A fast-response ultrasonic transducer resonant frequency tracking system and device

Abstract

The application relates to a fast-response ultrasonic transducer resonance frequency tracking system and device, and belongs to the technical field of ultrasonic frequency tracking. The system comprises a full-bridge inverter module with a power switching device of SiC-MOSFET, a signal acquisition and processing module for collecting and processing voltage phase difference and current phase difference, and a microcontroller module for training a machine learning model, building a response time model to predict a target frequency and system response time, and controlling and tracking the transducer frequency according to the phase difference. The application uses SiC-MOSFET with faster switching speed to replace the traditional IGBT as the power switching device of the inverter circuit, uses the machine learning model and the response time model to respectively predict the target frequency and the system response time, reacts in advance, and uses the real-time phase difference calculation method and the fast-response control strategy to realize the fast response and fast tracking of the ultrasonic transducer resonance frequency in metal welding, and the fast response speed of the ultrasonic transducer resonance frequency in metal welding reaches the microsecond level.

Description

Technical Field

[0001] This invention relates to the field of ultrasonic frequency tracking technology, and in particular to a fast-response ultrasonic transducer resonant frequency tracking system and device. Background Technology

[0002] Ultrasonic metal welding technology is considered an environmentally friendly and efficient welding technique with broad application prospects in fields such as automotive battery cell welding and electronic device assembly. This technology utilizes an ultrasonic power source to generate high-frequency electrical energy, which is then converted into high-frequency vibrational mechanical energy by a transducer, thereby achieving microscopic welding of the metal components. However, in actual operation, variations in welding load and wear and tear on the components themselves can cause changes in the transducer's impedance characteristics, leading to a shift in the system's resonant frequency. This results in unstable output power from the ultrasonic transducer, affecting the welding effect. Therefore, the frequency tracking function of the ultrasonic power source is crucial.

[0003] Existing ultrasonic power frequency tracking technology mainly adjusts the phase difference between the voltage and current across the transducer by changing the driving frequency, so that the transducer operates near the resonant frequency.

[0004] With advancements in frequency tracking technology, the speed of frequency tracking algorithms has increased to the millisecond or even microsecond level. At this point, the factors affecting system frequency tracking performance are no longer limited by the frequency tracking algorithm itself, but primarily focus on the response latency of the entire system's hardware circuitry. If the system's hardware response speed is too slow, even with a fast tracking algorithm, the entire system cannot achieve its optimal performance. Summary of the Invention

[0005] In view of this, it is necessary to provide a fast-response ultrasonic transducer resonant frequency tracking system and device to solve the technical problem of poor resonant frequency tracking speed caused by the slow hardware response speed of ultrasonic transducers in metal welding in the prior art.

[0006] To address the above problems, this invention provides a fast-response ultrasonic transducer resonant frequency tracking system, comprising:

[0007] The full-bridge inverter module and transducer are interconnected, and the power switching device of the full-bridge inverter module is SiC-MOSFET;

[0008] The signal acquisition and processing module connected to the transducer is used to acquire the voltage signal and current signal of the transducer, determine the voltage phase difference between the voltage signal and the preset reference signal, and the current phase difference between the current signal and the preset reference signal, wherein the preset reference signal is a signal with the same angular frequency as the voltage signal and the current signal;

[0009] The microcontroller module connected to the signal acquisition and processing module is used to train a pre-built machine learning model to obtain a first target model, taking the operating frequency, input power, equivalent impedance of the transducer, and load parameters as inputs and the target frequency of the transducer as outputs; it is also used to train a pre-built machine learning model to obtain a second target model, taking the driving frequency as input and the system response time as output; and to control the full-bridge inverter module based on the system response time and target frequency obtained from the first and second target models to achieve rapid tracking of the resonant frequency.

[0010] In some possible implementations, the signal acquisition and processing module includes a voltage signal acquisition and processing module, a current signal acquisition and processing module, and a judgment module, wherein,

[0011] The voltage signal acquisition and processing module is used to obtain the voltage phase difference between the transducer's voltage signal and the reference signal based on the real-time phase difference calculation method.

[0012] The current signal acquisition and processing module is used to obtain the current phase difference between the transducer's current signal and the reference signal based on the real-time phase difference calculation method.

[0013] The judgment module is used to compare the absolute value of the difference between the voltage phase difference and the current phase difference with a preset error threshold. If the absolute value is greater than the preset error threshold, the microcontroller module is activated to control the transducer frequency. If the absolute value is less than the preset error threshold, no action is taken.

[0014] In some possible implementations, the microcontroller module divides the system response time into six stages: the time for the drive signal to undergo phase-shift control processing, the time for the gate isolation drive module to process the signal, the switching delay time of the switching devices in the full-bridge inverter circuit, the transducer response time, the signal processing response time, and the time from microcontroller calculation and processing to issuing the drive signal. The microcontroller module establishes and trains a prediction model for these six stages to predict the relationship between the drive frequency and the system response time. Based on the target frequency obtained through machine learning, the system delay time is predicted, allowing the microcontroller to issue the drive signal in advance, thereby achieving a fast response of the resonant frequency tracking system.

[0015] In some possible implementations, it also includes: a drive signal generation module, the input of which is connected to the output of the microcontroller module, and the output of which is connected to the full-bridge inverter module through the power switching device;

[0016] The drive signal generation module is used to generate drive signals based on the control signals of the microcontroller module to control the opening and closing of the power switching device.

[0017] In some possible implementations, the drive signal generation module includes an isolated power supply module, a gate drive module, and a phase shift control module;

[0018] The phase-shift control module has its input terminal connected to the output terminal of the microcontroller module, the gate drive module has its input terminal connected to the output terminal of the phase-shift control module, the output terminal of the gate drive module has its output terminal connected to the input terminal of the full-bridge inverter module, and the output terminal of the isolation power supply module has its input terminal connected to the gate drive module. The isolation power supply module is used to isolate and protect the gate drive module, and the phase-shift control module is used to control the multiple outputs of the microcontroller module to enable and disable triggering.

[0019] In some possible implementations, the isolated power supply module employs magnetic coupling isolation.

[0020] In some possible implementations, a filtering and rectifying module is also included, which receives three-phase power and inputs the three-phase power into the full-bridge inverter module after passing through the filtering and rectifying module.

[0021] In some possible implementations, the filtering and rectifying module includes an interconnected lightning protection filtering module and a three-phase rectifier module.

[0022] In some possible implementations, an impedance matching module is also included between the full-bridge inverter module and the transducer to block signal reflection from the transducer.

[0023] The present invention also provides an ultrasonic transducer resonant frequency tracking device, including the ultrasonic transducer resonant frequency tracking system described above.

[0024] The beneficial effects of this invention are:

[0025] The system provided in this application includes a full-bridge inverter module with SiC-MOSFET power switching devices, a signal acquisition and processing module for acquiring and processing voltage and current phase differences, and a microcontroller module for training a machine learning model to predict the target frequency, constructing and training the model to predict the system response time, and controlling and tracking the transducer frequency based on the phase difference. This application achieves rapid response and tracking of the resonant frequency of the metal-welded ultrasonic transducer by using faster-switching SiC-MOSFETs instead of traditional IGBTs as the power switching devices in the inverter circuit, predicting the target frequency and system response time through machine learning and response time prediction models, reacting in advance, and employing real-time phase difference calculation methods and fast-response control strategies. This enables the frequency tracking algorithm to achieve microsecond-level speeds. Attached Figure Description

[0026] Figure 1A system architecture diagram of an embodiment of the fast-response ultrasonic transducer resonant frequency tracking system provided by the present invention;

[0027] Figure 2 This is a schematic diagram illustrating the principle of the signal acquisition and processing module in this embodiment of the invention for processing voltage phase difference and current phase difference;

[0028] Figure 3 This is a schematic diagram of the transducer phase difference control process in an embodiment of the present invention;

[0029] Figure 4 This is a schematic diagram of the structure of the microcontroller, drive signal generation module, and full-bridge inverter module in an embodiment of the present invention;

[0030] Figure 5 This is a schematic diagram of the delay stages of the ultrasonic transducer resonant frequency tracking system in an embodiment of the present invention;

[0031] Figure 6 This is a flowchart of an embodiment of frequency tracking control of ultrasonic transducer resonance in this invention. Detailed Implementation

[0032] Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which form part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not intended to limit the scope of the present invention.

[0033] A specific embodiment of the present invention discloses a fast-response ultrasonic transducer resonant frequency tracking system, such as... Figure 1 As shown, it includes: a full-bridge inverter module and a transducer connected to each other, wherein the power switching device of the full-bridge inverter module is a SiC-MOSFET;

[0034] The signal acquisition and processing module connected to the transducer is used to acquire the voltage signal and current signal of the transducer, determine the voltage phase difference between the voltage signal and the preset reference signal, and the current phase difference between the current signal and the preset reference signal, wherein the preset reference signal is a signal with the same angular frequency as the voltage signal and the current signal;

[0035] The microcontroller module connected to the signal acquisition and processing module is used to train a pre-built machine learning model to obtain a first target model, taking the operating frequency, input power, equivalent impedance of the transducer, and load parameters as inputs and the target frequency of the transducer as outputs; it is also used to train a pre-built machine learning model to obtain a second target model, taking the driving frequency as input and the system response time as output; and to control the full-bridge inverter module based on the system response time and target frequency obtained from the first and second target models to achieve rapid tracking of the resonant frequency.

[0036] It should be noted that the first objective model is used to predict the target frequency, and the second objective model is used to predict the system response time. Furthermore, the machine learning models corresponding to the first and second objective models can be neural network models, convolutional neural network models, etc.

[0037] In practice, the full-bridge inverter module provides AC power at a specific frequency. The phase difference between the current and voltage signals at both ends of the transducer and the reference signal is obtained using a cross-correlation function. The microcontroller module generates a control signal based on the phase difference to achieve fast real-time control of the transducer.

[0038] Furthermore, the microcontroller module collects the transducer's operating frequency, input power, equivalent impedance, and load parameters to train a machine learning model, builds and trains a predictive model of the driving frequency and system response time, predicts the system response time and target frequency, and can react in advance and respond quickly.

[0039] Compared to existing technologies, this application replaces traditional IGBTs with faster-switching SiC-MOSFETs as the power switching devices in the inverter circuit. It uses a machine learning model to predict the transducer's target frequency, and then, based on a mathematical model of the driving frequency and system response time, predicts the system response time to react in advance. Furthermore, it employs a real-time phase difference calculation method and a fast-response control strategy to achieve rapid response and tracking of the resonant frequency of the metal-welded ultrasonic transducer, enabling the frequency tracking algorithm to reach microsecond-level speeds. Simultaneously, using SiC-MOSFETs increases the system's upper operating frequency limit and reduces energy loss and heat generation. The machine learning model allows the system to shorten response time outside of the frequency tracking algorithm. This device is designed considering the overall system response time, better showcasing the speed of existing frequency tracking algorithms.

[0040] In a further preferred embodiment, the signal acquisition and processing module includes a voltage signal acquisition and processing module, a current signal acquisition and processing module, and a judgment module, wherein,

[0041] The voltage signal acquisition and processing module is used to obtain the voltage phase difference between the transducer's voltage signal and the reference signal based on the real-time phase difference calculation method.

[0042] The current signal acquisition and processing module is used to obtain the current phase difference between the transducer's current signal and the reference signal based on the real-time phase difference calculation method.

[0043] The judgment module is used to compare the absolute value of the difference between the voltage phase difference and the current phase difference with a preset error threshold. If the absolute value is greater than the preset error threshold, the microcontroller module is activated to control the transducer frequency. If the absolute value is less than the preset error threshold, no action is taken.

[0044] It should be noted that there are many methods for calculating phase difference with a short computation time, and no specific method is used here. The example of calculating phase difference using the cross-correlation function is given.

[0045] In a specific implementation, a reference signal is introduced when calculating the phase to reduce the influence of noise. S ( t ), and compare them with the sampling voltage respectively. U ( t and sampling current signal I ( t The calculation is performed as follows: Figure 2 and Figure 3 As shown.

[0046] Reference signal S ( t ), sampling voltage U ( t and sampling current signal I ( t The expression is as follows:

[0047]

[0048]

[0049]

[0050] in, m 1( t )and m 2( t ) represent the noise of the sampled voltage signal and the sampled current signal, respectively.

[0051] Let A1 be S ( t )and U ( t The cross-correlation function of ), where A2 is S ( t )and I ( t If the cross-correlation function of A1 and A2 is given, then A1 and A2 are:

[0052]

[0053]

[0054] in,S The amplitude of the reference signal, I The amplitude of the sampled current signal. U This represents the amplitude of the sampled voltage signal.

[0055] Let the sampling frequency be... M Then the discrete expressions for A1 and A2 are:

[0056]

[0057]

[0058] in, K The value of the sampling point. N This represents the total number of sampling points.

[0059] Based on the values ​​calculated after sampling, A1 and A2 are obtained. Then, the voltage phase difference between the sampled voltage and the sampled current is... and current phase difference They are respectively:

[0060]

[0061]

[0062] The judgment module is used to calculate the voltage phase difference. and current phase difference The absolute value of the difference Compared with the preset error threshold If the error exceeds the preset error threshold, a comparison will be made. Then it is controlled by the microcontroller module. If the error is less than the preset error threshold... If no action is taken, the phase difference can be kept within the allowable error range.

[0063]

[0064] As can be seen from the above formula, in the process of calculating the phase difference, the noise signal m 1( t )and m 2( t The impact of strong electromagnetic radiation is relatively small and can be ignored. Therefore, this method can effectively reduce the influence of strong electromagnetic radiation and improve the anti-interference capability of the device. Furthermore, since this application only requires calculating two inverse trigonometric functions to determine the phase difference between the voltage and current across the transducer, the calculation is simple, consumes less computing resources of the microcontroller, and improves the computing speed.

[0065] In some embodiments of the present invention, the microprocessor module includes a training set acquisition module, which is used to acquire the transducer's operating frequency, input power, equivalent impedance of the transducer, and load parameters, and to establish a training set.

[0066] In some embodiments of the present invention, the system further includes: a drive signal generation module, wherein the input of the drive signal generation module is connected to the output of the microcontroller module, and the output of the drive signal generation module is connected to the full-bridge inverter module through the power switching device;

[0067] The drive signal generation module is used to generate drive signals based on the control signals of the microcontroller module to control the opening and closing of the power switching device.

[0068] Specifically, this embodiment of the invention uses a target frequency prediction method based on a machine learning model. Please refer to [link / reference needed]. Figure 6 The target frequency is the resonant frequency of the transducer under fixed conditions at a certain moment.

[0069] Let the dataset for machine learning be (x i ,y i ), where x i =[x i1, x i2, x i3, x i4 ] is the input feature vector, y i This is the target frequency. The input layer here consists of four parameters: the transducer's operating frequency, input power, equivalent impedance, and load parameters.

[0070] Machine learning can build various models. For example, one type is the decision tree prediction model, which generates a tree structure by progressively splitting the data. Each node represents an object, branching paths represent possible attribute values, and leaf nodes correspond to specific target frequencies.

[0071] This model uses the mean squared error M. SE As a criterion for splitting:

[0072]

[0073]

[0074] In the formula |N i |For dataset N i Number of in-sample points; y ik Represents dataset N i The delay time of the k-th sample. Represents dataset N i The average of the predictions for all samples; Fj The feature set is randomly selected each time a node is split; t represents the threshold of the feature, which is used to determine the split.

[0075] For the new input feature vector a i0 The prediction is made using this model, expressed as follows:

[0076] y io =f(a i0 )

[0077] In the formula y io This represents the prediction result of the decision tree model for the target frequency.

[0078] Based on the input information, the system predicts the time delay required before the transducer operates and reacts in advance. With known input, the system can predict the frequency the device needs to change based on the constructed model, and control the drive signal in advance based on the predicted delay time, allowing the transducer to enter resonance more quickly when it starts operating.

[0079] It should be noted that common control strategies are used to obtain the target frequency during model training, while model prediction is used in actual work.

[0080] In a further preferred embodiment, a drive signal generation module is also included. The input of the drive signal generation module is connected to the output of the microcontroller module, and the output of the drive signal generation module is connected to the full-bridge inverter module through a power switching device, for generating a drive signal based on the control signal of the microcontroller module.

[0081] In a further preferred embodiment, the drive signal generation module of this application includes an isolated power supply module, a gate drive module, and a phase shift control module. The input terminal of the phase shift control module is connected to the output terminal of the controller module, the input terminal of the gate drive module is connected to the output terminal of the phase shift control module, the output terminal of the gate drive module is connected to the input terminal of the full-bridge inverter module via a power switching device, and the output terminal of the isolated power supply module is connected to the input terminal of the gate drive module. The phase shift control module mainly consists of a phase shift control chip and its peripheral circuitry. Alternatively, a microcontroller module can be used to directly generate four drive signals, which are then level-converted before being input to the gate drive module. The isolated power supply module is used to isolate and protect the gate drive module, and the phase shift control module is used to control the on / off triggering of the multiple outputs of the microcontroller module.

[0082] The power switching device of this application uses SiC-MOSFET. Compared with the traditional IGBT, the SiC-MOSFET of this application has a higher switching speed and a lower on-resistance. Therefore, the power switching device of this application can further reduce the delay response of the entire device.

[0083] The applicant further discovered that when using SiC-MOSFETs to replace IGBTs, the higher switching frequency and potentially higher voltage of SiC-MOSFETs necessitate stronger gate isolation. Additional protection circuitry may be required to prevent device damage and ensure long-term reliability. Furthermore, differences in characteristics such as gate capacitance and threshold voltage may necessitate adjustments to control and stability considerations. Therefore, the gate drive module circuitry of the power switching device must be redesigned.

[0084] The main functions of the gate drive module are to drive, isolate, and provide appropriate protection for power switching devices to ensure their efficient and safe operation. For example... Figure 4 As shown, the gate drive module circuit in this application includes: a signal input circuit, a fault handling circuit, and a power amplifier circuit. The output of the phase shift control module is connected to the signal input circuit, and after passing through the isolation power supply module, it is connected to the power switching device through the power amplifier circuit. The fault handling circuit ensures that the power switching device will not experience performance degradation or equipment damage due to gate drive problems during operation; it may include, for example, overcurrent protection, overvoltage protection, and overtemperature protection. The isolation power supply module provides safe isolation and effective drive signal transmission. The power amplifier circuit converts the received level signal and provides the current and voltage required to drive the device. In some possible implementations, the isolation power supply module employs magnetic coupling isolation.

[0085] In this circuit, one factor affecting response speed is the isolation method. Currently, there are three common isolation methods: optocoupler isolation, magnetic coupling isolation, and capacitive isolation. The appropriate isolation method should be selected based on the specific circumstances during implementation.

[0086] Optical coupling technology achieves isolation by transmitting light through a transparent insulating layer. Magnetic coupling isolation utilizes the current change at the isolation front end to cause a current change on the other side of the isolation via a coil. Capacitive isolation devices often employ on-offkeying (OOK) modulation, where the transmitter sends a high-frequency signal to represent one digital state, and no signal is sent to represent another. After signal conditioning, the signal is transmitted through a buffer. The applicant has researched these three isolation methods, as shown in Table 1 below.

[0087] Table 1: Performance Comparison of Three Isolation Methods

[0088]

[0089] Therefore, this application employs magnetic coupling isolation, which features short transmission delay, high reliability, and high data transmission rate, as the isolation method for the drive circuit. By using SiC-MOSFETs and their drive circuit, the faster switching speed of SiC-MOSFETs and the shorter gate drive delay further improve the overall device response speed.

[0090] In a further preferred embodiment, a filtering and rectification module is also included, through which the three-phase power is fed into the full-bridge inverter module.

[0091] In a further preferred embodiment, the filtering and rectifying module includes a lightning protection filter module and a three-phase rectifier module connected in sequence. The input terminal of the lightning protection filter module is connected to the received three-phase power, and the output terminal of the lightning protection filter module is connected to the three-phase rectifier module. The lightning protection filter module is used to prevent damage to the equipment caused by lightning strikes, electrostatic discharge, and other overvoltage events, and to protect the equipment from harmonic interference and other power quality problems in the power installation.

[0092] The response time of a transducer depends on the properties of the transducer itself. This device can process the acquired voltage and current signals to obtain the transducer's turn-on delay. Specifically, it tests the transducer under different frequencies and power changes while the microcontroller generates a reference signal for comparison. When the signal at both ends of the transducer changes, the time difference between the transducer and the reference signal is calculated, which is the transducer's response time.

[0093] In a further preferred embodiment, the fast-response ultrasonic transducer resonant frequency tracking system also includes an impedance matching module located between the full-bridge inverter module and the transducer to optimize energy transmission efficiency.

[0094] Based on the above design, the working process of the fast-response ultrasonic transducer resonant frequency tracking device of this application is as follows: The device is started, the transducer operates at its initial frequency by default, and the voltage signal across the transducer, the current signal flowing through the transducer, and the reference signal are collected. The collected signals are then sent to the microcontroller to calculate the phase difference between the voltage and the reference signal. φ 1 and the phase difference between the current and the reference signal φ 2; Judgment φ 1- φ 2. Is the value within the error range set? If yes, maintain the current frequency; if no, adjust the frequency through the control algorithm.

[0095] In summary, in the device of this application, if Figure 5As shown, the overall response time is divided into 6 stages. When the microcontroller module does not output a control signal, the phase-shift control module has already generated a drive signal. Therefore, the time when the phase-shift control module generates the signal is taken as the start. The time after phase-shift control processing is t01, the time after gate isolation drive module processing is t02, the switching delay time of the switching devices in the full-bridge inverter circuit is t03, the transducer response time is t04, the signal processing response time is t05, and the time from microcontroller calculation and processing to issuing the drive signal is t06.

[0096] A driving frequency and response time model is established and trained for the above six stages to predict the system delay time. Based on the model, the microcontroller sends the driving signal in advance, thereby realizing the rapid response of the ultrasonic transducer resonant frequency tracking system.

[0097] A second aspect of this application provides an ultrasonic transducer resonant frequency tracking device, which includes the ultrasonic transducer resonant frequency tracking system.

[0098] Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by a computer program instructing related hardware, and the program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.

[0099] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A fast response ultrasonic transducer resonance frequency tracking system, characterized by, The system comprises: a full-bridge inverter module and a transducer connected to each other, wherein the power switch device of the full-bridge inverter module is a SiC-MOSFET; a signal acquisition and processing module connected to the transducer, for acquiring a voltage signal and a current signal of the transducer, determining a voltage phase difference between the voltage signal and a preset reference signal, and a current phase difference between the current signal and the preset reference signal, wherein the preset reference signal is a signal with the same angular frequency as the voltage signal and the current signal; a microcontroller module connected to the signal acquisition and processing module, for taking the acquired working frequency, input power, equivalent impedance of the transducer, and parameters of the load as inputs, taking a target frequency of the transducer as an output, training a pre-constructed machine learning model to obtain a first target model; and for taking a driving frequency as an input, taking a system response time as an output, training a pre-constructed machine learning model to obtain a second target model; and for controlling the full-bridge inverter module based on the system response time and the target frequency obtained from the first target model and the second target model, so as to realize fast tracking of the resonant frequency; the signal acquisition and processing module comprises a voltage signal acquisition and processing module, a current signal acquisition and processing module, and a judgment module, wherein the voltage signal acquisition and processing module is configured to obtain the voltage phase difference between the voltage signal of the transducer and the reference signal based on a real-time phase difference calculation method; the current signal acquisition and processing module is configured to obtain the current phase difference between the current signal of the transducer and the reference signal based on the real-time phase difference calculation method; the judgment module is configured to compare an absolute value of the difference between the voltage phase difference and the current phase difference with a preset error threshold value, and if the absolute value is greater than the preset error threshold value, the microcontroller module is started to control the frequency of the transducer, and if the absolute value is less than the preset error threshold value, no processing is performed.

2. The fast responding ultrasonic transducer resonance frequency tracking system of claim 1, wherein, the microcontroller module comprises a training set acquisition module, which is configured to acquire the working frequency, input power, equivalent impedance of the transducer, and parameters of the load, and establish a training set.

3. The fast-response ultrasonic transducer resonant frequency tracking system according to claim 2, characterized in that the microcontroller module divides the response time of the system into six stages, which are, respectively, a time for driving signal processing through phase shift control, a time for gate isolation driving module processing, a switching delay time of the switch device in the full-bridge inverter circuit, a transducer response time, a signal processing response time, and a time for microcontroller calculation processing to driving signal output; the microcontroller module establishes and trains a prediction model to predict the relationship between the driving frequency and the system response time; and according to the target frequency obtained through machine learning, the delay time of the system is predicted, so that the microcontroller outputs the driving signal in advance, so as to realize fast response of the resonant frequency tracking system.

4. The fast responding ultrasonic transducer resonance frequency tracking system of claim 1, wherein, Further comprising: a driving signal generation module, wherein the input of the driving signal generation module is connected to the output of the microcontroller module, and the output of the driving signal generation module is connected to the full-bridge inverter module through the power switch device. The driving signal generation module is configured to generate a driving signal based on a control signal of the microcontroller module to control the on-off of the power switch device.

5. The fast responding ultrasonic transducer resonance frequency tracking system of claim 4, wherein, The driving signal generation module comprises an isolation power supply module, a gate drive module and a phase shift control module. The input end of the phase shift control module is connected with the output end of the microcontroller module, the input end of the gate drive module is connected with the output end of the phase shift control module, the output end of the gate drive module is connected with the input end of the full-bridge inverter module, the output end of the isolation power supply module is connected with the input end of the gate drive module, the isolation power supply module is configured to isolate and protect the gate drive module, and the phase shift control module is configured to trigger and control the on-off of the multi-way output of the microcontroller module.

6. The quick responding ultrasonic transducer resonance frequency tracking system of claim 1, wherein, The filter rectifier module is further configured to receive three-phase power and input the three-phase power to the full-bridge inverter module after filtering and rectifying.

7. The fast responding ultrasonic transducer resonance frequency tracking system of claim 6, wherein, The filter rectifier module comprises a lightning protection filter module and a three-phase rectifier module connected with each other. 8.The fast-response ultrasonic transducer resonant frequency tracking system of claim 1, further comprising an impedance matching module disposed between the full-bridge inverter module and the transducer, and configured to optimize energy transmission efficiency.

9. A fast responding ultrasonic transducer resonance frequency tracking device, characterized by, The fast-response ultrasonic transducer resonant frequency tracking system of any one of claims 1-8.

Images