Self-adaptive bias voltage optimization method based on MOS (Metal Oxide Semiconductor) tube and related equipment

Abstract

The invention relates to a self-adaptive bias voltage optimization method and related equipment based on an MOS (Metal Oxide Semiconductor) transistor, which are applied to a gate drive circuit and comprise the following steps of: monitoring a target MOS transistor in a working state of the gate drive circuit in real time to obtain a working parameter of the target MOS transistor; based on a preset bias voltage demand prediction model, performing bias voltage demand prediction on the target MOS tube based on the working parameters to obtain a bias voltage demand value; performing digital signal processing on the bias voltage demand value so as to convert the bias voltage demand value into a control signal; the control signal is applied to the grid electrode of the target MOS tube so as to dynamically adjust the working state of the target MOS tube, and the technical problem that under extreme conditions such as high temperature or heavy load, fixed bias voltage may cause overheating of the MOS tube or increase of switching loss is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of MOS tubes, and in particular to a method for optimizing an adaptive bias voltage based on a MOS tube and related equipment. Background Art

[0002] With the rapid development of electronic technology, high performance and low power consumption have become key indicators for electronic device design. Especially in portable devices and Internet of Things (IoT) applications, the demand for extending battery life is increasing, which has prompted a deeper study of power management technology. As one of the most basic and commonly used semiconductor components in modern electronic devices, MOS tubes play a vital role in power management and signal transmission. However, the traditional use of MOS tubes often uses a fixed bias voltage. Although this approach is simple and direct, it is difficult to adapt to the optimal performance requirements under different working conditions. Especially in the case of large load changes or unstable working environment, the fixed bias voltage may lead to problems such as reduced efficiency and increased power consumption.

[0003] In existing solutions, although some technologies have tried to adjust the bias voltage of MOS tubes through external circuits in order to achieve better performance. However, these solutions usually require additional hardware support, which not only increases the complexity and cost of the system, but also has a slow response speed and cannot achieve real-time adjustment in the true sense. In addition, due to the lack of an effective bias voltage demand prediction model, existing adaptive bias voltage adjustment technologies often rely on empirical data or simple feedback mechanisms, which limits their applicability in complex and changing application scenarios. Therefore, it is particularly important to develop a method that can automatically adjust the bias voltage of MOS tubes according to the actual working state.

[0004] To this end, an adaptive bias voltage optimization method based on MOS tube is proposed, which aims to solve the shortcomings of traditional fixed bias voltage schemes and existing adaptive adjustment technologies. By real-time monitoring of the working parameters of the MOS tube and using the pre-established bias voltage demand prediction model, the optimal bias voltage demand value can be accurately predicted. Subsequently, the predicted value is processed by digital signal, converted into a control signal and applied to the gate of the MOS tube, thereby realizing dynamic adjustment of its working state. This method can not only improve the efficiency and stability of the MOS tube under various working conditions, but also effectively reduce unnecessary energy consumption, which is of great significance for promoting the development of electronic equipment towards a more energy-saving and intelligent direction. Summary of the invention

[0005] The main purpose of the present invention is to provide an adaptive bias voltage optimization method based on MOS tube and related equipment, which solves the technical problem that under extreme conditions, such as high temperature or heavy load, a fixed bias voltage may cause MOS tube overheating or increased switching loss.

[0006] To achieve the above object, the present invention provides a method for optimizing an adaptive bias voltage based on a MOS tube, which is applied to a gate drive circuit and includes the following steps: Real-time monitoring of the target MOS tube under the working state of the gate drive circuit is performed to obtain the working parameters of the target MOS tube; Based on a preset bias voltage requirement prediction model, and based on the operating parameters, a bias voltage requirement prediction is performed on the target MOS tube to obtain a bias voltage requirement value; Performing digital signal processing on the bias voltage requirement value to convert it into a control signal; The control signal is applied to the gate of the target MOS transistor to dynamically adjust the working state of the target MOS transistor.

[0007] Furthermore, the real-time monitoring of the target MOS tube in the working state of the gate drive circuit to obtain the working parameters of the target MOS tube includes: Synchronously sampling the gate voltage, drain voltage and drain current of the target MOS tube in the working state of the gate drive circuit to obtain an original electrical signal data set, wherein the circuit working state includes the turning on, amplification and cutting off of the target MOS tube; Performing time series processing on the original electrical signal data set to obtain switching frequency and duty cycle information of the target MOS tube; Scanning the surface temperature of the target MOS tube to calculate the junction temperature data of the target MOS tube; Using a sliding window method, based on the switching frequency, duty cycle information and junction temperature data, the operating parameter statistics of the target MOS tube in different time periods are calculated; wherein the operating parameter statistics include an average gate-source voltage, an average drain-source voltage and an average drain current; A standard bias voltage is set based on the original electrical signal data set, switching frequency, duty cycle information, junction temperature data and parameter statistics, and the original electrical signal data set, switching frequency, duty cycle information, junction temperature data, parameter statistics and standard bias voltage are used as operating parameters of the target MOS tube.

[0008] Further, scanning the surface temperature of the target MOS tube to calculate the junction temperature data of the target MOS tube includes: Performing non-contact temperature measurement on the surface of the target MOS tube by using an infrared thermal imager to obtain a surface temperature distribution diagram of the target MOS tube; Using Fourier transform algorithm, performing temperature frequency domain analysis on the surface temperature distribution diagram to extract the temperature distribution frequency component; By using a reverse heat conduction model, an internal heat distribution inversion calculation is performed on the target MOS tube based on the temperature distribution frequency component to obtain an internal temperature distribution diagram; The junction temperature data of the target MOS tube is calculated based on the internal temperature distribution diagram using a thermal resistance network model.

[0009] Furthermore, the bias voltage requirement prediction model training process includes: Step S31, obtaining historical operating parameters, preprocessing the historical operating parameters to obtain a feature vector set, and dividing the feature vector set into a training set and a standard bias voltage corresponding to the training set; wherein the training set includes an original electrical signal data set, a switching frequency, a duty cycle information, a junction temperature data, and parameter statistics; Step S32, inputting the training set into a preset initial bias voltage demand prediction model for calculation to obtain a bias voltage prediction value; Step S33, comparing and calculating the standard bias voltage with the bias voltage prediction value to calculate a bias voltage error; Step S34, determining whether the bias voltage error is within a preset bias voltage error range, and if the bias voltage error is within the preset bias voltage error range, determining the initial bias voltage demand prediction model as the bias voltage demand prediction model; Step S35, if the bias voltage error exceeds the preset bias voltage error range, the parameters in the initial bias voltage requirement prediction model are adjusted, and steps S31 to S34 are repeated until the bias voltage output by the initial bias voltage requirement prediction model is within the preset range to obtain a bias voltage requirement prediction model.

[0010] Furthermore, the step of performing digital signal processing on the bias voltage requirement value to convert it into a control signal includes: quantizing the continuous bias voltage requirement value to obtain a discrete quantized value; Performing digital signal processing on the discrete quantized value to obtain a processed digital signal; Performing signal encoding on the processed digital signal to obtain an encoded control signal; Inputting the coded control signal into a preset digital-to-analog converter for signal conversion to obtain an analog control signal; The analog control signal is used as a control signal.

[0011] Further, applying the control signal to the gate of the target MOS transistor to dynamically adjust the working state of the target MOS transistor includes: Transmitting the control signal to the gate drive circuit to obtain a transmission control signal; Amplifying and shaping the transmission control signal through a gate drive circuit to obtain an amplified control signal; Performing overvoltage and overcurrent protection on the amplified control signal through an amplifier in a gate drive circuit to obtain a protection control signal; Electrically isolating the protection control signal to obtain an isolation control signal; Performing signal buffering processing on the isolation control signal to obtain a buffered control signal; The buffer control signal is applied to the gate of the target MOS transistor to dynamically adjust the working state of the target MOS transistor.

[0012] Furthermore, the amplifier in the gate drive circuit performs overvoltage and overcurrent protection on the amplified control signal to obtain a protection control signal, including: Performing spectrum decomposition on the amplification control signal to obtain a spectrum component set; Constructing a signal energy graph based on the spectral component set to obtain signal energy information; Determine whether the signal energy information has risk signal characteristics, and if so, extract characteristics of the signal energy information to obtain a risk characteristic signal; Analyzing the type of the risk characteristic signal by a preset analysis algorithm, and formulating a protection control signal strategy based on the type of the risk characteristic signal; Through the protection control signal strategy, the amplifier in the gate drive circuit is used to perform overvoltage and overcurrent protection on the amplified control signal to obtain a protection control signal.

[0013] The present invention also provides an adaptive bias voltage optimization device based on a MOS tube, which is applied to a gate drive circuit and includes: A monitoring module, used to monitor the target MOS tube in the working state of the gate drive circuit in real time to obtain the working parameters of the target MOS tube; A prediction module, configured to predict the bias voltage requirement of the target MOS tube based on the working parameters based on a preset bias voltage requirement prediction model, and obtain a bias voltage requirement value; A conversion module, used for performing digital signal processing on the bias voltage requirement value to convert it into a control signal; The applying module is used to apply the control signal to the gate of the target MOS tube to dynamically adjust the working state of the target MOS tube.

[0014] The present invention also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the above methods when executing the computer program.

[0015] The present invention also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of any of the above-mentioned methods are implemented.

[0016] The MOS tube-based adaptive bias voltage optimization method provided by the present invention comprises the following steps: real-time monitoring of a target MOS tube in a working state of a gate drive circuit to obtain working parameters of the target MOS tube; based on a preset bias voltage demand prediction model, predicting the bias voltage demand of the target MOS tube based on the working parameters to obtain a bias voltage demand value; performing digital signal processing on the bias voltage demand value to convert it into a control signal; applying the control signal to the gate of the target MOS tube to dynamically adjust the working state of the target MOS tube. Through the above technical means, the technical problem that a fixed bias voltage may cause overheating of the MOS tube or increase in switching loss under extreme conditions, such as high temperature or heavy load, is solved, and the adaptive bias voltage optimization method can make a rapid response according to the actual working conditions (such as temperature, current, etc.) of the MOS tube, avoiding the risk of overheating or damage caused by a fixed bias voltage, and improving the stability and reliability of the MOS tube and the entire circuit. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic diagram of the steps of a method for optimizing an adaptive bias voltage based on a MOS tube in one embodiment of the present invention; Figure 2 It is a structural block diagram of an adaptive bias voltage optimization device based on a MOS tube in one embodiment of the present invention; Figure 3 It is a schematic block diagram of the structure of a computer device according to an embodiment of the present invention.

[0018] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0020] like Figure 1 As shown, Figure 1It is a schematic diagram of the steps of a method for optimizing a MOS-based adaptive bias voltage in one embodiment of the present invention; In one embodiment of the present invention, a method for optimizing an adaptive bias voltage based on a MOS tube is provided, which is applied to a gate drive circuit and includes the following steps: Step S1, real-time monitoring of the target MOS tube in the working state of the gate driving circuit is performed to obtain the working parameters of the target MOS tube.

[0021] Specifically, the process of real-time monitoring of the target MOS tube under the working state of the gate drive circuit and obtaining the working parameters of the target MOS tube first involves the construction of the monitoring system. In this process, a series of high-precision sensors need to be installed or built-in detection circuits need to be used to continuously track the key working parameters of the MOS, such as temperature, current, voltage, etc. These sensors or detection circuits must have sufficient sensitivity and fast response capabilities to ensure that the changes in the MOS state can be captured in a timely and accurate manner. For example, in the power management system of an electric vehicle, when the vehicle is in an acceleration state, the current provided by the battery to the motor will suddenly increase. At this time, if this change can be monitored in real time through the built-in current sensor and quickly fed back to the control system, the gate drive voltage can be adjusted in time to ensure that the MOS can still maintain efficient operation under such transient conditions and avoid overheating or increased switching losses. At the same time, the temperature sensor also plays a vital role, because the performance of the MOS is greatly affected by temperature, especially in a high temperature environment, excessive junction temperature will cause the performance of the MOS to degrade or even damage. Therefore, by monitoring the operating temperature of the MOS in real time and sending it as an input signal to a microcontroller or a dedicated chip together with other operating parameters, comprehensive monitoring of the MOS working state can be achieved, providing accurate data support for subsequent intelligent analysis and dynamic adjustment. In this way, even in complex and changeable application environments such as electric vehicles, it can ensure that MOS is always in the best working state, thereby improving the energy efficiency and reliability of the entire system.

[0022] Step S2: Based on a preset bias voltage requirement prediction model and based on the operating parameters, a bias voltage requirement prediction is performed on the target MOS tube to obtain a bias voltage requirement value.

[0023] Specifically, based on the preset bias voltage demand prediction model, the bias voltage demand prediction of the target MOS tube is performed based on the working parameters to obtain the bias voltage demand value. This process is a key link in the adaptive bias voltage optimization method. In the specific implementation, it is first necessary to establish a prediction model that can reflect the working characteristics of MOS. This model is usually based on a large amount of experimental data and theoretical analysis, and can accurately predict the optimal bias voltage required by MOS under different working conditions. For example, in the power management system of an electric vehicle, when the system detects that the vehicle is accelerating, the current and temperature sensors respectively monitor the increase in current and the rise in temperature, these real-time acquired working parameters will be immediately transmitted to the microcontroller or dedicated chip. Then, the microcontroller will use the preset bias voltage demand prediction model, combined with the current working parameters, to calculate the bias voltage value required for MOS to achieve the best working state under the current conditions. This model may include characteristic parameters such as on-resistance, switching time and loss of MOS at different temperatures and different currents, as well as the relationship between them, so that it can comprehensively consider the influence of multiple factors and make more accurate predictions. In this way, even in conditions with frequent transient changes such as vehicle acceleration, the system can quickly and accurately adjust the bias voltage of the MOS to ensure that it always maintains the best working state, which not only improves the overall efficiency of the system, but also enhances the reliability and safety of the system. In addition, this model-based prediction method can also be flexibly adjusted according to different MOS models and application scenarios, further improving the adaptability and versatility of the system.

[0024] Step S3, performing digital signal processing on the bias voltage requirement value to convert it into a control signal.

[0025] Specifically, the process of performing digital signal processing on the bias voltage requirement value to convert it into a control signal is an important step in realizing MOS adaptive bias voltage optimization. In specific implementation, first, the bias voltage requirement value obtained from the prediction model is an analog signal or an unprocessed value, which needs to be converted into a digital signal suitable for processing by a microcontroller or a dedicated chip through digital signal processing technology. For example, in the power management system of an electric vehicle, when the system predicts the bias voltage value required by the MOS based on the current working parameters, this value will be transmitted to the signal processing unit. Next, the signal processing unit will perform a series of digital processing operations on the bias voltage requirement value, including but not limited to analog-to-digital conversion (ADC), filtering, quantization, etc., to eliminate noise interference and improve the accuracy and stability of the signal. After these processing, the digital signal can more accurately reflect the optimal bias voltage requirement of the MOS under the current working conditions.

[0026] Subsequently, these digital signals will be further converted into specific control signals to adjust the actual output voltage of the gate drive circuit. This conversion process is usually completed by the digital signal processor (DSP) inside the microcontroller, which converts the digital signal into a pulse width modulation (PWM) signal or other form of control signal that can directly control the drive circuit by executing a preset algorithm. For example, during the acceleration of an electric vehicle, if the prediction model calculates that the bias voltage of the MOS needs to be increased to reduce switching losses, the signal processing unit will generate a corresponding PWM signal, and the output voltage of the gate drive circuit will be changed by adjusting the duty cycle of the PWM signal, thereby realizing dynamic adjustment of the MOS bias voltage. This process requires not only a high degree of real-time performance, but also the accuracy of signal conversion to ensure that the MOS is always in the optimal working state. Through such digital signal processing and control signal conversion, the system can automatically and efficiently adjust the bias voltage of the MOS under various complex working conditions, which not only improves the overall efficiency of the power management system, but also enhances the reliability and adaptability of the system. For example, when electric vehicles are driving at high speed for a long time or frequently accelerating and decelerating, this adaptive bias voltage optimization method can effectively reduce energy loss and extend battery life, while protecting MOS from damage due to overheating or overload, ensuring safe and stable operation of the vehicle.

[0027] Step S4: applying the control signal to the gate of the target MOS transistor to dynamically adjust the working state of the target MOS transistor.

[0028] Specifically, the control signal is applied to the gate of the target MOS tube to dynamically adjust the working state of the target MOS tube. This is the last step of the adaptive bias voltage optimization method and the most critical link. In this process, the bias voltage requirement value that has been processed by digital signals and converted into a control signal needs to be effectively transmitted to the gate of the MOS to achieve accurate control of its working state. Specifically, when the power management system of the electric vehicle predicts the optimal bias voltage required by the MOS based on the current working parameters, and converts this requirement value into a control signal through digital signal processing, these control signals will be sent to the gate drive circuit. The gate drive circuit is responsible for amplifying and converting these control signals into appropriate voltage levels, and then directly applying them to the gate of the MOS. For example, during the acceleration of the electric vehicle, the system detects changes in current and temperature, and calculates through a prediction model that the bias voltage of the MOS needs to be increased to reduce switching losses. At this time, the control signal (such as a PWM signal) generated by the signal processing unit will be transmitted to the gate drive circuit. After receiving these control signals, the gate drive circuit will adjust the output voltage according to the requirements of the signal and apply the new bias voltage to the gate of the MOS. This process requires very high real-time and precision to ensure that the MOS can respond quickly to transient changes and maintain the optimal working state. In this way, even under complex working conditions such as vehicle acceleration, deceleration or climbing, the bias voltage of the MOS can be dynamically adjusted according to actual needs, thereby improving the overall efficiency and reliability of the power management system. In addition, this dynamic adjustment mechanism can not only optimize the performance of the MOS under different working conditions, but also extend the service life of the MOS to a certain extent. For example, under high temperature or heavy load conditions, by appropriately increasing the bias voltage, the switching loss of the MOS can be reduced and its overheating damage can be prevented. Under light load or low temperature conditions, appropriately reducing the bias voltage can reduce unnecessary energy consumption and further improve the energy efficiency of the system. Therefore, applying the control signal to the gate of the MOS can not only achieve precise control of the working state of the MOS, but also improve the performance and stability of the entire power management system, ensuring that electric vehicles can operate safely and efficiently under various complex working conditions.

[0029] In a specific embodiment, the real-time monitoring of the target MOS tube in the working state of the gate driving circuit to obtain the working parameters of the target MOS tube includes: Synchronously sampling the gate voltage, drain voltage and drain current of the target MOS tube in the working state of the gate drive circuit to obtain an original electrical signal data set, wherein the circuit working state includes the turning on, amplification and cutting off of the target MOS tube; Performing time series processing on the original electrical signal data set to obtain switching frequency and duty cycle information of the target MOS tube; Scanning the surface temperature of the target MOS tube to calculate the junction temperature data of the target MOS tube; Using a sliding window method, based on the switching frequency, duty cycle information and junction temperature data, the operating parameter statistics of the target MOS tube in different time periods are calculated; wherein the operating parameter statistics include an average gate-source voltage, an average drain-source voltage and an average drain current; A standard bias voltage is set based on the original electrical signal data set, switching frequency, duty cycle information, junction temperature data and parameter statistics, and the original electrical signal data set, switching frequency, duty cycle information, junction temperature data, parameter statistics and standard bias voltage are used as operating parameters of the target MOS tube.

[0030] Specifically, the target MOS tube under the working state of the gate drive circuit is monitored in real time to obtain the working parameters of the target MOS tube, including synchronous sampling of the gate voltage, drain voltage and drain current of the target MOS tube under the working state of the gate drive circuit to obtain the original electrical signal data set, and the circuit working state includes the opening, amplification and cutoff of the target MOS tube. In the power management system of the electric vehicle, this process first needs to be implemented by a high-precision sensor or a built-in detection circuit. For example, when the electric vehicle is in an acceleration state, the system will start the sensor to synchronously sample the gate voltage, drain voltage and drain current of the MOS. These sensors must have high precision and fast response capabilities to ensure that the electrical signals of the MOS in different working states can be captured in real time and accurately. After these electrical signals are synchronously collected, an original electrical signal data set containing the gate voltage, drain voltage and drain current is formed. Next, the original electrical signal data set is time-series processed to obtain the switching frequency and duty cycle information of the target MOS tube. This process is usually completed by a microcontroller or a dedicated chip, and the switching frequency and duty cycle of the MOS are extracted by performing time domain analysis on the collected electrical signals. For example, during the acceleration of an electric vehicle, the system will calculate the switching frequency and duty cycle of the MOS based on the collected electrical signal data set, which is crucial to understanding the working state of the MOS. The switching frequency reflects the number of times the MOS switches in a unit time, while the duty cycle indicates the proportion of the conduction time of the MOS in each cycle. These two parameters are of great significance for evaluating the working efficiency and loss of the MOS. At the same time, the surface temperature of the target MOS tube is scanned to calculate the junction temperature data of the target MOS tube. This process is usually implemented by temperature detection devices such as infrared temperature sensors or thermocouples. For example, in the power management system of an electric vehicle, when the system detects that the operating temperature of the MOS has changed, the temperature sensor is started to scan the surface temperature of the MOS, and the junction temperature data of the MOS is calculated through a thermal model. The junction temperature data is an important parameter for evaluating the thermal performance of the MOS. Excessive junction temperature will cause the performance of the MOS to degrade or even damage it, so real-time monitoring of the junction temperature data is of great significance for protecting the MOS. Then, the sliding window method is used to calculate the statistical values ​​of the operating parameters of the target MOS tube in different time periods based on the switching frequency, duty cycle information and junction temperature data. The operating parameter statistics include the average gate-source voltage, the average drain-source voltage and the average drain current. The sliding window method is a commonly used data processing method. By performing statistical analysis on the data within a certain time window, short-term fluctuations can be smoothed out and more stable operating parameters can be extracted. For example, during the acceleration of an electric vehicle, the system uses the sliding window method to process the switching frequency, duty cycle information and junction temperature data of the MOS, and calculates the average gate-source voltage, average drain-source voltage and average drain current in different time periods.These statistical values ​​can more accurately reflect the performance changes of MOS in actual work and provide a basis for subsequent optimization control. Finally, the standard bias voltage is set based on the original electrical signal data set, switching frequency, duty cycle information, junction temperature data and parameter statistics, and the original electrical signal data set, switching frequency, duty cycle information, junction temperature data, parameter statistics and standard bias voltage are used as the working parameters of the target MOS tube. This process is usually completed by the control algorithm in the microcontroller, and the optimal bias voltage of MOS under the current working conditions is calculated by comprehensively analyzing all the above data. For example, during the acceleration process of an electric vehicle, the system will set a standard bias voltage based on all the collected data to ensure that MOS is always in the optimal working state during the acceleration process. This standard bias voltage will be stored in the system as the basic data for subsequent control. Through this series of real-time monitoring and data analysis, the system can fully and accurately grasp the working state of MOS and provide reliable data support for dynamically adjusting the bias voltage of MOS. This can not only improve the efficiency and reliability of the power management system, but also ensure that MOS is always in the best working state in a complex and changeable application environment, extend its service life and reduce maintenance costs.

[0031] In a specific embodiment, scanning the surface temperature of the target MOS tube to calculate the junction temperature data of the target MOS tube includes: Performing non-contact temperature measurement on the surface of the target MOS tube by using an infrared thermal imager to obtain a surface temperature distribution diagram of the target MOS tube; Using Fourier transform algorithm, performing temperature frequency domain analysis on the surface temperature distribution diagram to extract the temperature distribution frequency component; By using a reverse heat conduction model, an internal heat distribution inversion calculation is performed on the target MOS tube based on the temperature distribution frequency component to obtain an internal temperature distribution diagram; The junction temperature data of the target MOS tube is calculated based on the internal temperature distribution diagram using a thermal resistance network model.

[0032] Specifically, the surface temperature scanning of the target MOS tube to calculate the junction temperature data of the target MOS tube includes performing non-contact temperature measurement on the surface of the target MOS tube by an infrared thermal imager to obtain a surface temperature distribution map of the target MOS tube. In the power management system of an electric vehicle, this process first requires the use of an infrared thermal imager to perform non-contact temperature measurement on the surface of the MOS. The infrared thermal imager can capture the temperature distribution of the MOS surface in real time and accurately, and generate a detailed surface temperature distribution map. For example, during the acceleration process of an electric vehicle, the system will start the infrared thermal imager to scan the surface of the MOS, record the temperature values ​​at each position, and form a high-resolution temperature distribution map. This temperature distribution map can intuitively display the temperature changes on the surface of the MOS and provide basic data for subsequent temperature analysis. Next, the surface temperature distribution map is subjected to temperature frequency domain analysis using the Fourier transform algorithm to extract the temperature distribution frequency component. Fourier transform is a commonly used signal processing technology that can convert time domain signals into frequency domain signals, thereby extracting the frequency components in the signal. In this example, the system converts the collected surface temperature distribution map into a frequency domain signal, and extracts the frequency components of the temperature distribution through the Fourier transform algorithm. These frequency components reflect the spatial variation of temperature, which is of great significance for understanding the heat conduction process inside the MOS. For example, through Fourier transform, the system can identify the main frequency components in the surface temperature distribution of the MOS, which may correspond to the heat conduction paths and rates of different regions inside the MOS. Then, through the inverse heat conduction model, the internal heat distribution of the target MOS tube is inverted and calculated based on the temperature distribution frequency components to obtain the internal temperature distribution map. The inverse heat conduction model is a mathematical model based on physical principles. Through the known surface temperature distribution and heat conduction equation, the temperature distribution inside the MOS is inverted and calculated. In the power management system of electric vehicles, the system uses the inverse heat conduction model, combined with the temperature distribution frequency components extracted by Fourier transform, to perform an inversion calculation of the internal heat distribution. This process can infer the temperature distribution of each part inside the MOS and generate an internal temperature distribution map. For example, through the inverse heat conduction model, the system can calculate the hot spot location and temperature gradient inside the MOS. This information is very important for evaluating the thermal performance of the MOS and potential overheating risks. Finally, the junction temperature data of the target MOS tube is calculated based on the internal temperature distribution map using the thermal resistance network model. The thermal resistance network model is a simplified but effective heat conduction model. By abstracting the heat conduction path inside the MOS into a series of thermal resistances and thermal capacitances, the junction temperature of the MOS can be easily calculated. In this example, the system uses the thermal resistance network model, combined with the internal temperature distribution map, to calculate the junction temperature data of the MOS. Junction temperature data is a key parameter for evaluating the thermal performance of the MOS. Excessively high junction temperature can cause the performance of the MOS to degrade or even damage it.For example, through the thermal resistance network model, the system can calculate the junction temperature of MOS under the current working conditions. If the junction temperature exceeds the preset safety threshold, the system will adjust the bias voltage in time to reduce the temperature of MOS and ensure its safe operation. Through this series of temperature scanning and analysis processes, the system can fully and accurately grasp the temperature distribution of MOS and provide reliable data support for dynamically adjusting the bias voltage of MOS. This can not only improve the efficiency and reliability of the power management system, but also ensure that MOS is always in the best working state in complex and changing application environments, extend its service life and reduce maintenance costs. For example, during the acceleration process of electric vehicles, the system can adjust the bias voltage of MOS in time through this series of temperature monitoring and analysis to ensure that it remains efficient and safe under high load conditions.

[0033] In a specific embodiment, the process of training the bias voltage requirement prediction model includes: Step S31, obtaining historical operating parameters, preprocessing the historical operating parameters to obtain a feature vector set, and dividing the feature vector set into a training set and a standard bias voltage corresponding to the training set; wherein the training set includes an original electrical signal data set, a switching frequency, a duty cycle information, a junction temperature data, and parameter statistics; Step S32, inputting the training set into a preset initial bias voltage demand prediction model for calculation to obtain a bias voltage prediction value; Step S33, comparing and calculating the standard bias voltage with the bias voltage prediction value to calculate a bias voltage error; Step S34, determining whether the bias voltage error is within a preset bias voltage error range, and if the bias voltage error is within the preset bias voltage error range, determining the initial bias voltage demand prediction model as the bias voltage demand prediction model; Step S35, if the bias voltage error exceeds the preset bias voltage error range, the parameters in the initial bias voltage requirement prediction model are adjusted, and steps S31 to S34 are repeated until the bias voltage output by the initial bias voltage requirement prediction model is within the preset range to obtain a bias voltage requirement prediction model.

[0034] Specifically, the process of training the bias voltage demand prediction model includes multiple steps, which together ensure that the model can accurately predict the optimal bias voltage of MOS under different working conditions. First, step S31 involves obtaining historical operating parameters, preprocessing the historical operating parameters, obtaining a feature vector set, and dividing the feature vector set into a training set and a standard bias voltage corresponding to the training set. In the power management system of an electric vehicle, this process first requires extracting various operating parameters of MOS from historical operating data, including original electrical signal data sets, switching frequency, duty cycle information, junction temperature data, and parameter statistics. These data need to be preprocessed, such as removing noise, filling missing values, and normalization, to ensure the quality and consistency of the data. The preprocessed data will be constructed into a feature vector set, which will be divided into a training set and a validation set. The training set includes all the above-mentioned feature vectors, and the standard bias voltage is based on the best bias voltage value known in the historical data. Next, step S32 inputs the training set into a preset initial bias voltage demand prediction model for calculation to obtain a bias voltage prediction value. This process usually uses machine learning algorithms, such as neural networks, support vector machines, or decision trees. The initial bias voltage demand prediction model is an untrained model, and its internal parameters need to be adjusted through the data of the training set. For example, in the power management system of an electric vehicle, the system inputs the feature vectors in the training set into the initial bias voltage demand prediction model, and the model calculates the corresponding bias voltage prediction value based on these feature vectors. Then, step S33 compares and calculates the bias voltage prediction value based on the standard bias voltage to calculate the bias voltage error. This process is to evaluate the prediction accuracy of the model. The system compares the bias voltage prediction value output by the model with the corresponding standard bias voltage in the training set to calculate the error between the two. The error can be measured by indicators such as mean square error (MSE) and mean absolute error (MAE). For example, in the power management system of an electric vehicle, the system calculates the error between the bias voltage prediction value and the standard bias voltage of each sample, and summarizes these error values ​​to obtain the overall error index. Next, step S34 determines whether the bias voltage error is within the preset bias voltage error range. If the bias voltage error is within the preset bias voltage error range, the initial bias voltage demand prediction model is determined as the bias voltage demand prediction model. The preset bias voltage error range is a threshold used to determine whether the model's prediction is accurate enough. If the error is within an acceptable range, it means that the model has been trained to a satisfactory level and can be directly used in practical applications. For example, in the power management system of an electric vehicle, if the error index of the model is lower than the preset threshold, the system will determine the current initial bias voltage demand prediction model as the final bias voltage demand prediction model, which is used to predict the optimal bias voltage of the MOS in real time.If the bias voltage error exceeds the preset bias voltage error range, then step S35 is entered to adjust the parameters in the initial bias voltage demand prediction model, and steps S31 to S34 are repeated until the bias voltage output by the initial bias voltage demand prediction model is within the preset range, so as to obtain the bias voltage demand prediction model. This process is an iterative training process of the model, and the prediction error is gradually reduced by continuously adjusting the parameters of the model until the preset error range is met. For example, in the power management system of an electric vehicle, if the error index of the model exceeds the preset threshold, the system will adjust the parameters of the model, such as the learning rate, weight, etc., and then retrain and verify until the prediction error of the model reaches an acceptable range. This process may require multiple iterations until the prediction performance of the model is optimal. Through this series of training and optimization processes, the system can establish an accurate bias voltage demand prediction model, which can predict the optimal bias voltage of MOS in real time under different working conditions, thereby improving the efficiency and reliability of the power management system. For example, during the acceleration of an electric vehicle, the system can adjust the bias voltage of the MOS in real time according to the current working parameters through the trained bias voltage demand prediction model, ensuring that it can still maintain efficient and safe working conditions under high load conditions. This can not only improve the overall performance of the system, but also extend the service life of the MOS and reduce maintenance costs.

[0035] In a specific embodiment, the step of performing digital signal processing on the bias voltage requirement value to convert it into a control signal includes: quantizing the continuous bias voltage requirement value to obtain a discrete quantized value; Performing digital signal processing on the discrete quantized value to obtain a processed digital signal; Performing signal encoding on the processed digital signal to obtain an encoded control signal; Inputting the coded control signal into a preset digital-to-analog converter for signal conversion to obtain an analog control signal; The analog control signal is used as a control signal.

[0036] Specifically, the bias voltage requirement value is subjected to digital signal processing to be converted into a control signal, including quantizing the continuous bias voltage requirement value to obtain a discrete quantized value. In the power management system of an electric vehicle, this process first requires converting the continuous bias voltage requirement value obtained from the prediction model into a discrete digital form. For example, when the system predicts the bias voltage required by the MOS based on the current operating parameters, the bias voltage requirement value is a continuous analog signal. In order to facilitate subsequent digital signal processing, an analog-to-digital converter (ADC) is required to convert this continuous analog signal into a discrete digital signal. The quantization process usually involves mapping a continuous voltage value to a set of discrete values, for example, mapping a voltage range of 0 to 5V to an integer value of 0 to 255. In this way, the continuous bias voltage requirement value is converted into a discrete quantized value. Next, the discrete quantized value is subjected to digital signal processing to obtain a processed digital signal. This process is usually performed by a microcontroller or a digital signal processor (DSP). The purpose of digital signal processing is to further optimize and filter to improve the accuracy and stability of the signal. For example, the system can use a digital filter (such as a low-pass filter or a high-pass filter) to remove noise and interference from the quantized value. In addition, other digital signal processing operations, such as signal smoothing, de-jittering, etc., can be performed to ensure that the final control signal can accurately reflect the bias voltage requirements of the MOS. For example, during the acceleration process of an electric vehicle, the system digitally filters the received discrete quantized values ​​to eliminate fluctuations caused by sensor noise or external interference, thereby obtaining a more stable processed digital signal. Then, the processed digital signal is signal encoded to obtain an encoded control signal. Signal encoding is the conversion of the processed digital signal into a format suitable for transmission and control. For example, the system can use a pulse width modulation (PWM) encoding method to convert the processed digital signal into a series of signals with different pulse widths. These pulse signals can be directly used to control the gate drive circuit of the MOS. The duty cycle of the PWM signal can reflect the required value of the bias voltage, and by adjusting the duty cycle, the output voltage of the gate drive circuit can be accurately controlled. For example, in the power management system of an electric vehicle, the system encodes the processed digital signal into a PWM signal, which can accurately reflect the bias voltage required by the MOS. Finally, the coded control signal is input into a preset digital-to-analog converter for signal conversion to obtain an analog control signal. The function of the digital-to-analog converter (DAC) is to convert digital signals into analog signals for the use of the driving circuit. For example, the encoded PWM signal is input into the DAC, which converts these digital signals into continuous analog voltage signals. These analog voltage signals can be directly applied to the gate of the MOS to achieve precise control of the MOS bias voltage.For example, during the acceleration of an electric vehicle, the system will input the encoded PWM signal into the DAC, which will convert these digital signals into continuous analog voltage signals, which will be applied to the gate of the MOS to achieve dynamic adjustment of the MOS bias voltage. Through this series of digital signal processing and conversion processes, the system can accurately convert the continuous bias voltage requirement value obtained from the prediction model into a control signal suitable for MOS gate drive. This not only ensures that the MOS is always in the best working state under different working conditions, but also improves the overall efficiency and reliability of the power management system. For example, during the acceleration of an electric vehicle, the system can adjust the bias voltage of the MOS in real time through this series of signal processing and conversion to ensure that it remains efficient and safe under high load conditions, thereby improving the performance and stability of the entire power management system.

[0037] In a specific embodiment, applying the control signal to the gate of the target MOS transistor to dynamically adjust the working state of the target MOS transistor includes: Transmitting the control signal to the gate drive circuit to obtain a transmission control signal; Amplifying and shaping the transmission control signal through a gate drive circuit to obtain an amplified control signal; Performing overvoltage and overcurrent protection on the amplified control signal through an amplifier in a gate drive circuit to obtain a protection control signal; Electrically isolating the protection control signal to obtain an isolation control signal; Performing signal buffering processing on the isolation control signal to obtain a buffered control signal; The buffer control signal is applied to the gate of the target MOS transistor to dynamically adjust the working state of the target MOS transistor.

[0038] Specifically, the control signal is applied to the gate of the target MOS tube to dynamically adjust the working state of the target MOS tube, including transmitting the control signal to the gate drive circuit to obtain a transmission control signal. In the power management system of the electric vehicle, this process first requires the analog control signal output from the digital-to-analog converter to be transmitted to the gate drive circuit. For example, after the system obtains the analog control signal through digital signal processing and digital-to-analog conversion, these signals will be transmitted to the gate drive circuit. The purpose of transmitting the control signal is to ensure that the control signal can reach the gate drive circuit without loss, providing a reliable basis for subsequent signal processing. Next, the transmission control signal is amplified and shaped by the gate drive circuit to obtain an amplified control signal. The gate drive circuit usually includes an amplifier and a shaping circuit for enhancing the amplitude of the control signal and improving the waveform quality of the signal. For example, during the acceleration process of the electric vehicle, the gate drive circuit will amplify the received transmission control signal to ensure that the signal has sufficient driving capability to control the gate of the MOS. At the same time, the shaping circuit will ensure that the waveform of the signal meets the working requirements of the MOS, for example, the edge of the PWM signal is made steeper to reduce switching losses. Then, the amplifier in the gate drive circuit performs overvoltage and overcurrent protection on the amplified control signal to obtain a protection control signal. Overvoltage and overcurrent protection are important measures to ensure the safe operation of MOS. The amplifier in the gate drive circuit is usually equipped with overvoltage and overcurrent protection functions, which can automatically cut off or limit the signal when the control signal exceeds the safety range. For example, if the system detects that the gate voltage is too high or the current is too large, the amplifier will immediately take protective measures to prevent the MOS from being damaged by overvoltage or overcurrent. The protection control signal ensures that the MOS will not be subjected to excessive voltage or current shocks under any circumstances, thereby improving the reliability and safety of the system. Next, the protection control signal is electrically isolated to obtain an isolation control signal. Electrical isolation is to prevent electrical interference between the high-voltage side and the low-voltage side to ensure the safe operation of the system. The gate drive circuit usually uses isolation elements such as optocouplers or transformers to achieve electrical isolation. For example, in the power management system of an electric vehicle, the gate drive circuit will isolate the protection control signal through an optocoupler to ensure that the signal on the high-voltage side will not interfere with the control circuit on the low-voltage side. The isolation control signal not only improves the safety of the system, but also enhances the anti-interference ability of the system. Then, the isolation control signal is buffered to obtain a buffered control signal. The purpose of the signal buffering is to further enhance the driving capability and stability of the signal. The buffer in the gate drive circuit can provide additional current driving capability to ensure that the signal can reliably drive the gate of the MOS. For example, the buffer will further amplify and stabilize the received isolation control signal to ensure that the amplitude and waveform of the signal meet the working requirements of the MOS.The buffer control signal can ensure that the gate voltage of the MOS remains stable under different working conditions, thereby improving the working performance of the MOS. Finally, the buffer control signal is applied to the gate of the target MOS tube to dynamically adjust the working state of the target MOS tube. This process is the final step of the entire control link. By applying the buffer control signal to the gate of the MOS, accurate control of the MOS bias voltage can be achieved. For example, during the acceleration of an electric vehicle, the system will apply the buffer control signal to the gate of the MOS to dynamically adjust the bias voltage of the MOS to ensure that it remains efficient and safe under high load conditions. Through this series of signal processing and control, the system can adjust the working state of the MOS in real time and accurately, improve the overall efficiency and reliability of the power management system, extend the service life of the MOS, and reduce maintenance costs. Through this series of steps, the system can obtain the bias voltage requirement value from the prediction model, after digital signal processing, transmission, amplification, protection, isolation and buffering, and finally accurately apply it to the gate of the MOS to achieve dynamic adjustment of the working state of the MOS. This can not only ensure that the MOS is always in the best working state under different working conditions, but also improve the performance and stability of the entire power management system. For example, during the acceleration of an electric vehicle, the system can adjust the bias voltage of the MOS in real time through this series of control steps to ensure that it remains efficient and safe under high load conditions, thereby improving the performance and reliability of the entire power management system.

[0039] In a specific embodiment, the step of performing overvoltage and overcurrent protection on the amplified control signal through an amplifier in the gate drive circuit to obtain a protection control signal includes: Performing spectrum decomposition on the amplification control signal to obtain a spectrum component set; Constructing a signal energy graph based on the spectral component set to obtain signal energy information; Determine whether the signal energy information has risk signal characteristics, and if so, extract characteristics of the signal energy information to obtain a risk characteristic signal; Analyzing the type of the risk characteristic signal by a preset analysis algorithm, and formulating a protection control signal strategy based on the type of the risk characteristic signal; Through the protection control signal strategy, the amplifier in the gate drive circuit is used to perform overvoltage and overcurrent protection on the amplified control signal to obtain a protection control signal.

[0040] Specifically, the amplifier in the gate drive circuit performs overvoltage and overcurrent protection on the amplified control signal to obtain a protection control signal, including performing spectrum decomposition on the amplified control signal to obtain a spectrum component set. In the power management system of the electric vehicle, this process first requires spectrum analysis of the amplified control signal output from the gate drive circuit amplifier. Spectral decomposition is a signal processing technology that can convert time domain signals into frequency domain signals to reveal the frequency composition of the signal. For example, the system uses a fast Fourier transform (FFT) algorithm to perform spectrum decomposition on the amplified control signal to obtain a spectrum component set containing different frequency components. These spectrum component sets can help the system identify high-frequency noise, harmonics and other potential interference components in the signal. Next, a signal energy graph is constructed based on the spectrum component set to obtain signal energy information. The signal energy graph is a visualization tool that can display the energy distribution of the signal at different frequencies. By constructing the signal energy graph, the system can more intuitively understand the energy distribution of the amplified control signal and identify possible abnormal signals. For example, during the acceleration of the electric vehicle, the system constructs a signal energy graph based on the spectrum component set to analyze the energy distribution of the signal at different frequencies. If the energy at certain frequencies is found to be abnormally high, it may indicate that there is noise or interference in the signal. Then, it is determined whether the signal energy information has risk signal characteristics. If so, the signal energy information is feature extracted to obtain a risk feature signal. This process involves analyzing the abnormal energy distribution in the signal energy diagram to identify possible risk signal characteristics. For example, the system will set some thresholds. If the energy at a certain frequency exceeds the preset threshold, the system will consider that the signal at this frequency may be risky. Once the risk signal characteristics are detected, the system will further perform feature extraction to extract specific features such as frequency, amplitude and duration to form a risk feature signal. For example, if the system detects that the energy at a certain high frequency band is abnormally high, the detailed features of the frequency band may be extracted for subsequent analysis. Next, the type of the risk feature signal is analyzed by a preset analysis algorithm, and a protection control signal strategy is formulated based on the type of the risk feature signal. The preset analysis algorithm can identify specific fault types such as overvoltage, overcurrent, high-frequency noise, etc. based on the extracted risk feature signal. For example, the system will use a classification algorithm (such as a support vector machine, a decision tree, etc.) to classify the risk feature signal and determine its specific type. Once the type of risk characteristic signal is determined, the system will formulate a corresponding protection control signal strategy based on the preset protection strategy. For example, if an overvoltage signal is detected, the system will formulate a strategy to reduce the gate drive voltage; if an overcurrent signal is detected, the system will formulate a strategy to limit the gate drive current. Finally, through the protection control signal strategy, the amplifier in the gate drive circuit is used to perform overvoltage and overcurrent protection on the amplified control signal to obtain a protection control signal.This process is to apply the formulated protection control signal strategy to the actual gate drive circuit, and appropriately adjust the amplified control signal through the amplifier to achieve overvoltage and overcurrent protection. For example, if the system detects an overvoltage signal, it will reduce the gate drive voltage through the amplifier to prevent the MOS from being damaged due to excessive gate voltage. If an overcurrent signal is detected, the gate drive current will be limited by the amplifier to prevent the MOS from overheating due to excessive current. Through these protection measures, the system can ensure that the MOS is always in a safe working state under various working conditions. Through this series of steps, the system can comprehensively analyze and protect the amplified control signal to ensure that the MOS is always in the best and safe working state under different working conditions. For example, during the acceleration process of electric vehicles, the system can timely detect and handle possible overvoltage and overcurrent risks through spectrum decomposition, signal energy graph construction, risk feature signal extraction, analysis algorithm and protection control signal strategy, ensuring that the MOS still maintains an efficient and safe working state under high load conditions, thereby improving the performance and reliability of the entire power management system.

[0041] The above describes the adaptive bias voltage optimization method based on MOS tube in the embodiment of the present invention. The following describes the adaptive bias voltage optimization device based on MOS tube in the embodiment of the present invention. Figure 2 In one embodiment of the present invention, an adaptive bias voltage optimization device based on a MOS tube includes: The monitoring module 21 is used to monitor the target MOS tube in the working state of the gate driving circuit in real time to obtain the working parameters of the target MOS tube; A prediction module 22, configured to predict the bias voltage requirement of the target MOS tube based on the preset bias voltage requirement prediction model and the operating parameters to obtain a bias voltage requirement value; A conversion module 23, used for performing digital signal processing on the bias voltage requirement value to convert it into a control signal; The applying module 24 is used to apply the control signal to the gate of the target MOS transistor to dynamically adjust the working state of the target MOS transistor.

[0042] In this embodiment, for the specific implementation of each unit in the above device embodiment, please refer to the above method embodiment, which will not be repeated here.

[0043] Reference Figure 3 The present invention also provides a computer device in an embodiment, wherein the internal structure of the computer device can be as follows: Figure 3As shown. The computer device includes a processor, a memory, a display screen, an input device, a network interface and a database connected through a system bus. Among them, the processor designed by the computer is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0044] Those skilled in the art will understand that Figure 3 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0045] An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the above method is implemented. It can be understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0046] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided by the present invention and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double-speed data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM.

[0047] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0048] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for optimizing an adaptive bias voltage based on a MOS tube, characterized in that: Applied to the gate drive circuit, including the following steps: Real-time monitoring of the target MOS tube under the working state of the gate drive circuit is performed to obtain the working parameters of the target MOS tube; Based on a preset bias voltage requirement prediction model, and based on the operating parameters, a bias voltage requirement prediction is performed on the target MOS tube to obtain a bias voltage requirement value; Performing digital signal processing on the bias voltage requirement value to convert it into a control signal; The control signal is applied to the gate of the target MOS transistor to dynamically adjust the working state of the target MOS transistor.

2. The method for optimizing the adaptive bias voltage based on MOS tube according to claim 1, characterized in that: The real-time monitoring of the target MOS tube in the working state of the gate drive circuit to obtain the working parameters of the target MOS tube includes: Synchronously sampling the gate voltage, drain voltage and drain current of the target MOS tube in the working state of the gate drive circuit to obtain an original electrical signal data set, wherein the circuit working state includes the turning on, amplification and cutting off of the target MOS tube; Performing time series processing on the original electrical signal data set to obtain switching frequency and duty cycle information of the target MOS tube; Scanning the surface temperature of the target MOS tube to calculate the junction temperature data of the target MOS tube; Using a sliding window method, based on the switching frequency, duty cycle information and junction temperature data, the operating parameter statistics of the target MOS tube in different time periods are calculated; wherein the operating parameter statistics include an average gate-source voltage, an average drain-source voltage and an average drain current; A standard bias voltage is set based on the original electrical signal data set, switching frequency, duty cycle information, junction temperature data and parameter statistics, and the original electrical signal data set, switching frequency, duty cycle information, junction temperature data, parameter statistics and standard bias voltage are used as operating parameters of the target MOS tube.

3. The method for optimizing the MOS transistor-based adaptive bias voltage according to claim 2, characterized in that: Scanning the surface temperature of the target MOS tube to calculate the junction temperature data of the target MOS tube includes: Performing non-contact temperature measurement on the surface of the target MOS tube by using an infrared thermal imager to obtain a surface temperature distribution diagram of the target MOS tube; Using Fourier transform algorithm, performing temperature frequency domain analysis on the surface temperature distribution diagram to extract the temperature distribution frequency component; By using a reverse heat conduction model, an internal heat distribution inversion calculation is performed on the target MOS tube based on the temperature distribution frequency component to obtain an internal temperature distribution diagram; The junction temperature data of the target MOS tube is calculated based on the internal temperature distribution diagram using a thermal resistance network model.

4. The method for optimizing the adaptive bias voltage based on MOS tube according to claim 1, characterized in that: The process of training the bias voltage requirement prediction model includes: Step S31, obtaining historical operating parameters, preprocessing the historical operating parameters to obtain a feature vector set, and dividing the feature vector set into a training set and a standard bias voltage corresponding to the training set; wherein the training set includes an original electrical signal data set, a switching frequency, a duty cycle information, a junction temperature data, and parameter statistics; Step S32, inputting the training set into a preset initial bias voltage demand prediction model for calculation to obtain a bias voltage prediction value; Step S33, comparing and calculating the standard bias voltage with the bias voltage prediction value to calculate a bias voltage error; Step S34, determining whether the bias voltage error is within a preset bias voltage error range, and if the bias voltage error is within the preset bias voltage error range, determining the initial bias voltage demand prediction model as the bias voltage demand prediction model; Step S35, if the bias voltage error exceeds the preset bias voltage error range, the parameters in the initial bias voltage requirement prediction model are adjusted, and steps S31 to S34 are repeated until the bias voltage output by the initial bias voltage requirement prediction model is within the preset range to obtain a bias voltage requirement prediction model.

5. The method for optimizing the adaptive bias voltage based on MOS tube according to claim 1, characterized in that: The step of performing digital signal processing on the bias voltage requirement value to convert it into a control signal includes: quantizing the continuous bias voltage requirement value to obtain a discrete quantized value; Performing digital signal processing on the discrete quantized value to obtain a processed digital signal; Performing signal encoding on the processed digital signal to obtain an encoded control signal; Inputting the coded control signal into a preset digital-to-analog converter for signal conversion to obtain an analog control signal; The analog control signal is used as a control signal.

6. The method for optimizing the adaptive bias voltage based on MOS tube according to claim 1, characterized in that: The step of applying the control signal to the gate of the target MOS transistor to dynamically adjust the working state of the target MOS transistor includes: Transmitting the control signal to the gate drive circuit to obtain a transmission control signal; Amplifying and shaping the transmission control signal through a gate drive circuit to obtain an amplified control signal; Performing overvoltage and overcurrent protection on the amplified control signal through an amplifier in a gate drive circuit to obtain a protection control signal; Electrically isolating the protection control signal to obtain an isolation control signal; Performing signal buffering processing on the isolation control signal to obtain a buffered control signal; The buffer control signal is applied to the gate of the target MOS transistor to dynamically adjust the working state of the target MOS transistor.

7. The method for optimizing the adaptive bias voltage based on MOS tube according to claim 6, characterized in that: The step of performing overvoltage and overcurrent protection on the amplified control signal through an amplifier in the gate drive circuit to obtain a protection control signal comprises: Performing spectrum decomposition on the amplification control signal to obtain a spectrum component set; Constructing a signal energy graph based on the spectral component set to obtain signal energy information; Determine whether the signal energy information has risk signal characteristics, and if so, extract characteristics of the signal energy information to obtain a risk characteristic signal; Analyzing the type of the risk characteristic signal by a preset analysis algorithm, and formulating a protection control signal strategy based on the type of the risk characteristic signal; Through the protection control signal strategy, the amplifier in the gate drive circuit is used to perform overvoltage and overcurrent protection on the amplified control signal to obtain a protection control signal.

8. An adaptive bias voltage optimization device based on MOS tube, characterized in that: Applied to gate drive circuits, including: A monitoring module, used to monitor the target MOS tube in the working state of the gate drive circuit in real time to obtain the working parameters of the target MOS tube; A prediction module, configured to predict the bias voltage requirement of the target MOS tube based on the working parameters based on a preset bias voltage requirement prediction model, and obtain a bias voltage requirement value; A conversion module, used for performing digital signal processing on the bias voltage requirement value to convert it into a control signal; The applying module is used to apply the control signal to the gate of the target MOS tube to dynamically adjust the working state of the target MOS tube.

9. A computer device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images