Intelligent control method, system and device for gate voltage of gallium nitride power transistor

Abstract

The application discloses a kind of gate voltage intelligent control method, system and device of gallium nitride power transistor, it is related to communication technical field, by collecting the power supply voltage and power supply current of gallium nitride power transistor, and acquisition detection signal;Average direct current power is calculated based on power supply parameter, average output power is calculated based on detection signal, and then average dissipation power is obtained;With average dissipation power as input, based on the equivalent thermal network model of the thermal dynamic relationship between dissipation power and junction temperature that is pre-set, the junction temperature of gallium nitride power transistor is estimated in real time;With the first frequency monitoring junction temperature change, when exceeding preset threshold, generate gate voltage compensation signal according to voltage change slope limit mode;Meanwhile, with the second frequency higher than the first frequency, drain electrode electric parameter and temperature state are monitored, when meeting abnormal event condition, negative adjustment or shutdown operation is executed to gate voltage, and compensation adjustment is frozen.The application can realize accurate, stable gate voltage compensation.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of communication technology, and in particular to a gate voltage intelligent control method, system and device of a gallium nitride power transistor. BACKGROUND

[0002] GaNHEMT (Gallium Nitride High Electron Mobility Transistor) power amplifier is widely used in modern communication systems due to its high power and high efficiency. However, its large thermal resistance leads to a significant rise in semiconductor junction temperature under high duty cycle or continuous working conditions, which in turn causes problems such as static operating point drift, gain variation and linearity deterioration. In order to ensure stable performance of the power amplifier, the gate bias voltage must be accurately temperature-compensated.

[0003] Currently, the gate bias voltage of GaNHEMT is temperature-compensated by using the ambient temperature direct compensation method: the ambient temperature around the power amplifier is directly collected by a temperature sensor, the temperature signal is input to a microcontroller after analog-to-digital conversion, and the microcontroller calculates the required gate voltage adjustment amount through a pre-set temperature-gate voltage compensation curve (usually obtained by polynomial fitting or table lookup), and then outputs the adjusted gate voltage to the gate of the power transistor through a digital-to-analog converter. The defects of the current scheme mainly come from two reasons: first, due to the hysteresis and deviation caused by thermal resistance between the ambient temperature and the actual working temperature of the power transistor semiconductor junction, the compensation based on the ambient temperature cannot accurately reflect the real change of the junction temperature, resulting in insufficient compensation accuracy; second, the compensation loop has a fast response speed, and its bandwidth easily covers the envelope frequency of the radio frequency signal, causing the compensation action to couple with the radio frequency envelope, resulting in gain fluctuation and even system instability.

[0004] Therefore, the prior art cannot achieve accurate and stable gate voltage compensation and fast and safe protection based on the real junction temperature of the power transistor while avoiding radio frequency envelope interference. SUMMARY

[0005] The gallium nitride power transistor gate voltage intelligent control method, system and device provided by the embodiments of the present application can achieve accurate and stable gate voltage compensation and fast and safe protection based on the real junction temperature of the power transistor while avoiding radio frequency envelope interference.

[0006] An embodiment of the present application provides a gallium nitride power transistor gate voltage intelligent control method, comprising:

[0007] The power supply voltage and current of the gallium nitride power transistor are collected, and the radio frequency output signal of the gallium nitride power transistor is power-detected to obtain a detected signal;

[0008] calculate an average DC power based on the supply voltage and the supply current, and calculate an average output power based on the detection signal;

[0009] calculate an average dissipation power of the gallium nitride power transistor based on the average DC power and the average output power;

[0010] estimate a junction temperature of the gallium nitride power transistor in real time based on a preset equivalent thermal network model reflecting a thermal dynamic relationship between a dissipation power and the junction temperature, with the average dissipation power as an input;

[0011] monitor a change in the junction temperature at a first frequency, and generate a gate voltage compensation signal according to a preset voltage change slope limitation mode to compensate and adjust a gate voltage of the gallium nitride power transistor when a change amount of the junction temperature detected exceeds a preset threshold;

[0012] monitor a drain electrical parameter and a temperature state of the gallium nitride power transistor at a second frequency higher than the first frequency to obtain a monitoring result;

[0013] in response to the monitoring result satisfying a preset abnormal event condition, perform a negative adjustment operation or an off level operation on the gate voltage of the gallium nitride power transistor, and freeze the compensation adjustment until the monitoring result does not satisfy the preset abnormal event condition.

[0014] As an improvement of the above scheme, the calculation of the average DC power based on the supply voltage and the supply current, and the calculation of the average output power based on the detection signal, comprises:

[0015] perform a first low-pass filtering process with a first preset time constant on the supply voltage and the supply current to obtain a first filtered signal;

[0016] calculate the average DC power based on a filtered supply voltage and a filtered supply current in the first filtered signal;

[0017] perform a second low-pass filtering process with a second preset time constant on the detection signal to obtain a second filtered signal;

[0018] calculate the average output power based on the second filtered signal.

[0019] As an improvement of the above scheme, the estimation of the junction temperature of the gallium nitride power transistor in real time based on the preset equivalent thermal network model reflecting the thermal dynamic relationship between the dissipation power and the junction temperature, comprises:

[0020] input the average dissipated power into the equivalent thermal network model containing a thermal resistance parameter and a thermal capacitance parameter; wherein the equivalent thermal network model outputs the junction temperature estimated in real time at the current moment by a first-order thermal dynamic relationship of resistance-capacitance, according to the average dissipated power at the current moment, the junction temperature estimated at the previous moment, and an ambient temperature, wherein a time constant of change of the junction temperature is determined by a product of the thermal resistance parameter and the thermal capacitance parameter.

[0021] As an improvement of the above scheme, the change of the junction temperature is monitored at the first frequency, and when the change amount of the junction temperature is detected to exceed a preset threshold, a gate voltage compensation signal is generated according to a preset voltage change slope limiting mode, comprising:

[0022] The estimated junction temperature is periodically acquired at the first frequency;

[0023] The difference between the junction temperature acquired in the current period and the junction temperature acquired in the last period is calculated as the change amount of the junction temperature;

[0024] When the change amount exceeds the preset threshold, an initial compensation voltage value is calculated according to the change amount and a preset temperature-voltage compensation coefficient, and the change rate of the initial compensation voltage value is limited to be not greater than a preset change rate safety value, so as to generate the gate voltage compensation signal.

[0025] As an improvement of the above scheme, the drain electrical parameter and temperature state of the gallium nitride power transistor are monitored in real time at a second frequency higher than the first frequency, and a monitoring result is obtained, comprising:

[0026] The drain voltage instantaneous value and the drain current instantaneous value of the gallium nitride power transistor are collected at the second frequency higher than the first frequency;

[0027] Based on the collected drain voltage instantaneous value, it is judged whether there is an overvoltage spike event;

[0028] Based on the collected drain current instantaneous value, the change rate thereof is calculated and it is judged whether there is an overcurrent change rate event;

[0029] A temperature detection signal from a temperature sensor is received, and it is judged whether there is an overtemperature event;

[0030] Based on the judgment results of the overvoltage spike event, the overcurrent change rate event and the overtemperature event, the monitoring result is generated.

[0031] As an improvement of the above-mentioned scheme, the response to the monitoring result satisfying the preset abnormal event condition, the gate voltage of the gallium nitride power transistor is executed negative adjustment operation or off level operation, and the compensation adjustment is frozen until the monitoring result does not satisfy the preset abnormal event condition, comprising:

[0032] When the monitoring result satisfies the preset abnormal event condition, a control signal is generated;The abnormal event condition includes at least one of the occurrence of the overvoltage peak event, the overcurrent rate event or the overtemperature event;

[0033] According to the control signal, the gate voltage of the gallium nitride power transistor is driven to a preset negative voltage off level;

[0034] At the same time of generating the control signal, the adjustment effect of the gate voltage compensation signal on the gate voltage is blocked to freeze the compensation adjustment until the monitoring result indicates that all abnormal events disappear.

[0035] As an improvement of the above-mentioned scheme, the first frequency is 2Hz, and the second frequency is 1kHz-10kHz.

[0036] As an improvement of the above-mentioned scheme, before collecting the supply voltage and supply current of the gallium nitride power transistor, and power detection of the radio frequency output signal of the gallium nitride power transistor to obtain the detection signal, the method further comprises:

[0037] After the system power is turned on, the gate voltage of the gallium nitride power transistor is set to a negative voltage safety bias level, and the drain supply voltage of the gallium nitride power transistor is gradually increased from zero to a target working voltage;

[0038] After confirming that the drain supply voltage is stable at the target working voltage, the gate voltage is adjusted from the negative voltage safety bias level to an initial working bias point with a controlled slope.

[0039] Another embodiment of the application provides a gallium nitride power transistor gate voltage intelligent control system, comprising:

[0040] The acquisition module is used for acquiring the supply voltage and supply current of the gallium nitride power transistor, and performing power detection on the radio frequency output signal of the gallium nitride power transistor to obtain a detection signal;

[0041] The first calculation module is used for calculating the average direct current power based on the supply voltage and the supply current, and calculating the average output power based on the detection signal;

[0042] The second calculation module is used for calculating the average dissipation power of the gallium nitride power transistor based on the average direct current power and the average output power.

[0043] an estimation module, configured to estimate a junction temperature of the gallium nitride power transistor in real time based on a preset equivalent thermal network model reflecting a thermal dynamic relationship between dissipated power and junction temperature, with the average dissipated power as input;

[0044] a compensation module, configured to monitor a change of the junction temperature at a first frequency, and generate a gate voltage compensation signal according to a preset voltage change slope limit mode to compensate and adjust a gate voltage of the gallium nitride power transistor when detecting that a change amount of the junction temperature exceeds a preset threshold;

[0045] a monitoring module, configured to monitor a drain electrical parameter and a temperature state of the gallium nitride power transistor at a second frequency higher than the first frequency to obtain a monitoring result;

[0046] an operation module, configured to perform a negative adjustment operation or an off level operation on the gate voltage of the gallium nitride power transistor and freeze the compensation adjustment until the monitoring result does not satisfy a preset abnormal event condition, in response to the monitoring result satisfying the preset abnormal event condition.

[0047] Another embodiment of the present application provides a gallium nitride power transistor gate voltage intelligent control device, which comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and the processor implements the gallium nitride power transistor gate voltage intelligent control method of the above-mentioned embodiments of the present application when executing the computer program.

[0048] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0049] Firstly, by collecting the supply voltage and supply current and power detection of the radio frequency output signal, the average dissipation power is calculated, which is directly related to the internal heat generation of the device, providing an accurate input for subsequent junction temperature estimation that is independent of the ambient temperature, thereby solving the problem of insufficient compensation accuracy; secondly, the average dissipation power is input into the equivalent thermal network model to estimate the junction temperature, and the model simulates the thermal inertia of the device through the thermal resistance and heat capacity parameters, so that the estimated junction temperature changes smoothly and truly reflects the thermal dynamic process; on this basis, a dual-frequency monitoring and control method is adopted: a lower first frequency is used to monitor the change of the estimated junction temperature, and only when the change exceeds the preset threshold, a gate voltage compensation signal is generated according to the preset voltage change slope limiting method to perform slow compensation, and this low-frequency and slope-limited compensation method avoids the coupling of compensation action and radio frequency envelope; at the same time, a higher second frequency is used to monitor the drain electrical parameters and temperature state, and when the monitoring results meet the preset abnormal event conditions, a negative adjustment operation or a shutdown operation is quickly executed and the slow compensation is frozen, realizing the fast response and protection of transient abnormal events. Therefore, the embodiment of the present application can realize accurate and stable gate voltage compensation based on the estimated real junction temperature on the premise of avoiding the interference of the radio frequency envelope, and has the ability of fast and safe protection. BRIEF DESCRIPTION OF DRAWINGS

[0050] Figure 1 is a flowchart of a gate voltage intelligent control method of a gallium nitride power transistor provided by an embodiment of the present application;

[0051] Figure 2 is an architecture diagram of a gate voltage intelligent control system provided by an embodiment of the present application;

[0052] Figure 3 is a fast and slow dual-time domain control flowchart provided by an embodiment of the present application;

[0053] Figure 4 is a power-on and temperature compensation coordination timing diagram provided by an embodiment of the present application;

[0054] Figure 5 is a structure diagram of a gate voltage intelligent control system of a gallium nitride power transistor provided by an embodiment of the present application;

[0055] Figure 6 is a structure diagram of a gate voltage intelligent control device of a gallium nitride power transistor provided by an embodiment of the present application. DETAILED DESCRIPTION

[0056] With reference to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described, obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work belong to the scope of protection of the present application.

[0057] Referring to Figure 1 It is a flowchart of a gate voltage intelligent control method of a gallium nitride power transistor provided by an embodiment of the present application. The gate voltage intelligent control method of the gallium nitride power transistor is executed by an intelligent gate voltage control system, and includes:

[0058] S10, collecting the supply voltage and supply current of the gallium nitride power transistor, and performing power detection on the radio frequency output signal of the gallium nitride power transistor to obtain a detected signal;

[0059] S11, calculating the average direct current power based on the supply voltage and the supply current, and calculating the average output power based on the detected signal;

[0060] S12, calculating the average dissipation power of the gallium nitride power transistor based on the average direct current power and the average output power;

[0061] S13, taking the average dissipation power as input, and based on a preset equivalent thermal network model reflecting the thermal dynamic relationship between the dissipation power and the junction temperature, real-time estimating the junction temperature of the gallium nitride power transistor;

[0062] S14, monitoring the change of the junction temperature at a first frequency, and when detecting that the change amount of the junction temperature exceeds a preset threshold, generating a gate voltage compensation signal according to a preset voltage change slope limiting mode to compensate and adjust the gate voltage of the gallium nitride power transistor;

[0063] S15, monitoring the drain electrode parameter and temperature state of the gallium nitride power transistor at a second frequency higher than the first frequency to obtain a monitoring result;

[0064] S16, in response to the monitoring result satisfying a preset abnormal event condition, performing a negative adjustment operation or a turn-off level operation on the gate voltage of the gallium nitride power transistor, and freezing the compensation adjustment until the monitoring result does not satisfy the preset abnormal event condition.

[0065] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0066] Firstly, by collecting the supply voltage and supply current and power detection of the radio frequency output signal, the average dissipation power is calculated, which is directly related to the internal heat generation of the device, providing an accurate input for subsequent junction temperature estimation that is independent of the ambient temperature, thereby solving the problem of insufficient compensation accuracy; secondly, the average dissipation power is input into the equivalent thermal network model to estimate the junction temperature, and the model simulates the thermal inertia of the device through the thermal resistance and heat capacity parameters, so that the estimated junction temperature changes smoothly and truly reflects the thermal dynamic process; on this basis, a dual-frequency monitoring and control mode is adopted: a lower first frequency is used to monitor the change of the estimated junction temperature, and only when the change exceeds the preset threshold, a gate voltage compensation signal is generated according to the preset voltage change slope limiting mode for slow compensation, and this low-frequency and slope-limited compensation mode avoids the coupling of compensation action and radio frequency envelope; at the same time, a higher second frequency is used to monitor the drain electrical parameters and temperature state, and when the monitoring results meet the preset abnormal event conditions, a negative adjustment operation or a shutdown operation is quickly executed and the slow compensation is frozen, realizing the rapid response and protection of transient abnormal events. Therefore, the embodiment of the present application can realize accurate and stable gate voltage compensation based on the estimated real junction temperature while avoiding the interference of the radio frequency envelope, and has the ability of fast and safe protection.

[0067] Referring to Figure 2 , which shows a hardware structure block diagram of an intelligent gate voltage control system provided by an embodiment of the present application. The system takes a microprocessor as a core controller, and its peripherals include an analog-to-digital converter (ADC), a digital-to-analog converter (DAC), a non-volatile memory, a multiplexer (MUX), and a temperature detection module and a drain voltage control module. The specific signal acquisition and control path is as follows: the supply voltage (drain voltage) of the gallium nitride power amplifier (GaNPA) is connected to the ADC after being input through a precision voltage detection module and an RC low-pass filter network; its supply current is connected to the ADC after being input through a precision current detection module and an RC low-pass filter network; and a part of the radio frequency output signal of the GaNPA is coupled by a coupler, detected by a detector with temperature compensation, and then connected to the ADC after being filtered by an RC low-pass filter to obtain a detection voltage signal. The microprocessor executes the algorithm of the method of the present application according to the collected signals, and the output gate voltage control signal of the microprocessor is connected to the gate of the GaNPA through the DAC and possibly through the MUX selection (arbitration between fast loop and slow loop outputs), finally completing the closed-loop control.

[0068] As one of the examples, the average direct current power is calculated based on the supply voltage and the supply current, and the average output power is calculated based on the detection signal, which includes:

[0069] The supply voltage and the supply current are subjected to first low-pass filtering processing with a first preset time constant to obtain a first filtered signal;

[0070] The average DC power is calculated based on the filtered supply voltage and the filtered supply current in the first filtered signal;

[0071] The second low-pass filtering with a second preset time constant is performed on the detected signal to obtain a second filtered signal;

[0072] The average output power is calculated based on the second filtered signal.

[0073] In the embodiment, the first filtered signal representing the steady-state DC energy is obtained by performing RC low-pass filtering with a millisecond time constant on the supply voltage and the supply current, which can effectively filter out the high-frequency switching noise, ripple and radio frequency leakage components in the supply circuit. Then, the average DC power is calculated based on the signal, and the detected signal from the temperature compensation detector is low-pass filtered with a time constant of the same order of magnitude to completely filter out the low-frequency fluctuations related to the radio frequency envelope, thereby obtaining the second filtered signal reflecting only the average radio frequency output energy. Then, the average dissipation power is obtained by subtracting the two strongly filtered power values, which can accurately strip the influence of transient electrical fluctuations and radio frequency modulation and purely reflect the energy converted into heat inside the device. Therefore, the embodiment provides a stable, accurate and radio frequency interference-free power driving source for subsequent junction temperature estimation based on thermal dynamics through double strong low-pass filtering, which fundamentally avoids the coupling of the envelope signal into the temperature compensation loop.

[0074] Specifically, the working process of the embodiment is as follows: combined with Figure 2The precise voltage detection module and the precise current detection module sample the drain voltage Vdd and the drain current Idc of the GaN PA in real time, respectively. The two signals are first filtered by an RC low-pass filter network with a time constant τ = 200 ms (hardware filtering) to preliminarily filter out high-frequency noise. The filtered analog signals are sampled by an ADC in the microprocessor at a moderate frequency (e.g., 1 kHz). In the microprocessor, digital low-pass filtering (software filtering) can be further performed, with a cutoff frequency matching the hardware filtering, to finally obtain highly smoothed filtered supply voltage Vdc_f and current Idc_f. The average DC power Pdc is calculated in real time by the formula Pdc = Vdc_f * Idc_f. At the same time, the RF signal output by the GaN PA is sampled by a coupler and converted into a DC voltage signal proportional to the instantaneous value of the RF power by a detector with temperature compensation. This detector voltage signal also passes through an RC low-pass filter with a time constant τ = 200 ms to filter out any residual envelope fluctuations and obtain a smoothed detector voltage Vdet_f. The average output power Pout is obtained by table lookup calculation using a detector calibration curve (relating Vdet_f and power values) pre-stored in a non-volatile memory. Finally, the average dissipation power Pdiss is derived as Pdiss = Pdc - Pout and serves as the core input of the junction temperature estimation unit.

[0075] As one of the examples, the equivalent thermal network model based on a preset thermal dynamic relationship between dissipation power and junction temperature is used to estimate the junction temperature of the GaN power transistor in real time, including:

[0076] The average dissipation power is input into the equivalent thermal network model containing thermal resistance parameters and thermal capacity parameters; wherein, according to the average dissipation power at the current time, the junction temperature estimated at the previous time, and the ambient temperature, the equivalent thermal network model calculates and outputs the junction temperature estimated in real time at the current time through a first-order resistance-capacitance thermal dynamic relationship, wherein the change time constant of the junction temperature is determined by the product of the thermal resistance parameters and the thermal capacity parameters.

[0077] In the embodiment, the calculated average dissipated power is first input into a first-order RC equivalent thermal network model as a driving source; then, the model calculates the estimated junction temperature at the current time by solving the discretized thermal balance equation according to the current dissipated power, the historical junction temperature estimated in the last control period, and the ambient temperature obtained through the temperature sensor; then, since the heat capacity parameter in the model simulates the thermal mass of the device, the change of the output junction temperature presents inherent inertia, the response speed is determined by the thermal time constant, the change is smooth and significantly lags behind the rapid fluctuation of the electrical power; finally, the smooth and slowly changing junction temperature output effectively distinguishes the instantaneous power fluctuation caused by the radio frequency signal or the load change from the real thermal change caused by the continuous accumulation of energy. Therefore, the embodiment realizes accurate and dynamic mapping from electrical quantity to thermal state quantity by introducing the equivalent thermal network model with clear physical meaning, and provides a feedback signal that truly reflects the thermal dynamics of the device for subsequent slow temperature compensation.

[0078] Specifically, the working process of the embodiment is as follows: the junction temperature estimation unit runs in the microprocessor at a fixed period (such as 1 ms). The core parameters of the equivalent thermal network model are the thermal resistance Rθ from the junction to the environment and the thermal capacity Cθ, and the product τ = Rθ * Cθ is the thermal time constant. These parameters can be obtained in advance through the device data manual or experimental calibration and stored in the non-volatile memory. In each calculation period, the unit reads the current average dissipated power Pdiss(k) and the ambient temperature Tamb from the temperature detection module. The model iteratively calculates according to the differential relationship of the first-order RC network: Tj(k) = Tj(k-1) + Δt * [Pdiss(k) - (Tj(k-1) - Tamb) / Rθ] / Cθ, where Tj(k-1) is the estimated junction temperature of the last period, and Δt is the calculation period. Through this calculation, no matter how Pdiss jumps, the output of Tj(k) is always exponentially smoothed to approach its steady-state value Tamb + Pdiss * Rθ. For example, when the dissipated power jumps from 0 to 22 W, the junction temperature Tj will not jump immediately, but will start to rise slowly with the time constant τ, and finally tend to the steady-state value, which truly simulates the actual physical change process of the chip junction temperature.

[0079] As one of the examples, the first frequency is monitored to detect the change of the junction temperature, and when the change of the junction temperature is detected to exceed a preset threshold, a gate voltage compensation signal is generated according to a preset voltage change slope limiting mode, which includes:

[0080] The estimated junction temperature is periodically acquired at the first frequency;

[0081] The difference between the junction temperature acquired in the current period and the junction temperature acquired in the last period is calculated as the change of the junction temperature;

[0082] When the change amount exceeds a preset threshold, an initial compensation voltage value is calculated according to the change amount and a preset temperature-pressure compensation coefficient, and a change rate of the initial compensation voltage value is limited to be not greater than a preset change rate safety value, so as to generate the gate voltage compensation signal.

[0083] In the embodiment, a junction temperature is periodically woken up and read at a first extremely low frequency (e.g., 2 Hz), which is much lower than a typical frequency of a radio frequency envelope, and any high-frequency fluctuation in the junction temperature estimation can be naturally filtered out. Then, a junction temperature difference between adjacent monitoring periods is calculated and compared with a smaller threshold (e.g., 2.5°C), and only when the thermal change accumulates to a certain degree, an effective temperature drift requiring compensation is determined, so as to avoid unnecessary frequent adjustment caused by small fluctuations in the model or slight changes in the ambient temperature. Then, a theoretical required gate voltage adjustment amount is calculated according to the effective junction temperature change amount and a pre-labeled temperature-pressure compensation coefficient. Finally, a strict slope limit (e.g., ≤1 mV / ms) is implemented on the output of the adjustment amount, so as to ensure that the voltage compensation signal finally applied to the gate is an extremely slow and smooth ramp signal. Therefore, the embodiment constructs an ultra-narrow band slow loop with an equivalent bandwidth less than 1 Hz through the three mechanisms of "low-frequency sampling, threshold hysteresis, and slope limiting output", completely blocks the response of the temperature compensation action to signals in the frequency range of the radio frequency envelope, and thus eliminates the risk of gain fluctuation or self-oscillation caused by the temperature compensation loop intervention in the radio frequency domain.

[0084] Specifically, the working process of the embodiment is as follows: combined with Figure 3The slow loop temperature compensation control unit (<1 Hz) is independent of the main control cycle, and is triggered once every 0.5 seconds (corresponding to 2 Hz) by a low-power timer of a microprocessor. In the interrupt service program, the unit reads the latest estimated junction temperature value Tj_current from the junction temperature estimation unit. The unit internally maintains a reference junction temperature value Tj_ref, which is updated after each successful compensation. Calculate ΔT = Tj_current - Tj_ref. If |ΔT| is less than the preset threshold ΔTth (for example, 2-3°C), the unit does not perform any operation and directly exits. If |ΔT| exceeds the threshold, compensation calculation is triggered: first, read the temperature-voltage compensation coefficient kt (unit: mV / °C) for this specific GaN transistor from the non-volatile memory, which is obtained by performing a temperature scan of the device in a temperature chamber from -40°C to +100°C, accurately measuring and recording the gate voltage change under constant static current every 1-2°C, and then linear fitting. Calculate the theoretical compensation voltage ΔVg_target = kt * ΔT. Subsequently, the unit does not directly output ΔVg_target, but compares it with the current output compensation voltage Vg_comp_current, and gradually adjusts Vg_comp_current to ΔVg_target with a slope not exceeding the preset change rate safety value (such as 1 mV / ms). For example, the output change of Vg_comp in each 0.5 second period does not exceed 0.5 V. This slow updating Vg_comp is the gate voltage compensation signal.

[0085] As one of the examples, the second frequency is higher than the first frequency, and the drain electrical parameters and temperature state of the GaN power transistor are monitored in real time to obtain a monitoring result, including:

[0086] The drain voltage instantaneous value and the drain current instantaneous value of the GaN power transistor are collected at a second frequency higher than the first frequency.

[0087] Based on the collected drain voltage instantaneous value, it is judged whether there is an overvoltage spike event.

[0088] Based on the collected drain current instantaneous value, the change rate is calculated and it is judged whether there is an overcurrent change rate event.

[0089] Receive the temperature detection signal from the temperature sensor, and judge whether there is an over-temperature event.

[0090] Based on the judgment results of the overvoltage spike event, the overcurrent change rate event and the over-temperature event, the monitoring result is generated.

[0091] In this embodiment, the drain voltage and current are first sampled in kHz level high second frequency, which can capture the voltage peak and current mutation with extremely short duration, providing high timeliness of original data for fast protection; then, the drain voltage of each sampling point is compared in real time, and once it exceeds the set absolute voltage threshold, it is marked as an overvoltage peak event, which can effectively prevent the risk of breakdown caused by load mutation or switch transient; at the same time, the differential operation is performed on the continuous drain current sampling value, the change rate is calculated in real time, and compared with the preset safe change rate threshold, so as to identify the abnormal current growth representing possible load short circuit or serious mismatch; then, the direct temperature reading from the independent temperature sensor is integrated to provide a final temperature safety line; finally, the three parallel judgment results are integrated to generate a unified monitoring result word containing multiple fault information. Therefore, through the high-frequency, multi-dimensional and parallel real-time monitoring network, a front-end sensing system with sensitive response and comprehensive coverage is constructed, which provides accurate and reliable input for microsecond-level fast protection decision.

[0092] Specifically, the working process of the embodiment is as follows: Figure 2 and Figure 3 The fast loop protection detection control unit (1-10 kHz) is usually realized by high priority timer interrupt of microprocessor or dedicated hardware comparison circuit, and operates at a frequency of 1 kHz to 10 kHz. In each high-speed sampling interrupt: 1) the drain voltage raw signal Vds_raw and the drain current raw signal Id_raw which are not subjected to strong low-pass filtering (or only subjected to simple filtering with wide bandwidth) are quickly sampled by ADC to retain transient characteristics. 2) Overvoltage judgment: Vds_raw is compared with the preset overvoltage protection threshold V_OVP (for example, 32V for 28V power supply) in real time, and if Vds_raw>V_OVP, the overvoltage flag bit is set. 3) Overcurrent change rate judgment: the current value Id_last of the last sampling is recorded, and the current change rate dI / dt=(Id_raw-Id_last) / Δt is calculated, where Δt is the high-speed sampling interval. dI / dt is compared with the preset critical change rate threshold di / dt_th (for example, 10A / ms), and if it exceeds, the overcurrent change rate flag bit is set. This method is particularly sensitive to instantaneous short circuit. 4) Over-temperature judgment: the temperature value T_sensor from the on-board temperature sensor (such as a thermistor) is read, compared with the preset device maximum safe junction temperature or shell temperature threshold T_OTP (for example, 125℃), and if it exceeds, the over-temperature flag bit is set. The three flag bits are combined into a byte or word of monitoring result status word for subsequent arbitration logic.

[0093] As one of the examples, the response to the monitoring result satisfying the preset abnormal event condition is to perform a negative adjustment operation or an off-level operation on the gate voltage of the gallium nitride power transistor, and freeze the compensation adjustment until the monitoring result does not satisfy the preset abnormal event condition, which includes:

[0094] When the monitoring result satisfies the preset abnormal event condition, a control signal is generated; the abnormal event condition includes at least one of the occurrence of the overvoltage spike event, the overcurrent rate event or the overtemperature event;

[0095] According to the control signal, the gate voltage of the gallium nitride power transistor is driven to a preset negative voltage off level;

[0096] At the same time of generating the control signal, the adjustment effect of the gate voltage compensation signal on the gate voltage is blocked to freeze the compensation adjustment until the monitoring result indicates that all abnormal events disappear.

[0097] In this embodiment, when the monitoring result indicates that any abnormal event occurs, a high-priority hardware or software control signal is immediately generated; then, the signal sets the gate drive voltage to a deep negative voltage (such as -5V) through the most direct path (such as controlling MUX switching or directly setting the DAC output value), and forcibly turns off the power transistor within a few microseconds, thereby cutting off the fault energy path at the fastest speed to achieve hard protection; then, at the same time of triggering the off action, the control signal is synchronously sent to the arbitration and freezing logic unit, which immediately issues a freezing instruction to the slow loop temperature compensation control unit to suspend the internal compensation calculation and output update, and keeps the original output value unchanged; finally, this freezing state will be maintained until the monitoring result confirms that all abnormal conditions have been cleared, and the system is confirmed to be stable after a preset safe recovery delay, and then the freezing is released. Therefore, through the instantaneous cooperation mechanism of "fast hardware off" and "logic freezing slow loop", this embodiment not only realizes the microsecond-level response to transient faults to protect the safety of the device, but also ensures that the slow temperature compensation loop will not make incorrect or conflicting adjustments due to abnormal states during the fault and recovery period, thereby maintaining the determinacy and stability of the entire control system.

[0098] Specifically, the working process of this embodiment is as follows: in combination with Figure 2 and Figure 3 , the arbitration and freezing logic unit receives the monitoring result from the fast loop protection unit. Once the monitoring result word shows that any abnormal flag bit is set, the arbitration logic immediately (usually within a few CPU clock cycles) performs the following actions: 1) generate a fast off signal: the arbitration logic controls Figure 1The arbitration logic controls a multiplexer (MUX) in the slow loop to switch its output from the "normal compensated gate voltage" channel from the slow loop to a fixed "safe-off gate voltage" reference source (corresponding to -5V). Meanwhile, the output voltage can also be pulled down directly by writing to the DAC register. This operation has the highest priority and the fastest response time. 2) Freeze instruction generation: the arbitration logic sends a clear "freeze" signal to the slow loop temperature compensation control unit. Upon receiving this signal, the slow loop unit suspends its 0.5-second periodic interrupt or skips the compensation calculation and output update part in the interrupt service program, and all internal states (such as Tj_ref and Vg_comp_current) remain unchanged at the moment before freezing. 3) Recovery process: when the fast loop unit continuously monitors the results for multiple times (for example, 10 consecutive sampling periods) and all flags are cleared, the arbitration logic considers that the abnormal state has been resolved. However, it does not immediately release the freeze, but starts a 200ms to 500ms configurable recovery delay timer. After the timer ends, the arbitration logic cancels the freeze instruction and controls the MUX to switch back to the normal channel. The slow loop unit resumes operation and continues to work from the state before freezing, thereby achieving smooth and disturbance-free recovery.

[0099] As an example, the first frequency is 2Hz, and the second frequency is 1kHz-10kHz. Specifically, the slow loop temperature compensation module operates with a period of 0.5 seconds, and the equivalent bandwidth is much lower than the typical radio frequency envelope frequency; the fast loop protection unit operates at a frequency of 1kHz to 10kHz, ensuring that transient abnormalities can be captured in time.

[0100] As an example, before collecting the supply voltage and supply current of the gallium nitride power transistor and performing power detection on the radio frequency output signal of the gallium nitride power transistor to obtain a detected signal, the method further comprises:

[0101] After the system power is turned on, the gate voltage of the gallium nitride power transistor is set to a negative voltage safety bias level, and the drain supply voltage of the gallium nitride power transistor is gradually increased from zero to a target working voltage;

[0102] After confirming that the drain supply voltage is stable at the target working voltage, the gate voltage is adjusted from the negative voltage safety bias level to an initial working bias point at a controlled slope.

[0103] In this embodiment, first, at the initial power-on moment, the gate voltage is set to the deep negative voltage of the safety-off state to ensure that the power transistor is absolutely off in the unknown state and prevent misdirecting on; then, on the premise that the device has been reliably turned off, the controllable drain power module is instructed to slowly establish the drain voltage, which completely avoids the excessive current impact that may be caused by loading the drain voltage when the gate control is not ready; then, after the drain voltage has been stably established to the target value, the gate voltage is lifted from the off negative voltage to the preset static working point at a controlled and slow slope, which realizes the smooth establishment of the static working current and avoids the current surge caused by the step of the gate voltage; finally, after the static working point is stably established, the intelligent temperature compensation and fast loop protection functions are enabled, and the system enters the steady-state operation. Therefore, this embodiment provides reliable start-up security for the GaNHEMT, which is sensitive to bias and relatively weak in overcurrent resistance, by specifying and executing a strict and time-sequenced power-on bias sequence, which fundamentally prevents potential damage risks in the power-on process.

[0104] Specifically, the working process of this embodiment is as follows: please refer to the power-on and temperature compensation coordination timing diagram shown in Figure 4 After the system main power is turned on (at T0 moment), the microprocessor initializes and first executes the power-on safety sequence: 1) set the safety gate voltage: the microprocessor sets the gate voltage Vgs of the GaN PA to -5V (safety gate voltage) through the DAC and the driving circuit, and maintains it. At this time, the transistor is in a completely off state. 2) slowly rise the drain voltage: the microprocessor enables the drain voltage control module (such as a programmable power supply or a PMIC with enable and soft start functions), instructs it to output voltage from 0V to gradually rise to the target working voltage Vdd_target (such as +28V or +48V) at a certain slope (for example, 1V / ms). This process corresponds to the T1 to T2 stage in the figure. 3) confirm and stabilize the drain voltage: the microprocessor monitors the drain voltage through the ADC and confirms that it has reached and stabilized at Vdd_target. 4) controlled opening of the gate voltage: at T2 moment, the drain voltage has stabilized. The microprocessor starts to adjust the DAC output at a preset and slow slope (for example, 1mV / µs) to gradually adjust the gate voltage Vgs from -5V to the target static working point (for example, -2.65V). This process corresponds to the T2 to T3 stage, which ensures the smooth establishment of the static current. 5) enter steady-state operation: at T3 moment, Vgs reaches the target working point and stabilizes. Then, at T4 moment, the microprocessor starts the main control loop described in the background art and enables all intelligent control functions such as low-frequency energy measurement, junction temperature estimation, slow loop temperature compensation and fast loop protection, and the system enters the full-function running state.

[0105] Please refer to Figure 5, is a structural schematic view of a gate voltage intelligent control system of a gallium nitride power transistor provided by an embodiment of the present application. The gate voltage intelligent control system of the gallium nitride power transistor comprises:

[0106] The acquisition module 10 is configured to acquire a supply voltage and a supply current of the gallium nitride power transistor, and perform power detection on a radio frequency output signal of the gallium nitride power transistor to obtain a detected signal;

[0107] The first calculation module 11 is configured to calculate an average direct current power based on the supply voltage and the supply current, and calculate an average output power based on the detected signal;

[0108] The second calculation module 12 is configured to calculate an average dissipation power of the gallium nitride power transistor based on the average direct current power and the average output power;

[0109] The estimation module 13 is configured to take the average dissipation power as an input, and estimate a junction temperature of the gallium nitride power transistor in real time based on a preset equivalent thermal network model reflecting a thermal dynamic relationship between a dissipation power and the junction temperature;

[0110] The compensation module 14 is configured to monitor a change of the junction temperature at a first frequency, and generate a gate voltage compensation signal according to a preset voltage change slope limiting mode to compensate and adjust a gate voltage of the gallium nitride power transistor when a change amount of the junction temperature is detected to exceed a preset threshold value;

[0111] The monitoring module 15 is configured to monitor a drain electrical parameter and a temperature state of the gallium nitride power transistor at a second frequency higher than the first frequency to obtain a monitoring result;

[0112] The operation module 16 is configured to perform a negative adjustment operation or an off level operation on the gate voltage of the gallium nitride power transistor and freeze the compensation adjustment until the monitoring result does not satisfy a preset abnormal event condition in response to the monitoring result satisfying the preset abnormal event condition.

[0113] Compared with the prior art, the embodiment of the present application has the following beneficial effects:

[0114] First, by collecting the supply voltage and current and performing power detection on the RF output signal, the average dissipation power is calculated. This power is directly related to the internal heat generation of the device, providing an accurate input for subsequent junction temperature estimation that is independent of ambient temperature, thus solving the problem of insufficient compensation accuracy. Second, the average dissipation power is input into an equivalent thermal network model to estimate the junction temperature. This model simulates the thermal inertia of the device through thermal resistance and thermal capacity parameters, making the estimated junction temperature change smooth and gradual, truly reflecting the thermal dynamic process. Based on this, a dual-frequency monitoring and control method is adopted: the lower first frequency monitors the estimated junction temperature change, and only when the change exceeds a preset threshold is a gate voltage compensation signal generated according to a preset voltage change slope limit method for slow compensation. This low-frequency, slope-limited compensation method avoids the coupling between the compensation action and the RF envelope. At the same time, the higher second frequency monitors the drain electrical parameters and temperature status. When the monitoring results meet the preset abnormal event conditions, a negative adjustment operation or a level shutdown operation is quickly executed and the slow compensation is frozen, realizing rapid response and protection against transient abnormal events. Therefore, the embodiments of the present invention can achieve accurate and stable gate voltage compensation based on the estimated true junction temperature while avoiding radio frequency envelope interference, and have fast safety protection capabilities.

[0115] It is understood that the specific implementation of the above-described intelligent control system for the gate voltage of gallium nitride power transistors can be referred to the relevant content of the above-described intelligent control method for the gate voltage of gallium nitride power transistors, and will not be repeated here.

[0116] See Figure 6 This is a schematic diagram of a smart gate voltage control device for gallium nitride power transistors according to an embodiment of the present invention. The smart gate voltage control device for gallium nitride power transistors in this embodiment includes: a processor 100, a memory 101, and a computer program stored in the memory 101 and executable on the processor 100, such as a smart gate voltage control program for gallium nitride power transistors. When the processor 100 executes the computer program, it implements the steps in the above-described embodiments of the smart gate voltage control methods for gallium nitride power transistors. Alternatively, when the processor 100 executes the computer program, it implements the functions of each module / unit in the above-described device embodiments.

[0117] For example, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the intelligent gate voltage control device for gallium nitride power transistors.

[0118] The gate voltage intelligent control device of the gallium nitride power transistor can be a main control board or a main controller of a power amplifier, etc. The gate voltage intelligent control device of the gallium nitride power transistor can include, but is not limited to, a processor, a memory. Those skilled in the art can understand that the schematic diagram is only an example of the gate voltage intelligent control device of the gallium nitride power transistor, and does not constitute a limitation on the gate voltage intelligent control device of the gallium nitride power transistor, and can include more or fewer components than the diagram, or combine certain components, or different components, for example, the gate voltage intelligent control device of the gallium nitride power transistor can also include an input and output device, a network access device, a bus, etc.

[0119] The processor can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc. The processor is the control center of the gate voltage intelligent control device of the gallium nitride power transistor, and connects various parts of the gate voltage intelligent control device of the gallium nitride power transistor through various interfaces and lines.

[0120] The memory can be used to store the computer programs and / or modules, and the processor realizes various functions of the gate voltage intelligent control device of the gallium nitride power transistor by running or executing the computer programs and / or modules stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required by a function (such as a sound playing function, an image playing function, etc.), etc.; and the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, for example, a hard disk, a memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one disk storage device, a flash memory device, or other volatile solid-state memory devices.

[0121] The modules / units of the gate voltage intelligent control device of the gallium nitride power transistor are stored in a computer readable storage medium if they are realized in the form of software function units and sold or used as independent products. Based on this understanding, all or part of the processes in the above-mentioned embodiment methods can also be completed by a computer program instructing related hardware. The computer program can be stored in a computer readable storage medium. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or some intermediate forms, etc. The computer readable medium can include any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the content included in the computer readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer readable medium does not include electrical carrier signals and telecommunication signals.

[0122] It should be noted that the above-described device embodiments are only schematic, and the units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, i.e., they can be located in one place or distributed on multiple network units. Part or all of the modules can be selected to achieve the purpose of the embodiment according to actual needs. In addition, the connection relationship between the modules in the device embodiment provided by the present application indicates that there is a communication connection between them, which can be realized as one or more communication buses or signal lines. Those skilled in the art can understand and implement it without creative labor.

[0123] The above is the preferred embodiment of the present application. It should be noted that those skilled in the art can make several improvements and refinements without departing from the principles of the present application, and these improvements and refinements are also considered within the scope of protection of the present application.

Claims

1. A method for intelligent control of the gate voltage of a gallium nitride power transistor, characterized in that, include: The power supply voltage and current of the gallium nitride power transistor are collected, and the radio frequency output signal of the gallium nitride power transistor is subjected to power detection to obtain the detection signal. The average DC power is calculated based on the supply voltage and the supply current, and the average output power is calculated based on the detection signal. The average power dissipation of the gallium nitride power transistor is calculated based on the average DC power and the average output power. Using the average power dissipation as input, and based on a preset equivalent thermal network model that reflects the thermal dynamic relationship between power dissipation and junction temperature, the junction temperature of the gallium nitride power transistor is estimated in real time. The junction temperature change is monitored at a first frequency, and when the change in junction temperature exceeds a preset threshold, a gate voltage compensation signal is generated according to a preset voltage change slope limiting method to compensate and adjust the gate voltage of the gallium nitride power transistor. The drain electrical parameters and temperature status of the gallium nitride power transistor are monitored at a second frequency higher than the first frequency to obtain the monitoring results; In response to the monitoring result meeting a preset abnormal event condition, a negative adjustment operation or a shutdown level operation is performed on the gate voltage of the gallium nitride power transistor, and the compensation adjustment is frozen until the monitoring result does not meet the preset abnormal event condition.

2. The intelligent gate voltage control method for gallium nitride power transistors as described in claim 1, characterized in that, The calculation of average DC power based on the supply voltage and the supply current, and the calculation of average output power based on the detection signal, include: The supply voltage and the supply current are subjected to a first low-pass filter with a first preset time constant to obtain a first filtered signal; The average DC power is calculated based on the filtered supply voltage and filtered supply current in the first filtered signal. The detected signal is subjected to a second low-pass filter with a second preset time constant to obtain a second filtered signal. The average output power is calculated based on the second filtered signal.

3. The intelligent gate voltage control method for gallium nitride power transistors as described in claim 1, characterized in that, The method, based on a preset equivalent thermal network model reflecting the thermal dynamic relationship between power dissipation and junction temperature, estimates the junction temperature of the gallium nitride power transistor in real time, including: The average power dissipation is input into the equivalent thermal network model, which includes thermal resistance and thermal capacity parameters. The equivalent thermal network model calculates the current estimated junction temperature based on the average power dissipation at the current moment, the junction temperature estimated at the previous moment, and the ambient temperature through the thermal dynamic relationship of first-order resistance and capacitance. The time constant of the junction temperature change is determined by the product of the thermal resistance and thermal capacity parameters.

4. The intelligent gate voltage control method for gallium nitride power transistors as described in claim 1, characterized in that, The step of monitoring the junction temperature change at a first frequency, and generating a gate voltage compensation signal according to a preset voltage change slope limiting method when the detected junction temperature change exceeds a preset threshold, includes: The estimated junction temperature is periodically acquired at a first frequency; The difference between the junction temperature obtained in the current cycle and the junction temperature obtained in the previous cycle is calculated as the change in junction temperature. When the change exceeds a preset threshold, an initial compensation voltage value is calculated based on the change and a preset temperature and pressure compensation coefficient. The rate of change of the initial compensation voltage value is then limited to ensure that it does not exceed a preset safe rate of change value, thereby generating the gate voltage compensation signal.

5. The intelligent gate voltage control method for gallium nitride power transistors as described in claim 1, characterized in that, The method involves real-time monitoring of the drain electrical parameters and temperature state of the gallium nitride power transistor at a second frequency higher than the first frequency, to obtain monitoring results, including: The instantaneous values ​​of the drain voltage and drain current of the gallium nitride power transistor are acquired at a second frequency higher than the first frequency. Based on the collected instantaneous value of the drain voltage, determine whether an overvoltage spike event exists; Based on the instantaneous value of the drain current collected, its rate of change is calculated and it is determined whether an overcurrent rate of change event exists. Receive temperature detection signals from temperature sensors and determine whether an over-temperature event has occurred; The monitoring results are generated based on the judgment results of the overpressure spike event, the overcurrent rate change event, and the overtemperature event.

6. The intelligent gate voltage control method for gallium nitride power transistors as described in claim 5, characterized in that, In response to the monitoring result meeting a preset abnormal event condition, the step of performing a negative adjustment operation or a shutdown operation on the gate voltage of the gallium nitride power transistor, and freezing the compensation adjustment until the monitoring result no longer meets the preset abnormal event condition, includes: When the monitoring results meet the preset abnormal event conditions, a control signal is generated; the abnormal event conditions include at least one of the following: the occurrence of the overpressure spike event, the overcurrent rate of change event, or the overtemperature event; According to the control signal, the gate voltage of the gallium nitride power transistor is driven to a preset negative voltage turn-off level; While generating the control signal, the adjustment effect of the gate voltage compensation signal on the gate voltage is blocked to freeze the compensation adjustment until the monitoring results indicate that all abnormal events have disappeared.

7. The intelligent gate voltage control method for gallium nitride power transistors as described in claim 5, characterized in that, The first frequency is 2Hz, and the second frequency is 1kHz-10kHz.

8. The intelligent gate voltage control method for gallium nitride power transistors as described in claim 1, characterized in that, Before acquiring the supply voltage and supply current of the gallium nitride power transistor and performing power detection on the RF output signal of the gallium nitride power transistor to obtain the detected signal, the method further includes: After the system power is turned on, the gate voltage of the gallium nitride power transistor is set to a negative safety bias level, and the drain supply voltage of the gallium nitride power transistor is gradually increased from zero to the target operating voltage. After confirming that the drain supply voltage is stable at the target operating voltage, the gate voltage is adjusted from the negative safety bias level to the initial operating bias point with a controlled slope.

9. A smart control system for the gate voltage of a gallium nitride power transistor, characterized in that, include: The acquisition module is used to acquire the supply voltage and supply current of the gallium nitride power transistor, and to perform power detection on the radio frequency output signal of the gallium nitride power transistor to obtain the detection signal. The first calculation module is used to calculate the average DC power based on the power supply voltage and the power supply current, and to calculate the average output power based on the detection signal. The second calculation module is used to calculate the average dissipation power of the gallium nitride power transistor based on the average DC power and the average output power. The estimation module is used to estimate the junction temperature of the gallium nitride power transistor in real time, based on the average dissipation power as input and a preset equivalent thermal network model that reflects the thermal dynamic relationship between dissipation power and junction temperature. The compensation module is used to monitor the change in junction temperature at a first frequency, and when the change in junction temperature is detected to exceed a preset threshold, generate a gate voltage compensation signal according to a preset voltage change slope limiting method to compensate and adjust the gate voltage of the gallium nitride power transistor. The monitoring module is used to monitor the drain electrical parameters and temperature status of the gallium nitride power transistor at a second frequency higher than the first frequency, and obtain the monitoring results. The operation module is configured to, in response to the monitoring result meeting a preset abnormal event condition, perform a negative adjustment operation or a shutdown level operation on the gate voltage of the gallium nitride power transistor, and freeze the compensation adjustment until the monitoring result no longer meets the preset abnormal event condition.

10. A smart control device for the gate voltage of a gallium nitride power transistor, characterized in that, The system includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor, when executing the computer program, implements the intelligent gate voltage control method for gallium nitride power transistors as described in any one of claims 1 to 8.

Images