Wireless power transmission method and system

Abstract

The present invention provides a wireless power transmission method and system, belonging to the field of power technology. First, environmental parameters in the transmission path are collected, and then a constructed prediction model is used to predict the transmission efficiency under different frequency channels. Based on the prediction results, a deep reinforcement learning algorithm is used to dynamically calculate the optimal multi-frequency resonance frequency combination and power allocation scheme to adapt to the current transmission environment. The transmitting end dynamically adjusts the values of inductance and capacitance components according to the calculated optimal configuration to ensure that the resonance condition matches the selected frequency, and transmits the alternating electromagnetic field according to the power allocation scheme. The receiving end captures and efficiently absorbs this energy through a resonant circuit, converting it into direct current power for use by the load. The combined power is reasonably distributed according to the load demand. The present invention can monitor the transmission efficiency and environmental changes in real time, and dynamically adjust the parameters through a closed-loop feedback mechanism to ensure efficient and stable wireless power transmission.

Description

Technical Field

[0001] The present invention belongs to the field of electric power technology, and particularly relates to a wireless power transmission method and system. Background Art

[0002] Traditional power transmission relies on physical wires to transmit electrical energy, which has limitations in certain application scenarios, such as crossing complex terrain, powering mobile devices, and environments where it is difficult to lay cables. With the development of science and technology, wireless power transmission technology has gradually become a research hotspot, aiming to overcome the inconvenience of wired transmission and achieve more flexible, safe, and efficient energy distribution. Early attempts at wireless power transmission, such as the Tesla coil, demonstrated that electricity can be transmitted without contact, but the efficiency and practicality were limited. In recent years, with the advancement of technologies such as electromagnetic induction, radio frequency, resonance, and microwaves, the efficiency and feasibility of wireless power transmission have been significantly improved.

[0003] However, existing wireless power transmission systems still face challenges, including transmission efficiency significantly affected by distance, obstacles, and environmental factors, as well as a lack of intelligent adjustment of frequency and power configurations. Therefore, it is particularly important to develop efficient and stable wireless power transmission methods and systems that can adapt to environmental changes in real time and dynamically optimize transmission parameters. Summary of the Invention

[0004] To address the above technical issues, the present invention provides a wireless power transmission method and system that optimizes transmission efficiency by monitoring and adjusting environmental parameters in real time. It also dynamically adjusts frequency and power through an intelligent scheduling algorithm to ensure efficient and stable power transmission.

[0005] The present invention provides a wireless power transmission method, the method comprising:

[0006] Step S1: collecting environmental parameters in the wireless power transmission path;

[0007] Step S2: construct a prediction model to predict the expected transmission efficiency under different frequency channels;

[0008] Step S3: Calculating the optimal multi-frequency resonance frequency combination and the corresponding power distribution scheme;

[0009] Step S4: The transmitter dynamically adjusts the values of the inductor and capacitor components according to the optimal operating frequency combination and the adaptive tuning network;

[0010] Step S5: The transmitting end transmits an alternating electromagnetic field to the receiving end according to the power allocation scheme; the receiving end captures the alternating electromagnetic field emitted by the transmitting end and absorbs energy through the resonant circuit;

[0011] Step S6: The receiving end integrates the energy received from each frequency channel and combines it, and intelligently distributes the combined electric energy to the target load according to the load demand.

[0012] Optionally, the constructing of a prediction model to predict expected transmission efficiency under different frequency channels specifically includes:

[0013] Define a feature extraction subnetwork that takes an input layer and a list indicating the number of neurons in each layer, and then creates a series of fully connected layers, each using a ReLU activation function.

[0014] Define the attention mechanism layer, convert the input into a probability distribution through a fully connected layer (using Softmax activation); multiply these probabilities with the original input through a dot product operation;

[0015] Build a prediction model, create multiple input layers based on the input shape; create a feature extraction subnetwork for each input layer; use a concatenation layer to merge the outputs of all subnetworks into a single vector; add an attention mechanism layer to dynamically emphasize the feature combination that the model focuses on; use the concatenation layer again to combine the original merged features with the features processed by the attention mechanism; add a fully connected layer to further process these features; create an output layer for each frequency channel, use the Sigmoid activation function, and each output corresponds to the prediction of a frequency channel; compile the model, and the loss function is the mean squared error.

[0016] Optionally, the calculation of the optimized multi-frequency resonance frequency combination and the corresponding power distribution scheme specifically includes:

[0017] Initialize the deep reinforcement learning agent class for learning and decision-making on the optimal configuration of frequency and power; define the state space size; define the action space size; use a double-ended queue to set up the experience replay buffer to store each interaction data; set the discount factor, initial exploration rate, minimum exploration rate, exploration rate decay rate, and learning rate; build the deep Q network model, which is implemented by the construction function;

[0018] Build a neural network function for estimating action values. The input is the current environment state (such as transmission distance and attenuation), and the output is the value of each action. The model consists of three fully connected layers. The first and second layers use the ReLU activation function to increase nonlinear processing capabilities. The last layer uses a linear activation function to output the expected value of each action (i.e., Q value).

[0019] Save the agent's experience (state, action, reward) to the memory replay, that is, store the data of an interaction in the experience replay buffer;

[0020] The action selected according to the current strategy (exploration or exploitation) is randomly selected when the random number is less than the current exploration rate; otherwise, the action corresponding to the maximum Q value predicted by the model is selected as the expected optimal action;

[0021] Randomly extract a batch of experiences from memory for learning, and optimize the network weights by updating the target Q value to minimize the mean square error between the predicted Q value and the target Q value. Specifically,

[0022] Randomly extract a certain number of samples from the experience replay buffer; the samples contain the state of the past exploration, the actions performed, the rewards obtained, the next state, and whether it has ended;

[0023] Loop through each extracted sample and unpack each element in the sample;

[0024] Calculate the target value. If this experience is a terminal state, the target value is the immediate reward obtained; if it is not a terminal state, the target value is the discounted value of the immediate reward plus the expected future reward;

[0025] Use the current model to predict the Q-values of all actions in a given state;

[0026] Assign the calculated target value to the Q value corresponding to the action taken;

[0027] Use the updated target Q value as the label to train the model once and optimize its weights to reduce the gap between the predicted Q value and the target Q value;

[0028] Update the initial exploration rate;

[0029] Outputting an instruction set including a selected multi-frequency resonant frequency combination and a corresponding power allocation plan.

[0030] Optionally, the transmitting end dynamically adjusts the values of the inductor and capacitor components according to the optimal operating frequency combination and the adaptive tuning network, specifically including:

[0031] After calculating the best frequency combination, the command is sent to the controller of the tuning network through the digital interface. The command contains the target frequency f target ; The controller is based on the target frequency f target , calculate the current capacitance C current or inductor L current The amount of adjustment to match the target frequency;

[0032] Based on the calculated adjustment amount, the controller dynamically adjusts the physical parameters of the variable capacitor or variable inductor through motor control or electronic drive to achieve the calculated capacitance C new or inductor L new value;

[0033] Use a high-frequency precision frequency sensor to monitor the tuned resonant frequency f measured and with the target frequency f target By comparison, if there is a deviation e, the capacitance or inductance is adjusted through a closed-loop control algorithm, which is expressed as:

[0034]

[0035] Where u(t) is the control signal; K b , K i and K d are the proportional, integral and derivative gains respectively.

[0036] Optionally, the transmitting end transmits an alternating electromagnetic field to the receiving end according to the power allocation scheme; the receiving end captures the alternating electromagnetic field emitted by the transmitting end and absorbs energy through a resonant circuit, specifically including:

[0037] According to the calculated frequency and the corresponding power distribution, for each calculated frequency f α , the frequency synthesizer generates the corresponding alternating current signal, which is expressed as:

[0038]

[0039] Where, I α (t) is the current of the αth frequency component; A α is the amplitude; φ α is the phase angle; the calculated frequencies are f1, f2, ..., f n ; Power distribution is P allocated,1 , P allocated,2 ,...,P allocated,n ;

[0040] According to the power distribution scheme, the current signal of each frequency is amplified. Considering the amplification efficiency, the actual output power is expressed as:

[0041] P out，c =η PA,c ·P allocated,c

[0042] Where, P out，c is the actual output power of the cth channel after amplification; η PA，c is the power amplifier efficiency of the cth channel;

[0043] Pre-distortion technology is used before transmission. The receiving end is equipped with multiple resonant circuits corresponding to the transmitting end frequency. Each circuit is optimized for a specific frequency. By adjusting the values of capacitors and inductors, the resonant frequency of the circuit is ensured to match the frequency of the received electromagnetic wave.

[0044] Adjust the Q value of the resonant circuit to improve the energy absorption efficiency, which is expressed as:

[0045]

[0046] Where, ∝ is proportional to; f i is the transmission frequency; η abs，r is the absolute efficiency of the rth resonant circuit; f 0，r is the center frequency of the rth resonant circuit; Q r is the quality factor of the rth resonant circuit; f BW，r is the bandwidth of the rth resonant circuit;

[0047] The AC power absorbed in the resonant circuit is converted into DC power through a rectifier bridge circuit, which can be expressed as:

[0048]

[0049] Where η conv is the conversion efficiency; P DC is the converted DC power; P AC is the absorbed AC power.

[0050] The present invention also provides a wireless power transmission system, comprising:

[0051] An environmental parameter collection module, used to collect environmental parameters in the wireless power transmission path;

[0052] The transmission efficiency prediction module is used to build a prediction model to predict the transmission efficiency of different frequency channels;

[0053] Power scheme allocation module, used to calculate the optimized multi-frequency resonance frequency combination and the corresponding power allocation scheme;

[0054] An electrical component adjustment module is used to dynamically adjust the values of inductance and capacitance components at the transmitter based on the optimal operating frequency combination and adaptive tuning network;

[0055] The transmitter-receiver matching module is used for the transmitter to transmit an alternating electromagnetic field to the receiver according to the power allocation scheme; the receiver captures the alternating electromagnetic field emitted by the transmitter and absorbs energy through the resonant circuit;

[0056] The energy intelligent distribution module is used to integrate the energy received from each frequency channel at the receiving end and intelligently distribute the combined electric energy to the target load according to the load demand.

[0057] Optionally, the transmission efficiency prediction module specifically includes:

[0058] Define a feature extraction subnetwork that takes an input layer and a list indicating the number of neurons in each layer, and then creates a series of fully connected layers, each using a ReLU activation function.

[0059] Define the attention mechanism layer, convert the input into a probability distribution through a fully connected layer (using Softmax activation); multiply these probabilities with the original input through a dot product operation;

[0060] Build a prediction model, create multiple input layers based on the input shape; create a feature extraction subnetwork for each input layer; use a concatenation layer to merge the outputs of all subnetworks into a single vector; add an attention mechanism layer to dynamically emphasize the feature combination that the model focuses on; use the concatenation layer again to combine the original merged features with the features processed by the attention mechanism; add a fully connected layer to further process these features; create an output layer for each frequency channel, use the Sigmoid activation function, and each output corresponds to the prediction of a frequency channel; compile the model, and the loss function is the mean squared error.

[0061] Optionally, the power solution allocation module specifically includes:

[0062] Initialize the deep reinforcement learning agent class for learning and decision-making on the optimal configuration of frequency and power; define the state space size; define the action space size; use a double-ended queue to set up the experience replay buffer to store each interaction data; set the discount factor, initial exploration rate, minimum exploration rate, exploration rate decay rate, and learning rate; build the deep Q network model, which is implemented by the construction function;

[0063] Build a neural network function for estimating action values. The input is the current environment state (such as transmission distance and attenuation), and the output is the value of each action. The model consists of three fully connected layers. The first and second layers use the ReLU activation function to increase nonlinear processing capabilities. The last layer uses a linear activation function to output the expected value of each action (i.e., Q value).

[0064] Save the agent's experience (state, action, reward) to the memory replay, that is, store the data of an interaction in the experience replay buffer;

[0065] The action selected according to the current strategy (exploration or exploitation) is randomly selected when the random number is less than the current exploration rate; otherwise, the action corresponding to the maximum Q value predicted by the model is selected as the expected optimal action;

[0066] Randomly extract a batch of experiences from memory for learning, and optimize the network weights by updating the target Q value to minimize the mean square error between the predicted Q value and the target Q value. Specifically,

[0067] Randomly extract a certain number of samples from the experience replay buffer; the samples contain the state of the past exploration, the actions performed, the rewards obtained, the next state, and whether it has ended;

[0068] Loop through each extracted sample and unpack each element in the sample;

[0069] Calculate the target value. If this experience is a terminal state, the target value is the immediate reward obtained; if it is not a terminal state, the target value is the discounted value of the immediate reward plus the expected future reward;

[0070] Use the current model to predict the Q-values of all actions in a given state;

[0071] Assign the calculated target value to the Q value corresponding to the action taken;

[0072] Use the updated target Q value as the label to train the model once and optimize its weights to reduce the gap between the predicted Q value and the target Q value;

[0073] Update the initial exploration rate;

[0074] Outputting an instruction set including a selected multi-frequency resonant frequency combination and a corresponding power allocation plan.

[0075] Optionally, the electrical component adjustment module specifically includes:

[0076] After calculating the best frequency combination, the command is sent to the controller of the tuning network through the digital interface. The command contains the target frequency f target ; The controller is based on the target frequency f target , calculate the current capacitance C current or inductor L current The amount of adjustment to match the target frequency;

[0077] Based on the calculated adjustment amount, the controller dynamically adjusts the physical parameters of the variable capacitor or variable inductor through motor control or electronic drive to achieve the calculated capacitance C new or inductor L new value;

[0078] Use a high-frequency precision frequency sensor to monitor the tuned resonant frequency f measured and with the target frequency f target By comparison, if there is a deviation e, the capacitance or inductance is adjusted through a closed-loop control algorithm, which is expressed as:

[0079]

[0080] Where u(t) is the control signal; K b , K i and Kd are the proportional, integral and derivative gains respectively.

[0081] Optionally, the transmitting and receiving matching module specifically includes:

[0082] According to the calculated frequency and the corresponding power distribution, for each calculated frequency f α , the frequency synthesizer generates the corresponding alternating current signal, which is expressed as:

[0083]

[0084] Where, I α (t) is the current of the αth frequency component; A α is the amplitude; φ α is the phase angle; the calculated frequencies are f1, f2, ..., f n ; Power distribution is P allocated,1 , P allocated,2 ,...,P allocated,n ;

[0085] According to the power distribution scheme, the current signal of each frequency is amplified. Considering the amplification efficiency, the actual output power is expressed as:

[0086] P out，c =η PA,c ·P allocated,c

[0087] Where, P out，c is the actual output power of the cth channel after amplification; η PA，c is the power amplifier efficiency of the cth channel;

[0088] Pre-distortion technology is used before transmission. The receiving end is equipped with multiple resonant circuits corresponding to the transmitting end frequency. Each circuit is optimized for a specific frequency. By adjusting the values of capacitors and inductors, the resonant frequency of the circuit is ensured to match the frequency of the received electromagnetic wave.

[0089] Adjust the Q value of the resonant circuit to improve the energy absorption efficiency, which is expressed as:

[0090]

[0091] Where, ∝ is proportional to; f i is the transmission frequency; η abs，r is the absolute efficiency of the rth resonant circuit; f 0，r is the center frequency of the rth resonant circuit; Q r is the quality factor of the rth resonant circuit; f BW，r is the bandwidth of the rth resonant circuit;

[0092] The AC power absorbed in the resonant circuit is converted into DC power through a rectifier bridge circuit, which can be expressed as:

[0093]

[0094] Where η conv is the conversion efficiency; P DC is the converted DC power; P AC is the absorbed AC power.

[0095] Compared with the prior art, the present invention has the following beneficial effects:

[0096] Through the deep reinforcement learning algorithm, the present invention enables the system to autonomously learn and optimize frequency and power configuration, automatically adapt to changes in the transmission environment without human intervention, and improve the system flexibility and response speed; the prediction model and optimization algorithm ensure that the best frequency combination and power allocation scheme can be found in various environments, improving energy transmission efficiency and reducing losses; the closed-loop feedback control mechanism continuously monitors and adjusts the system status to ensure efficient and stable power transmission even in the face of environmental interference or load changes, enhancing the system's reliability and continuous power supply capability; the system design is suitable for a variety of application scenarios, including but not limited to home, industry, mobile device charging, and remote monitoring equipment power supply, broadening the application scenarios of wireless power transmission. BRIEF DESCRIPTION OF THE DRAWINGS

[0097] Figure 1 is a flow chart of a wireless power transmission method of the present invention;

[0098] Figure 2 This is a structural diagram of a wireless power transmission system of the present invention. DETAILED DESCRIPTION

[0099] The present invention will be further described below with reference to specific implementation cases and drawings, but the present invention is not limited to these embodiments.

[0100] Example 1

[0101] like Figure 1 As shown, the present invention discloses a wireless power transmission method, the method comprising:

[0102] Step S1: Collect environmental parameters in the wireless power transmission path.

[0103] Step S2: Construct a prediction model to predict the expected transmission efficiency under different frequency channels.

[0104] Step S3: Calculate the optimal multi-frequency resonance frequency combination and the corresponding power distribution scheme.

[0105] Step S4: The transmitter dynamically adjusts the values of the inductor and capacitor components according to the optimal operating frequency combination and the adaptive tuning network.

[0106] Step S5: the transmitting end transmits an alternating electromagnetic field to the receiving end according to the power allocation scheme; the receiving end captures the alternating electromagnetic field emitted by the transmitting end and absorbs energy through the resonant circuit.

[0107] Step S6: The receiving end integrates the energy received from each frequency channel and combines it, and intelligently distributes the combined electric energy to the target load according to the load demand.

[0108] The following is a detailed discussion of each step

[0109] Step S1: Collect environmental parameters in the wireless power transmission path.

[0110] Step S1 specifically includes:

[0111] Design and deployment of environmental perception units, including:

[0112] Environmental perception units typically integrate multiple sensors and signal processing devices, including but not limited to:

[0113] Distance sensors (such as lidar, ultrasonic sensors, or infrared sensors) measure the real-time distance between the transmitter and receiver, and identify and track the location of obstacles.

[0114] Material identification sensors (such as near-infrared spectrometers or material identification radars) analyze the type and material of obstacles. This is because obstacles of different materials and shapes have different absorption, reflection, and diffraction effects on electromagnetic waves, affecting the transmission path and efficiency, and are crucial to frequency selection.

[0115] The load status monitor is connected to the receiving end or directly integrated into the receiving device to monitor the load's real-time power consumption, voltage, current and other parameters to ensure that the power supply matches the demand.

[0116] Environmental condition sensors (such as temperature, humidity, and air pressure sensors) monitor environmental factors that may affect the propagation of electromagnetic waves and device performance.

[0117] These sensing units are installed at the transmitter, receiver, and key locations along the transmission path to ensure comprehensive coverage and real-time monitoring of all relevant parameters.

[0118] Data collection and preliminary processing, including:

[0119] The environmental perception unit continuously collects data from the above-mentioned sensors to achieve continuous monitoring of the transmission environment.

[0120] The collected data undergoes preliminary verification and filtering to remove outliers or noise to ensure data accuracy. The raw data also needs to be normalized to make it suitable for the input format of subsequent algorithms.

[0121] Data transmission and synchronization, including:

[0122] The processed environmental parameter data are packaged into a unified format for easy transmission and analysis.

[0123] Use low-power wireless communication technologies (such as Wi-Fi, Bluetooth, Zigbee, or proprietary wireless protocols) to send data to a central processing unit or intelligent scheduling system.

[0124] Ensuring the time synchronization of all sensing units allows the system to accurately understand the spatiotemporal relationships of the data, which is crucial for dynamically adjusting transmission parameters.

[0125] Through the above steps, real-time environmental parameter monitoring provides real-time and accurate environmental information for the entire wireless power transmission system, which is the basis for realizing intelligent and efficient energy transmission.

[0126] Step S2: Construct a prediction model to predict the expected transmission efficiency under different frequency channels.

[0127] Step S2 specifically includes:

[0128] Information about the current transmission path is collected through environmental perception technologies (such as sensors, cameras, RF signal feedback, etc.). This may include but is not limited to changes in transmission distance (due to movement or changes in obstacle position), the degree of electromagnetic interference in the surrounding environment, the material of the obstacle and its absorption or reflection characteristics of electromagnetic waves of specific frequencies, and the specific requirements of the receiving end load (such as changes in power requirements).

[0129] Based on historical data, a prediction model is built, combined with machine learning algorithms, to predict the expected transmission efficiency under different frequency channels based on the input environmental parameters, including:

[0130] The TensorFlow library and the layers and models modules in Keras are imported to build and compile the neural network model, expressed as import tensorflow as tf; from tensorflow.keras import layers,models.

[0131] Define a feature extraction subnetwork that receives an input layer and a list indicating the number of neurons in each layer, and then creates a series of fully connected layers (Dense layers), each of which uses the ReLU activation function; this process is performed for each input feature vector to extract higher-level features from the original input, represented as defcreate_feature_subnetwork(input_layer, layer_sizes):; x = input_layer; for size in layer_sizes: x = layers.Dense(size, activation = 'relu')(x); return x.

[0132] Define the attention mechanism layer, convert the input into a probability distribution (using Softmax activation) through a fully connected layer, these probabilities represent the importance of each input feature; multiply these probabilities with the original input through a dot product operation to highlight important features, expressed as def create_attention_layer(inputs,attention_size):attention_probs = layers.Dense(attention_size,activation = 'soft max',name = 'attention_vec')(inputs); attention_mul = layers.multiply([inputs,attention_probs]); return attention_mul.

[0133] Build a prediction model and create multiple input layers according to the input shape (corresponding to different input sources), represented as input_layers = [layers.Input(shape = (shape,)) for shape in input_shapes]; create a feature extraction subnetwork for each input layer, represented as subnets = [create_feature_subnetwork(input_layer, [128, 128]) for input_layer in input_layers]; use the concat layer to merge the outputs of all subnetworks into a single vector, which helps the model understand the interaction between different inputs, represented as merged = layers.concatenate(subnets); add an attention mechanism layer to dynamically emphasize the feature combination that the model pays attention to, represented as attention_output = create_attention_layer(merged, 128); use the concat layer again to combine the original merged features and the features processed by the attention mechanism, represented as merged_with_attention = layers.concatenate([merged, attention_output]); add a fully connected layer These features are further processed as dense_output = layers.Dense(256, activation = 'relu')(merged_with_attention); an output layer is created for each frequency channel, using the Sigmoid activation function, where each output corresponds to the prediction of a frequency channel, as final_outputs = [layers.Dense(1, activation = 'sigmoid')(dense_output)for_inrange(5)]; the model is compiled, with the loss function being MSE (mean squared error), as model = models.Model(inputs = input_layers, outputs = final_outputs).

[0134] In this embodiment, the prediction model includes multiple input layers, which process different types of input data (such as distance, electromagnetic interference, obstacle material, etc.), feature extraction sub-network, attention mechanism layer, fusion layer and output layer (generating prediction efficiency for each frequency channel).

[0135] Step S3: Calculate the optimal multi-frequency resonance frequency combination and the corresponding power distribution scheme.

[0136] Step S3 specifically includes:

[0137] Evaluate the suitability of each potential resonant frequency (in a preset range of frequencies) for energy transmission in the current environment, taking into account factors such as frequency penetration, attenuation due to environmental influences, and coupling efficiency with the receiving end; determine one or more frequency combinations that can minimize energy loss and improve transmission efficiency. For the selected frequency combinations, calculate the optimal power ratio to be allocated to each frequency channel, specifically including:

[0138] Import libraries and initializations, such as NumPy, TensorFlow, and deque in the collections module, as well as the random library, which are the basis for building and training DQN models.

[0139] Initialize the deep reinforcement learning agent class for learning and decision-making on the optimal configuration of frequency and power, expressed as def__init__(self,state_size,action_size):; define the state space size (the number of environment parameters), expressed as self.state_size = state_size; define the action space size (such as the number of possible frequency and power combinations), expressed as self.action_size = action_size; use a double-ended queue (deque) to set up the experience replay buffer for storing each interaction data, expressed as self.memory = deque(maxlen = 2000); set the discount factor (used to calculate the current value of future rewards) : The default value is 0.995, which is the previous value), represented by self.gamma = 0.95; set the initial exploration rate (determines whether to take random actions or follow the model's prediction frequency), represented by self.epsilon = 1.0; set the minimum exploration rate (to prevent complete cessation of exploration), represented by self.epsilon_min = 0.01; set the exploration rate decay rate (reduce the exploration rate after each learning), represented by self.epsilon_decay = 0.995; set the learning rate (affects the amplitude of model parameter updates), represented by self.learning_rate = 0.001; build the deep Q network model, represented by self.model = self._build_model(), which is implemented by the construction function.

[0140] Build a neural network function for estimating action value. The input is the current environment state (such as transmission distance, attenuation, etc.), and the output is the value of each possible action (i.e., expected return). The model consists of three fully connected layers. The first and second layers use the ReLU activation function to increase nonlinear processing capabilities. The last layer uses a linear activation function to output the expected value of each action (i.e., Q value), expressed as def_build_model(self): model = models.Sequential(); model.add(Dense(24, input_dim = self.state_size, activation = 'relu')); model.add(Dense(24, activation = 'relu')); model el.add(Dense(self.action_size, activation = 'linear')); model.compile(loss = 'mse', optimizer = keras.optimizers.Adam(lr = self.learning_rate)); return model.

[0141] Save the agent's experience (state, action, reward, etc.) to the memory replay, that is, store the data of an interaction in the experience replay buffer for subsequent learning, expressed as defremember(self,state,action,reward,next_state,done):self.memory.append((state,action,reward,next_state,done)).

[0142] The action selected according to the current strategy (exploration or exploitation) is randomly selected when the random number is less than the current exploration rate; otherwise, the action corresponding to the maximum Q value predicted by the model is selected, that is, the expected optimal action, which is expressed as defact(self, state): if np.random.rand() <= self.epsilon: return random.rand range(self.action_size); act_values = self.model.predict(state); return np.argmax(act_values[0]).

[0143] Randomly extract a batch of experiences from memory for learning, and optimize the network weights by updating the target Q value (taking into account future rewards) to minimize the mean squared error (MSE) between the predicted Q value and the target Q value, specifically including:

[0144] Randomly extract batch_size samples from the experience replay buffer self.memory. These samples contain information about the state (state), action (action), reward (reward), next state (next_state), and whether the exploration was completed (done). They are represented as minibatch = random.sample(self.memory, batch_size).

[0145] Each extracted sample is processed in a loop, and each element in the sample is unpacked, expressed as for state, action, reward, next_state, done in minibatch:.

[0146] Calculate the target value (target). If this experience is a terminal state (done is True), the target value is the immediate reward (reward) obtained; if it is not a terminal state, the target value, in addition to the immediate reward, will also add the discounted value of the expected future reward, that is, the Q value calculated using the Bellman equation; self.gamma is used here as the discount factor to predict the maximum Q value of all possible actions in the next state, expressed as target = reward if done else reward + self.gamma × np.amax(self.model.predict(np.array([next_state]))[0]).

[0147] Use the current model to predict the Q value of all actions in a given state, expressed as target_f = self.model.predict(state).

[0148] The calculated target value target is assigned to the Q value corresponding to the action taken. This is for targeted update, i.e., TD error correction in Q-learning, expressed as target_f[0][action]=target.

[0149] Use the updated target Q value target_f as the label to train the model once (single iteration, epochs=1), and optimize its weights to reduce the gap between the predicted Q value and the target Q value. verbose=0 means that the log information of the training process is not displayed, which is expressed as self.model.fit(state,target_f,epochs=1,verbose=0).

[0150] Update the important parameter epsilon in the exploration and exploitation strategy. Epsilon usually controls the probability of randomly selecting an action. It is initially high to encourage exploration and gradually decreases as learning progresses to increase the utilization of the known best strategy. The value of epsilon is reduced by multiplying the current epsilon by the decay factor self.epsilon_decay, and ensuring that it does not fall below the preset minimum value self.epsilon_min to prevent the exploration probability from falling to zero, expressed as self.epsilon = max(self.epsilon_min, self.epsilon×self.epsilon_decay).

[0151] By defining the size of the state space and action space, a reinforcement learning agent class object is instantiated to guide the frequency and power adjustment of the wireless power transmission system. By continuously interacting with the environment, accumulating experience, and performing batch learning when appropriate, the DQN agent gradually learns to select the frequency and power configuration strategy that is most conducive to energy transmission efficiency and system stability under different environmental conditions. This is particularly suitable for solving problems such as wireless power transmission that require decision-making in complex and dynamic environments.

[0152] In this embodiment, the state space and action space define the framework for the interaction between the agent and the environment. The state space indicates how many state variables can be used to describe the system or environment. These state parameters may include, but are not limited to: transmission distance (the distance from the transmitter to the receiver), obstacle information (such as the number, location, type of obstacles and their absorption or reflection characteristics for electromagnetic waves of specific frequencies), environmental attenuation (the attenuation of electromagnetic waves in the air or when passing through obstacles), receiving end coupling efficiency (the receiving end's ability to receive wireless power), load demand (the power level currently required by the receiving end), system energy consumption (the current energy consumption level), system temperature (to prevent the risk of overheating), and other environmental factors, such as temperature and humidity, which may affect the efficiency of wireless transmission.

[0153] The action space refers to the number of different actions that an agent (in this case, the wireless power transmission system) can take in a given state. It indicates that different combinations of frequency and power configurations are designed, and the agent can choose one of these configurations as an action. Each action may represent a different frequency selection (each frequency may correspond to one or more specific wireless power transmission channels). The power allocation scheme also includes how the total output power is distributed across the selected frequencies to achieve optimal transmission efficiency and system stability. Due to the complexity of actual operations and the limitations of computational resources, directly setting an extremely large action space may be impractical and detrimental to model convergence. Hierarchical or combined strategies are used to reduce the action space. The first method is frequency band optimization, which divides the available frequencies into several bands and selects an optimal frequency within each band. This decomposes the problem into selecting a frequency band and adjusting the power within the selected band. For example, if there are three frequency bands, each with three power levels, then the action space = 3 (frequency band selection) × 3 (power level) = 9. The second method precomputes a series of efficient frequency-power combinations rather than exhaustively enumerating all possible combinations. If 10 efficient combinations are identified after screening, the action space = 10.

[0154] In this embodiment, deep reinforcement learning (DRL) is used to optimize frequency selection and power allocation. Deep reinforcement learning is a method that allows a model to learn optimal strategies by interacting with the environment. It is suitable for decision-making problems in dynamic environments, such as frequency and power optimization of wireless power transmission systems. The above-mentioned prediction model solves the problem of how to predict the transmission efficiency under different conditions, and deep reinforcement learning, after obtaining these prediction capabilities, solves the problem of how to make the best frequency and power adjustment strategy based on these predictions to achieve the highest efficiency.

[0155] The algorithm ultimately outputs a set of instructions, including the selected multi-frequency resonant frequency combination and the corresponding power allocation plan. These instructions are used to adjust the transmission parameters of the transmitter and the reception and processing settings of the receiver. Specifically, they include:

[0156] After intelligent frequency and power optimization, the system algorithm (intelligent scheduling algorithm) generates a set of specific execution instructions. These instructions form the core of the optimization solution. These instructions not only include the most efficient multi-frequency resonant frequency combination under the current environment, but also specify the power allocation details to be used on each frequency channel. Specifically:

[0157] Frequency combinations list one or several optimal frequencies that have been proven to achieve optimal electromagnetic wave coupling, reduce energy loss, and improve transmission efficiency under current environmental conditions; each frequency corresponds to a specific wireless power transmission channel, ensuring that energy can be efficiently transmitted through multiple paths.

[0158] The power allocation plan allocates the most appropriate power output for each selected frequency channel based on an evaluation of each channel's efficiency, total energy consumption constraints, and the actual load demand. The goal is to ensure that the load's power requirements are met to the greatest extent possible without sacrificing system stability.

[0159] The optimization plan is output in the form of an instruction set, which is convenient for the system control unit to directly interpret and convert into actual control commands for the hardware, such as adjusting the frequency synthesizer and power amplifier settings on the transmitter side, as well as the matching network and energy converter configuration on the receiver side.

[0160] Based on the instructions output by the algorithm, the hardware components make corresponding adjustments and implement new frequency and power strategies. At the same time, the system continuously monitors the actual transmission effect. Through a closed-loop feedback mechanism, if it detects a decrease in efficiency or changes in the environment, the algorithm will restart the optimization calculation and dynamically adjust the strategy to ensure that the transmission efficiency and stability are always maintained at the best state. Specifically,

[0161] Hardware adjustment: Based on the optimization plan, the hardware components of the transmitter (such as the RF frequency source and power amplifier) and the hardware of the receiver (such as the resonant circuit and rectifier) will be automatically adjusted to the specified frequency and power settings; this step ensures that the physical level of the system matches the optimal configuration calculated by the algorithm.

[0162] System configuration and software level will also be updated accordingly to ensure that the control system can accurately track and manage the current frequency and power settings, and prepare for subsequent feedback control.

[0163] Real-time monitoring: The system's built-in sensors and monitoring modules continuously track the performance indicators of wireless power transmission, such as transmission efficiency, energy consumption, temperature, and changes in environmental parameters (new obstacles or distance changes).

[0164] Performance evaluation uses algorithms to quickly analyze collected real-time data to assess whether the current configuration's transmission efficiency meets expectations and whether there are any changes that may affect efficiency or stability.

[0165] Closed-loop feedback: Once performance degradation or environmental changes exceeding the threshold are detected, the algorithm will automatically trigger the re-optimization process, re-execute environmental parameter monitoring and intelligent scheduling algorithm calculations, generate new frequency and power optimization plans, and instruct hardware adjustments.

[0166] Step S4: After receiving the instruction, the transmitter dynamically adjusts the values of the inductor and capacitor components according to the optimal operating frequency combination and the adaptive tuning network.

[0167] Step S4 specifically includes:

[0168] After calculating the best frequency combination, the digital interface (such as serial communication protocol such as I 2 C, SPI or network interface) sends instructions to the controller of the tuning network, which contains the target frequency f target .

[0169] The controller is based on the target frequency f target , calculate the current capacitance C current or inductor L current The amount of adjustment to match the target frequency includes:

[0170] Assume that the capacitance and inductance before adjustment are C current and L current , then:

[0171]

[0172] Read the current capacitance C current or inductor L current , if f current (current frequency) and f target If there is a difference, adjust to the target frequency f target , solve one of the following equations to find the adjusted C new or L new , adjust the other parameter and keep it unchanged;

[0173]

[0174] If you choose to adjust the capacitor C current ,but:

[0175]

[0176] If you choose to adjust the capacitor L current ,but:

[0177]

[0178] Based on the calculated adjustment amount, the controller dynamically adjusts the physical parameters of the variable capacitor or variable inductor through motor control (such as a stepper motor or servo motor) or electronic drive (such as a piezoelectric drive) to achieve the calculated capacitance C new or inductor L new value.

[0179] The variable capacitor adjusts the capacitance by changing the distance or area between the two plates. The capacitance formula is:

[0180] C=εA / d

[0181] Where ε is the dielectric constant; A is the plate area; d is the distance between the plates; dynamic adjustment is achieved through motor drive or micro-electromechanical system (MEMS).

[0182] Variable inductors change the inductance by adjusting the number of coil turns, coil diameter, or the magnetic permeability of the core material. The inductance formula is:

[0183] L=μN 2 S / l

[0184] Where μ is the magnetic permeability; N is the number of turns; S is the cross-sectional area of the coil; l is the length of the coil; and dynamic adjustment is achieved using technologies such as magnetorheological fluid or shape memory alloy.

[0185] Use a high-frequency precision frequency sensor (such as a PLL or FLL circuit) to monitor the tuned resonant frequency f measured and with the target frequency f target By comparison, if there is a deviation e, the capacitance or inductance is adjusted through a closed-loop control algorithm (such as PID control), which is expressed as:

[0186]

[0187] e=f measured -f target

[0188] Where u(t) is the control signal; K b , K i and K d They are proportional, integral and differential gains respectively; the role of the proportional part is to provide an immediate corrective action, the size of which is directly proportional to the size of the current error, and is intended to respond quickly and begin to correct the system deviation; the integral part integrates the error signal e(t) to accumulate past errors to eliminate steady-state errors; the differential part is the differential of the error signal e(t), that is, to calculate the error change rate, which is used to predict and suppress rapid changes in the system.

[0189] In this embodiment, the transmitter receives instructions from the intelligent scheduling controller and adjusts the operating state of each resonant coil through the multi-frequency transmission module based on the optimal operating frequency combination. At the same time, the adaptive tuning network is used to adjust the impedance characteristics of the transmitter in real time to ensure that the resonant conditions match the selected frequency and allocate the corresponding power to each frequency channel.

[0190] The implementation process of step S4 is as follows: after the intelligent scheduling algorithm determines the frequency combination, the instruction is transmitted to the controller of the tuning network through the digital interface; based on the target frequency, the controller calculates the amount by which the current capacitance or inductance value needs to be adjusted to match the target frequency; by controlling the motor or electronic drive, the physical parameters of the variable capacitor and variable inductor are adjusted to achieve the calculated capacitance / inductance value; the tuned resonant frequency is monitored using a sensor and compared with the target frequency; if there is a deviation, further fine-tuning is performed through the closed-loop control algorithm until the desired frequency match is achieved; after the system enters a stable state, the transmission efficiency and stability are continuously monitored to ensure that the tuning effect meets the performance indicators.

[0191] Step S5: The transmitting end transmits an alternating electromagnetic field to the receiving end according to the power allocation scheme; the receiving end captures the alternating electromagnetic field emitted by the transmitting end, absorbs energy through the resonant circuit, and converts the received electromagnetic energy into electrical energy.

[0192] Step S5 specifically includes:

[0193] According to the calculated frequency and the corresponding power distribution, for each calculated frequency f α , the frequency synthesizer generates the corresponding alternating current signal, which is expressed as:

[0194]

[0195] Where, I α (t) is the current of the αth frequency component; A α is the amplitude; φ α is the phase angle; the calculated frequencies are f1, f2, ..., f n ; Power distribution is P allocated,1 , P allocated,2 ,...,P allocated,n .

[0196] According to the power distribution scheme, the current signal of each frequency is amplified. Considering the amplification efficiency, the actual output power is expressed as:

[0197] P out，c =η PA，c ·P allocated,c

[0198] Where, P out，c is the actual output power of the cth channel after amplification; η PA，c is the power amplifier efficiency of the cth channel; the frequency signal is implicit in the power amplification process. After the multi-frequency signal is synthesized, the current signal of each frequency component needs to be amplified according to the power allocation scheme.

[0199] In the actual signal amplification process, the current signal I α(t) represents the time domain expression of each frequency component, while P calculated in the power allocation scheme is allocated,c The power level of each frequency component is defined; the working goal of the amplifier is to allocated,c To adjust its gain or working state, so that the amplified signal power is as close to the allocated value as possible; this formula directly focuses on the relationship between the output power and the theoretical allocated power and the amplifier efficiency, and it implies the amplifier's response to the current signal I α (t) The processing process is to adjust the gain so that the output power reaches the expected value without directly changing the current amplitude of the signal. α (t).

[0200] Pre-distortion technology is used before transmission to ensure the correct waveform of the signal after nonlinear amplification. The amplified signal needs to be transmitted precisely and synchronously to ensure the superposition effect of multi-frequency signals. The purpose of waveform control and synchronous transmission is to ensure that all amplified frequency components can be perfectly superimposed in time and phase, based on the frequency and power configuration carefully designed in the previous steps, as well as the characteristics of the amplified signal.

[0201] The receiving end is equipped with multiple resonant circuits corresponding to the transmitting end frequency, each circuit is optimized for a specific frequency (i.e. f1, f2, ..., f n ), by adjusting the values of capacitors and inductors to ensure that the circuit resonant frequency matches the frequency of the received electromagnetic wave, thereby achieving maximum energy transfer.

[0202] Adjust the Q value of the resonant circuit to improve the energy absorption efficiency, which is expressed as:

[0203]

[0204] Where, ∝ is proportional to; f i is the transmission frequency; η abs，r is the absolute efficiency of the rth resonant circuit, an indicator of energy conversion efficiency, that is, how much input energy is effectively used rather than lost; f 0，r is the center frequency of the rth resonant circuit, that is, the frequency point at which the resonant circuit is designed to work optimally; Q r is the quality factor of the rth resonant circuit; the higher the quality factor, the better the selectivity of the circuit, the more effective it is in amplifying signals near the center frequency, and the better it suppresses signals far from the center frequency; a high Q value usually means higher working efficiency; f BW，r is the bandwidth of the rth resonant circuit, which refers to the width of the frequency range to which the circuit can effectively respond. In a resonant system, the bandwidth is usually related to the degree of deviation from the center frequency, and a narrow bandwidth means higher selectivity.

[0205] The above describes the absolute efficiency η abs，rand frequency offset f i -f o，r and quality factor Q r and bandwidth f BW，r is closely related; specifically, when the actual operating frequency f i Close to the resonant center frequency f 0，r , and the bandwidth f BW，r Relatively small (i.e. good selectivity), and the quality factor Q r When it is high, the expression in the denominator will be small, causing the value of the entire fraction to be close to 1, thus making the efficiency η abs，r On the contrary, if the frequency deviates far from the center frequency, the bandwidth is wide, or the quality factor is low, the efficiency is reduced.

[0206] When a resonant circuit resonates with an electromagnetic wave, it effectively absorbs the electromagnetic energy and converts it into current in the circuit; this current is then sent to an energy converter (such as a rectifier bridge or DC-DC converter) and converted into direct current power to supply the load or for storage.

[0207] The AC power absorbed in the resonant circuit is converted into DC power through a rectifier bridge circuit, which can be expressed as:

[0208]

[0209] Where η conv is the conversion efficiency; P DC is the converted DC power; P AC is the absorbed AC power.

[0210] The above describes the energy conversion efficiency from AC to DC, the resonant circuit is optimized by the quality factor Q r Improved energy absorption efficiency η abs,r The absorbed AC power is then converted into DC power through circuits such as a rectifier bridge, which directly depends on the quality of the AC power absorbed in the previous step, reflecting the effective conversion process from high-frequency electromagnetic energy to usable DC power.

[0211] Sensors are used to monitor receiving efficiency and environmental changes, and the resonant frequency and circuit parameters are adjusted in real time to maintain the optimal receiving state. A communication link is established with the transmitter to exchange receiving status information and further optimize the transmission parameters. The DC power after energy conversion becomes the basis for evaluating system performance. Through sensor monitoring and feedback mechanisms, the receiver can adjust the resonant frequency and circuit parameters according to the actual conversion efficiency, and even communicate with the transmitter to optimize the transmission parameters. This feedback loop ensures that the system can dynamically adapt to environmental changes and continuously optimize energy transmission efficiency.

[0212] Step S6: The receiving end integrates the energy received from each frequency channel and combines it, and intelligently distributes the combined power to the target load according to the load demand to ensure stable power supply.

[0213] Step S6 specifically includes:

[0214] The DC power obtained by each channel after energy conversion is P DC，1 , P DC，2 ,...,P DC，n , power combining aggregates these independent DC powers into a single output power P total , expressed as:

[0215]

[0216] Power combining can be achieved in a variety of ways, such as using DC-DC converters (e.g., parallel Buck or Boost converters) or simple parallel circuits to maintain voltage matching and current balance, avoiding power loss or safety issues caused by mismatch.

[0217] Evaluate the instantaneous power demand of the target load by monitoring the voltage V load and current I load , calculate the required power P Load =V load ×I Load .

[0218] According to P Load and currently available P total , proportional allocation is performed, that is, the power contribution of each channel is consistent with the total power demand, or allocation is performed taking into account factors such as efficiency maximization, battery charge status (such as in energy storage applications), load priority, etc.

[0219] Monitor the actual power P distributed to the load actual and demand P Load The output of each channel is adjusted accordingly to ensure that the load is stably powered.

[0220] The subsequent system continuously monitors actual transmission efficiency and environmental changes, feeding real-time data into the intelligent scheduling algorithm. It dynamically adjusts parameters such as operating frequency and power allocation as needed, forming a closed-loop control to maintain optimal transmission efficiency. Specifically, it includes:

[0221] Use sensors (such as power meters and efficiency measurement modules) to monitor the actual wireless power transmission efficiency in real time and evaluate the conversion and utilization efficiency of energy from the transmitter to the receiver; including but not limited to changes in distance, the presence or removal of obstacles, and changes in temperature and humidity, which may affect the efficiency and stability of wireless power transmission.

[0222] The real-time data collected includes transmission efficiency, changes in environmental parameters, load status (such as changes in power demand), etc., which are aggregated to the central processor or cloud platform through the data acquisition system.

[0223] After receiving the new data, the intelligent scheduling algorithm reruns and recalculates the optimal frequency combination, power allocation, and other parameters based on the current environmental conditions, previous transmission efficiency data, and the new load requirements. This involves complex optimization algorithms such as genetic algorithms, particle swarm optimization, or machine learning models to find the optimal solution under the current conditions.

[0224] Based on the results of the algorithm recalculation, the system dynamically adjusts the operating frequency, power allocation scheme, and even the resonant circuit parameters (if necessary) of the transmitter; this includes adjusting the output of the transmit frequency synthesizer and power amplifier, and notifying the receiver to adjust the resonant circuit.

[0225] After adjustments are made, the system enters the monitoring phase again, forming a continuous closed-loop feedback loop. New efficiency and environmental data is collected and compared with the adjusted system performance to assess the effectiveness of the adjustments. If the expected performance is not achieved, the system triggers the adjustment process again until transmission efficiency is maximized.

[0226] Example 2

[0227] like Figure 2 As shown, the present invention discloses a wireless power transmission system, the system comprising:

[0228] The environmental parameter collection module 10 is used to collect environmental parameters in the wireless power transmission path.

[0229] The transmission efficiency prediction module 20 is used to construct a prediction model to predict the transmission efficiency of different frequency channels.

[0230] The power solution allocation module 30 is used to calculate the optimal multi-frequency resonance frequency combination and the corresponding power allocation solution.

[0231] The electrical component adjustment module 40 is used for dynamically adjusting the values of the inductor and capacitor components at the transmitting end according to the optimal operating frequency combination and the adaptive tuning network.

[0232] The transmitting and receiving matching module 50 is used for the transmitting end to transmit an alternating electromagnetic field to the receiving end according to the power allocation scheme; the receiving end captures the alternating electromagnetic field emitted by the transmitting end and absorbs energy through the resonant circuit.

[0233] The energy intelligent distribution module 60 is used to integrate the energy received from each frequency channel at the receiving end and combine it, and intelligently distribute the combined electric energy to the target load according to the load demand.

[0234] As an optional implementation manner, the transmission efficiency prediction module 20 of the present invention specifically includes:

[0235] Define the feature extraction subnetwork, which takes an input layer and a list indicating the number of neurons in each layer, and then creates a series of fully connected layers, each using the ReLU activation function.

[0236] Define the attention mechanism layer, convert the input into a probability distribution through a fully connected layer (using Softmax activation); multiply these probabilities with the original input through a dot product operation.

[0237] Build a prediction model, create multiple input layers based on the input shape; create a feature extraction subnetwork for each input layer; use a concatenation layer to merge the outputs of all subnetworks into a single vector; add an attention mechanism layer to dynamically emphasize the feature combination that the model focuses on; use the concatenation layer again to combine the original merged features with the features processed by the attention mechanism; add a fully connected layer to further process these features; create an output layer for each frequency channel, use the Sigmoid activation function, and each output corresponds to the prediction of a frequency channel; compile the model, and the loss function is the mean squared error.

[0238] As an optional implementation manner, the power scheme allocation module 30 of the present invention specifically includes:

[0239] Initialize the deep reinforcement learning agent class for learning and decision-making on the optimal configuration of frequency and power; define the state space size; define the action space size; use a double-ended queue to set up an experience replay buffer to store the data of each interaction; set the discount factor, initial exploration rate, minimum exploration rate, exploration rate decay rate, and learning rate; and build a deep Q network model, which is implemented by the construction function.

[0240] Construct a neural network function for estimating the value of an action. The input is the current environment state (such as transmission distance and attenuation), and the output is the value of each action. The model consists of three fully connected layers. The first and second layers use the ReLU activation function to increase nonlinear processing capabilities. The last layer uses a linear activation function to output the expected value of each action (i.e., Q value).

[0241] Save the agent's experience (state, action, reward) to the memory replay, that is, store the data of an interaction in the experience replay buffer.

[0242] The action selected according to the current strategy (exploration or exploitation) is randomly selected when the random number is less than the current exploration rate; otherwise, the action corresponding to the maximum Q value predicted by the model is selected as the expected optimal action.

[0243] Randomly extract a batch of experiences from memory for learning, and optimize the network weights by updating the target Q value to minimize the mean square error between the predicted Q value and the target Q value. Specifically,

[0244] A certain batch of samples are randomly extracted from the experience replay buffer; the samples contain the state of the past exploration, the actions performed, the rewards obtained, the next state, and information about whether it has ended.

[0245] Each extracted sample is processed in a loop to unpack each element in the sample.

[0246] Calculate the target value. If this experience is a terminal state, the target value is the immediate reward obtained; if it is not a terminal state, the target value is the discounted value of the immediate reward plus the expected future reward.

[0247] Use the current model to predict the Q-values of all actions in a given state.

[0248] Assign the calculated target value to the Q value corresponding to the action taken.

[0249] Using the updated target Q value as the label, the model is trained once and its weights are optimized to reduce the gap between the predicted Q value and the target Q value.

[0250] Update the initial exploration rate.

[0251] Outputting an instruction set including a selected multi-frequency resonant frequency combination and a corresponding power allocation plan.

[0252] As an optional embodiment, the electrical component adjustment module 40 of the present invention specifically includes:

[0253] After calculating the best frequency combination, the command is sent to the controller of the tuning network through the digital interface. The command contains the target frequency f target ; The controller is based on the target frequency f target , calculate the current capacitance C current or inductor L current Adjust the amount to match the target frequency.

[0254] Based on the calculated adjustment amount, the controller dynamically adjusts the physical parameters of the variable capacitor or variable inductor through motor control or electronic drive to achieve the calculated capacitance C new or inductor L new value.

[0255] Use a high-frequency precision frequency sensor to monitor the tuned resonant frequency f measured and with the target frequency f target By comparison, if there is a deviation e, the capacitance or inductance is adjusted through a closed-loop control algorithm, which is expressed as:

[0256]

[0257] Where u(t) is the control signal; K b , K i and K d are the proportional, integral and derivative gains respectively.

[0258] As an optional implementation, the transmitting and receiving matching module 50 of the present invention specifically includes:

[0259] According to the calculated frequency and the corresponding power distribution, for each calculated frequency f α , the frequency synthesizer generates the corresponding alternating current signal, which is expressed as:

[0260]

[0261] Where, I α (t) is the current of the αth frequency component; A α is the amplitude; φ α is the phase angle; the calculated frequencies are f1, f2, ..., f n ; Power distribution is P allocated,1 , P allocated,2 ,...,P allocated,n .

[0262] According to the power distribution scheme, the current signal of each frequency is amplified. Considering the amplification efficiency, the actual output power is expressed as:

[0263] P out，c =η PA,c ·P allocated,c

[0264] Where, P out，c is the actual output power of the cth channel after amplification; η PA,c is the power amplifier efficiency of the cth channel.

[0265] Pre-distortion technology is used before transmission. The receiving end is equipped with multiple resonant circuits corresponding to the transmitting end frequency. Each circuit is optimized for a specific frequency. By adjusting the values of capacitors and inductors, the circuit resonant frequency is ensured to match the frequency of the received electromagnetic wave.

[0266] Adjust the Q value of the resonant circuit to improve the energy absorption efficiency, which is expressed as:

[0267]

[0268] Where, ∝ is proportional to; f i is the transmission frequency; η abs，r is the absolute efficiency of the rth resonant circuit; f0，r is the center frequency of the rth resonant circuit; Q r is the quality factor of the rth resonant circuit; f BW，r is the bandwidth of the rth resonant circuit.

[0269] The AC power absorbed in the resonant circuit is converted into DC power through a rectifier bridge circuit, which can be expressed as:

[0270]

[0271] Where η conv is the conversion efficiency; P DC is the converted DC power; P AC is the absorbed AC power.

[0272] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A wireless power transmission method, characterized in that: The method comprises: Step S1: collecting environmental parameters in the wireless power transmission path; Step S2: construct a prediction model to predict the expected transmission efficiency under different frequency channels; Step S3: Calculating the optimal multi-frequency resonance frequency combination and the corresponding power distribution scheme; Step S4: The transmitter dynamically adjusts the values of the inductor and capacitor components according to the optimal operating frequency combination and the adaptive tuning network; Step S5: The transmitting end transmits an alternating electromagnetic field to the receiving end according to the power allocation scheme; the receiving end captures the alternating electromagnetic field emitted by the transmitting end and absorbs energy through the resonant circuit; Step S6: The receiving end integrates the energy received from each frequency channel and combines it, and intelligently distributes the combined energy to the target load according to the load demand; The transmitting end transmits an alternating electromagnetic field to the receiving end according to the power allocation scheme; the receiving end captures the alternating electromagnetic field emitted by the transmitting end and absorbs energy through a resonant circuit, specifically including: According to the calculated frequency and the corresponding power distribution, for each calculated frequency , the frequency synthesizer generates the corresponding alternating current signal, which is expressed as: ； Where, For the A current with frequency components; is the amplitude; is the phase angle; the calculated frequency is ; Power distribution is ; According to the power distribution scheme, the current signal of each frequency is amplified. Considering the amplification efficiency, the actual output power is expressed as: ； Where, For the The actual output power of each channel after amplification; For the Power amplifier efficiency of each channel; Pre-distortion technology is used before transmission. The receiving end is equipped with multiple resonant circuits corresponding to the transmitting end frequency. Each circuit is optimized for a specific frequency. By adjusting the values of capacitors and inductors, the resonant frequency of the circuit is ensured to match the frequency of the received electromagnetic wave. Adjust the Q value of the resonant circuit to improve the energy absorption efficiency, which is expressed as: ； Where, is proportional to; is the transmitting frequency; For the The absolute efficiency of a resonant circuit; For the The center frequency of the resonant circuit; For the The quality factor of a resonant circuit; For the The bandwidth of a resonant circuit; The AC power absorbed in the resonant circuit is converted into DC power through a rectifier bridge circuit, which can be expressed as: ； Where, is the conversion efficiency; is the converted DC power; is the absorbed AC power.

2. The wireless power transmission method according to claim 1, wherein: The construction of the prediction model to predict the expected transmission efficiency under different frequency channels specifically includes: Define a feature extraction subnetwork that takes an input layer and a list indicating the number of neurons in each layer, and then creates a series of fully connected layers, each using a ReLU activation function. Define an attention mechanism layer, convert the input into a probability distribution through a fully connected layer; multiply these probabilities with the original input through a dot product operation; Build a prediction model, create multiple input layers based on the input shape; create a feature extraction subnetwork for each input layer; use a concatenation layer to merge the outputs of all subnetworks into a single vector; add an attention mechanism layer to dynamically emphasize the feature combination that the model focuses on; use the concatenation layer again to combine the original merged features with the features processed by the attention mechanism; add a fully connected layer to further process these features; create an output layer for each frequency channel, use the Sigmoid activation function, and each output corresponds to the prediction of a frequency channel; compile the model, and the loss function is the mean squared error.

3. The wireless power transmission method according to claim 1, wherein: The calculation of the optimized multi-frequency resonant frequency combination and the corresponding power distribution scheme specifically includes: Initialize the deep reinforcement learning agent class for learning and decision-making on the optimal configuration of frequency and power; define the state space size; define the action space size; use a double-ended queue to set up the experience replay buffer to store each interaction data; set the discount factor, initial exploration rate, minimum exploration rate, exploration rate decay rate, and learning rate; build the deep Q network model, which is implemented by the construction function; Build a neural network function for estimating action values. The input is the current environment state, and the output is the value of each action. The model consists of three fully connected layers. The first and second layers use the ReLU activation function to increase nonlinear processing capabilities. The last layer uses a linear activation function to output the expected value of each action, that is, the Q value. Save the agent's experience to memory replay, that is, store the data of an interaction in the experience replay buffer; According to the action selected by the current strategy, when the random number is less than the current exploration rate, the action is randomly selected; otherwise, the action corresponding to the maximum Q value predicted by the model is selected as the expected optimal action; Randomly extract a batch of experiences from memory for learning, and optimize the network weights by updating the target Q value to minimize the mean square error between the predicted Q value and the target Q value. Specifically, Randomly extract a certain number of samples from the experience replay buffer; the samples contain the state of the past exploration, the actions performed, the rewards obtained, the next state, and whether it has ended; Loop through each extracted sample and unpack each element in the sample; Calculate the target value. If this experience is a terminal state, the target value is the immediate reward obtained; if it is not a terminal state, the target value is the discounted value of the immediate reward plus the expected future reward; Use the current model to predict the Q-values of all actions in a given state; Assign the calculated target value to the Q value corresponding to the action taken; Use the updated target Q value as the label to train the model once and optimize its weights to reduce the gap between the predicted Q value and the target Q value; Update the initial exploration rate; Outputting an instruction set including a selected multi-frequency resonant frequency combination and a corresponding power allocation plan.

4. The wireless power transmission method according to claim 1, wherein: The transmitter dynamically adjusts the values of the inductor and capacitor components according to the optimal operating frequency combination and the adaptive tuning network, specifically including: After calculating the best frequency combination, the command is sent to the controller of the tuning network through the digital interface. The command contains the target frequency. ; The controller is based on the target frequency , calculate the current capacitance or inductor The amount of adjustment to match the target frequency; Based on the calculated adjustment amount, the controller dynamically adjusts the physical parameters of the variable capacitor or variable inductor through motor control or electronic drive to achieve the calculated capacitance. or inductor value; Use a high-frequency precision frequency sensor to monitor the tuned resonant frequency and with the target frequency Compare, if there is a deviation , the capacitance or inductance is adjusted through a closed-loop control algorithm, which is expressed as: ； Where, is the control signal; 、 and are the proportional, integral and derivative gains respectively.

5. A wireless power transmission system, characterized in that: The system comprises: An environmental parameter collection module, used to collect environmental parameters in the wireless power transmission path; The transmission efficiency prediction module is used to build a prediction model to predict the transmission efficiency of different frequency channels; Power scheme allocation module, used to calculate the optimized multi-frequency resonance frequency combination and the corresponding power allocation scheme; An electrical component adjustment module is used to dynamically adjust the values of inductance and capacitance components at the transmitter based on the optimal operating frequency combination and adaptive tuning network; The transmitter-receiver matching module is used for the transmitter to transmit an alternating electromagnetic field to the receiver according to the power allocation scheme; the receiver captures the alternating electromagnetic field emitted by the transmitter and absorbs energy through the resonant circuit; The energy intelligent distribution module is used to integrate the energy received from each frequency channel at the receiving end and intelligently distribute the combined electric energy to the target load according to the load demand; The transmitting and receiving matching module specifically includes: According to the calculated frequency and the corresponding power distribution, for each calculated frequency , the frequency synthesizer generates the corresponding alternating current signal, which is expressed as: ； Where, For the A current with frequency components; is the amplitude; is the phase angle; the calculated frequency is ; Power distribution is ; According to the power distribution scheme, the current signal of each frequency is amplified. Considering the amplification efficiency, the actual output power is expressed as: ； Where, For the The actual output power of each channel after amplification; For the Power amplifier efficiency of each channel; Pre-distortion technology is used before transmission. The receiving end is equipped with multiple resonant circuits corresponding to the transmitting end frequency. Each circuit is optimized for a specific frequency. By adjusting the values of capacitors and inductors, the resonant frequency of the circuit is ensured to match the frequency of the received electromagnetic wave. Adjust the Q value of the resonant circuit to improve the energy absorption efficiency, which is expressed as: ； Where, is proportional to; is the transmitting frequency; For the The absolute efficiency of a resonant circuit; For the The center frequency of the resonant circuit; For the The quality factor of a resonant circuit; For the The bandwidth of a resonant circuit; The AC power absorbed in the resonant circuit is converted into DC power through a rectifier bridge circuit, which can be expressed as: ； Where, is the conversion efficiency; is the converted DC power; is the absorbed AC power.

6. The wireless power transmission system according to claim 5, wherein: The transmission efficiency prediction module specifically includes: Define a feature extraction subnetwork that takes an input layer and a list indicating the number of neurons in each layer, and then creates a series of fully connected layers, each using a ReLU activation function. Define an attention mechanism layer, convert the input into a probability distribution through a fully connected layer; multiply these probabilities with the original input through a dot product operation; Build a prediction model, create multiple input layers based on the input shape; create a feature extraction subnetwork for each input layer; use a concatenation layer to merge the outputs of all subnetworks into a single vector; add an attention mechanism layer to dynamically emphasize the feature combination that the model focuses on; use the concatenation layer again to combine the original merged features with the features processed by the attention mechanism; add a fully connected layer to further process these features; create an output layer for each frequency channel, use the Sigmoid activation function, and each output corresponds to the prediction of a frequency channel; compile the model, and the loss function is the mean squared error.

7. The wireless power transmission system according to claim 5, wherein: The power solution allocation module specifically includes: Initialize the deep reinforcement learning agent class for learning and decision-making on the optimal configuration of frequency and power; define the state space size; define the action space size; use a double-ended queue to set up the experience replay buffer to store each interaction data; set the discount factor, initial exploration rate, minimum exploration rate, exploration rate decay rate, and learning rate; build the deep Q network model, which is implemented by the construction function; Build a neural network function for estimating action values. The input is the current environment state, and the output is the value of each action. The model consists of three fully connected layers. The first and second layers use the ReLU activation function to increase nonlinear processing capabilities. The last layer uses a linear activation function to output the expected value of each action, that is, the Q value. Save the agent's experience to memory replay, that is, store the data of an interaction in the experience replay buffer; According to the action selected by the current strategy, when the random number is less than the current exploration rate, the action is randomly selected; otherwise, the action corresponding to the maximum Q value predicted by the model is selected as the expected optimal action; Randomly extract a batch of experiences from memory for learning, and optimize the network weights by updating the target Q value to minimize the mean square error between the predicted Q value and the target Q value. Specifically, Randomly extract a certain number of samples from the experience replay buffer; the samples contain the state of the past exploration, the actions performed, the rewards obtained, the next state, and whether it has ended; Loop through each extracted sample and unpack each element in the sample; Calculate the target value. If this experience is a terminal state, the target value is the immediate reward obtained; if it is not a terminal state, the target value is the discounted value of the immediate reward plus the expected future reward; Use the current model to predict the Q-values of all actions in a given state; Assign the calculated target value to the Q value corresponding to the action taken; Use the updated target Q value as the label to train the model once and optimize its weights to reduce the gap between the predicted Q value and the target Q value; Update the initial exploration rate; Outputting an instruction set including a selected multi-frequency resonant frequency combination and a corresponding power allocation plan.

8. The wireless power transmission system according to claim 5, wherein: The electrical component adjustment module specifically includes: After calculating the best frequency combination, the command is sent to the controller of the tuning network through the digital interface. The command contains the target frequency. ; The controller is based on the target frequency , calculate the current capacitance or inductor The amount of adjustment to match the target frequency; Based on the calculated adjustment amount, the controller dynamically adjusts the physical parameters of the variable capacitor or variable inductor through motor control or electronic drive to achieve the calculated capacitance. or inductor value; Use a high-frequency precision frequency sensor to monitor the tuned resonant frequency and with the target frequency Compare, if there is a deviation , the capacitance or inductance is adjusted through a closed-loop control algorithm, which is expressed as: ； Where, is the control signal; 、 and are the proportional, integral and derivative gains respectively.

Images