application relates to the field of radio transmission technology, and in particular to a method and system for adaptive frequency tracking of wireless power transmission based on deep learning. Background Art

[0002] Wireless power transmission (WPT) technology is widely used in electric transportation, medical devices, the Internet of Things, and other fields. WPT transmits electrical energy to a receiver via a radio magnetic field, enabling contactless power supply. Currently, most WPT systems rely on fixed frequencies or simple adaptive adjustment strategies to match the resonant frequencies of the transmitter and receiver.

[0003] However, in practical applications, fixed-frequency transmission schemes often fail to achieve efficient and stable power transmission due to the uncertainty and volatility of the receiving-end load. Load variations can cause the receiving-end resonant frequency to drift, reducing the system's energy transmission efficiency and potentially even causing resonant mismatch. This is particularly true when multiple devices are sharing wireless power transmission. The coupling effects between multiple receiving ends can cause resonant frequency interference, impacting the overall system transmission performance.

[0004] In summary, the prior art has a technical problem in that it is difficult to perform fine-grained adjustments based on real-time load changes due to the uncertainty and volatility of loads at multiple receiving ends and frequency coupling. Summary of the Invention

[0005] The purpose of this application is to provide a deep learning-based adaptive frequency tracking method and system for wireless power transmission, so as to solve the technical problem in the prior art that it is difficult to make fine-grained adjustments according to real-time load changes due to the uncertainty and volatility of multiple receiving end loads and frequency coupling.

[0006] In view of the above problems, the present application provides a method and system for adaptive frequency tracking of wireless power transmission based on deep learning.

[0007] In the first aspect, the present application provides a deep learning-based adaptive frequency tracking method for wireless power transmission, which is implemented by a deep learning-based adaptive frequency tracking system for wireless power transmission, wherein the deep learning-based adaptive frequency tracking method for wireless power transmission includes: collecting and analyzing the real-time transmission scenario of a wireless power transmission platform shared by multiple devices to determine the transmitting end and multiple receiving ends; receiving multiple power demand information of the multiple receiving ends; calling a multi-frequency resonance analysis model to perform a dynamic matching analysis of the resonant frequency under the influence of load fluctuations on the multiple power demand information, and generating a dynamic transmission parameter sequence of the transmitting end; performing a coupling effect analysis of the resonant frequency on the multiple receiving ends, establishing a coupling effect network, optimizing the dynamic transmission parameter sequence, and generating a target dynamic transmission parameter sequence; controlling the wireless power transmission platform shared by multiple devices to transmit power with the target dynamic transmission parameter sequence, and at the same time analyzing the impedance state of the multiple receiving ends in real time to perform stable tracking of the resonant frequency.

[0008] In the second aspect, the present application also provides a deep learning-based wireless power transmission adaptive frequency tracking system for executing the deep learning-based wireless power transmission adaptive frequency tracking method as described in the first aspect, wherein the deep learning-based wireless power transmission adaptive frequency tracking system includes: a scene analysis module, which is used to collect and analyze the real-time transmission scene of a wireless power transmission platform shared by multiple devices, and determine the transmitting end and multiple receiving ends; a demand receiving module, which is used to receive multiple power demand information of the multiple receiving ends; a frequency matching module, which is used to call a multi-frequency resonance analysis model to perform dynamic matching analysis of the resonant frequency under the influence of load fluctuations on the multiple power demand information, and generate a dynamic transmission parameter sequence of the transmitting end; a coupling analysis module, which is used to perform coupling effect analysis of the resonant frequency on the multiple receiving ends, establish a coupling effect network, optimize the dynamic transmission parameter sequence, and generate a target dynamic transmission parameter sequence; a transmission module, which is used to control the wireless power transmission platform shared by multiple devices to transmit power with the target dynamic transmission parameter sequence, and at the same time analyze the impedance state of the multiple receiving ends in real time to perform stable tracking of the resonant frequency.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages: The real-time transmission scenario of a multi-device shared wireless power transmission platform is collected and analyzed to determine the transmitter and multiple receivers; multiple power demand information of the multiple receivers is received; a multi-frequency resonance analysis model is called to perform dynamic matching analysis of the resonant frequency under the influence of load fluctuations on the multiple power demand information to generate a dynamic transmission parameter sequence of the transmitter; a coupling effect analysis of the resonant frequency of the multiple receivers is performed, a coupling effect network is established, the dynamic transmission parameter sequence is optimized, and a target dynamic transmission parameter sequence is generated; the target dynamic transmission parameter sequence is used to control the multi-device shared wireless power transmission platform to transmit electric energy, and at the same time, the impedance state of the multiple receivers is analyzed in real time to track the resonant frequency stably. By collecting the real-time transmission scenario of a multi-device shared wireless power transmission platform and analyzing it, the power demand information of the receiver is combined with the multi-frequency resonance analysis model and coupling effect analysis to optimize the dynamic transmission parameter sequence, and then perform real-time impedance state analysis and resonant frequency stably tracking, efficient frequency matching and energy transmission under dynamic load and frequency fluctuations are achieved, achieving the technical effect of improving the stability and efficiency of wireless power transmission in a multi-device shared environment.

[0010] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, which can be implemented in accordance with the contents of the description, and to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically listed below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in this application or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely exemplary, and a person of ordinary skill in the art can obtain other drawings based on the provided drawings without creative work.

[0012] Figure 1 This is a flowchart of the deep learning-based wireless power transmission adaptive frequency tracking method of this application.

[0013] Figure 2 This is a structural diagram of the deep learning-based wireless power transmission adaptive frequency tracking system of this application.

[0014] Description of the accompanying drawings: scene analysis module 11, demand receiving module 12, frequency matching module 13, coupling analysis module 14, transmission module 15. DETAILED DESCRIPTION

[0015] This application solves the technical problem in the prior art of difficulty in making fine-grained adjustments based on real-time load changes due to the uncertainty and volatility of multiple receiving-end loads and frequency coupling by providing a method and system for adaptive frequency tracking of wireless power transmission based on deep learning. By collecting and analyzing the real-time transmission scenarios of a wireless power transmission platform shared by multiple devices, the power demand information of the receiving end is optimized, combined with a multi-frequency resonance analysis model and coupling effect analysis, and then performing real-time impedance state analysis and stable resonant frequency tracking, efficient frequency matching and energy transmission under dynamic loads and frequency fluctuations are achieved, achieving the technical effect of improving the stability and efficiency of wireless power transmission in a multi-device shared environment.

[0016] Below, the technical solutions in this application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of this application, rather than all the embodiments of this application. It should be understood that this application is not limited to the example embodiments described herein. Based on the embodiments of this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application. It should also be noted that, for the convenience of description, only the parts related to this application, rather than all of them, are shown in the accompanying drawings.

[0017] For example, see the attached Figure 1 The present application provides a method for adaptive frequency tracking of wireless power transmission based on deep learning, wherein the method is applied to a system for adaptive frequency tracking of wireless power transmission based on deep learning, and the method comprises the following steps: Step 1: Collect and analyze real-time transmission scenarios of multiple devices sharing the wireless power transmission platform to determine the transmitter and multiple receivers.

[0018] Specifically, a multi-device shared wireless power transmission platform usually consists of a transmitter and a receiver. The transmitter is responsible for transmitting electrical energy to the receiver via radio waves. In order to achieve efficient and stable power transmission, it is first necessary to collect real-time transmission scenarios and parse these data to determine the specific status of each device. In this step, the collection of real-time transmission scenarios usually involves the use of existing sensors or radio wave detection equipment to obtain the working status, power requirements, and relative position of each device with other devices. These sensors may include but are not limited to power meters, signal receivers, and transmission bandwidth monitoring modules. The data obtained through these devices can help parse out a transmitter and multiple receivers that need to transmit electrical energy, providing accurate basic data for subsequent dynamic matching and coupling effect analysis.

[0019] Step 2: Receive multiple power requirement information of the multiple receiving ends.

[0020] Specifically, the power demand data of each receiving end is received through sensors or smart devices. Usually, each receiving end is equipped with a power sensor or power management module that can monitor the power value required by each receiving end in real time. The power demand information usually includes the charging power, voltage, current, etc. required by each receiving end. For example, suppose there are two receiving ends in a platform: receiving ends A and B. Receiver A may be a high-power device (such as a charging device for an electric tricycle) with a relatively large power demand, which may reach 100W, while receiving end B may be a low-power device (such as a low-power sensor) with a power demand of only 20W. This data can be transmitted to the central processing unit of the wireless power transmission platform shared by multiple devices through existing wireless communication protocols (such as Wi-Fi, Zigbee, Bluetooth, etc.) as power demand information, thereby providing key input for subsequent resonant frequency analysis and dynamic transmission parameter optimization.

[0021] Step three: calling a multi-frequency resonance analysis model to perform dynamic matching analysis of the resonant frequencies under the influence of load fluctuations on the multiple power demand information, and generating a dynamic transmission parameter sequence of the transmitting end.

[0022] Specifically, in wireless power transmission systems, dynamic matching analysis of the resonant frequency under the influence of load fluctuations based on the power demand information of multiple receiving ends is a key step in ensuring the efficient operation of the system. Since the power demand of different receiving ends may fluctuate during use, traditional fixed-frequency transmission methods are often unable to cope with such dynamic changes. Therefore, it is necessary to use a multi-frequency resonance analysis model to analyze the power demand information of each receiving end and adjust the transmission parameters of the transmitting end in real time based on the influence of load fluctuations to generate a dynamic transmission parameter sequence. The core of this process is to combine the power demand of each receiving end with the load fluctuation factor through the multi-frequency resonance analysis model to calculate highly adaptable resonant frequency and power transmission parameters to ensure that each receiving end can obtain optimal power transmission under different power demands, thereby achieving more efficient and stable wireless power transmission.

[0023] Step 4: Analyze the coupling effect of the resonant frequencies of the multiple receiving ends, establish a coupling effect network, optimize the dynamic transmission parameter sequence, and generate a target dynamic transmission parameter sequence.

[0024] Specifically, in wireless power transmission, coupling effects may exist between the resonant frequencies of multiple receivers, affecting the overall frequency stability and energy transmission efficiency of the system. Therefore, to achieve efficient energy distribution, it is necessary to analyze the coupling effects of the resonant frequencies of these receivers and, based on the analysis results, optimize the dynamic transmission parameters to generate a target dynamic transmission parameter sequence.

[0025] First, the purpose of coupling effect analysis is to understand how the resonant frequencies of multiple receivers interact when operating. In a multi-device shared wireless power transmission platform, the resonant frequencies of each receiver are not completely independent. Their frequency responses may interact due to factors such as electromagnetic interference and physical structural similarities. For example, if the resonant frequencies of receivers A and B are close, they may interfere with each other during transmission, resulting in reduced energy transfer efficiency and even energy loss or stability issues. To analyze these coupling effects, a coupling effect network must be established. This is a graph model in which each node represents a receiver and each edge represents the coupling effect relationship between receivers. This network allows the system to quantitatively analyze the degree of frequency coupling between receivers and the impact of this coupling. Once the coupling effect network is established, dynamic transmission parameters are optimized. The goal is to reduce negative coupling effects between receivers and ensure that each receiver can stably receive power within its resonant frequency range without interference from other receivers. The optimization process typically involves the following aspects: Frequency adjustment: By precisely adjusting the transmit frequency, mutual interference between the resonant frequencies of the receivers can be avoided. For example, if the frequency coupling effect between receivers A and B is strong, the frequency of the transmitter can be appropriately adjusted to increase the difference in the resonant frequencies of the two receivers, thereby reducing interference. Power allocation: Based on the results of the coupling effect analysis, the power allocation can be adjusted to ensure that each receiver obtains sufficient power while avoiding unnecessary interference caused by excessive power output. During the optimization process, existing optimization algorithms such as particle swarm optimization (PSO) and genetic algorithm (GA) may be used. By simulating the adjustment schemes of different transmit parameters, the optimal dynamic transmit parameter sequence is selected to generate a target dynamic transmit parameter sequence. This sequence represents the transmit power and frequency settings that can achieve optimal energy transmission under the resonant frequency coupling effect of multiple receivers. This target sequence not only ensures the stable charging needs of each receiver, but also improves the overall efficiency of the system and reduces losses in power transmission. For example, in the experiment, it is assumed that the resonant frequencies of receivers A and B are relatively close, and the coupling effect between them is strong, resulting in their power requirements not being fully met. After discovering this problem through coupling effect analysis, the transmission frequency is adjusted from the original 250kHz to 275kHz, and the power output is increased to ensure that the two receivers can receive power stably. At the same time, the power allocation of other receivers is also optimized. The final target dynamic transmission parameter sequence will include the adjusted transmission frequency and power value, ensuring that all receivers can transmit power under stable conditions. This process not only helps to reduce interference, but also improves the overall performance of the wireless power transmission system, especially in a multi-device shared environment, ensuring that each receiver can obtain stable and efficient power transmission.

[0026] Step 5: Control the multi-device shared wireless power transmission platform to transmit power using the target dynamic transmission parameter sequence, and simultaneously analyze the impedance states of the multiple receiving ends in real time to track the stable resonant frequency.

[0027] Specifically, the target dynamic transmission parameter sequence includes optimized transmission frequency and power, which can achieve balanced and stable energy distribution in a multi-receiver transmission environment. First, the target dynamic transmission parameter sequence is used to control the wireless power transmission platform shared by multiple devices. During the power transmission process, the impedance state of multiple receivers is also analyzed in real time. Impedance changes are an important factor affecting the resonant frequency in wireless power transmission. This is because impedance changes directly affect the energy exchange efficiency between the receiver and the transmitter. Changes in the load, environment, and operating state of the receiver itself can all cause impedance changes, which will affect its resonant frequency, potentially leading to a decrease in energy transmission efficiency or even unstable transmission. For example, if the load on the receiver changes, causing its impedance to fluctuate, if the transmitter parameters are not adjusted in time, the resonant frequency of the receiver may shift, affecting the energy transmission of other receivers. To address this situation, the impedance state of multiple receivers is monitored in real time, and their current impedance data is obtained through impedance sensors. This data is then input into an impedance optimization network for analysis and feedback.

[0028] The impedance optimization network calculates the current impedance trend of each receiver and adjusts the transmitter's power and frequency accordingly to stabilize the resonant frequency, ensuring that each receiver receives power at the optimal frequency. This dynamic impedance state tracking and frequency adjustment mechanism significantly improves stability and reliability under dynamic load conditions, ensuring that multiple devices sharing the wireless power transmission platform can efficiently and stably transmit power under different operating conditions, maximizing power transmission efficiency and avoiding instability during transmission.

[0029] Furthermore, step three of this application includes: Based on the multiple power demand information, historical matching resonant frequencies of the multiple receiving ends are retrieved to establish multiple frequency matching spaces; the resonant frequency impact analysis under the influence of load fluctuations is performed on the multiple frequency matching spaces to establish multiple optimized frequency matching spaces; and a transmission parameter control particle is randomly selected in the multiple optimized frequency matching spaces to perform power distribution balance optimization for the multiple receiving ends, thereby generating a dynamic transmission parameter sequence for the transmitting end, wherein the dynamic transmission parameters include transmission power and frequency.

[0030] Specifically, when performing adaptive frequency tracking of wireless power transmission, it is first necessary to retrieve historical matching resonant frequencies based on the power demand information received from multiple receiving ends. The goal is to understand the frequency requirements of each receiving end under different loads by analyzing historical data, thereby establishing multiple frequency matching spaces. The frequency matching space is established based on the relationship between the historical power requirements and resonant frequencies of each receiving end, reflecting the frequency characteristics of different receiving ends under various load conditions. In actual operation, historical matching resonant frequency retrieval relies on a large-scale data set, which contains the frequency response data of each receiving end under multiple load states. By retrieving this data, a frequency matching space can be established for each receiving end. For example, assuming that the frequency of receiving end A is 250kHz at low load and 300kHz at high load, a matching space containing multiple frequency points is established for receiving end A to adapt to the power requirements under different loads.

[0031] Next, the impact of load fluctuations on the resonant frequency of multiple frequency matching spaces is analyzed, generating multiple optimized frequency matching spaces. Load fluctuations can be caused by many factors, such as changes in the receiver's internal load and external environmental interference. To ensure effective response to these fluctuations, the resonant frequency changes under different load fluctuation scenarios are simulated and their impact on each receiver's frequency matching space is analyzed. Based on the fluctuation information, the frequency matching space is adjusted to ensure frequency stability. The optimized frequency matching space reflects the impact of load fluctuations on frequency. To achieve optimal power transmission efficiency, a transmit parameter control particle is randomly selected from each of the multiple optimized frequency matching spaces to perform power allocation balance optimization. This process is similar to the particle swarm optimization (PSO) algorithm, in which the control particle represents different transmit power and frequency combinations. The optimal power allocation solution is determined by simulating and analyzing the performance of each control particle. By evaluating the match between the power requirements of multiple receivers and the actual transmission efficiency, the most suitable control particle is ultimately selected to generate a dynamic transmit parameter sequence for the transmitter, which includes the transmit power and frequency that adapts to the power requirements of each receiver. The dynamic transmit parameter sequence is used in real time to control the transmitter's power output to ensure the efficiency and stability of wireless power transmission. This approach not only enables real-time power demand changes to be addressed, but also optimizes energy distribution in a multi-device shared environment, reducing energy waste and improving overall system performance. For example, in an experiment, assuming that the power demand of receiver a fluctuates between 100W and 120W, the demand of receiver b fluctuates less, maintaining at around 20W, while the demand of receiver c fluctuates more, perhaps between 30W and 60W. After historical matching frequency retrieval and load fluctuation analysis, it may be decided to adjust the transmit power to 115W and the frequency to 270kHz to meet the needs of receivers a and c while avoiding excessive power to receiver B. This optimized dynamic transmit parameter sequence will improve overall efficiency and ensure that each receiver has a stable power supply.

[0032] Furthermore, the present application further comprises the following steps: Extract a first frequency matching space corresponding to the first receiving end, and randomly extract a first frequency matching parameter in the first frequency matching space; use the first frequency matching parameter to predict the load fluctuation trend of the first receiving end during wireless power transmission to generate a first load fluctuation trend; activate the fluctuation optimization module to dynamically optimize the first frequency matching parameter according to the first load fluctuation trend, generate a first optimized frequency matching space, and add it to the multiple optimized frequency matching spaces.

[0033] Specifically, the steps of analyzing the resonant frequency impact under the influence of load fluctuations on the multiple frequency matching spaces and establishing multiple optimized frequency matching spaces are as follows: the first receiving end generally refers to any one of the multiple receiving ends, and the first frequency matching space refers to the frequency matching space corresponding to the first receiving end in the multiple frequency matching spaces. First, a frequency matching parameter is randomly extracted from the established first frequency matching space. These frequency matching parameters represent the optimal frequency at which the receiving end can effectively receive electrical energy under specific power demand and load conditions. Next, based on the selected first frequency matching parameter, the load fluctuation trend of the first receiving end is predicted to obtain a first load fluctuation trend. The first load fluctuation trend reflects the changing pattern of the power demand of the receiving end under different operating conditions, thereby predicting how the load fluctuation of the first receiving end will affect the resonant frequency over a period of time in the future. For example, if the power demand of the first receiving end fluctuates due to changes in the external ambient temperature, the fluctuation amplitude and frequency change trend over the next few seconds or minutes can be predicted based on historical data. Specifically, a relationship model between load fluctuation data and harmonic frequency can be constructed based on an existing intelligent learning model based on the historical fluctuation data of the receiving end. The first frequency matching parameter is then input into the relationship model to perform load fluctuation prediction and generate the first load fluctuation trend.

[0034] Once the load fluctuation trend is predicted, the fluctuation optimization module is activated to dynamically optimize the first frequency matching parameters based on the first load fluctuation trend. The fluctuation optimization module uses the first load fluctuation trend to adjust the frequency matching parameters to ensure that the first receiving end can continuously and stably receive power despite load fluctuations. The optimized frequency matching parameters may be adjusted based on the predicted fluctuations, for example, adjusting the frequency from 250 kHz to 275 kHz to better adapt to changes caused by load fluctuations. Specifically, the fluctuation optimization module can be constructed based on a mapping relationship between historical load fluctuation data and historical frequency adjustment data to identify frequency adjustment data (e.g., adjustment amplitude) corresponding to the first load fluctuation trend. The first frequency matching parameters are dynamically optimized and then constructed using the optimized frequency matching parameters. The first optimized frequency matching space is then added to the multiple optimized frequency matching spaces. After fluctuation optimization, a wider range of load fluctuations can be accommodated, improving the stability of wireless power transmission. This ensures that each receiving end can dynamically adjust the frequency based on load fluctuations, ensuring overall performance in a multi-receiving end shared environment. The resulting multiple optimized frequency matching spaces provide more possible frequency solutions, thereby achieving optimal power allocation and frequency tracking, and improving overall power transmission efficiency.

[0035] Furthermore, the present application further comprises the following steps: Randomly selecting a transmission parameter control particle in each of the multiple optimized frequency matching spaces to generate a first group of control particles; performing power allocation balance analysis on multiple receiving ends on the first group of control particles to generate a first balance visibility graph; continuing to generate a second group of control particles in the multiple optimized frequency matching spaces up to an Nth group of control particles, analyzing and generating second balance visibility graphs up to an Nth balance visibility graph; determining multiple charging priority identifiers of the multiple receiving ends based on the real-time transmission scenario; performing matching analysis on the multiple charging priority identifiers with the first balance visibility graph, the second balance visibility graph, up to the Nth balance visibility graph, and generating a group of control particles with the highest matching degree as the dynamic transmission parameter sequence.

[0036] Specifically, a transmission parameter control particle is randomly selected from each of the multiple optimized frequency matching spaces to optimize power distribution balance for multiple receiving ends. The specific process of generating a dynamic transmission parameter sequence for the transmitting end includes: first, randomly selecting an optimized frequency from each of the multiple optimized frequency matching spaces; then, based on a historical transmission database, filtering transmitter control parameters, including transmit frequency and power, that meet the selected multiple optimized frequencies to form a first set of control particles. A power distribution balance analysis is then performed on the selected first set of control particles across multiple receiving ends. The goal is to evaluate the power distribution effect of this set of control particles on different receiving ends. To achieve balance, the transmit power and frequency are adjusted based on the power requirements of the receiving ends to ensure that each receiving end receives the required power without overload or waste. This analysis generates a first visual balance graph, which is a chart showing power distribution balance and can intuitively reflect the power reception status of different receiving ends. For example, in the graph, the ratio of the power requirements to the actual allocated power of multiple receiving ends is represented by different colors or markers, thereby helping to identify whether there is uneven or biased power distribution.

[0037] After completing the balance analysis for the first set of control particles, the second, third, and Nth sets of control particles are generated in multiple optimized frequency matching spaces, where N is an integer greater than 1. Each set of control particles undergoes the same power allocation balance analysis and generates a corresponding balance visualization graph. As the number of control particles increases, more transmit power and frequency combinations can be evaluated. The balance visualization graph compares the performance of different solutions, gradually identifying the most optimal transmit parameters. Next, based on real-time transmission scenarios, charging priority identifiers are determined for multiple receivers. These identifiers reflect the urgency of each receiver's charging needs. For example, if receiver A's battery level is low, it might be marked as high priority, while receiver B might be marked as low priority. Real-time scenario data (such as the receiver's battery status and power demand fluctuations) is used to determine the charging priority of each receiver, or the administrator can directly set the priority.

[0038] Finally, the charging priority identifiers of multiple receivers are matched against the balance visualization graphs. This matching analysis evaluates the charging priority performance of each control particle group, and the group with the highest matching score is generated as the final dynamic transmit parameter sequence. This process ensures that the transmitter's power and frequency allocation are optimized based on the priority requirements of the real-time transmission scenario, achieving optimal power distribution and transmission efficiency. For example, in an experiment, receiver A has a high priority, receiver B has a medium priority, and receiver C has a low priority. The balance visualization graph of the first group of control particles is analyzed to ensure that the high-priority receiver A receives the required power first, while preventing the low-priority receiver C from excessively consuming system resources. Finally, the matching analysis determines the most appropriate transmit power and frequency combination, forming an optimized dynamic transmit parameter sequence to ensure efficient and stable power transmission in a multi-receiver environment. This process not only ensures balanced power distribution but also adjusts transmit parameters according to real-time scenario requirements to maximize overall performance.

[0039] Furthermore, step 4 of this application includes: Determine the anti-interference configuration between the multiple receiving ends; in combination with the anti-interference configuration, model the wireless power transmission between the transmitting end and the multiple receiving ends to generate a twin power transmission model; use the twin power transmission model to perform transmission tests on the multiple receiving ends, and analyze the resonant frequency coupling effects of the multiple receiving ends based on the test results to generate the coupling effect network, wherein the nodes of the coupling effect network represent the individual receiving ends, and the edges represent the coupling effect relationships.

[0040] Specifically, in a multi-device shared wireless power transmission system, anti-interference configuration is a key factor in ensuring stable system operation. This is especially true when multiple receivers are operating simultaneously. Each receiver may be subject to interference from other receivers, environmental factors, or signals from the transmitter. Therefore, designing a sound anti-interference strategy is crucial. Based on this, the current anti-interference configuration between each receiver is determined, including the installed shields and the receiver locations. Wireless power transmission is then modeled for the transmitter and multiple receivers using the anti-interference configuration. Through simulation or actual experiments, the electromagnetic coupling between the transmitter and each receiver is analyzed, as well as the performance of each receiver under various interference conditions. This modeling process is conventional and requires consideration of multiple variables, such as transmit power, frequency, receiver power requirements, distance, and environmental influences. Ultimately, this information is used to generate a twin power transmission model. The twin power transmission model is a digital, virtual model that closely resembles the actual system. This model allows for the simulation of wireless power transmission in a virtual environment, the impact of different configurations on transmission efficiency, and the prediction of the performance of different receivers under various conditions.

[0041] After generating the twin power transmission model, the model is used to conduct transmission tests on multiple receiving ends. By simulating different operating conditions and interference scenarios, the transmission performance data of each receiving end under different environments can be obtained, such as the variable speed data of load and resonant frequency, which are used as test results. On this basis, the resonant frequency coupling effect analysis of multiple receiving ends is performed. The goal of the coupling effect analysis is to identify the mutual influence between the receiving ends, especially the interaction in frequency. For example, the load change of one receiving end may affect the operating frequency of another receiving end, thereby reducing the overall energy transmission efficiency. Based on existing correlation analysis methods, such as gray correlation analysis, the resonant frequency coupling relationship between each receiving end is analyzed. Finally, the generated coupling effect network will represent each receiving end as a node, and the coupling relationship between the receiving ends is connected by edges. The coupling effect network can help to intuitively understand the interaction between the receiving ends, identify which receiving ends have stronger coupling effects, and which receiving ends are susceptible to interference.

[0042] By analyzing the coupling effect network, frequency and power allocation strategies can be further optimized to reduce the negative impact of the coupling effect on system performance and improve the stability and efficiency of wireless power transmission. For example, suppose during testing, the resonant frequency coupling effect between receivers A and B is strong, and both experience heating, causing frequency drift. In the generated coupling effect network, the edges between the nodes of these two receivers will show a strong coupling relationship. By analyzing this network, it is possible to decide whether to adjust the operating frequencies of the two receivers or optimize the power allocation between them to minimize the negative impact of the coupling effect, ensure overall performance, reduce interference between receivers, optimize the wireless power transmission process, and improve the efficiency and stability of power transmission in a multi-device shared environment.

[0043] Furthermore, step five of this application includes: Historical impedance change characteristics and corresponding optimization parameters of the transmitting end power and frequency are collected to train the impedance optimization network; multiple impedance change data of the multiple receiving ends are collected in real time through the impedance sensor; the multiple impedance change data are input into the impedance optimization network for feedback optimization, and the optimized power and frequency are used to stably track the resonant frequency.

[0044] Specifically, when multiple devices share wireless power transmission, impedance changes at the receiving end have a crucial impact on power transmission performance. This requires real-time tracking of the receiving end's impedance state and dynamic adjustment of the transmitting end's power and frequency based on these changes. To this end, historical impedance change characteristics and corresponding transmitter optimization parameters are collected, and this data is used to train an impedance optimization network based on existing machine learning models, such as neural network models. This network can provide feedback and optimize transmitter parameters based on real-time impedance data, thereby achieving stable tracking of the resonant frequency and ensuring the efficiency and stability of wireless power transmission.

[0045] First, historical impedance variation characteristics and their corresponding transmitter power and frequency optimization parameters are collected to provide data for training the impedance optimization network. This historical data includes the impedance variations of each receiver under different operating conditions, along with the corresponding transmit power and frequency settings. This data helps understand the relationship between impedance fluctuation trends and transmitter parameters under varying load and environmental conditions. For example, when the receiver load changes, the impedance may shift, causing the receiver's resonant frequency to drift. This historical data helps the impedance optimization network understand this variation pattern and provides a reference for real-time adjustment of transmit parameters. Based on this historical data, an impedance optimization network is trained. This network is a deep learning model that learns the relationship between impedance variation and transmitter power and frequency, enabling it to automatically predict how to adjust transmitter parameters for optimal power transfer given a given impedance variation. During training, the network continuously optimizes its parameters to minimize the error between the predicted power and frequency and the actual optimization requirements, enabling more precise parameter adjustments in subsequent operations.

[0046] When entering the real-time operation state, the impedance sensor will collect impedance change data from multiple receiving ends in real time. This data reflects the impedance state of each receiving end under the current load and working environment. For example, the impedance of the receiving end may fluctuate due to factors such as load changes, temperature fluctuations or electromagnetic interference. These real-time collected impedance change data will be input into the trained impedance optimization network for feedback optimization. The impedance optimization network calculates how to adjust the power and frequency of the transmitting end based on the real-time data to ensure that each receiving end can operate stably within its resonant frequency range.

[0047] For example, if the load at a certain receiving end fluctuates, causing its impedance to change significantly, the frequency and power adjustment parameters required by the receiving end can be calculated through the impedance optimization network. Through continuous feedback optimization, the power and frequency of the transmitting end can be adjusted to ensure that energy transmission is in a stable and efficient state, significantly improving the adaptability and transmission performance of the wireless power transmission platform shared by multiple devices in complex environments.

[0048] Furthermore, the present application further comprises the following steps: The multi-device shared wireless power transmission platform is connected to a parasitic monitoring module, and the parasitic parameters of the multi-device shared wireless power transmission platform are monitored by the parasitic monitoring module, wherein the parasitic parameters include at least leakage magnetic parameters and parasitic capacitance parameters; frequency offset impact analysis is performed based on the leakage magnetic parameters and the parasitic capacitance parameters to generate a frequency offset; and feedback control of power transmission is performed using the frequency offset.

[0049] Specifically, in a multi-device shared wireless power transmission platform, parasitic effects are a significant factor affecting stability and energy transmission efficiency. To monitor these parasitic effects in real time, a parasitic monitoring module is deployed. This module collects and analyzes parasitic parameters, analyzes the impact of frequency offset based on these parameters, and ultimately optimizes power transmission through a feedback control mechanism to improve system stability and transmission efficiency. First, the parasitic monitoring module monitors the parasitic parameters of the multi-device shared wireless power transmission platform, including magnetic flux leakage and parasitic capacitance. Magnetic flux leakage primarily refers to the magnetic flux leakage effect caused by incomplete electromagnetic field coupling during wireless power transmission. High magnetic flux leakage can lead to energy loss and electromagnetic interference (EMI) to surrounding electronic devices. Parasitic capacitance refers to the non-ideal capacitance generated by coupling between the coil, transmission path, and the environment. This parasitic capacitance can affect the resonant frequency, causing resonant matching offsets and, in turn, impacting power transmission efficiency. To ensure system stability, the parasitic monitoring module must measure these parasitic parameters in real time. For example, a high-precision magnetic field sensor can be used to configure the parasitic monitoring module.

[0050] Next, based on the leakage magnetic parameters and parasitic capacitance parameters collected by the parasitic monitoring module, a frequency offset impact analysis is performed to calculate the frequency offset. Ideally, the transmitting frequency should match the resonant frequency of the receiving end to achieve the most efficient energy transmission, but due to the influence of parasitic parameters, the actual resonant frequency may shift. For example, the design resonant frequency of the transmitting end is 250kHz, but due to the influence of parasitic capacitance, the measured actual operating frequency may shift to 245kHz or 255kHz. This offset will reduce the energy transmission efficiency and may even cause the receiving end to be unable to effectively receive electrical energy. Therefore, it is necessary to calculate the frequency offset, that is, the difference between the actual operating frequency and the theoretical resonant frequency. For example, if the measured frequency offset is +5kHz, it means that the transmitting end needs to make corresponding adjustments to compensate for this deviation. Specifically, historical leakage magnetic parameters, historical parasitic capacitance parameters and corresponding frequency offset data are collected as training data, and based on existing machine learning models, such as neural network models, frequency offset recognition models are trained. This is a common technical means used by those skilled in the art and will not be expanded here.

[0051] After calculating the frequency offset, a feedback control mechanism is used to dynamically adjust the frequency to ensure optimal wireless power transmission. The basic principle of feedback control is to utilize a closed-loop control system to make real-time corrections to the transmitter's operating frequency based on measured parasitic parameters and frequency offset. For example, if a change in parasitic capacitance causes the resonant frequency to shift by +5 kHz, the transmit frequency may be automatically lowered by 5 kHz to realign with the receiver's resonant frequency. By monitoring parasitic parameters in real time, calculating the frequency offset, and adjusting transmitter parameters based on feedback control, frequency drift caused by parasitic effects can be effectively suppressed, ensuring the stability and efficiency of wireless power transmission.

[0052] In summary, the deep learning-based adaptive frequency tracking method for wireless power transmission provided in this application has the following technical effects: The real-time transmission scenario of a multi-device shared wireless power transmission platform is collected and analyzed to determine the transmitter and multiple receivers; multiple power demand information of the multiple receivers is received; a multi-frequency resonance analysis model is called to perform dynamic matching analysis of the resonant frequency under the influence of load fluctuations on the multiple power demand information to generate a dynamic transmission parameter sequence of the transmitter; a coupling effect analysis of the resonant frequency of the multiple receivers is performed, a coupling effect network is established, the dynamic transmission parameter sequence is optimized, and a target dynamic transmission parameter sequence is generated; the target dynamic transmission parameter sequence is used to control the multi-device shared wireless power transmission platform to transmit electric energy, and at the same time, the impedance state of the multiple receivers is analyzed in real time to track the resonant frequency stably. By collecting the real-time transmission scenario of a multi-device shared wireless power transmission platform and analyzing it, the power demand information of the receiver is combined with the multi-frequency resonance analysis model and coupling effect analysis to optimize the dynamic transmission parameter sequence, and then perform real-time impedance state analysis and resonant frequency stably tracking, efficient frequency matching and energy transmission under dynamic load and frequency fluctuations are achieved, achieving the technical effect of improving the stability and efficiency of wireless power transmission in a multi-device shared environment.

[0053] In the second embodiment, based on the same inventive concept as the wireless power transmission adaptive frequency tracking method based on deep learning in the above embodiment, this application also provides a wireless power transmission adaptive frequency tracking system based on deep learning, please refer to the attached Figure 2 , the wireless power transmission adaptive frequency tracking system based on deep learning includes: The scenario analysis module 11 is used to collect and analyze the real-time transmission scenario of multiple devices sharing the wireless power transmission platform to determine the transmitting end and multiple receiving ends.

[0054] The demand receiving module 12 is configured to receive a plurality of power demand information of the plurality of receiving ends.

[0055] The frequency matching module 13 is configured to call a multi-frequency resonance analysis model to perform dynamic matching analysis of the resonance frequencies under the influence of load fluctuations on the multiple power demand information, and generate a dynamic transmission parameter sequence of the transmitting end.

[0056] The coupling analysis module 14 is configured to analyze the coupling effect of the resonant frequencies of the multiple receiving ends, establish a coupling effect network, optimize the dynamic transmission parameter sequence, and generate a target dynamic transmission parameter sequence.

[0057] The transmission module 15 is used to control the multi-device shared wireless power transmission platform to transmit power using the target dynamic transmission parameter sequence, and simultaneously analyze the impedance states of the multiple receiving ends in real time to track the stable resonant frequency.

[0058] Furthermore, the frequency matching module 13 in the deep learning-based wireless power transmission adaptive frequency tracking system is further configured to: Based on the multiple power demand information, historical matching resonant frequencies of the multiple receiving ends are retrieved to establish multiple frequency matching spaces; the resonant frequency impact analysis under the influence of load fluctuations is performed on the multiple frequency matching spaces to establish multiple optimized frequency matching spaces; and a transmission parameter control particle is randomly selected in the multiple optimized frequency matching spaces to perform power distribution balance optimization for the multiple receiving ends, thereby generating a dynamic transmission parameter sequence for the transmitting end, wherein the dynamic transmission parameters include transmission power and frequency.

[0059] Furthermore, the frequency matching module 13 in the deep learning-based wireless power transmission adaptive frequency tracking system is further configured to: Extract a first frequency matching space corresponding to the first receiving end, and randomly extract a first frequency matching parameter in the first frequency matching space; use the first frequency matching parameter to predict the load fluctuation trend of the first receiving end during wireless power transmission to generate a first load fluctuation trend; activate the fluctuation optimization module to dynamically optimize the first frequency matching parameter according to the first load fluctuation trend, generate a first optimized frequency matching space, and add it to the multiple optimized frequency matching spaces.

[0060] Furthermore, the frequency matching module 13 in the deep learning-based wireless power transmission adaptive frequency tracking system is further configured to: Randomly selecting a transmission parameter control particle in each of the multiple optimized frequency matching spaces to generate a first group of control particles; performing power allocation balance analysis on multiple receiving ends on the first group of control particles to generate a first balance visibility graph; continuing to generate a second group of control particles in the multiple optimized frequency matching spaces up to an Nth group of control particles, analyzing and generating second balance visibility graphs up to an Nth balance visibility graph; determining multiple charging priority identifiers of the multiple receiving ends based on the real-time transmission scenario; performing matching analysis on the multiple charging priority identifiers with the first balance visibility graph, the second balance visibility graph, up to the Nth balance visibility graph, and generating a group of control particles with the highest matching degree as the dynamic transmission parameter sequence.

[0061] Furthermore, the deep learning-based wireless power transmission adaptive frequency tracking system further includes a parasitic monitoring feedback module, which is configured to: The multi-device shared wireless power transmission platform is connected to a parasitic monitoring module; the parasitic parameters of the multi-device shared wireless power transmission platform are monitored by the parasitic monitoring module, wherein the parasitic parameters include at least leakage magnetic parameters and parasitic capacitance parameters; frequency offset impact analysis is performed based on the leakage magnetic parameters and the parasitic capacitance parameters to generate a frequency offset; and feedback control of power transmission is performed using the frequency offset.

[0062] Furthermore, the coupling analysis module 14 in the deep learning-based wireless power transmission adaptive frequency tracking system is further used to: Determine the anti-interference configuration between the multiple receiving ends; in combination with the anti-interference configuration, model the wireless power transmission between the transmitting end and the multiple receiving ends to generate a twin power transmission model; use the twin power transmission model to perform transmission tests on the multiple receiving ends, and analyze the resonant frequency coupling effects of the multiple receiving ends based on the test results to generate the coupling effect network, wherein the nodes of the coupling effect network represent the individual receiving ends, and the edges represent the coupling effect relationships.

[0063] Furthermore, the transmission module 15 in the deep learning-based wireless power transmission adaptive frequency tracking system is further configured to: Historical impedance change characteristics and corresponding optimization parameters of the transmitting end power and frequency are collected to train the impedance optimization network; multiple impedance change data of the multiple receiving ends are collected in real time through the impedance sensor; the multiple impedance change data are input into the impedance optimization network for feedback optimization, and the optimized power and frequency are used to stably track the resonant frequency.

[0064] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. Figure 1The deep learning-based adaptive frequency tracking method for wireless power transmission and the specific examples in Example 1 are also applicable to the deep learning-based adaptive frequency tracking system for wireless power transmission in this embodiment. Through the detailed description of the deep learning-based adaptive frequency tracking method for wireless power transmission, those skilled in the art can clearly understand the deep learning-based adaptive frequency tracking system for wireless power transmission in this embodiment. Therefore, for the sake of brevity, they will not be described in detail here. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple. For relevant details, please refer to the method description.

[0065] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one 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 present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

[0066] Obviously, those skilled in the art may make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.

Claims

1. A wireless power transmission adaptive frequency tracking method based on deep learning, characterized in that: include: Collect and analyze real-time transmission scenarios of multiple devices sharing a wireless power transmission platform to determine the transmitter and multiple receivers; receiving a plurality of power requirement information of the plurality of receiving ends; Calling a multi-frequency resonance analysis model to perform a dynamic matching analysis of the resonant frequency under the influence of load fluctuation on the multiple power demand information to generate a dynamic transmission parameter sequence of the transmitting end; Performing coupling effect analysis on the resonant frequencies of the multiple receiving ends, establishing a coupling effect network, optimizing the dynamic transmission parameter sequence, and generating a target dynamic transmission parameter sequence; The target dynamic transmission parameter sequence is used to control the multi-device shared wireless power transmission platform to transmit power, and the impedance states of the multiple receiving ends are analyzed in real time to track the stable resonant frequency.

2. The method for adaptive frequency tracking of wireless power transmission based on deep learning according to claim 1, wherein: Calling a multi-frequency resonance analysis model to perform dynamic matching analysis of the resonant frequency under the influence of load fluctuation on the multiple power demand information to generate a dynamic transmission parameter sequence of the transmitting end, including: Performing historical matching resonant frequency retrieval on the multiple receiving ends based on the multiple power demand information to establish multiple frequency matching spaces; Performing resonant frequency impact analysis on the multiple frequency matching spaces under the influence of load fluctuations to establish multiple optimized frequency matching spaces; A transmission parameter control particle is randomly selected in each of the multiple optimized frequency matching spaces to perform power allocation balance optimization of multiple receiving ends, and a dynamic transmission parameter sequence of the transmitting end is generated, wherein the dynamic transmission parameters include transmission power and frequency.

3. The method for adaptive frequency tracking of wireless power transmission based on deep learning according to claim 2, wherein: Performing resonant frequency impact analysis on the multiple frequency matching spaces under the influence of load fluctuations to establish multiple optimized frequency matching spaces includes: Extracting a first frequency matching space corresponding to the first receiving end, and randomly extracting a first frequency matching parameter in the first frequency matching space; Predicting a load fluctuation trend of the first receiving end during wireless power transmission using the first frequency matching parameter to generate a first load fluctuation trend; The fluctuation optimization module is activated to dynamically optimize the first frequency matching parameter according to the first load fluctuation trend, generate a first optimized frequency matching space, and add the first optimized frequency matching space to the multiple optimized frequency matching spaces.

4. The method for adaptive frequency tracking of wireless power transmission based on deep learning according to claim 2, wherein: Randomly selecting a transmission parameter control particle in each of the multiple optimized frequency matching spaces to perform power allocation balance optimization on multiple receiving ends, and generating a dynamic transmission parameter sequence for the transmitting end, including: Randomly selecting an emission parameter control particle in each of the multiple optimized frequency matching spaces to generate a first group of control particles; Performing power distribution balance analysis on multiple receiving ends of the first group of control particles to generate a first balance visualization graph; Continue to generate a second group of control particles in the plurality of optimized frequency matching spaces until an Nth group of control particles, and analyze and generate a second uniformity visualization graph until an Nth uniformity visualization graph; Determining multiple charging priority identifiers of the multiple receiving terminals based on the real-time transmission scenario; Matching analysis is performed on the multiple charging priority identifiers and the first balance visual map, the second balance visual map, and up to the Nth balance visual map to generate a group of control particles with the highest matching degree as the dynamic emission parameter sequence.

5. The method for adaptive frequency tracking of wireless power transmission based on deep learning according to claim 1, wherein: The multi-device shared wireless power transmission platform is connected to a parasitic monitoring module; Monitoring parasitic parameters of the multi-device shared wireless power transmission platform by the parasitic monitoring module, wherein the parasitic parameters include at least magnetic leakage parameters and parasitic capacitance parameters; Performing frequency offset impact analysis based on the magnetic flux leakage parameter and the parasitic capacitance parameter to generate a frequency offset; Feedback control of power transmission is performed using the frequency offset.

6. The method for adaptive frequency tracking of wireless power transmission based on deep learning according to claim 1, wherein: Performing coupling effect analysis of the resonant frequencies of the multiple receiving ends to establish a coupling effect network includes: determining an anti-interference configuration among the plurality of receiving ends; In combination with the anti-interference configuration, wireless power transmission modeling is performed on the transmitting end and the multiple receiving ends to generate a twin power transmission model; The twin power transmission model is used to perform transmission tests on multiple receiving ends, and the resonant frequency coupling effects of the multiple receiving ends are analyzed based on the test results to generate the coupling effect network, wherein the nodes of the coupling effect network represent the respective receiving ends, and the edges represent the coupling effect relationships.

7. The method for adaptive frequency tracking of wireless power transmission based on deep learning according to claim 1, wherein: Analyzing the impedance states of the multiple receiving ends in real time to track the stable resonant frequency includes: Collect historical impedance change characteristics and the corresponding transmitter power and frequency optimization parameters to train the impedance optimization network; collecting multiple impedance change data of the multiple receiving ends in real time through an impedance sensor; The plurality of impedance change data are input into the impedance optimization network for feedback optimization, and the resonant frequency is stably tracked using the optimized power and frequency.

8. A wireless power transmission adaptive frequency tracking system based on deep learning, characterized in that: The steps for implementing the deep learning-based adaptive frequency tracking method for wireless power transmission according to any one of claims 1 to 7 include: The scenario analysis module is used to collect and analyze the real-time transmission scenarios of multiple devices sharing the wireless power transmission platform to determine the transmitter and multiple receivers; A demand receiving module, configured to receive a plurality of power demand information of the plurality of receiving ends; A frequency matching module is configured to call a multi-frequency resonance analysis model to perform dynamic matching analysis of the resonant frequencies under the influence of load fluctuations on the multiple power demand information, and generate a dynamic transmission parameter sequence for the transmitting end; A coupling analysis module, configured to analyze the coupling effect of the resonant frequencies of the multiple receiving ends, establish a coupling effect network, optimize the dynamic transmission parameter sequence, and generate a target dynamic transmission parameter sequence; The transmission module is used to control the multi-device shared wireless power transmission platform to transmit power using the target dynamic transmission parameter sequence, and simultaneously analyze the impedance states of the multiple receiving ends in real time to track the stable resonant frequency.

Images