Method for deconstructing parasitic parameters of MOSFET power device and computer equipment
By establishing the equivalent circuit of the MOSFET power device and calculating the actual transfer function of the MOSFET power device using the least squares method, the problem of large error in parasitic parameter analysis in the prior art is solved, and the accurate deconstruction of the parasitic parameters of the MOSFET power device is realized and the Improve efficiency.
Patent Information
- Application Number
- CN202411211202.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-30
- Publication Date
- 2024-11-19
AI Technical Summary
The existing parasitic parameter analysis methods of MOSFET power devices have problems such as large calculation errors and difficult to guarantee model accuracy, especially when the actual current path of MOSFET power devices is difficult to accurately obtain and the measurement error of experimental observation methods is large.
By obtaining the physical structure of the power device to be tested, establishing an equivalent circuit, determining the theoretical transfer function of the input voltage and the output voltage, obtaining the measured data, and using the least squares method to calculate the actual transfer function, and then calculating the parasitic parameters to reduce the error.
The accurate deconstruction of the parasitic parameters of MOSFET power devices is achieved, which reduces calculation errors and improves the accuracy and efficiency of the model.
Smart Images

Figure CN118981995A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of power electronics, and in particular to a method for deconstructing parasitic parameters of a MOSFET power device and a computer device. Background Art
[0002] Wave energy is a renewable energy source with large reserves, high energy density and wide distribution range. It plays an important role in solving the energy crisis and promoting the construction of a new power system. The DC collection system is the key equipment for wave energy power generation. Power devices, as the core components of the DC collection system, determine its efficiency and quality. However, the manufacturing process, physical materials and structural characteristics of power devices will cause parasitic parameters, including parasitic inductance, parasitic capacitance and parasitic resistance, which will reduce the performance of power devices, increase losses, and even cause device failures. Therefore, in-depth research on the parasitic parameters of power devices is crucial to optimize device design and manufacturing processes and improve the performance and efficiency of power devices.
[0003] The existing parasitic parameter analysis methods of power devices mainly include computational solution method and experimental observation method. The computational solution method usually uses finite element simulation software to perform three-dimensional modeling of the device and calculates the parasitic parameters based on the physical properties of the material. However, since the actual current path of the MOSFET power device is difficult to obtain accurately, it has a great impact on the accuracy of the parasitic parameter value finally obtained; the experimental observation method mainly uses the built hardware steady-state circuit or observes the transient response characteristic curve for parameter analysis. Among them, the steady-state circuit measurement method proposes various equivalent circuit models for different power modules, but the analysis of these models is mainly limited to the simulation verification of the equivalent circuit model, lacks actual measurement comparison verification, and the accuracy of the model is difficult to ensure; the transient response method requires that the sampling frequency of the voltage and current probes needs to meet 3 times the rise time of the measurement signal, otherwise a 5% measurement error will be generated, and the error is still large. Summary of the invention
[0004] The embodiment of the present invention provides a method for deconstructing parasitic parameters of a MOSFET power device and a computer device, aiming to accurately represent the characteristics of parasitic parameters in the MOSFET power device and reduce the calculation error of the parasitic parameters.
[0005] The deconstruction method provided by the present invention comprises the following steps:
[0006] S1, obtaining a physical structure of a power device to be tested, and determining an equivalent circuit of the power device according to the physical structure;
[0007] S2, determining the theoretical transfer function of the input voltage and output voltage of the power device to be tested according to the equivalent circuit;
[0008] S3, obtaining measured data of input voltage and output voltage of the power device to be tested;
[0009] S4, calculating the actual transfer function of the power device to be tested by using the least square method according to the theoretical transfer function and the measured data;
[0010] S5, calculating and obtaining parasitic parameters of the power device to be tested according to the theoretical transfer function and the actual transfer function.
[0011] In some embodiments, the power device to be tested includes a half-bridge circuit, the half-bridge circuit includes an upper tube and a lower tube, the upper tube and the lower tube are both MOSFETs, and in step S1, the equivalent circuit includes:
[0012] An RC circuit, comprising n RC branches, wherein the n RC branches are connected in parallel, wherein n is a positive integer;
[0013] Any RC branch includes an equivalent resistor and an equivalent capacitor, and the equivalent resistor and the equivalent capacitor are connected in series;
[0014] The equivalent resistances in the n RC branches are respectively denoted as R1, R2, ..., R n The equivalent capacitances in the n RC branches are respectively denoted as C1, C2, ..., C n .
[0015] In some embodiments, the equivalent circuit further includes:
[0016] The first RL circuit includes the upper tube drain parasitic resistance R L1 and the upper tube drain parasitic inductance L1, the upper tube drain parasitic resistance R L1 Connected in series with the upper tube drain parasitic inductance L1;
[0017] The second RL circuit includes an equivalent series parasitic resistance R of the upper tube and the lower tube. L2 and the equivalent series parasitic inductance L2, the equivalent series parasitic resistance R L2 connected in series with the equivalent series parasitic inductance L2;
[0018] The first RL circuit is connected in series to one side of the RC circuit, and the second RL circuit is connected in series to the other side of the RC circuit.
[0019] In some embodiments, step S2 includes:
[0020] According to the equivalent circuit, a transfer function expression of the input voltage and the output voltage is obtained, and the transfer function expression is:
[0021]
[0022] Where u0 is the output voltage, u DS is the input voltage, Z c is the total impedance of the RC circuit, s is a complex variable; the total impedance Z of the RC circuit c It is expressed as:
[0023]
[0024] In the formula, Z i is the impedance of the RC branch, expressed as:
[0025]
[0026] The transfer function expression is generalized to obtain the theoretical transfer function G circuit (s):
[0027]
[0028] In the formula, a1, ..., a n-1 、a n and b1, …, b n 、b n+1 They represent the coefficients after the common denominator.
[0029] In some embodiments, step S3 includes:
[0030] Power on the power device to be tested and obtain discrete waveforms of input voltage and output voltage through an oscilloscope;
[0031] The voltage waveform of the discrete waveform diagram is smoothed by using a low-pass filter to obtain the measured data.
[0032] In some embodiments, the low-pass filter is a Butterworth low-pass filter, and the Butterworth low-pass filter is used to convert discrete data of the discrete waveform diagram into a continuous signal.
[0033] In some embodiments, step S4 includes:
[0034] Based on the theoretical transfer function, the value of n is selected, and step response tests are performed on the input voltage and the output voltage respectively to obtain voltage waveforms close to the measured data, and the corresponding voltage waveform data is recorded as analysis group data;
[0035] The actual transfer function is calculated using the least square method based on the analysis group data and the measured data.
[0036] In some embodiments, step S5 includes:
[0037] According to the theoretical transfer function and the actual transfer function, the residual of the equation group is minimized by using the fmincon function to calculate the parasitic parameters of the power device to be tested.
[0038] In some embodiments, the deconstruction method further comprises the following steps:
[0039] S6, building a simulation reproduction platform, constructing the equivalent circuit according to the values of the parasitic parameters, so as to obtain the output voltage curve of the equivalent circuit.
[0040] According to some embodiments, the present invention also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the method for deconstructing parasitic parameters of a MOSFET power device provided in any of the aforementioned embodiments are implemented.
[0041] It can be seen from the above technical solutions that the present invention has the following beneficial effects:
[0042] According to the laws of physics, an equivalent circuit of a MOSFET power device is reasonably established, and the equivalent circuit includes various parasitic parameters that theoretically exist in the MOSFET power device; a specific input voltage / output voltage theoretical transfer function model is constructed according to the equivalent circuit, and based on the input voltage and output voltage waveforms actually measured by the hardware, and the input voltage and output voltage waveforms simulated by a step response test in the software according to the theoretical transfer function model, the theoretical transfer function model is solved using the least squares method, and then the actual transfer function model is determined, and the parasitic parameter values of the MOSFET power device are deconstructed by combining the theoretical and actual transfer function models; compared with the prior art, this solution constructs an equivalent circuit including various parasitic parameters, and based on actual measurement data, the actual transfer function is obtained using the principle of the least squares method, thereby reducing the error in solving the parasitic parameters. BRIEF DESCRIPTION OF THE DRAWINGS
[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.
[0044] Figure 1 Flow charts of some embodiments of the method for deconstructing parasitic parameters of a MOSFET power device of the present invention;
[0045] Figure 2 A power module circuit diagram and its n-order equivalent circuit diagram in some embodiments of the present invention;
[0046] Figure 3 for Figure 2 The waveform diagram of the measured input voltage and output voltage of the embodiment;
[0047] Figure 4 for Figure 2 The waveform diagram of the input voltage before and after filtering of the embodiment;
[0048] Figure 5 for Figure 2 The waveform diagram of the output voltage of the embodiment before and after filtering;
[0049] Figure 6 A power module circuit diagram and a second-order equivalent circuit diagram thereof in some embodiments of the present invention;
[0050] Figure 7 Input voltage and output voltage waveform diagrams of the analysis group and the verification group for some embodiments of the present invention;
[0051] Figure 8 Waveform diagrams of measured output curves and simulated output curves of some embodiments of the present invention;
[0052] Fig. 9 The actual output curve, system identification output curve, and voltage waveform diagram of the output curve reproduced by Simulink software of some embodiments of the present invention are shown. DETAILED DESCRIPTION
[0053] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0054] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system described above can refer to the corresponding process in the method embodiment, and will not be repeated here.
[0055] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and drawings of the present application are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein, for example. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0056] It should be understood that in this application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.
[0057] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
[0058] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
[0059] Existing parasitic parameter analysis of power devices mainly includes computational solution method and experimental observation method.
[0060] The modeling accuracy of the three-dimensional model and the simulation accuracy of the current path in the computational solution method have an important impact on the analysis of parasitic parameters. In actual experiments, the actual current path of MOSFET power devices is difficult to obtain accurately. In addition, the influence of environmental factors must also be considered when studying the working conditions of power devices. At this time, Simplorer and Q3D need to be used for joint simulation. However, when the physical structure of the package becomes complex, the calculation time of this method is long and the convergence is poor. In addition, the computational solution method usually cannot obtain special simulation tools and proprietary information of package design, so it is difficult to accurately deconstruct the parasitic parameters.
[0061] The experimental observation method mainly includes steady-state circuit measurement method and transient response method. On the one hand, the hardware steady-state circuit experiment has high requirements for experimental equipment. On the other hand, researchers have proposed equivalent circuit models for different types of power modules, but the analysis of these models is mainly limited to simulation verification of equivalent circuit models. There is a lack of experimental comparison verification, and the accuracy of the model is difficult to guarantee. The basic principle of the transient response method is to analyze the dynamic response characteristic curve of the output voltage or output current in the switching transient to obtain the parasitic inductance value. The key is the accurate sampling of the input and output signals in the hardware experiment. However, the transient response method requires that the sampling frequency of the voltage and current probes needs to meet 3 times the rise time of the measurement signal, otherwise a 5% measurement error will be generated; in the transient response method, some researchers use the differential method to analyze the parasitic parameter values of SiC-MOSFET power devices, but this method regards the inductor current change rate between the two MOSFETs as a linear change, and ignores the oscillation between the parasitic capacitance and parasitic inductance in the SiC-MOSFET half-bridge loop, so the parasitic parameter error obtained by this method is large; there are also studies that use the integral method to calculate the parasitic inductance of the power device, but do not fully consider the influence of interlayer contact and boundary conditions on the parasitic inductance, and also have a large error. Therefore, there is still a lack of accurate deconstruction methods for the specific values of MOSFET parasitic parameters in the existing technical solutions.
[0062] The present invention proposes a method for deconstructing parasitic parameters of a MOSFET power device and a computer device, aiming to solve the above-mentioned problem.
[0063] See also Figure 1 In one embodiment, a method for deconstructing parasitic parameters of a MOSFET power device comprises the following steps:
[0064] S1, obtaining a physical structure of a power device to be tested, and determining an equivalent circuit of the power device according to the physical structure;
[0065] S2, determining the theoretical transfer function of the input voltage and output voltage of the power device to be tested according to the equivalent circuit;
[0066] S3, obtaining measured data of input voltage and output voltage of the power device to be tested;
[0067] S4, calculating the actual transfer function of the power device to be tested by using the least square method according to the theoretical transfer function and the measured data;
[0068] S5, calculating and obtaining parasitic parameters of the power device to be tested according to the theoretical transfer function and the actual transfer function.
[0069] In this embodiment, the power device is a key component in the field of power electronics designed to handle high voltage and / or high current applications, including diodes, transistors (such as MOSFET, IGBT, GTO, etc.), thyristors, and emerging wide bandgap semiconductor devices (such as SiC-MOSFET and GaN HEMT); according to the physical structure characteristics of the specific power device to be tested, a reasonable equivalent circuit including various parasitic parameters is obtained.
[0070] In this embodiment, the transfer function is determined based on the topological structure and the working mode of the circuit. In a power electronic converter, the transfer function is usually used to describe the dynamic relationship between the output voltage and the input voltage.
[0071] In this embodiment, parasitic parameters generally include parasitic resistance, parasitic capacitance and parasitic inductance. Parasitic parameters are generally caused by the structure or circuit layout of the device itself, and sometimes by connecting wires or packaging. Parasitic resistance is the additional resistance in the actual circuit due to conductors, connection points, contact surfaces, etc., parasitic capacitance is the unexpected capacitance effect between two conductors or between a conductor and the ground, and parasitic inductance is the unexpected inductance effect caused by the arrangement of wires, joints or other components. For MOSFET, a single MOSFET can be regarded as consisting of a main MOS tube and a reverse parallel body diode. Considering the parasitic parameters, the main MOS tube can be equivalent to a switch with parasitic resistance, parasitic capacitance and parasitic inductance at the same time. The body diode also has parasitic parameters, but because the body diode of the MOSFET has a very high inflection point voltage, it will hardly conduct when connected in parallel with an external freewheeling diode. Therefore, the influence of the body diode can be ignored in the modeling process.
[0072] In this embodiment, after obtaining the theoretical transfer function model, the input voltage and output voltage data obtained by hardware experimental measurement are analyzed, and the required data matrix, parameter vector and output vector are constructed by the least squares method to determine the actual transfer function. The least squares method is a statistical method used to fit a data set to a specific mathematical model. The basic idea is to find a set of parameters that minimizes the sum of squared errors between the observed data and the model prediction value, so as to find the model parameters that are closest to the real data.
[0073] The embodiment of the present invention constructs a specific theoretical transfer function model of input voltage / output voltage through an equivalent circuit. It only needs to design a reasonable equivalent circuit including parasitic parameters to construct a theoretical function including parasitic parameters. Then, based on experimental measurement data, the actual transfer function is solved by the least squares method to achieve a smaller calculation error. Then, the specific values of the parasitic parameters can be further calculated by combining the theoretical transfer function and the actual transfer function.
[0074] Based on the above embodiments, SiC-MOSFET power devices are used as the power devices to be tested, and the technical solutions of the present invention are explained in detail through the following embodiments. Among many power device materials, silicon carbide (SiC) has attracted extensive attention from researchers because of its wide bandgap, high thermal conductivity, low on-resistance and other characteristics, and its performance is better than that of traditional silicon materials in high temperature environments; the parasitic parameters in SiC-MOSFET power devices mainly include the stray inductance and resistance of the wires, and the inter-electrode parasitic capacitance of the switch tube. These parameters directly affect the switching characteristics, power loss, control characteristics, service life, working mode and other performance of the device; systematic research and optimization of the parasitic parameters of SiC-MOSFET power devices have great practical significance for improving the efficiency and reliability of power electronic devices.
[0075] In some embodiments, see Figure 2 , Figure 2 is a power module circuit diagram and its n-order equivalent circuit diagram, wherein the power device to be tested includes a half-bridge circuit, the half-bridge circuit includes an upper tube and a lower tube, the upper tube and the lower tube are both MOSFETs, and in step S1, the equivalent circuit includes:
[0076] An RC circuit, comprising n RC branches, wherein the n RC branches are connected in parallel, wherein n is a positive integer;
[0077] Any RC branch includes an equivalent resistor and an equivalent capacitor, and the equivalent resistor and the equivalent capacitor are connected in series;
[0078] The equivalent resistances in the n RC branches are respectively denoted as R1, R2, ..., R n The equivalent capacitances in the n RC branches are respectively denoted as C1, C2, ..., C n .
[0079] The equivalent circuit also includes:
[0080] The first RL circuit includes the upper tube drain parasitic resistance R L1 and the upper tube drain parasitic inductance L1, the upper tube drain parasitic resistance R L1 Connected in series with the upper tube drain parasitic inductance L1;
[0081] The second RL circuit includes an equivalent series parasitic resistance R of the upper tube and the lower tube. L2 and the equivalent series parasitic inductance L2, the equivalent series parasitic resistance R L2 connected in series with the equivalent series parasitic inductance L2;
[0082] The first RL circuit is connected in series to one side of the RC circuit, and the second RL circuit is connected in series to the other side of the RC circuit.
[0083] In this embodiment, since the upper and lower tubes of the power half-bridge module (half-bridge circuit) are alternately turned on, and the drain-source parasitic capacitance can be regarded as a short circuit when turned on, the half-bridge module can be regarded as a simplified RLC circuit, that is, the drain-source parasitic capacitance of the SiC-MOSFET is equivalent to a structure of n RC series circuits connected in parallel with each other, and is respectively connected in series with the upper tube drain parasitic resistance and inductance, and the equivalent series parasitic resistance and inductance of the upper tube source and the lower tube drain; the input voltage of the half-bridge module is defined as the voltage u between the upper tube drain and the lower tube source DS The output voltage is the voltage u0 between the series output of the upper tube source and the lower tube drain and the lower tube source.
[0084] In this embodiment, the power half-bridge module is a commonly used power electronic module, which is usually composed of two power switching devices, which are respectively called an upper tube and a lower tube. They are connected in series to form a half-bridge structure, thereby realizing voltage control of the load.
[0085] In this embodiment, the variation of the drain-source parasitic capacitance of the MOSFET with frequency is relatively complex. The use of a multi-order RC model can better approximate the frequency characteristics of the actual capacitance, more accurately reflect the nonlinear effect of the capacitance, and thus more easily adjust the model parameters to match the actual measurement data.
[0086] Furthermore, in one embodiment, step S2 includes:
[0087] According to the equivalent circuit, a transfer function expression of the input voltage and the output voltage is obtained, and the transfer function expression is:
[0088]
[0089] In formula (1), u0 is the output voltage, u DS is the input voltage, Z c is the total impedance of the RC circuit, s is a complex variable; the total impedance Z of the RC circuit c It is expressed as:
[0090]
[0091] In formula (2), Z i is the impedance of the RC branch, expressed as:
[0092]
[0093] The transfer function expression is generalized to obtain the theoretical transfer function G circuit (s):
[0094]
[0095] In formula (4), (a1, ..., a n-1 、a n ) and (b1, …, b n 、b n+1 ) represent the coefficients of each term after the common denominator.
[0096] In this embodiment, the volt-ampere relationship of electronic components can usually be found in circuit analysis. The resistor is represented by U=RI, the inductor is represented by U=sLI, and the capacitor is represented by I=sCU, where s is the complex variable of the Laplace transform.
[0097] This embodiment describes the input voltage u in the equivalent circuit in a simple and intuitive way by establishing a theoretical transfer function model. DS The characteristics between the output voltage u0, in the subsequent steps, various coefficients (a1, ..., a n-1 , a n ) and (b1, …, b n , b n+1 ), the parasitic parameter values can be further obtained.
[0098] Furthermore, in one embodiment, step S3 includes:
[0099] Power on the power device to be tested and obtain the discrete waveforms of input voltage and output voltage through an oscilloscope, such as Figure 3 As shown, Figure 3 It is the waveform diagram of the measured input voltage and output voltage;
[0100] The voltage waveform of the discrete waveform diagram is smoothed by using a low-pass filter to obtain the measured data. Figure 4 and Figure 5 , Figure 4 is the waveform of the input voltage before and after filtering. Figure 5 This is the waveform of the output voltage before and after filtering.
[0101] In this embodiment, the low-pass filter is a Butterworth low-pass filter, which is used to convert the discrete data of the discrete waveform diagram into a continuous signal, thereby removing high-frequency noise and interference in the input voltage and the output voltage to obtain a smoother voltage waveform. The Butterworth low-pass filter is a classic analog filter design with a flat passband response and a monotonically decreasing stopband response. The order of the Butterworth low-pass filter determines the steepness of its transition band. The higher the order, the steeper the transition band. The general mathematical expression of the Butterworth low-pass filter used in this embodiment is as follows:
[0102]
[0103] In formula (5), z represents the complex frequency in the discrete time signal, and the coefficients (a1, a2, ..., a m ) and (b1, b2, …, b n ) can be calculated using the butter function in Matlab software.
[0104] When using the Butterworth low-pass filter to smooth the voltage waveform, the cutoff frequency is set to half of the sampling frequency and used as a normalized benchmark. The filtering effect is evaluated by comparing the waveform and spectrum before and after filtering and calculating the signal-to-noise ratio of the signal.
[0105] This embodiment obtains the measured input voltage and output voltage curves of the SiC-MOSFET power device to be tested through hardware experiments, and smoothes the waveforms through filtering, thereby providing effective data preparation for subsequent steps.
[0106] Furthermore, in one embodiment, step S4 includes:
[0107] Based on the theoretical transfer function, the value of n is selected, and step response tests are performed on the input voltage and the output voltage respectively to obtain voltage waveforms close to the measured data, and the corresponding voltage waveform data is recorded as analysis group data;
[0108] The actual transfer function is calculated using the least square method based on the analysis group data and the measured data.
[0109] In one achievable method, according to the theoretical transfer function model shown in equation (4), n=2 is selected, such as Figure 6 As shown, Figure 6 The corresponding power module circuit diagram and its second-order equivalent circuit diagram when n=2 are used to solve the actual transfer function of the power half-bridge module. The specific steps are as follows:
[0110] Substituting n=2 into equation (4), we obtain:
[0111]
[0112] From formula (4), we can see that a0=b0=1 in formula (6).
[0113] In this embodiment, for analysis and verification, two experimental groups are selected, namely, an analysis group obtained by simulating and analyzing the theoretical transfer function through Matlab software, and a verification group consisting of the measured data obtained in step S3; according to the transfer function shown in formula (6), the input voltage u is respectively calculated in Matlab software. DS The step response test of the output voltage u0 is performed, the sampling time interval is set to 0.4 nanoseconds, the total number of samples is set to 18000, and the output waveform is observed to obtain the input voltage curve and the corresponding output voltage curve closest to the measured data, which are recorded as the analysis group data. DS Waveform diagram Figure 7 As shown (the blue line represents the analysis group, and the red line represents the verification group), it can be found that the waveforms of the two groups are highly consistent, which proves the accuracy and reliability of the transfer function model proposed in this embodiment.
[0114] After obtaining the analysis group and verification group data of input voltage and output voltage, substitute them into the transfer function model of formula (6) and solve the actual transfer function of the SiC-MOSFET power module designed by this design scheme by the least squares method as shown below:
[0115]
[0116] After obtaining the actual transfer function, the input voltage data of the verification group is substituted into formula (7) to calculate the output voltage frequency characteristic curve (simulated output curve), and compared with the actual measured output voltage frequency characteristic curve (measured output curve), as shown in Figure 8 After calculation, the measured output curve of the verification group and the output voltage waveform calculated based on the actual transfer function have a fitting degree of 92.4%.
[0117] Furthermore, the fitting degree of the output voltage frequency characteristic curve and the measured output curve is calculated based on different orders (i.e., n values), as shown in Table 1. According to Table 1, it is feasible to construct the actual transfer function of the SiC-MOSFET power module using the least squares principle, and the proposed n-order SiC-MOSFET power module equivalent circuit can accurately reflect the characteristics of the SiC-MOSFET half-bridge circuit after considering parasitic parameters. The fitting degree of the output voltage calculated by the equivalent circuit model of each order and the actual output voltage exceeds 90%, among which the fitting degree is the highest when n=2.
[0118]
[0119] Table 1 Comparison of curve fitting of equivalent circuit models of different orders
[0120] Further, in one embodiment, step S5 includes:
[0121] According to the theoretical transfer function and the actual transfer function, the residual of the equation group is minimized by using the fmincon function to calculate the parasitic parameters of the power device to be tested.
[0122] Combining equation (6) and equation (7), we get equation (8) as follows:
[0123]
[0124] Among them, the theoretical transfer function G circuit The coefficient a0=b0=1 in the equation, and the other coefficients (a1, a2) and (b1, b2, b3) can be obtained from the derivation process of equations (1) to (4):
[0125]
[0126] In this embodiment, there are 8 parasitic parameters in the power module (R1, R2, R L1 , R L2 , C1, C2, L1, L2) need to be determined, but formula (9) only provides 5 equations, which is not enough to solve all the parameters.
[0127] To solve this problem, this embodiment sets additional conditions to assist in solving the problem. By referring to the data sheet of SiC-MOSFET and analyzing the characteristic curves of parasitic parameters, the possible range of each parameter can be proposed. Then, the fmincon function in Matlab is used to minimize the residual of the system of equations, that is, to make the solution of the system of equations as close to zero as possible; the fmincon function uses the Sequential Quadratic Programming (SQP) method, which is an iterative optimization algorithm that finds the minimum value of the objective function that satisfies the constraints in each iteration; after optimizing the problem, the following inequality (10) is obtained:
[0128]
[0129] In this embodiment, the fmincon function calculates the gradient and Hessian matrix of the objective function and constraints in each iteration, then uses numerical methods to find the optimal solution, and uses line search technology to determine the step size and direction of the iteration. The iteration stops when one of the following conditions is met: (1) the maximum number of iterations is reached; (2) the function value change is lower than the set threshold; (3) the parameter change is lower than the set threshold; (4) the gradient norm is lower than the set threshold. The fmincon function allows the gradient, Hessian matrix, and step size and direction of the objective function and constraints to be considered during the solution process, so as to find the global or local optimal solution; finally, the values of the various parasitic parameters of the SiC-MOSFET power module can be determined as follows:
[0130]
[0131] By comparing the parasitic parameter reference values at different temperatures provided in the SiC-MOSFET data sheet with the SiC-MOSFET power module parasitic parameters calculated in this embodiment, as shown in Table 2, it can be seen that the drain-source parasitic capacitance C DS The calculation error (calculated based on the equivalent capacitance values C1 and C2 of the second-order equivalent circuit) is only 3.3%. This error also needs to take into account the difference between the reference temperature provided in the data sheet and the experimental environment temperature. DS The calculated value is also within the reference value range. Similarly, the error of parasitic inductance is kept within 10%. The calculated parasitic resistance R DS , R L1 and R L2 It is consistent with the range of reference values; these results show that the parasitic parameter analysis method of SiC-MOSFET power devices proposed in this scheme is feasible and accurate.
[0132]
[0133] Table 2 Comparison of calculated parasitic parameter values with reference values
[0134] Furthermore, in one embodiment, after obtaining the specific values of the parasitic parameters of the SiC-MOSFET power device by calculation, the following steps are also included:
[0135] S6, building a simulation reproduction platform, constructing the equivalent circuit according to the values of the parasitic parameters, so as to obtain the output voltage curve of the equivalent circuit.
[0136] Specifically, a simulation reproduction platform was built using Simulink software, the values of various parasitic parameters were substituted to construct an equivalent circuit, the input port was connected to a 64V step response power supply, and the output voltage was observed on an oscilloscope.
[0137] In order to conduct comparative analysis, the measured output voltage waveform after filtering (actual output), the output voltage waveform obtained by system identification using the actual transfer function model step response (system identification output), and the reproduced circuit output voltage waveform (Simulink reproduced output) with various parasitic parameter values substituted are shown in a figure for comparison, as shown in the figure below: Fig. 9 As shown. Fig. 9 It can be seen that after the input step response, the actual output voltage of the half-bridge circuit based on the SiC-MOSFET power device rises from 0 to about 62.5V and remains stable, with a fluctuation value of no more than ±3% and a response time of 80ns; the actual transfer function model obtained through system identification also maintains stability at 62.5V when the input step response is increased, and its fluctuation range is smaller than that of the actual output voltage curve, and the response time is close to 80ns; the specific values of each parasitic parameter are obtained through the parasitic parameter analysis method of this scheme, and the output voltage curve obtained by substituting it into the Simulink equivalent circuit for simulation and reproduction can also gradually increase the voltage to the target value after the input step response and remain stable, and the fluctuation range is smaller than that of the actual output voltage curve, and the response time is almost equal to the actual response time. Therefore, this step verifies the accuracy and effectiveness of this scheme by comparing the voltage waveforms.
[0138] In the above embodiments, in the steps of respectively performing step response tests on the input voltage and the output voltage, and in the step of calculating the parasitic parameters of the power device to be tested using the fmincon function, the calculation and analysis software used is Matlab.
[0139] The present application also provides a computer device according to some embodiments. The computer device may include a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the steps of the deconstruction method of the parasitic parameters of the MOSFET power device provided in any of the above embodiments can be implemented. Therefore, the technical effects that can be achieved by the deconstruction method of the parasitic parameters of the MOSFET power device provided in the above embodiments can be achieved by the computer device, and will not be repeated here.
[0140] A person skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory.
[0141] It can be seen from the above embodiments that the method for deconstructing the parasitic parameters of a MOSFET power device proposed in the present invention establishes a reasonable n-order equivalent circuit of a SiC-MOSFET power device through the laws of physics, and constructs a specific theoretical transfer function model of input voltage / output voltage based on the equivalent circuit; based on the input voltage and output voltage waveforms actually measured by the hardware, and the input voltage and output voltage data simulated by the step response test in the software according to the theoretical transfer function model, the least squares method is used to construct the data matrix, parameter vector and output vector required for calculation to determine the actual transfer function; finally, the parasitic parameter value of the SiC-MOSFET power device is calculated by combining the theoretical transfer function and the actual transfer function with constraints, wherein the drain-source parasitic capacitance C DS The calculation error is only 3.3%, and the calculated values of the remaining parasitic parameters are also in line with the reference value range.
[0142] Compared with the existing technology, this solution can more accurately represent the characteristics of parasitic parameters in the actual circuit of SiC-MOSFET power devices, and effectively reduce the error of solving parasitic parameters based on measured data using the least squares method. This solution uses SiC-MOSFET power devices as an example to provide an effective and accurate method for deconstructing the parasitic parameters of power devices, and provides new ideas for parasitic parameter analysis and design of semiconductor materials in the new era, which has great practical significance for improving the efficiency and reliability of power electronic devices.
[0143] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.
Claims
1. A method for deconstructing parasitic parameters of a MOSFET power device, characterized in that: The following steps are involved: S1, obtaining a physical structure of a power device to be tested, and determining an equivalent circuit of the power device according to the physical structure; S2, determining the theoretical transfer function of the input voltage and output voltage of the power device to be tested according to the equivalent circuit; S3, obtaining measured data of input voltage and output voltage of the power device to be tested; S4, calculating the actual transfer function of the power device to be tested by using the least square method according to the theoretical transfer function and the measured data; S5, calculating and obtaining parasitic parameters of the power device to be tested according to the theoretical transfer function and the actual transfer function.
2. The deconstruction method according to claim 1, characterized in that: The power device to be tested includes a half-bridge circuit, the half-bridge circuit includes an upper tube and a lower tube, the upper tube and the lower tube are both MOSFETs, and in step S1, the equivalent circuit includes: An RC circuit, comprising n RC branches, wherein the n RC branches are connected in parallel, wherein n is a positive integer; Any RC branch includes an equivalent resistor and an equivalent capacitor, and the equivalent resistor and the equivalent capacitor are connected in series; The equivalent resistances in the n RC branches are respectively denoted as R1, R2, ..., R n The equivalent capacitances in the n RC branches are respectively denoted as C1, C2, ..., C n .
3. The deconstruction method according to claim 2, characterized in that: The equivalent circuit also includes: The first RL circuit includes the upper tube drain parasitic resistance R L1 and the upper tube drain parasitic inductance L1, the upper tube drain parasitic resistance R L1 Connected in series with the upper tube drain parasitic inductance L1; The second RL circuit includes an equivalent series parasitic resistance R of the upper tube and the lower tube. L2 and the equivalent series parasitic inductance L2, the equivalent series parasitic resistance R L2 connected in series with the equivalent series parasitic inductance L2; The first RL circuit is connected in series to one side of the RC circuit, and the second RL circuit is connected in series to the other side of the RC circuit.
4. The deconstruction method according to claim 3, characterized in that: The step S2 comprises: According to the equivalent circuit, a transfer function expression of the input voltage and the output voltage is obtained, and the transfer function expression is: Where u0 is the output voltage, u DS is the input voltage, Z c is the total impedance of the RC circuit, s is a complex variable; the total impedance Z of the RC circuit c It is expressed as: In the formula, Z i is the impedance of the RC branch, expressed as: The transfer function expression is generalized to obtain the theoretical transfer function G circuit (s): In the formula, a1, ..., a n-1 、a n and b1, …, b n 、b n+1 They represent the coefficients after the common denominator.
5. The deconstruction method according to claim 1, characterized in that: The step S3 comprises: Power on the power device to be tested and obtain discrete waveforms of input voltage and output voltage; The voltage waveform of the discrete waveform diagram is smoothed by using a low-pass filter to obtain the measured data.
6. The deconstruction method according to claim 5, characterized in that: The low-pass filter is a Butterworth low-pass filter, and the Butterworth low-pass filter is used to convert discrete data of the discrete waveform diagram into a continuous signal.
7. The deconstruction method according to claim 4, characterized in that: The step S4 comprises: Based on the theoretical transfer function, a step response test is performed on the input voltage and the output voltage respectively to obtain a voltage waveform close to the measured data, and the corresponding voltage waveform data is recorded as analysis group data; The actual transfer function is calculated using the least square method based on the analysis group data and the measured data.
8. The deconstruction method according to claim 7, characterized in that: The step S5 comprises: According to the theoretical transfer function and the actual transfer function, the residual of the equation group is minimized by using the fmincon function to calculate the parasitic parameters of the power device to be tested.
9. The deconstruction method according to claim 8, characterized in that: The following steps are also included: S6, building a simulation reproduction platform, constructing the equivalent circuit according to the values of the parasitic parameters, so as to obtain the output voltage curve of the equivalent circuit.
10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method for deconstructing parasitic parameters of a MOSFET power device according to any one of claims 1 to 9 are implemented.
Citation Information
Cited By
GaN HEMT device power driving parasitic parameter influence analysis system
CN119578336A
Semiconductor packaging structure health monitoring method and system based on embedded sensor
CN120177985A
Semiconductor packaging structure health monitoring method and system based on embedded sensors
CN120177985B
Electronic component testing system, method, equipment and medium
CN120629791A
