Intelligent rotary high frequency machining control system based on historical data for predictive control
By introducing a predictive control system based on historical data into the intelligent rotary high-frequency machining control system, and combining multiple sensors and neural network algorithms, real-time monitoring and fault diagnosis of the system status are realized, solving the stability and frequency tracking problems of the system under complex working conditions, and improving the stability and reliability of the system.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- JIANGSU BRANCH OF CHINA ACAD OF MASCH SCI & TECH GRP CO LTD
- Filing Date
- 2023-06-16
- Publication Date
- 2026-06-19
AI Technical Summary
Existing intelligent rotary high-frequency machining control systems cannot effectively monitor the power supply status when faced with factors such as load changes, tool wear, and transducer heating, leading to mechanical resonant frequency drift and affecting machining quality and system stability.
A predictive control system based on historical data is adopted, including a main control module, a rectifier and filter module, a power regulation module, a high-frequency inverter module, a high-frequency transformer module, an impedance matching module, and a feedback signal processing module. It combines Hall voltage sensors, Hall current sensors, thermistor temperature sensors, and a BP neural network algorithm to monitor the system status in real time and perform fault diagnosis.
It enables real-time monitoring and fault diagnosis of intelligent rotary high-frequency machining control system under complex working conditions, ensuring frequency tracking response speed and accurate and stable output signal, and improving the stability and reliability of the system.
Smart Images
Figure CN116586274B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the technical field of high-frequency vibration machining systems, specifically relating to an intelligent rotary high-frequency machining control system based on historical data for predictive control. Background Technology
[0002] The ultrasonic vibration system in ultrasonic composite machining mainly consists of two parts: a high-frequency machining control system and an ultrasonic transducer composed of a transducer, an amplitude transformer, and a tool head. The intelligent rotary high-frequency machining control system generates ultrasonic vibration signals, which are converted into high-frequency mechanical vibrations by the transducer structure in the ultrasonic transducer. These vibrations drive the tool head to machine the workpiece. The performance of the intelligent rotary high-frequency machining control system largely determines the machining quality of the workpiece. Almost all high-power ultrasonic transducers are affected by factors such as load changes, tool wear, and transducer heating during operation, causing the transducer's mechanical resonant frequency to drift. If the intelligent rotary high-frequency machining control system fails to output the correct electrical oscillation signal, the subsequent ultrasonic transducer will malfunction or even be damaged.
[0003] Current intelligent rotary high-frequency machining control systems employ single-chip control, focusing only on acquiring necessary current and voltage signals from the high-frequency electrical signals output by the system. A feedback signal processing module performs phase difference detection on these acquired signals to provide circuit status signals to the control module. Therefore, there is a lack of effective monitoring of the operating status of each module within the intelligent rotary high-frequency machining control system, making rapid and effective diagnosis impossible when the power supply malfunctions. Furthermore, in complex ultrasonic composite machining processes, the mechanical resonant frequency of the ultrasonic transducer undergoes complex changes, requiring the intelligent rotary high-frequency machining control system to track the resonant frequency in real time and accurately output the high-frequency electrical signals required by the ultrasonic transducer. This necessitates high-precision, high-sensitivity sampling circuits and feedback signal processing circuits, as well as a more accurate power supply control model. However, due to the lack of monitoring of the power supply's internal operating status and the limited input parameters for the control model, the stability and reliability of the intelligent rotary high-frequency machining control system's control model cannot be guaranteed.
[0004] Therefore, there is an urgent need to develop a new intelligent rotary high-frequency machining control system based on historical data for predictive control in order to solve the above problems. Summary of the Invention
[0005] The purpose of this invention is to provide an intelligent rotary high-frequency machining control system based on historical data for predictive control.
[0006] To address the aforementioned technical problems, this invention provides an intelligent rotary high-frequency machining control system based on historical data for predictive control. The system includes: a main control module, a rectifier and filter module, a power regulation module, a high-frequency inverter module, a high-frequency transformer module, an impedance matching module, and a feedback signal processing module. The rectifier and filter module, power regulation module, high-frequency inverter module, high-frequency transformer module, and impedance matching module are connected sequentially. The input terminal of the rectifier and filter module is connected to AC power, and the output terminal of the impedance matching module is connected to a transducer via a wireless coil. The power regulation module, high-frequency inverter module, and feedback signal processing module are electrically connected to the main control module, and the power regulation module, high-frequency inverter module, high-frequency transformer module, impedance matching module, and wireless coil are electrically connected to the feedback signal processing module. The circuit is connected as follows: the rectifier and filter module converts AC power into DC power, which is then regulated by the power regulation module and input to the high-frequency inverter module. The regulated DC power is converted into a high-frequency AC power signal by the high-frequency inverter module, boosted and tuned by the high-frequency transformer module and impedance matching circuit, and output to the piezoelectric transducer via the wireless coil. The feedback signal processing module samples the voltage, current, and temperature signals of the power regulation module, high-frequency inverter module, high-frequency transformer module, impedance matching module, and wireless coil, and performs bandpass filtering on the sampled data to convert it into corresponding digital quantities or pulse width signals, which are then isolated and input to the main control module. The main control module adjusts the voltage output of the power regulation module and the frequency output of the high-frequency inverter module according to the feedback signal.
[0007] Furthermore, the high-frequency inverter module adopts a full-bridge high-frequency inverter circuit composed of silicon carbide MOSFETs.
[0008] Furthermore, the high-frequency inverter module employs a high-frequency transformer to boost the high-frequency AC signal.
[0009] Furthermore, the feedback signal processing module includes: several Hall voltage sensors, several Hall current sensors, and several thermistor temperature sensors; each of the Hall voltage sensors is connected to the power regulation module, the high-frequency inverter module, the high-frequency transformer module, the impedance matching module, the feedback signal processing module, and the wireless coil to acquire the corresponding voltage signal; each of the Hall current sensors is connected to the power regulation module, the high-frequency inverter module, the high-frequency transformer module, the impedance matching module, the feedback signal processing module, and the wireless coil to acquire the corresponding current signal; each of the thermistor temperature sensors is connected to the power regulation module, the high-frequency inverter module, the wireless coil, and the piezoelectric transducer to acquire the corresponding temperature signal.
[0010] Furthermore, the feedback signal processing module also includes a bandpass filter circuit and a signal amplification circuit; each of the Hall voltage sensors, each of the Hall current sensors, and each of thermistor temperature sensors are connected to the bandpass filter circuit, that is, the sampled data is bandpass filtered and amplified by the bandpass filter circuit and the signal amplification circuit.
[0011] Furthermore, the feedback signal processing module also includes an impedance angle detection circuit; the sampled data after bandpass filtering and amplification generates a corresponding pulse width signal through the impedance angle detection circuit.
[0012] Furthermore, the feedback signal processing module also includes a peak detection circuit and an analog-to-digital converter; the sampled data after bandpass filtering and amplification generates corresponding digital quantities through the peak detection circuit and the analog-to-digital converter.
[0013] Furthermore, the feedback signal processing module also includes: an optocoupler isolation circuit; the digital quantity or pulse width signal generated based on the sampled data is sent to the main control module via the optocoupler isolation circuit.
[0014] Furthermore, the main control module includes an ARM chip and an FPGA chip; the ARM chip processes the sampled data and exchanges data through the FSMC bus of the FPGA chip; the FPGA chip controls the power regulation module and the high-frequency inverter module through DDS output and PWM wave output.
[0015] Furthermore, the required power and ideal resonant frequency are input into the ARM chip, and the FPGA chip preprocesses the sampled data; the FPGA chip inputs the processed data into the ARM chip for logical operations, and outputs corresponding control signals using the trained BP neural network algorithm model; the ARM chip outputs the control signals to the FPGA chip so that the FPGA chip adjusts the output of the power regulation module and the high-frequency inverter module.
[0016] Furthermore, the trained BP neural network algorithm model includes an input layer, a hidden layer, and an output layer; the input layer, hidden layer, and output layer are connected sequentially, each neuron in the input layer is fully connected to each neuron in the hidden layer, each neuron in the hidden layer is fully connected to each neuron in the output layer, and there are corresponding weight coefficients between the connections of the two neurons.
[0017] Furthermore, the threshold θ and weight W of each neuron are initialized; the voltage and current amplitude signals of the wireless coil, the voltage and current amplitude phase difference signals of the wireless coil, the ideal resonant frequency, and the piezoelectric transducer temperature in the training set are used as model inputs, and then... The calculated input is then fed into the hidden layer as its input; after passing through the activation function... The calculated signal is input to the output layer, and the input signal to the output layer is then processed... The calculated output is compared with the output current, output voltage, and output frequency in the training set; through... Calculate the relative error. The output signal in the training set is compared with a pre-set relative error limit. If the relative error is less than the limit, the next set of data is used for training; otherwise, the training proceeds according to the pre-set error limit. , , , Update the corresponding thresholds and weights, and retrain the dataset. The set learning rate.
[0018] The beneficial effects of this invention are that it enables real-time monitoring and fault diagnosis of the working status of the intelligent rotary high-frequency machining control system under complex working conditions, ensuring the response speed of frequency tracking and the accuracy and stability of the output signal, thereby improving the stability of the system.
[0019] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention.
[0020] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description
[0021] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.
[0022] Figure 1 This is a circuit diagram of the intelligent rotary high-frequency machining control system based on historical data predictive control according to the present invention.
[0023] Figure 2 This is a circuit diagram of the feedback signal processing module and the main control module of the present invention;
[0024] Figure 3 This is a flowchart of the feedback signal processing module of the present invention;
[0025] Figure 4 This is a control flowchart of the intelligent rotary high-frequency machining control system based on historical data predictive control according to the present invention.
[0026] Figure 5 This is a diagram of the BP neural network algorithm model of the present invention;
[0027] Figure 6 This is a flowchart illustrating the data processing workflow of the FPGA chip in this invention.
[0028] Figure 7 This is a flowchart of the BP neural network algorithm model of the present invention. Detailed Implementation
[0029] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0030] Example 1, in this example, as Figures 1 to 7 As shown, this embodiment provides an intelligent rotary high-frequency machining control system based on historical data for predictive control. It includes: a main control module, a rectifier and filter module, a power regulation module, a high-frequency inverter module, a high-frequency transformer module, an impedance matching module, and a feedback signal processing module. The rectifier and filter module, power regulation module, high-frequency inverter module, high-frequency transformer module, and impedance matching module are connected sequentially. The input terminal of the rectifier and filter module is connected to AC power, and the output terminal of the impedance matching module is connected to a transducer via a wireless coil. The power regulation module, high-frequency inverter module, and feedback signal processing module are electrically connected to the main control module, and the power regulation module, high-frequency inverter module, high-frequency transformer module, impedance matching module, and wireless coil are electrically connected to the feedback signal processing module. The rectifier and filter module converts AC power into DC power, which is then regulated by the power regulation module and input to the high-frequency inverter module. The regulated DC power is converted into a high-frequency AC power signal by the high-frequency inverter module, boosted and tuned by the high-frequency transformer module and impedance matching circuit, and output to the piezoelectric transducer via the wireless coil. The feedback signal processing module samples the voltage, current, and temperature signals from the power regulation module, high-frequency inverter module, high-frequency transformer module, impedance matching module, and wireless coil, and performs bandpass filtering on the sampled data to convert it into corresponding digital quantities or pulse width signals, which are then isolated and input to the main control module. The main control module adjusts the voltage output of the power regulation module and the frequency output of the high-frequency inverter module based on the feedback signal.
[0031] In this embodiment, energy transfer of the piezoelectric transducer during rotation is ensured by a wireless coil.
[0032] In this embodiment, the 220V AC power is sequentially boosted by a rectifier and filter module, a power regulation module, a high-frequency inverter module, and a high-frequency transformer module. Then, it is connected to the primary side of the wireless power transmission coil through an impedance matching module to form the main circuit of the intelligent rotary high-frequency machining control system. Each module of the main circuit of the intelligent rotary high-frequency machining control system is connected to the feedback signal processing module through a current and voltage sampling circuit. The feedback signal processing module is connected to the main control module to provide digital signals.
[0033] In this embodiment, the system can achieve real-time monitoring and fault diagnosis of the working status of the intelligent rotary high-frequency machining control system under complex working conditions, ensuring the response speed of frequency tracking and the accuracy and stability of the output signal, thereby improving the stability of the system.
[0034] In this embodiment, the high-frequency inverter module adopts a full-bridge high-frequency inverter circuit composed of silicon carbide MOSFETs, which can convert DC signals into high-frequency AC signals of a specific frequency.
[0035] In this embodiment, the high-frequency inverter module uses a high-frequency transformer to boost the high-frequency AC signal, thereby enabling high-frequency mechanical vibration of the transducer.
[0036] In this embodiment, the feedback signal processing module includes: a plurality of Hall voltage sensors, a plurality of Hall current sensors, and a plurality of thermistor temperature sensors; each of the Hall voltage sensors is respectively connected to the power regulation module, the high-frequency inverter module, the high-frequency transformer module, the impedance matching module, the feedback signal processing module, and the wireless coil to acquire the corresponding voltage signal; each of the Hall current sensors is respectively connected to the power regulation module, the high-frequency inverter module, the high-frequency transformer module, the impedance matching module, the feedback signal processing module, and the wireless coil to acquire the corresponding current signal; each of the thermistor temperature sensors is respectively connected to the power regulation module, the high-frequency inverter module, the wireless coil, and the piezoelectric transducer to acquire the corresponding temperature signal.
[0037] In this embodiment, the feedback signal processing module further includes a bandpass filter circuit and a signal amplification circuit; each of the Hall voltage sensors, each of the Hall current sensors, and each of thermistor temperature sensors are connected to the bandpass filter circuit, that is, the sampled data is bandpass filtered and amplified by the bandpass filter circuit and the signal amplification circuit.
[0038] In this embodiment, the feedback signal processing module further includes an impedance angle detection circuit; the sampled data after bandpass filtering and amplification generates a corresponding pulse width signal through the impedance angle detection circuit.
[0039] In this embodiment, the feedback signal processing module further includes a peak detection circuit and an analog-to-digital converter; the sampled data after bandpass filtering and amplification generates corresponding digital quantities through the peak detection circuit and the analog-to-digital converter.
[0040] In this embodiment, the feedback signal processing module further includes an optocoupler isolation circuit; the digital quantity or pulse width signal generated based on the sampled data is sent to the main control module via the optocoupler isolation circuit.
[0041] In this embodiment, the current and voltage signals of the power regulation module, high-frequency inverter module, high-frequency transformer module, impedance matching module, and the primary side of the wireless coil are sampled, and the sampled data is input to the feedback signal processing module. The feedback signal processing module performs bandpass filtering on the returned analog signals and converts them into corresponding digital quantities through a precision ADC, or converts the voltage and current into corresponding pulse width signals through an impedance angle detection circuit. After isolation, the signals are input to the main control module. The main control module adjusts the voltage output of the power regulation module and the frequency output of the high-frequency inverter module in real time based on the feedback voltage and current information. Temperature is detected by a thermistor temperature sensor. Since temperature detection is mainly used for over-temperature protection, the accuracy requirement is not high. Using a thermistor temperature sensor can reduce detection costs. The collected signals are used not only for functional parameter control but also for system diagnosis of abnormal conditions. During debugging, the basis and rules for judging system faults are determined through data collection and synthesis.
[0042] In this embodiment, the main control module includes an ARM chip and an FPGA chip; the ARM chip processes the sampled data and exchanges data through the FSMC bus of the FPGA chip; the FPGA chip controls the power regulation module and the high-frequency inverter module through DDS output and PWM wave output.
[0043] In this embodiment, the main control module consists of an FPGA chip and an ARM chip. By controlling the power regulation module and the high-frequency inverter module through DDS output and PWM wave output, it can realize real-time monitoring and fault diagnosis of the working status of the intelligent rotary high-frequency machining control system under complex working conditions, ensure the response speed of frequency tracking and the accuracy and stability of the output signal, and improve the stability of the system.
[0044] In this embodiment, the main control module adopts an MCU+FPGA architecture to form the control core. The ARM chip and FPGA chip achieve high-speed data exchange through the FSMC bus. The ARM chip is mainly responsible for processing feedback analog signals such as voltage and current, control logic, simple algorithm processing, and human-computer interaction. The FPGA chip realizes multi-task parallel processing and high-precision PWM signal generation and output, mainly handling phase detection and frequency tracking of the transducer's operating impedance angle, DDS algorithm processing and control signal output, power control algorithm calculation and control quantity output, serial communication, and other tasks. This architecture separates sequential logic tasks and high-speed parallel tasks, improving the system's response speed and reserving a large amount of scalability for future higher functions and performance requirements. The main control module is isolated from other modules through optocouplers, which can effectively protect the main control module and reduce external circuit signal interference under complex external operating conditions.
[0045] In this embodiment, the ARM chip is connected to the touch screen; the FPGA chip is connected to the machine tool CNC system via a serial communication module.
[0046] In this embodiment, the required power and ideal resonant frequency are input to the ARM chip, and the FPGA chip preprocesses the sampled data; the FPGA chip inputs the processed data into the ARM chip for logical operations, and outputs corresponding control signals using the trained BP neural network algorithm model; the ARM chip outputs the control signals to the FPGA chip so that the FPGA chip adjusts the output of the power regulation module and the high-frequency inverter module.
[0047] In this embodiment, the trained BP neural network algorithm model includes an input layer, a hidden layer, and an output layer; the input layer, hidden layer, and output layer are connected sequentially, each neuron in the input layer is fully connected to each neuron in the hidden layer, each neuron in the hidden layer is fully connected to each neuron in the output layer, and there are corresponding weight coefficients between the connections of the two neurons.
[0048] In this embodiment, the sampled data (the current and voltage signals of the wireless coil, the temperature of the piezoelectric transducer, the matching inductance value, and the resonant frequency) are used as the input to the BP neural network algorithm model; the output current, the output voltage amplitude, and the output signal frequency are used as the output of the BP neural network algorithm model.
[0049] In this embodiment, as Figure 5As shown, the data is input into the FPGA chip for storage. The wireless coil voltage and current amplitude signal, coil voltage and current phase difference signal, ideal resonant frequency, transducer temperature, and control system output current, output voltage, and output frequency are selected from historical data and the latest data collected 5 seconds before the power supply starts. Data fitting and similarity analysis are performed in the FPGA chip to exclude data with high similarity. The remaining data is normalized to form a BP neural network training set, which is then input into the ARM chip for model training.
[0050] In this embodiment, the threshold θ and weight W of each neuron are initialized; the voltage and current amplitude signals of the wireless coil, the voltage and current amplitude phase difference signals of the wireless coil, the ideal resonant frequency, and the piezoelectric transducer temperature in the training set are used as model inputs, and then... The calculated input is then fed into the hidden layer as its input; after passing through the activation function... The calculated signal is input to the output layer, and the input signal to the output layer is then processed... The calculated output is compared with the output current, output voltage, and output frequency in the training set; through... Calculate the relative error. The output signal in the training set is compared with a pre-set relative error limit. If the relative error is less than the limit, the next set of data is used for training; otherwise, the training proceeds according to the pre-set error limit. , , , Update the corresponding thresholds and weights, and retrain the dataset. The set learning rate.
[0051] In this embodiment, after the BP neural network algorithm model has been trained on each set of data in the training set, it has only completed one training iteration. Subsequent simulations are performed multiple times on the same training set to continuously ensure that the error for each set of data in the entire training set is less than the error limit. Once the BP neural network algorithm model is trained, the system output can be adjusted.
[0052] In summary, this invention enables real-time monitoring and fault diagnosis of the working status of the intelligent rotary high-frequency machining control system under complex working conditions, ensuring the response speed of frequency tracking and the accuracy and stability of the output signal, thereby improving the stability of the system.
[0053] All devices selected in this application (parts whose specific structures are not described) are general standard parts or parts known to those skilled in the art, and their structures and principles can be learned by those skilled in the art through technical manuals or conventional experimental methods. Furthermore, all software programs involved in this application are prior art, and this application does not involve any improvements to the software programs.
[0054] In the description of the embodiments of the present invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in the present invention based on the specific circumstances.
[0055] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.
[0056] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the shown or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.
[0057] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.
[0058] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.
[0059] Based on the above-described preferred embodiments of the present invention, and through the foregoing description, those skilled in the art can make various changes and modifications without departing from the inventive concept. The technical scope of this invention is not limited to the contents of the specification, but must be determined according to the scope of the claims.
Claims
1. An intelligent rotary high frequency machining control system for predictive control based on historical data, characterized in that, include: The main control module, rectifier and filter module, power regulation module, high-frequency inverter module, high-frequency transformer module, impedance matching module, and feedback signal processing module; in The rectifier and filter module, power regulation module, high-frequency inverter module, high-frequency transformer module, and impedance matching module are connected in sequence. The input terminal of the rectifier and filter module is connected to AC power, and the output terminal of the impedance matching module is connected to the transducer through a wireless coil. The power regulation module, high-frequency inverter module, feedback signal processing module and main control module are electrically connected; the power regulation module, high-frequency inverter module, high-frequency transformer module, impedance matching module, wireless coil and feedback signal processing module are electrically connected. The rectifier and filter module converts AC power into DC power signal, and the power regulation module adjusts the DC voltage input to the high-frequency inverter module. The DC signal after voltage regulation is converted into a high-frequency AC signal by a high-frequency inverter module, and then boosted and tuned by the high-frequency transformer module and impedance matching circuit before being output to the piezoelectric transducer through the wireless coil. The feedback signal processing module samples the voltage, current, and temperature signals of the power regulation module, high-frequency inverter module, high-frequency transformer module, impedance matching module, and wireless coil, and performs bandpass filtering on the sampled data to convert it into corresponding digital quantities or pulse width signals, which are then isolated and input to the main control module. as well as The main control module adjusts the voltage output of the power regulation module and the frequency output of the high-frequency inverter module; The main control module includes: an ARM chip and an FPGA chip; The ARM chip processes the sampled data, and the ARM chip exchanges data through the FSMC bus of the FPGA chip; The FPGA chip controls the power regulation module and the high-frequency inverter module through DDS output and PWM wave output. The required power and ideal resonant frequency are input into the ARM chip, and the FPGA chip preprocesses the sampled data. The FPGA chip inputs the processed data into the ARM chip for logical operations, and uses the trained BP neural network algorithm model to output corresponding control signals. The ARM chip outputs control signals to the FPGA chip, so that the FPGA chip can adjust the output of the power regulation module and the high-frequency inverter module. The trained BP neural network algorithm model includes: an input layer, a hidden layer, and an output layer; The input layer, hidden layer, and output layer are connected in sequence. Each neuron in the input layer is fully connected to each neuron in the hidden layer, and each neuron in the hidden layer is fully connected to each neuron in the output layer. There are corresponding weight coefficients between the connections of the two neurons. Initialize the threshold θ and weight W for each neuron; The voltage and current amplitude signals of the wireless coil, the voltage and current amplitude and phase difference signals of the wireless coil, the ideal resonant frequency, and the piezoelectric transducer temperature from the training set are used as model inputs. The calculated value is then input into the hidden layer as the input to the hidden layer. After excitation function The calculated signal is input to the output layer, and the input signal to the output layer is then processed... The calculated output is compared with the output current, output voltage, and output frequency in the training set. By calculating the relative error, is the output signal in the training set; Compared with a preset relative error limit, if the relative error is less than the preset limit, the next set of data is used for training; if the relative error is greater than the preset limit, the training proceeds according to the preset limit. , , , Update the corresponding thresholds and weights, and retrain the dataset. The set learning rate.
2. The intelligent rotary high-frequency machining control system based on historical data for predictive control as described in claim 1, characterized in that, The high-frequency inverter module adopts a full-bridge high-frequency inverter circuit composed of silicon carbide MOSFETs.
3. The intelligent rotary high-frequency machining control system based on historical data for predictive control as described in claim 1, characterized in that, The high-frequency inverter module uses a high-frequency transformer to boost the voltage of the high-frequency AC signal.
4. The intelligent rotary high-frequency machining control system based on historical data for predictive control as described in claim 1, characterized in that, The feedback signal processing module includes: several Hall voltage sensors, several Hall current sensors, and several thermistor temperature sensors. Each of the Hall voltage sensors is connected to the power regulation module, high-frequency inverter module, high-frequency transformer module, impedance matching module, feedback signal processing module, and wireless coil to acquire the corresponding voltage signal. Each of the Hall current sensors is connected to the power regulation module, high-frequency inverter module, high-frequency transformer module, impedance matching module, feedback signal processing module, and wireless coil to collect the corresponding current signal. Each of the thermistor temperature sensors is connected to a power regulation module, a high-frequency inverter module, a wireless coil, and a piezoelectric transducer to collect the corresponding temperature signal.
5. The intelligent rotary high-frequency machining control system based on historical data for predictive control as described in claim 4, characterized in that, The feedback signal processing module further includes: a bandpass filter circuit and a signal amplification circuit; Each of the Hall voltage sensors, Hall current sensors, and thermistor temperature sensors is connected to a bandpass filter circuit, i.e. The sampled data is bandpass filtered and amplified by the bandpass filter circuit and the signal amplification circuit.
6. The intelligent rotary high-frequency machining control system based on historical data for predictive control as described in claim 5, characterized in that, The feedback signal processing module further includes: an impedance angle detection circuit; The sampled data, after bandpass filtering and amplification, generates a corresponding pulse width signal through the impedance angle detection circuit. The feedback signal processing module further includes: a peak detection circuit and an analog-to-digital converter; The sampled data, after bandpass filtering and amplification, is used by the peak detection circuit and analog-to-digital converter to generate corresponding digital quantities. The feedback signal processing module further includes: an optocoupler isolation circuit; The digital quantity or pulse width signal generated based on the sampled data is sent to the main control module via the optocoupler isolation circuit.