Dual-pulse test method and system for power device based on multi-dimensional coupling effect

Abstract

This invention provides a dual-pulse testing method and system for power devices based on multi-dimensional coupling effects, belonging to the field of power semiconductor device testing technology. The method includes: controlling the input of a dual-pulse excitation signal to the control terminal of the power device under test; adjusting the driving resistor on the driving circuit from the dual-pulse excitation signal to the control terminal; controlling the execution of dual-pulse testing under various combinations of different temperature conditions and driving resistor values; acquiring the electrical switching signal and near-field electromagnetic interference signal of the power device under test in each dual-pulse test; and establishing a coupling relationship model between temperature conditions, driving resistor value, switching characteristic parameters of the power device under test, and electromagnetic interference characterization parameters of the power device under test based on the electrical switching signal and near-field electromagnetic interference signal acquired under various combinations. This invention, based on four-dimensional coupling effects, achieves multi-dimensional coupling analysis of the switching characteristics and electromagnetic interference of power devices.

Description

Technical Field

[0001] This invention relates to the field of power semiconductor device testing technology, and in particular to a dual-pulse testing method and system for power devices based on multi-dimensional coupling effects. Background Technology

[0002] Double-pulse testing is a standard method for evaluating the dynamic switching characteristics of power semiconductor devices, such as insulated-gate bipolar transistors (IGBTs) and silicon carbide metal-oxide-semiconductor field-effect transistors (MOSFETs). Traditional double-pulse testing primarily focuses on fundamental electrical parameters during the device's switching process under fixed ambient temperature and drive resistance conditions, such as turn-on delay time, turn-off delay time, and switching losses. During testing, typically only the collector-emitter voltage and collector current waveforms are collected and analyzed.

[0003] However, the operating states of power devices in practical applications are complex and variable. There is a strong coupling relationship between the device's switching characteristics, the generated electromagnetic interference, its operating junction temperature, and the gate resistance value in the drive circuit. Traditional testing methods isolate factors such as temperature, drive resistance, switching characteristics, and electromagnetic interference, or only consider the influence of a single variable, failing to systematically reveal and quantify the multidimensional coupling effects between them. This results in driver designs based on traditional test data often failing to maintain optimal performance across the entire operating temperature range and under different operating conditions. Summary of the Invention

[0004] This invention provides a power device double-pulse testing method and system based on multi-dimensional coupling effects. Based on the four-dimensional coupling effects of temperature, driving resistance, switching characteristics, and electromagnetic interference, it realizes multi-dimensional coupling analysis of the switching characteristics and electromagnetic interference of power devices.

[0005] In a first aspect, the present invention provides a dual-pulse testing method for power devices based on multi-dimensional coupling effects, comprising:

[0006] The system controls the input of a dual-pulse excitation signal to the control terminal of the power device under test; wherein the driving resistor on the driving line from the dual-pulse excitation signal to the control terminal is adjustable.

[0007] The power device under test is controlled to be under different temperature conditions, and a double-pulse test is executed under various combinations of different temperature conditions and different resistance values ​​of the driving resistor.

[0008] In each double-pulse test, the electrical switching signal of the power device under test and the near-field electromagnetic interference signal generated by the switching process of the power device under test are acquired.

[0009] Based on the electrical switching signals and near-field electromagnetic interference signals obtained under the various combinations, a coupling relationship model is established between the temperature conditions, the resistance value of the driving resistor, the switching characteristic parameters of the power device under test, and the electromagnetic interference characterization parameters of the power device under test.

[0010] In some embodiments, acquiring the electrical switching signal of the power device under test includes:

[0011] The collector-emitter voltage, collector current, and gate-emitter voltage of the power device under test are acquired; wherein the switching characteristic parameters are extracted from the signals of the collector-emitter voltage, the collector current, and the gate-emitter voltage.

[0012] In some embodiments, based on the electrical switching signal and the near-field electromagnetic interference signal obtained under the various combinations, a coupling relationship model is established between the temperature conditions, the resistance value of the driving resistor, the switching characteristic parameters of the power device under test, and the electromagnetic interference characterization parameters of the power device under test, including:

[0013] Extract at least one switching characteristic parameter from the electrical switching signal;

[0014] Extract at least one electromagnetic interference characterization parameter from the near-field electromagnetic interference signal;

[0015] Using the temperature conditions and the resistance of the driving resistor as input variables, and the extracted switching characteristic parameters and the electromagnetic interference characterization parameters as output variables, a coupling relationship model is established through a machine learning algorithm.

[0016] In some embodiments, after establishing a coupling relationship model between the temperature conditions, the resistance value of the driving resistor, the switching characteristic parameters of the power device under test, and the electromagnetic interference characterization parameters of the power device under test, the model further includes:

[0017] The adaptive driving strategy of the power device under test is obtained according to the coupling relationship model; wherein, the adaptive driving strategy defines the adjustment rule of the resistance value of the driving resistor as the temperature condition changes;

[0018] The resistance value of the drive resistor is controlled and adjusted based on the adaptive drive strategy and the temperature parameters of the power device under test acquired in real time.

[0019] In some embodiments, the adaptive driving strategy includes a strategy aimed at minimizing electromagnetic interference characterization parameters and a strategy aimed at minimizing switching losses.

[0020] In some embodiments, adjusting the resistance value of the drive resistor based on the adaptive drive strategy and the real-time acquired temperature parameters of the power device under test includes:

[0021] The resistance value of the drive resistor is controlled and adjusted based on the adaptive drive strategy, the temperature parameters of the power device under test acquired in real time, and the electrical switching signals of the power device under test acquired in real time.

[0022] Secondly, the present invention also provides a power device dual-pulse test system based on multi-dimensional coupling effect, comprising:

[0023] The dual-pulse excitation unit is used to generate a dual-pulse excitation signal and output it to the control terminal of the power device under test.

[0024] An adjustable drive unit is used to switch the drive resistor on the drive line that receives the dual-pulse excitation signal from the control terminal.

[0025] A thermal environment control and acquisition unit is used to provide different temperature conditions for the power device under test and acquire the temperature parameters of the power device under test;

[0026] The multi-dimensional data acquisition unit includes an electrical detection component for acquiring the electrical switching signal of the power device under test, and a near-field detection component for acquiring the near-field electromagnetic interference signal of the power device under test.

[0027] The control processing unit is communicatively connected to the dual-pulse excitation unit, the adjustable drive unit, the thermal environment control acquisition unit, and the multi-dimensional data acquisition unit, respectively, and is used to execute the power device dual-pulse test method based on multi-dimensional coupling effect as described in the first aspect.

[0028] In some embodiments, the adjustable drive unit includes multiple switching channels, each equipped with an analog switch and a resistor. The adjustable drive unit is used to switch the drive resistor connected to the drive line of the control terminal by controlling the conduction state of the analog switch.

[0029] In some embodiments, the thermal environment control acquisition unit includes:

[0030] A thermal environment control box is provided, in which the power device under test is placed and the thermal environment control box is used to provide different temperature conditions for the power device under test.

[0031] A temperature measuring component is used to measure the junction temperature of the power device under test.

[0032] In some embodiments, the near-field detection component is disposed at a preset position on the load lead of the power device under test.

[0033] This invention innovatively places temperature conditions, driving resistance, switching characteristics, and near-field electromagnetic interference within the same testing framework for synchronous and correlated measurement and analysis. By actively changing the combination of temperature and driving resistance and collecting data from all dimensions, it can systematically reveal multi-factor coupling effects that cannot be observed by traditional methods. The established coupling relationship model goes beyond simple single-factor analysis and can predict the performance of the power device under test under temperature-resistance combinations that are not directly tested, providing a foundation for a comprehensive understanding of the dynamic behavior of the power device under test. Attached Figure Description

[0034] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced one by one below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0035] Figure 1 This is a flowchart illustrating the dual-pulse testing method for power devices based on multi-dimensional coupling effects provided by the present invention.

[0036] Figure 2 This is a schematic diagram of the four-dimensional data acquisition and processing process in the power device dual-pulse testing method based on multi-dimensional coupling effect provided by the present invention.

[0037] Figure 3 This is a flowchart illustrating the adaptive driving algorithm in the power device dual-pulse testing method based on multi-dimensional coupling effect provided by this invention.

[0038] Figure 4 This is a schematic diagram showing the comparative verification test results of the adaptive drive resistor provided by this invention and the traditional fixed drive resistor.

[0039] Figure 5 This is a schematic diagram of the structure of the power device dual-pulse test system based on multi-dimensional coupling effect provided by the present invention.

[0040] Figure 6 This is a schematic diagram of the physical structure of the electronic device provided by the present invention. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0042] Double-pulse testing is a standard method for evaluating the switching characteristics of power devices and is widely used in the dynamic performance analysis of devices such as IGBTs (Insulated Gate Bipolar Transistors) and SiC MOSFETs (Metal-Oxide-Semiconductor Field-Effect Transistors). Traditional double-pulse testing mainly focuses on parameters such as switching losses, turn-on and turn-off times, lacking a systematic study of the coupling effects between temperature, drive resistance, switching characteristics, and electromagnetic interference (EMI). While some existing studies have addressed the influence of temperature or drive resistance on switching behavior, none have established multi-dimensional coupling models or implemented adaptive drive strategies based on measured data. Furthermore, existing testing systems suffer from insufficient signal acquisition accuracy and repeatability under high-temperature and high-noise environments.

[0043] Specifically, traditional dual-pulse test systems for power devices have inherent limitations. They focus solely on voltage and current waveforms during the switching process to extract basic parameters such as switching losses and delays, lacking systematic monitoring and analysis of electromagnetic interference (EMI), a critical performance characteristic. Furthermore, they fail to reveal the intrinsic coupling relationship between temperature, drive resistance, switching characteristics, and EMI. Tests are typically conducted at room temperature or only at a single high-temperature point, with a fixed drive resistance. This fails to reflect the evolution of the switching behavior and EMI characteristics of power devices under dynamic junction temperature changes in actual operating conditions, making it difficult to guide performance optimization across the entire temperature range in practical applications. The test system is merely an open-loop data acquisition device; conclusions drawn from the test data cannot provide real-time feedback and optimization of the device's drive conditions, resulting in a disconnect between testing and application.

[0044] To address the aforementioned technical issues, this invention proposes a dual-pulse testing method and system capable of characterizing the four-dimensional coupling effect of temperature, driving resistance, switching characteristics, and electromagnetic interference, and possessing adaptive optimization capabilities. It constructs a closed-loop testing system that can precisely control temperature, flexibly adjust resistance, synchronously acquire electrical and near-field electromagnetic interference signals, and has back-end intelligent data processing and modeling capabilities, thereby realizing multi-dimensional coupling analysis of the switching characteristics and electromagnetic interference of power devices.

[0045] Figure 1 This is a flowchart illustrating the power device double-pulse testing method based on multi-dimensional coupling effects provided by this invention. The power device double-pulse testing method based on multi-dimensional coupling effects can be executed by the power device double-pulse testing system based on multi-dimensional coupling effects provided in this embodiment of the invention. Figure 1 As shown, the power device double-pulse testing method based on multi-dimensional coupling effect includes the following steps:

[0046] S101, control the input of a dual-pulse excitation signal to the control terminal of the power device under test; wherein, the drive resistor on the drive line from the dual-pulse excitation signal to the control terminal is adjustable.

[0047] Specifically, the drive resistor refers to the resistor connected in series in the circuit that transmits the dual-pulse excitation signal to the control terminal of the power device under test (DDT), such as the gate. Its resistance value is adjustable and used to control the strength and speed of the dual-pulse excitation signal input to the control terminal of the DDT, directly affecting the switching dynamics of the DDT. First, the dual-pulse excitation signal is input to the control terminal of the DDT, and the drive resistor on the drive circuit transmitting the dual-pulse excitation signal is adjustable. Therefore, the resistance value of the drive resistor can be changed by external commands, thereby altering the edge speed and drive capability of the drive signal applied to the control terminal of the DDT.

[0048] S102. Control the power device under test to be under different temperature conditions, and control the execution of double pulse test under various combinations of different temperature conditions and different resistance values ​​of drive resistors.

[0049] Specifically, temperature conditions refer to the thermal state of the power device under test (DUT) during testing. This can be achieved by controlling the ambient temperature of the DUT, or by using other methods to bring the chip junction temperature of the DUT to and stabilize it at a specific value. Controlling the DUT to different temperature conditions can be done by placing it in a temperature-controlled chamber, or by using other heating and temperature control methods. The tests will be conducted under various test scenarios formed by combinations of different temperature conditions and different drive resistor values.

[0050] S103. In each double-pulse test, acquire the electrical switching signal of the power device under test and the near-field electromagnetic interference signal generated by the switching process of the power device under test.

[0051] Specifically, electrical switching signals refer to the voltage and current signals reflecting the switching process of the power device under test (DUT) during the double-pulse test, such as collector-emitter voltage, gate-emitter voltage, and collector current. Near-field electromagnetic interference (NEMI) signals refer to the electromagnetic field signals generated by the high-speed switching action of the DUT and its main power circuit during the double-pulse test, propagating in close space, and can be captured using a near-field probe. In each specific test scenario, i.e., a combination of temperature and driving resistor, a standard double-pulse test is executed. During this process, the method simultaneously performs two data acquisition tasks: first, acquiring the electrical switching signals of the DUT, such as voltage and current waveforms; and second, acquiring the near-field NEMI signals generated by this switching action.

[0052] In some embodiments, acquiring the electrical switching signals of the power device under test includes acquiring the collector-emitter voltage, collector current, and gate-emitter voltage of the power device under test; wherein the switching characteristic parameters are extracted from the signals of the collector-emitter voltage, collector current, and gate-emitter voltage.

[0053] Specifically, for three-terminal power devices such as insulated-gate bipolar transistors (IGBTs), the collector-emitter voltage refers to the voltage between its collector and emitter. For metal-oxide-semiconductor (MOSFETs), it corresponds to the voltage between the drain and source, reflecting the voltage across the main power circuit of the device under test. The gate-emitter voltage refers to the voltage between the gate and emitter of the device under test. It directly reflects the shape and quality of the drive signal. The collector current is the current flowing through the main power circuit of the device under test, i.e., the current from the collector to the emitter.

[0054] The specific operation of acquiring electrical switching signals involves simultaneously acquiring three key electrical quantities: collector-emitter voltage, collector current, and gate-emitter voltage. Collector-emitter voltage and collector current are the core signals for calculating switching losses and observing voltage and current overshoot and ringing. The gate-emitter voltage waveform contains information such as the rise and fall times of the drive signal, plateau voltage, Miller plateau duration, and gate ringing. This information is crucial for analyzing switching delay and understanding the interaction between the drive circuit and the internal capacitance of the power device under test. From these three simultaneously acquired signals, the required switching characteristic parameters can be extracted using signal processing algorithms. For example, the turn-on delay can be calculated by combining the rising edges of the gate-emitter voltage and collector current; the switching energy loss can be calculated by integrating the product of the collector-emitter voltage and collector current during the switching transient; and the turn-off overshoot peak value can be extracted from the collector-emitter voltage waveform.

[0055] Therefore, by synchronously acquiring collector-emitter voltage, collector current, and gate-emitter voltage, the embodiments of the present invention can completely capture the entire electrical response from drive input to main power output in a single switching event. This provides a complete data source for the subsequent accurate extraction of various switching characteristic parameters, such as delay, loss, and overshoot, ensuring that the established coupling relationship model has high accuracy and reliability and can truly reflect the comprehensive impact of changes in drive conditions on switching behavior.

[0056] S104. Based on the electrical switching signals and near-field electromagnetic interference signals obtained under various combinations, establish a coupling relationship model between temperature conditions, the resistance value of the driving resistor, the switching characteristic parameters of the power device under test, and the electromagnetic interference characterization parameters of the power device under test.

[0057] Specifically, the coupling relationship model refers to a model established through mathematical methods or algorithms that quantitatively describes the relationship between two input variables—temperature conditions and the resistance value of the driving resistor—and two output variables—switching characteristic parameters and electromagnetic interference characterization parameters. By repeatedly performing tests and data acquisition under various combinations of the aforementioned conditions, a dataset covering four dimensions—temperature, driving resistance, electrical response, and electromagnetic interference response—can be obtained. Finally, based on this dataset, a modeling algorithm is used to establish the coupling relationship model among these four components. This model can quantitatively reveal how the switching characteristic parameters of the power device under test, such as losses and time, and the electromagnetic interference characterization parameters, such as intensity and spectrum, change when temperature conditions or the driving resistance value changes.

[0058] Therefore, this invention innovatively places temperature conditions, driving resistance, switching characteristics, and near-field electromagnetic interference within the same test framework for synchronous and correlated measurement and analysis. By actively changing the combination of temperature and driving resistance and collecting data from all dimensions, it can systematically reveal multi-factor coupling effects that cannot be observed by traditional methods. The established coupling relationship model goes beyond simple single-factor analysis and can predict the performance of the power device under test under temperature-resistance combinations that are not directly tested, providing a foundation for a comprehensive understanding of the dynamic behavior of the power device under test.

[0059] In some embodiments, based on electrical switching signals and near-field electromagnetic interference signals obtained under various combinations, a coupling relationship model is established between temperature conditions, the resistance value of the driving resistor, the switching characteristic parameters of the power device under test, and the electromagnetic interference characterization parameters of the power device under test. This includes extracting at least one switching characteristic parameter from the electrical switching signal; extracting at least one electromagnetic interference characterization parameter from the near-field electromagnetic interference signal; and establishing the coupling relationship model using a machine learning algorithm, with temperature conditions and the resistance value of the driving resistor as input variables and the extracted switching characteristic parameters and electromagnetic interference characterization parameters as output variables.

[0060] Specifically, switching characteristic parameters refer to parameters extracted from electrical switching signals to quantify and describe the dynamic process of switching, such as turn-on delay time td(on), turn-off delay time td(off), turn-off loss E_off, and voltage overshoot V_spike. Electromagnetic interference characterization parameters refer to parameters extracted from near-field electromagnetic interference signals to quantify and describe the intensity of electromagnetic interference, such as ringing energy E_ring within a certain time window. Machine learning algorithms refer to computer algorithms that can automatically learn patterns and rules from data, such as physical information neural network algorithms, which can be used to establish complex nonlinear input-output mapping relationships. First, the electrical switching signals obtained under each test combination are processed to extract one or more parameters that can characterize the switching characteristics, such as turn-off loss E_off. Second, the near-field electromagnetic interference signals obtained in the same test are processed to extract one or more parameters that can characterize the intensity of electromagnetic interference generated by the switch, such as ringing energy E_ring. Thus, each test yielded a set of corresponding input variables: temperature condition value T, driving resistor value Rg, and output variables: switching characteristic parameter P_sw and electromagnetic interference parameter P_emi. Then, the numerous sets of data (T, Rg, P_sw, P_emi) from all test combinations were aggregated. Using temperature condition T and driving resistor value Rg as input variables and switching characteristic parameter P_sw and electromagnetic interference parameter P_emi as output variables, a physical information neural network algorithm was used to establish a coupling relationship model.

[0061] First, based on our understanding of the physics and experiments of power semiconductor devices, we determine the a priori physical relationships that reflect the intrinsic correlation between the switching characteristic parameter P_sw and the temperature condition value T and the driving resistance value Rg. For example, taking the switching characteristic parameter P_sw as the turn-off loss E_off and the junction temperature Tj as the temperature condition T, the a priori relationship for the turn-off loss E_off can be expressed as: E_off = A exp(-B Rg)+C Tj^2, where A, B, and C are physical parameters to be determined. This equation reflects the basic physical trend that the turn-off loss E_off decreases exponentially with the increase of the driving resistance Rg, and increases quadratically with the increase of the junction temperature Tj.

[0062] Then, a neural network is constructed with the aforementioned prior physical relation as its core physical constraint. During network training, the loss function consists of two parts: the first part is the error between the network's predicted value and the measured data; the second part is the network output, such as the deviation between the predicted E_off and the value calculated by the aforementioned prior physical relation. By minimizing the total loss function through an optimization algorithm, the network is forced to obey the physical rules defined by the prior relation while learning the patterns in the data.

[0063] Ultimately, the converged physical information neural network encodes a complete mapping from (Tj, Rg) to E_off, consistent with physical laws, through its connection weights and activation functions. This network model is the desired coupling relationship model, which not only accurately reflects the data but also ensures consistency between the predicted behavior and the device physics. The quantitative relationships implied in the model, such as the equivalent set of parameters A, B, and C, can be obtained by analyzing the network's response to specific inputs.

[0064] Similarly, taking the electromagnetic interference parameter P_emi as the ringing energy E_ring and the junction temperature Tj as the temperature condition T, the prior relation for the ringing energy E_ring can be expressed as: E_ring = D / [(Rg + E Tj)^F], where D, E, and F are physical parameters to be determined. This equation qualitatively describes the trend of ringing energy E_ring decaying with the combined effect of the driving resistance Rg and the junction temperature Tj.

[0065] When constructing the physical information neural network, the aforementioned prior physical relationships regarding switching loss E_off and electromagnetic interference energy E_ring can be introduced simultaneously as physical constraints for training the network. During training, the network simultaneously minimizes multiple outputs, such as the errors between the predicted E_off and E_ring and the measured data, as well as the deviations between these outputs and their respective calculated values ​​from the prior physical relationships. This process ensures that the final trained coupling model accurately captures data features as a whole, while the multidimensional mapping relationships learned internally strictly follow the physical laws defined by prior knowledge regarding the switching characteristic parameter P_sw and the electromagnetic interference parameter P_emi.

[0066] Therefore, the embodiments of the present invention extract the original massive waveform data into clear and quantitative parameter relationships. The model established by the machine learning algorithm can characterize complex coupling effects with concise mathematical forms or efficient prediction functions, making it possible to predict the performance of the power device under test at any temperature and with any driving resistance, and greatly reducing the workload of exhaustive testing.

[0067] In some embodiments, after establishing a coupling relationship model between temperature conditions, the value of the driving resistor, the switching characteristic parameters of the power device under test, and the electromagnetic interference characterization parameters of the power device under test, the method further includes obtaining an adaptive driving strategy for the power device under test based on the coupling relationship model; wherein, the adaptive driving strategy defines the adjustment rules for the value of the driving resistor as the temperature conditions change; and the value of the driving resistor is controlled and adjusted according to the adaptive driving strategy and the temperature parameters of the power device under test obtained in real time.

[0068] Specifically, the adaptive drive strategy refers to a control rule that dynamically adjusts the drive resistor value based on real-time operating conditions, such as temperature. This strategy originates from the aforementioned coupling relationship model and aims to ensure that the power device under test maintains the desired performance under changing operating conditions. Temperature parameters refer to parameters that reflect the real-time thermal state of the power device under test. These can be directly measured junction temperatures or temperatures such as heat sink temperatures and ambient temperatures used to estimate junction temperatures.

[0069] After establishing the coupling relationship model, this model is further applied to achieve adaptive driving. First, based on the established coupling relationship model, an adaptive driving strategy for the power device under test is derived or generated. Specifically, analysis of the model shows that for each specific temperature condition, there exists a value or range of values ​​for the driving resistor, which optimizes certain performance indicators, such as electromagnetic interference or switching losses. Therefore, the adaptive driving strategy is essentially a function or lookup table that defines the rules for how the optimal value of the driving resistor should be adjusted according to changes in real-time temperature parameters. Then, in actual operation or subsequent verification testing, the temperature parameters of the power device under test are acquired in real time, for example, by obtaining the junction temperature of the power device under test through an onboard temperature sensor. Based on this real-time temperature parameter, and by querying or calculating the aforementioned adaptive driving strategy, the optimal driving resistor value to be set at the current temperature can be obtained. Finally, through control commands, such as switching an analog switch, the driving resistor on the drive circuit is adjusted to this optimal value.

[0070] Therefore, the embodiments of the present invention realize a closed loop from test analysis to active control, transforming the profound insights gained from offline testing into intelligent strategies for online applications. By sensing the temperature in real time and adjusting the drive resistor accordingly, the power device under test can automatically adapt within the operating temperature range and always operate near the optimal or suboptimal operating point determined by the model. This effectively solves the problem that fixed parameter drives cannot balance performance and electromagnetic compatibility across the entire temperature range, and improves the overall reliability, efficiency, and electromagnetic compatibility level of the system.

[0071] In some embodiments, the adaptive driving strategy includes a strategy aimed at minimizing electromagnetic interference characterization parameters and a strategy aimed at minimizing switching losses.

[0072] Specifically, the strategy aimed at minimizing electromagnetic interference characterization parameters aims to minimize the intensity of electromagnetic interference generated by the power device under test (DUT) by adjusting the drive resistor at any temperature. The strategy aimed at minimizing switching losses aims to minimize the switching energy losses of the DUT by adjusting the drive resistor at any temperature, thereby improving system efficiency.

[0073] Based on the same coupling model, different optimization strategies can be derived, depending on the primary considerations during system design. If the system has extremely high electromagnetic compatibility requirements, such as in environments with a high density of sensitive electronic equipment, the strategy can be generated with minimizing electromagnetic interference as the objective. Specifically, in the model, for each temperature, find the driving resistor that minimizes the output variable, such as the electromagnetic interference characterization parameter. Establishing rules for the specific correspondences of these parameters yields the optimal electromagnetic interference strategy. If the system has extremely high conversion efficiency requirements, such as in applications pursuing maximum energy efficiency, the strategy can be generated with minimizing switching losses as the objective. Specifically, in the model, for each temperature, find the driving resistor that minimizes the output variable, such as the switching loss. Establishing rules for the specific correspondences of these parameters yields the efficiency-first strategy. For example, the driving resistor that minimizes electromagnetic interference has a larger resistance value, while the driving resistor that minimizes switching losses has a smaller resistance value; there is a trade-off between the two.

[0074] Therefore, by clearly defining two strategies with different objectives, the embodiments of the present invention enable the adaptive driving scheme to flexibly adapt to diverse application needs. Users or systems can choose to enable the corresponding strategy according to the main contradictions of the actual scenario, thereby making targeted optimization choices between low electromagnetic interference and high efficiency, enhancing the practicality and applicability of the method.

[0075] In some embodiments, the resistance value of the drive resistor is controlled and adjusted according to the adaptive drive strategy and the temperature parameters of the power device under test acquired in real time. This includes controlling and adjusting the resistance value of the drive resistor according to the adaptive drive strategy, the temperature parameters of the power device under test acquired in real time, and the electrical switching signal of the power device under test acquired in real time.

[0076] Specifically, the real-time acquired electrical switching signals of the power device under test refer to the electrical waveform signals measured and acquired in real time during the actual operation of the power device under test, such as real-time gate-emitter voltage and collector-emitter voltage. Desired characteristics refer to the waveform characteristic parameters that the electrical switching signals should possess to achieve a specific optimization objective, derived from the coupling relationship model. Examples include the decay time constant of gate voltage ringing and the allowable peak value of collector voltage overshoot. Specific optimization objectives include minimizing electromagnetic interference or switching losses.

[0077] Building upon the above embodiments, the adjustment process not only relies on real-time temperature parameters but also incorporates real-time acquired electrical switching signals as additional feedback input. First, based on the real-time temperature parameters and the adaptive drive strategy, a preliminary value for the drive resistor is set to achieve feedforward control. Then, a switching action is performed at this drive resistor value, and the electrical switching signals generated by this switching action are acquired in real-time, such as the gate voltage ringing amplitude and the collector voltage overshoot peak value. These real-time voltage waveforms directly determine the transient current and voltage change rates of the switch, which are the primary sources of electromagnetic interference. Therefore, the characteristics of the real-time voltage waveforms are a direct basis for judging the current level of electromagnetic interference.

[0078] Then, the characteristics of the real-time signal are compared with the expected characteristics. The expected characteristics are derived from the coupling relationship model. Specifically, when the model aims to minimize electromagnetic interference or switching losses, it can calculate a theoretically optimal drive resistor value for a given temperature condition. Simultaneously, the model can predict the theoretical electrical waveform characteristics that the switching process should produce under this optimal drive resistor value, i.e., the expected characteristics. If the real-time acquired waveform characteristics, such as ringing energy, deviate from the expected characteristics given by the model, it indicates that the actual switching state has not reached the theoretical optimum at the current temperature, possibly affected by factors such as parasitic parameters, device dispersion, or aging. In this case, the control algorithm can fine-tune the drive resistor value within the range allowed by the adaptive drive strategy. For example, if the actual ringing is greater than the expected value, the resistance is slightly increased to suppress ringing, and a new waveform is acquired for verification during the next switching operation. This constitutes a closed-loop feedback control loop of measurement, comparison, and adjustment, with the goal of continuously bringing the real-time waveform characteristics closer to the expected characteristics.

[0079] Therefore, this invention introduces real-time electrical signals as feedback input, upgrading adaptive control from temperature-based open-loop feedforward to closed-loop control combining model feedforward and real-time feedback. This effectively compensates for factors that the feedforward model may not fully cover or cannot perceive in real time, such as device parameter dispersion, changes in circuit parasitic parameters, aging drift, and load fluctuations. By forcing the real-time waveform of each switching action to approach the optimal waveform determined by the coupled model, the system can respond more quickly and accurately to dynamically changing operating conditions, thereby achieving a more robust and stable performance optimization effect than simply relying on temperature feedforward, further improving the system's electromagnetic compatibility and efficiency throughout its entire lifecycle and under complex operating conditions.

[0080] Figure 2 This is a schematic diagram of the four-dimensional data acquisition and processing process in the power device dual-pulse testing method based on multi-dimensional coupling effect provided by this invention. (See diagram for example.) Figure 2 As shown, the four-dimensional data acquisition and processing methods specifically include:

[0081] S201, Begin.

[0082] S202, System Calibration.

[0083] Specifically, system calibration and background noise measurement were performed. With no pulse signal, the waveforms of each oscilloscope channel, especially the near-field probe channel, were recorded, and the system's inherent noise floor was measured and used as the background subtraction benchmark for subsequent data. Simultaneously, the voltage and current probes were calibrated using a standard signal source to ensure amplitude and timing accuracy.

[0084] S203, Temperature traversal cycle begins.

[0085] S204, Resistor traversal loop begins.

[0086] S205, Repeat test loop begins.

[0087] S206, Send a double-pulse excitation signal.

[0088] S207, Synchronous waveform acquisition.

[0089] For example, the collector-emitter voltage Vce, collector current Ic, and gate-emitter voltage Vge of the power device under test can be acquired simultaneously, as well as the near-field probe signal Vnp.

[0090] S208, Save data.

[0091] S209, The repeat test loop ends.

[0092] S210, the resistor traversal loop ends.

[0093] S211, Temperature traversal loop ends.

[0094] Specifically, a full-dimensional fingerprint database is collected. The test script is executed automatically by the control processing unit. The script control loop sequence is as follows: set the temperature of the thermal environment control chamber, then wait for thermal equilibrium (this process takes more than 15 minutes), then set the drive resistor level, then send a dual-pulse excitation signal, trigger the oscilloscope to acquire data, and finally save the data.

[0095] S212, Batch import data.

[0096] S213, Extract switch parameters.

[0097] S214. Calculate electromagnetic interference indicators.

[0098] S215. Construct a four-dimensional fingerprint database.

[0099] S216. Establish a four-dimensional coupling model.

[0100] S217. Generate response surface.

[0101] Specifically, when saving data, the data can be named in a structured manner, such as Dev1_T75_Rg10_Rep3.mat, containing all dimensions of information such as device number, junction temperature, gate resistance, and number of repetitions. The gate resistance is the drive resistance mentioned above. Through this automated traversal test, massive amounts of raw waveform data can be obtained, forming the basis for the four-dimensional fingerprint library for subsequent modeling.

[0102] Then, key parameters are extracted and modeled. An automated processing script is written using MATLAB to read raw waveform files in batches. Regarding switch parameter extraction, the turn-on delay time td(on), turn-off delay time td(off), turn-off loss E_off, and voltage overshoot V_spike are automatically extracted from the collector-emitter voltage and collector current waveforms. Regarding electromagnetic interference (EMI) parameter extraction, the time-domain ringing energy E_ring is calculated as a quantitative indicator of EMI intensity. Regarding coupling model construction, a machine learning algorithm can be used, taking the driving resistor value Rg and junction temperature Tj as inputs, and the switch characteristic parameters and EMI parameters as outputs, to establish an accurate mapping model. For details, please refer to the corresponding description in the aforementioned embodiments.

[0103] S218, Derive the adaptive algorithm.

[0104] Specifically, adaptive algorithms are generated and verified. Based on the aforementioned model, an optimal adaptive algorithm for electromagnetic interference is derived. For example, the algorithm output is: Rg_opt(Tj) = 10 + 0.08 (Tj-25). This function indicates that the optimal gate resistance should increase linearly as the junction temperature rises to continuously suppress electromagnetic interference. This algorithm is embedded in the microcontroller firmware of the adjustable drive resistor driver board as a lookup table or formula, enabling it to automatically adjust the drive resistor value based on real-time temperature signals, such as junction temperatures detected by onboard NTC (Negative Temperature Coefficient Thermistor) or thermistor diodes.

[0105] S219, End.

[0106] Figure 3 This is a flowchart illustrating the adaptive driving algorithm in the power device dual-pulse testing method based on multi-dimensional coupling effects provided by this invention. Figure 3 As shown, the adaptive driving algorithm specifically includes:

[0107] S301, Begin.

[0108] Specifically, the system is powered on.

[0109] S302, System initialization.

[0110] Specifically, system initialization includes the initialization of the GPIO (General Purpose Input / Output) ports and GPIB (General Purpose Interface Bus) ports of the control processing unit, as well as the initialization of the analog-to-digital converter module in the control processing unit. The GPIO ports are used to control the analog switches in the adjustable drive unit and select different gate resistors. The GPIB port is the communication interface for controlling the dual-pulse excitation unit. The analog-to-digital converter module is used to acquire voltage signals, such as gate-emitter voltage, in real time for adaptive control.

[0111] S303, Obtain junction temperature and voltage values.

[0112] Specifically, the junction temperature can be obtained by reading the measurement signal from the temperature sensor, and the voltage value can be obtained using the analog-to-digital converter module.

[0113] S304. Determine if EMI optimization is the goal. If yes, proceed to S305; otherwise, proceed to S306.

[0114] S305, Call the optimal algorithm for electromagnetic interference.

[0115] Specifically, the adaptive driving strategy includes strategies aimed at minimizing electromagnetic interference characterization parameters.

[0116] S306, Invoke the efficiency-first algorithm.

[0117] Specifically, the adaptive driving strategy includes a strategy aimed at minimizing switching losses.

[0118] S307. Calculate the optimal driving resistor value.

[0119] Specifically, based on the aforementioned model and the specific algorithm employed, such as the electromagnetic interference optimization algorithm or the efficiency-first algorithm, the optimal driving resistor value is calculated. For example, the output of the electromagnetic interference optimization algorithm or the efficiency-first algorithm is: Rg_opt(Tj) = 10 + 0.08 (Tj-25).

[0120] S308 controls the adjustable drive resistor to switch the drive board to the corresponding resistance position.

[0121] S309, Delayed waiting.

[0122] Specifically, after a delay of one sampling period, the aforementioned initialization steps are executed repeatedly to match the physical process speed and create a clear control cycle.

[0123] Figure 4 This is a schematic diagram showing the comparative verification test results of the adaptive drive resistor provided by this invention and the traditional fixed drive resistor. Figure 4 The left ordinate represents the electromagnetic interference intensity, characterized by ringing energy E_ring, and the right ordinate represents the turn-off loss E_off. A comparative verification experiment is designed to compare the traditional fixed drive resistor value Rg strategy with the adaptive drive resistor value Rg strategy proposed in this invention. Figure 4 It can be seen that the ringing energy E_ring decreases by more than 60% and the turn-off loss E_off decreases by more than 20%, verifying the effectiveness of the adaptive algorithm provided in this embodiment of the invention from multiple dimensions such as electromagnetic interference suppression effect, switching loss, and system stability.

[0124] In summary, this invention, for the first time, simultaneously and quantitatively characterizes the coupling effects of four dimensions—temperature conditions, drive resistor value, switching characteristic parameters, and electromagnetic interference parameters—in testing, providing unprecedented insights into the dynamic behavior of devices. The quantitative model built upon massive amounts of data can not only explain existing phenomena but also predict device performance at untested temperature points and without tested drive resistors, significantly reducing the workload of additional testing. The generated adaptive drive algorithm can be directly applied to the drive design of actual products, achieving a closed loop of testing, modeling, and application, enabling power devices to automatically maintain optimal performance—i.e., low loss and low electromagnetic interference—across the entire operating temperature range. Through a series of measures, including thermal balance control, electromagnetic shielding, background noise cancellation, and repeated statistical analysis, the accuracy and repeatability of the collected data are ensured under harsh testing environments with high interference and high temperatures.

[0125] This invention also provides a dual-pulse test system for power devices based on multi-dimensional coupling effects. Figure 5 This is a schematic diagram of the structure of the power device dual-pulse test system based on multi-dimensional coupling effect provided by the present invention. Figure 5As shown, a power device dual-pulse test system based on multi-dimensional coupling effect includes a dual-pulse excitation unit 101, an adjustable drive unit 102, a thermal environment control and acquisition unit 103, a multi-dimensional data acquisition unit 104, and a control processing unit 105. The dual-pulse excitation unit 101 generates a dual-pulse excitation signal and outputs it to the control terminal of the power device under test 106. The adjustable drive unit 102 switches the drive resistance on the drive line that receives the dual-pulse excitation signal from the control terminal. The thermal environment control and acquisition unit 103 provides different temperature conditions for the power device under test 106 and acquires the thermal environment data of the power device under test 106. Temperature parameters; the multi-dimensional data acquisition unit 104 includes an electrical detection component 107 for acquiring the electrical switching signal of the power device under test 106, and a near-field detection component 108 for acquiring the near-field electromagnetic interference signal of the power device under test 106; the control processing unit 105 is communicatively connected to the dual-pulse excitation unit 101, the adjustable drive unit 102, the thermal environment control acquisition unit 103, and the multi-dimensional data acquisition unit 104, respectively, and is used to execute the power device dual-pulse test method based on multi-dimensional coupling effect as described in the above embodiments, thus possessing the beneficial effects described in the above embodiments, which will not be repeated here. For example, Figure 5 The solid lines in the diagram represent control signal flow, and the dashed lines represent data signal flow.

[0126] Specifically, the dual-pulse excitation unit 101 refers to a hardware module used to generate standard dual-pulse voltage or current signals, which can be implemented by a signal generator or a specific circuit. The adjustable drive unit 102 refers to a drive circuit module capable of receiving dual-pulse excitation signals and controllably changing the series resistance value at its output terminal; it can be implemented by an adjustable drive resistor drive board. The dual-pulse excitation signal is output to the control terminal of the power device under test 106 through the resistor connected to the adjustable drive unit 102. The thermal environment control and acquisition unit 103 refers to a device capable of providing a controllable temperature environment for the power device under test 106 and measuring temperature-related parameters of the power device under test 106. The multi-dimensional data acquisition unit 104 refers to a hardware collection containing various sensors and their signal conditioning circuits for synchronously acquiring different types of signals, such as electrical signals and electromagnetic field signals; for example, it can be a multi-channel digital oscilloscope. The control processing unit 105 refers to a computing and control device responsible for coordinating the work of each unit in the system, executing the test process, processing data, and running algorithms; such as a computer or microcontroller. Electrical detection component 107 refers to sensors used to sense electrical quantities such as voltage and current, such as voltage probes, current probes, differential probes, and combinations of isolation amplifiers and shunt resistors. Specifically, voltage probes are used to sense the gate-emitter voltage, current probes are used to sense the collector current, and differential probes are used to sense the collector-emitter voltage. Near-field detection component 108 refers to sensors used to sense the intensity of near-field electromagnetic fields, which can be near-field magnetic field probes or near-field electric field probes.

[0127] The dual-pulse excitation unit 101 is responsible for generating the standard dual-pulse sequence required for testing. The adjustable drive unit 102, located between the dual-pulse excitation unit 101 and the power device under test 106, functions as a variable resistance box, switching the actual resistance value in the drive circuit according to instructions. The thermal environment control and acquisition unit 103 creates and maintains various temperature conditions required for testing and monitors the junction temperature of the power device under test 106. The electrical detection component 107 in the multi-dimensional data acquisition unit 104 is responsible for capturing the voltage and current waveforms on the pins of the power device under test 106, while the near-field detection component 108 is responsible for capturing the near-field electromagnetic signals radiated by the switching action. All units are controlled by the control processing unit 105. The control processing unit 105 executes the power device dual-pulse test method based on multi-dimensional coupling effect as described in the above embodiments, instructing the thermal environment control and acquisition unit 103 to set the temperature, instructing the adjustable drive unit 102 to set the resistance, triggering the dual-pulse excitation unit 101 to emit test pulses, and simultaneously triggering the multi-dimensional data acquisition unit 104 to capture data. Subsequently, the control processing unit 105 reads all the data and executes algorithmic tasks such as signal processing, parameter extraction, model building, and strategy generation.

[0128] Therefore, the embodiments of the present invention form a complete and operable technical solution implementation platform, which integrates precise temperature control, flexible resistance switching, synchronous multi-dimensional data acquisition and powerful data processing capabilities, enabling systematic research on four-dimensional coupling effects and the realization of adaptive drive optimization.

[0129] In some embodiments, the adjustable drive unit 102 includes multiple switching channels, each equipped with an analog switch and a resistor. The adjustable drive unit 102 is used to switch the drive resistor on the drive line that receives the dual-pulse excitation signal to the control terminal by controlling the conduction state of the analog switch.

[0130] Specifically, the switching channel refers to an independent branch within the adjustable drive unit 102, each consisting of an analog switch and a resistor connected in series. Multiple such branches are connected in parallel between the input and output terminals of the dual-pulse excitation signal. An analog switch is an electronic switching device whose on / off state can be controlled by a digital signal, such as a metal-oxide-semiconductor field-effect transistor analog switch or a relay. The adjustable drive unit 102 contains multiple parallel switching channels. Each switching channel has a resistor of a specific resistance value and an analog switch connected in series. One end of all switching channels is connected to the input terminal of the dual-pulse excitation signal, and the other end is connected in parallel to the control terminal of the power device under test 106. When it is necessary to switch to a certain resistance value of the drive resistor, the control processing unit 105 sends a control signal to the adjustable drive unit 102. This control signal only closes the analog switch on the corresponding channel, while keeping the analog switches of all other channels open. At this time, the drive current will flow through the only conducting control signal, and the resistor connected in series with the control signal is connected to the drive circuit, becoming the currently effective drive resistor. By controlling the conduction of different analog switches, the driving resistor can be switched between different preset resistance values.

[0131] Therefore, the multi-channel analog switch switching structure adopted in this embodiment of the invention is a reliable circuit scheme for realizing fast, accurate, and programmable switching of the driving resistor. The structure is clear and the control is simple. By selecting high-precision resistors and low charge injection analog switches, the accuracy of the input resistor value can be guaranteed, and the introduction of additional interference by the switching action can be avoided, thereby reliably realizing the core function of adjustable driving conditions.

[0132] In some embodiments, the thermal environment control acquisition unit 103 includes a thermal environment control box 109, in which the power device under test 106 is placed, and the thermal environment control box 109 is used to provide different temperature conditions for the power device under test 106; and a temperature measuring component 110 is used to measure the junction temperature of the power device under test 106.

[0133] Specifically, the thermal environment control chamber 109 is a sealed or semi-sealed chamber capable of precisely controlling and regulating the temperature of its internal space, such as a high and low temperature test chamber or a high-temperature environment chamber equipped with a heating element 111. The temperature measuring component 110 is a device used to directly or indirectly measure the junction temperature of the chip in the power device under test 106, such as an infrared thermal imager. The power device under test 106 is installed or placed in the internal working space of the thermal environment control chamber 109. By controlling the heating element of the chamber, its internal ambient temperature can be stabilized at a set target value, thereby providing stable and repeatable temperature conditions for the power device under test 106. To obtain the actual junction temperature of the power device under test 106 more accurately, the temperature measuring component 110 is also provided. For example, during test intervals, an infrared thermal imager can be used to perform non-contact temperature measurement by looking through the infrared window on the chamber at the surface of the power device under test 106. This measured temperature data is used to verify the effectiveness of the environmental control and to provide accurate temperature input values ​​for establishing a coupling relationship model.

[0134] Therefore, this embodiment of the invention provides a stable and uniform temperature field through the thermal environment control box 109, ensuring the accuracy and repeatability of temperature conditions during testing. Meanwhile, the independent temperature measurement component 110 provides direct or precise indirect measurement of the key physical quantity, namely the junction temperature, avoiding errors caused by relying solely on ambient temperature. This makes the temperature variables in the established coupling model truly reliable, significantly improving the model's prediction accuracy and practical value.

[0135] In some embodiments, the near-field detection component 108 is disposed at a preset position on the load lead of the power device under test 106.

[0136] Specifically, the load lead refers to the connection line connecting the main power output terminal of the power device under test 106, such as the emitter or collector, to the non-inductive multilayer busbar (BUS) and the power inductive load (L), and is the main path for the switching current. The preset position refers to the fixed installation position of the near-field probe 108 relative to the load lead, determined in advance according to the test specifications, for example, 10 mm directly above it. Exemplarily, the main power output terminal of the power device under test 106 can be connected to the non-inductive multilayer busbar and the power inductive load to minimize parasitic inductance in the loop, which can be powered by, for example, a 400V DC power supply.

[0137] In double-pulse testing of power devices, the high-frequency current generated during switching primarily flows through the load leads. According to electromagnetic field theory, a changing current generates a changing magnetic field around it, and the magnetic field strength in the near-field region is directly related to the current magnitude and rate of change. Therefore, the load leads are a major source of magnetic field radiation from switching noise. A near-field detection component 108, used to collect near-field electromagnetic interference signals, is fixed at a preset position on the load leads. The geometric relationship between the probe and the radiation source remains constant in each test, ensuring the comparability of electromagnetic interference signals collected under different test combinations. The changes in their intensity accurately reflect the influence of different temperatures and driving resistors on the switching current waveform, rather than measurement differences caused by probe position variations.

[0138] Therefore, by fixing the detection component at a preset position near the load lead, this embodiment of the invention can effectively capture the high-frequency magnetic field radiation generated during switching, achieving standardization and repeatability of near-field electromagnetic interference measurement. This ensures that the electromagnetic interference characterization parameters extracted from the electromagnetic interference signal can serve as stable and reliable metrics for quantitative analysis and modeling. Without such standardized measurement settings, the randomness of electromagnetic interference data would increase significantly, making it difficult to establish an effective coupling relationship model.

[0139] The following is an overall description of the power device dual-pulse test system based on multi-dimensional coupling effects. In this system, the dual-pulse excitation unit is controlled by a script run by the control processing unit via a GPIB or USB (Universal Serial Bus) interface to generate dual-pulse excitation signals with adjustable parameters such as pulse width, interval, and amplitude. The core of the adjustable drive unit is a multi-level programmable drive resistor drive board. This board receives standard pulses from the dual-pulse excitation unit and switches the gate resistor connected to the drive circuit via a high-performance analog switch, such as a relay or MOSFET switch array. The resistor levels include at least 2Ω, 3Ω, 5Ω, 8Ω, 10Ω, 12Ω, 15Ω, and 20Ω to cover typical application ranges from emphasizing speed to strong vibration suppression, where emphasizing speed corresponds to low drive resistance and strong vibration suppression corresponds to high drive resistance. The power device under test is installed in a thermal environment control box, and its output is connected to a carefully designed non-inductive multilayer busbar and a power inductive load to minimize parasitic inductance in the circuit. This circuit is powered by a 400V DC power supply.

[0140] The multi-dimensional data acquisition unit includes electrical parameter acquisition and electromagnetic interference (EMI) parameter acquisition. Electrical parameter acquisition can utilize a high-voltage differential probe to measure the collector-emitter voltage of the power device under test; a high-frequency current probe or a non-inductive shunt resistor with an isolation amplifier to measure the collector current; and a passive voltage probe to measure the gate-emitter voltage. All electrical signals are input to a high-speed digital oscilloscope with more than four channels, a sampling rate of 1 GS / s, and a storage depth of 1 M points. EMI parameter acquisition can utilize a near-field magnetic field probe, such as the NFP-3 model, whose signal is also input to one channel of the high-speed digital oscilloscope.

[0141] In the thermal environment control and acquisition unit, the thermal environment control box is used to control the ambient temperature of the power device under test. Temperature measurement components, such as an infrared thermal imager, are aimed at the surface of the power device under test through the anti-reflective infrared window of the thermal environment control box to monitor and verify the chip junction temperature in real time after the pulse is generated, ensuring the accuracy of thermal boundary conditions. All devices are synchronized by a high-speed digital oscilloscope trigger signal to ensure data time alignment. The inner wall of the thermal environment control box is lined with copper foil and properly grounded. All outgoing signal lines are threaded through ferrite cores to suppress cavity resonance and common-mode interference, ensuring the accuracy of electromagnetic interference measurements.

[0142] The present invention will now be described in detail with reference to a preferred embodiment, taking a SiC MOSFET half-bridge module with a power device under test of 1200V / 600A as an example.

[0143] The first step is system setup.

[0144] The oscilloscope used is a Keysight Infiniium S-series, four-channel, with a bandwidth >1GHz, sampling rate ≥5GS / s, and storage depth ≥10M points. The collector-emitter voltage uses a Tektronix THDP0200 high-voltage differential probe, the collector current uses a Pearson 6595 current transformer, and the gate-emitter voltage uses a Tektronix TPP1000 passive voltage probe; the near-field probe is a Langer EMV-Technik RF-R 400-1. The adjustable drive resistor driver board uses a printed circuit board, the core controller is an STM32G4 series microcontroller unit, the analog switch is a TI TS5A23166 low-charge-injection dual-channel single-pole single-throw switch, and the resistor network consists of high-precision, low-inductive metal film resistors with resistance ranges of at least 2Ω, 3Ω, 5Ω, 8Ω, 10Ω, 12Ω, 15Ω, and 20Ω. The thermal environment control box uses an ESPEC temperature cycling test chamber, with openings in its side walls and gold-plated copper terminal blocks and an infrared quartz glass viewing window to ensure airtightness and infrared transmission. The load inductor is made of hollow copper tubing with an inductance of approximately 50μH and a rated current >500A.

[0145] The second step is the testing process.

[0146] The power device under test was mounted on a water-cooled substrate inside the thermal environment control box, and all probes were connected. An automated test script written in Python was run on the control processing unit. The script first set the thermal environment control box to 25°C and waited 15 minutes for thermal equilibrium to be reached. Then, the script sequentially set the drive resistors to 2Ω, 3Ω, 5Ω, 8Ω, 10Ω, 12Ω, 15Ω, and 20Ω. At each drive resistor, five double pulses were sent, with a pulse width of 10μs and an interval of 2μs. The oscilloscope acquired and saved waveform data from all channels with each trigger. After completion, the thermal environment control box was set to 75°C, and the above process was repeated. Finally, it was set to 125°C. During the 125°C test, a 1-minute cooling period was allowed after each pulse to prevent excessive junction temperature. A total of 105 sets of complete waveform data were acquired.

[0147] The third step is data processing and modeling.

[0148] Import all .mat files into MATLAB. For the Vce and Ic waveforms of each data set, automatically calculate td(on) and td(off) by setting voltage or current thresholds; and adjust Vce... During the off-time period, Ic is used to calculate E_off by numerical integration; V_spike is extracted by finding the first peak of the Vce waveform. For the near-field probe signal, a Hanning window is applied and a Fast Fourier Transform is performed to find the peak amplitude in the 30MHz to 300MHz frequency band, or E_ring is obtained by integrating the ringing attenuation envelope.

[0149] The fourth step is algorithm implementation and verification.

[0150] Based on the response surface of E_ring, the optimal algorithm for electromagnetic interference is determined to be Rg_opt = 10 + 0.08. (Tj-25). The algorithm was compiled and then programmed into the STM32 microcontroller of the adjustable drive resistor driver board using an ST-LINK debugger. The thermal environment control box was set to 100℃, and the load current was set to 150A. One set of tests used a fixed drive resistor of 10Ω, while the other set used adaptive mode. In this mode, the driver board estimated the junction temperature based on the heatsink temperature measured by the onboard NTC and dynamically adjusted the drive resistor. The measured values ​​showed that E_ring was lower in adaptive mode than in fixed resistance mode, turn-off loss E_off was lower, and voltage overshoot V_spike was significantly reduced, demonstrating the effectiveness and superiority of this invention.

[0151] Figure 6 This is a schematic diagram of the physical structure of the electronic device provided by the present invention. For example... Figure 6As shown, the electronic device may include a processor 601, a communications interface 602, a memory 603, and a communication bus 604. The processor 601, communications interface 602, and memory 603 communicate with each other via the communication bus 604. The processor 601 can call logic instructions from the memory 603 to execute a power device double-pulse test method based on multi-dimensional coupling effects, including:

[0152] The controller inputs a dual-pulse excitation signal to the control terminal of the power device under test; wherein the drive resistor on the drive line from the dual-pulse excitation signal to the control terminal is adjustable.

[0153] The device under test is controlled to operate under different temperature conditions, and a dual-pulse test is executed under various combinations of different temperature conditions and different resistance values ​​of the drive resistor.

[0154] In each double-pulse test, the electrical switching signal of the power device under test and the near-field electromagnetic interference signal generated by the switching process of the power device under test are acquired.

[0155] Based on the electrical switching signals and near-field electromagnetic interference signals obtained under various combinations, a coupling relationship model is established between temperature conditions, the resistance value of the driving resistor, the switching characteristic parameters of the power device under test, and the electromagnetic interference characterization parameters of the power device under test.

[0156] Furthermore, the logical instructions in the aforementioned memory 603 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0157] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the power device double-pulse testing method based on multi-dimensional coupling effect provided by the above methods, including:

[0158] The controller inputs a dual-pulse excitation signal to the control terminal of the power device under test; wherein the drive resistor on the drive line from the dual-pulse excitation signal to the control terminal is adjustable.

[0159] The device under test is controlled to operate under different temperature conditions, and a dual-pulse test is executed under various combinations of different temperature conditions and different resistance values ​​of the drive resistor.

[0160] In each double-pulse test, the electrical switching signal of the power device under test and the near-field electromagnetic interference signal generated by the switching process of the power device under test are acquired.

[0161] Based on the electrical switching signals and near-field electromagnetic interference signals obtained under various combinations, a coupling relationship model is established between temperature conditions, the resistance value of the driving resistor, the switching characteristic parameters of the power device under test, and the electromagnetic interference characterization parameters of the power device under test.

[0162] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the power device double-pulse testing method based on multi-dimensional coupling effects provided by the methods described above, including:

[0163] The controller inputs a dual-pulse excitation signal to the control terminal of the power device under test; wherein the drive resistor on the drive line from the dual-pulse excitation signal to the control terminal is adjustable.

[0164] The device under test is controlled to operate under different temperature conditions, and a dual-pulse test is executed under various combinations of different temperature conditions and different resistance values ​​of the drive resistor.

[0165] In each double-pulse test, the electrical switching signal of the power device under test and the near-field electromagnetic interference signal generated by the switching process of the power device under test are acquired.

[0166] Based on the electrical switching signals and near-field electromagnetic interference signals obtained under various combinations, a coupling relationship model is established between temperature conditions, the resistance value of the driving resistor, the switching characteristic parameters of the power device under test, and the electromagnetic interference characterization parameters of the power device under test.

[0167] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0168] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or some parts of embodiments.

[0169] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A double-pulse testing method for power devices based on multi-dimensional coupling effects, characterized in that, include: The system controls the input of a dual-pulse excitation signal to the control terminal of the power device under test; wherein the driving resistor on the driving line from the dual-pulse excitation signal to the control terminal is adjustable. The power device under test is controlled to be under different temperature conditions, and a double-pulse test is executed under various combinations of different temperature conditions and different resistance values ​​of the driving resistor. In each double-pulse test, the electrical switching signal of the power device under test and the near-field electromagnetic interference signal generated by the switching process of the power device under test are acquired. Based on the electrical switching signals and near-field electromagnetic interference signals obtained under the various combinations, a coupling relationship model is established between the temperature conditions, the resistance value of the driving resistor, the switching characteristic parameters of the power device under test, and the electromagnetic interference characterization parameters of the power device under test. Specifically, this includes: Extract at least one switching characteristic parameter from the electrical switching signal; Extract at least one electromagnetic interference characterization parameter from the near-field electromagnetic interference signal; Using the temperature conditions and the resistance of the driving resistor as input variables, and the extracted switching characteristic parameters and the electromagnetic interference characterization parameters as output variables, a coupling relationship model is established through a machine learning algorithm.

2. The method of claim 1, wherein, Acquiring the electrical switching signal of the power device under test includes: The collector-emitter voltage, collector current, and gate-emitter voltage of the power device under test are acquired; wherein the switching characteristic parameters are extracted from the signals of the collector-emitter voltage, the collector current, and the gate-emitter voltage.

3. The method of claim 1 or 2, wherein, After establishing the coupling relationship model between the temperature conditions, the resistance value of the driving resistor, the switching characteristic parameters of the power device under test, and the electromagnetic interference characterization parameters of the power device under test, the model further includes: The adaptive driving strategy of the power device under test is obtained according to the coupling relationship model; wherein, the adaptive driving strategy defines the adjustment rule of the resistance value of the driving resistor as the temperature condition changes; The resistance value of the drive resistor is controlled and adjusted based on the adaptive drive strategy and the temperature parameters of the power device under test acquired in real time.

4. The method of claim 3, wherein the method further comprises: The adaptive driving strategy includes a strategy aimed at minimizing electromagnetic interference characterization parameters and a strategy aimed at minimizing switching losses.

5. The power device double-pulse testing method based on multi-dimensional coupling effect according to claim 3, characterized in that, Based on the adaptive driving strategy and the real-time acquired temperature parameters of the power device under test, the resistance value of the driving resistor is controlled and adjusted, including: The resistance value of the drive resistor is controlled and adjusted based on the adaptive drive strategy, the temperature parameters of the power device under test acquired in real time, and the electrical switching signals of the power device under test acquired in real time.

6. A double-pulse test system for power devices based on multi-dimensional coupling effects, characterized in that, include: The dual-pulse excitation unit is used to generate a dual-pulse excitation signal and output it to the control terminal of the power device under test. An adjustable drive unit is used to switch the drive resistor on the drive line that receives the dual-pulse excitation signal from the control terminal. A thermal environment control and acquisition unit is used to provide different temperature conditions for the power device under test and acquire the temperature parameters of the power device under test; The multi-dimensional data acquisition unit includes an electrical detection component for acquiring the electrical switching signal of the power device under test, and a near-field detection component for acquiring the near-field electromagnetic interference signal of the power device under test. The control processing unit is communicatively connected to the dual-pulse excitation unit, the adjustable drive unit, the thermal environment control acquisition unit, and the multi-dimensional data acquisition unit, respectively, and is used to execute the power device dual-pulse test method based on multi-dimensional coupling effect as described in any one of claims 1-5.

7. The multi-dimension coupling effect based power device double-pulse test system of claim 6, wherein, The adjustable drive unit includes multiple switching channels, each equipped with an analog switch and a resistor. The adjustable drive unit is used to switch the drive resistor connected to the drive line of the control terminal by controlling the conduction state of the analog switch.

8. The multi-dimensional coupling effect based power device double-pulse test system of claim 6, wherein, The thermal environment control and acquisition unit includes: A thermal environment control box is provided, in which the power device under test is placed and the thermal environment control box is used to provide different temperature conditions for the power device under test. A temperature measuring component is used to measure the junction temperature of the power device under test.

9. The multi-dimension coupling effect based power device double-pulse test system according to any one of claims 6-8, wherein, The near-field detection component is positioned at a preset location on the load lead of the power device under test.

Images