A method for establishing an IV model of a GaN HEMT power device

Abstract

The present invention relates to a method for establishing an I-V model of a GaN HEMT power device. Based on a MET model and an Angelov model, starting from the meaning of each parameter of the model itself, a direct current I-V forward conduction characteristic model and a reverse conduction characteristic model of a GaN HEMT are improved and established respectively, and each parameter in the model has its meaning; each parameter of the model can be directly fitted and extracted by using 1stOpt and matlab and measured I-V data, without extracting the initial value of the parameter at a specific voltage, so that the final obtained direct current I-V model of the GaN HEMT is more accurate.

Description

Technical Field

[0001] The present invention relates to the technical field of power electronics, and in particular to a method for establishing an IV model of a GaN HEMT power device. Background Art

[0002] As one of the III-V group and wide bandgap semiconductor materials, gallium nitride (GaN) has the characteristics of high critical breakdown electric field, high electron mobility, high heat capacity, high thermal conductivity, low resistivity, etc. These characteristics enable semiconductor power devices based on GaN materials to meet the application scenarios of high power, high frequency and high temperature in both microwave radio frequency field and power electronic power conversion field. Thanks to the superior performance of GaN materials, in power electronic power conversion applications, GaN HEMT (gallium nitride high electron mobility transistor) can achieve extremely fast switching transients and low losses with its high switching frequency and low on-resistance, and can far exceed the efficiency level of silicon-based converters at the same frequency. Accordingly, with the development of computer and electronic information technology, electronic design is closely related to EDA technology, and semiconductor device models have become the link between device physical properties and circuit design. Therefore, in order to promote the application and development of GaN HEMT, it is necessary to improve the efficiency of GaN HEMT modeling and the accuracy of the model. The main problems of the existing voltage-current (IV) model of GaN HEMT power electronic power devices include:

[0003] 1. The PSpice model of GaN HEMT based on the Shichman-Hodges model needs to divide the IV equation into five working areas in order to characterize the behavior of the device in the linear region, saturation region and cut-off region, which not only brings inconvenience to parameter extraction, but also easily leads to simulation non-convergence problems in practical applications;

[0004] 2. In order to improve the convergence of PSpice model simulation, a non-segmented PSpice model of GaN HEMT with easy-to-converge simulation can be established based on Motorola Electric Thermal model (MET) and Pade approximation, but the form of rational function is introduced in the output characteristics:

[0005]

[0006] Due to the existence of inherent poles of rational functions, the model may potentially be divided by vertical asymptotes. In addition, the meaning of the parameters in the model is not given, which will be detrimental to the subsequent adjustment and optimization of the model.

[0007] 3. The parameters of the GaN MIS-HEMT PSpice model established based on the commercial model provided by the GaN HEMT device manufacturer give corresponding physical meanings, but the parameters of the model need to first extract the initial values ​​at a specific voltage, and then undergo secondary optimization to establish a relatively accurate model. Summary of the invention

[0008] The invention discloses a method for establishing an IV model of a GaN HEMT power device, aiming to solve the technical problems existing in the prior art.

[0009] The present invention adopts the following technical solution: a method for establishing an IV model of a GaN HEMT power device, comprising the following steps:

[0010] Based on the MET model and Angelov model, the initial GaN HEMT DC IV forward conduction characteristic model is established;

[0011] Obtain the drain-source current I in the initial GaN HEMT DC IV forward conduction characteristic model under different parameters ds The value of each parameter determines the effect of the initial GaN HEMT DC IV forward conduction characteristic model on the drain-source current I ds The weight of the impact;

[0012] Measuring the drain-source current I in the forward conduction characteristics of GaN HEMT ds , drain-source voltage V ds and gate-source voltage V gs and compare it with the initial GaN HEMT DC IV forward conduction characteristic model to optimize the initial GaN HEMT DC IV forward conduction characteristic model;

[0013] Establish an initial GaN HEMT DC IV reverse conduction characteristic model;

[0014] Measuring the drain-source current I in the reverse conduction characteristics of GaN HEMT ds , drain-source voltage V ds and gate-source voltage V gs and compare it with the initial GaN HEMT DC IV reverse conduction characteristic model to optimize the initial GaN HEMT DC IV reverse conduction characteristic model;

[0015] According to the forward conduction characteristic model and the reverse conduction characteristic model, a non-segmented GaN HEMT DC IV model is obtained;

[0016] The optimal values ​​of each parameter in the DC IV model are determined by nonlinear fitting.

[0017] As a preferred technical solution, in the step of establishing an initial GaN HEMT DC IV forward conduction characteristic model based on the MET model and the Angelov model, it also includes:

[0018] The transfer characteristic equation is established based on the MET model:

[0019]

[0020] The output characteristic equation is established based on the equation characterizing the output characteristic in the Angelov IV model:

[0021] I output =f output (V ds )=tanh(aV ds )(1+λV ds ).

[0022] As a preferred technical solution, K1 is the saturation current coefficient, b1 is the threshold voltage, c1 is the modulation parameter related to transconductance, a represents the slope of the IV curve in the knee voltage region; λ is the channel length modulation parameter, representing the slope of the IV curve in the saturation region.

[0023] As a preferred technical solution, the transfer characteristic equation is multiplied by the output characteristic equation to construct the initial GaN HEMT DC IV forward conduction characteristic model:

[0024]

[0025] Among them, K 11 is the saturation current modulation coefficient, V th1 is the threshold voltage, a 10 is the knee voltage region modulation coefficient, λ 11 is the channel length modulation factor.

[0026] As a preferred technical solution, in the measurement of the forward conduction characteristics of the GaN HEMT, the drain-source current I ds , drain-source voltage V ds and gate-source voltage V gs The step of comparing the value of , and optimizing the initial GaN HEMT DC IV forward conduction characteristic model with the initial GaN HEMT DC IV forward conduction characteristic model also includes:

[0027] Comparing the initial DC IV forward conduction characteristic model with the forward conduction characteristic of GaN HEMT, the gate-source voltage V gs The related saturation current modulation coefficient K 12 , trap effect modulation coefficients α1 and β1, second-order channel length modulation coefficient λ12 , the initial DC IV forward conduction characteristic model is optimized, and the optimized model is:

[0028]

[0029] As a preferred technical solution, the step of establishing the initial GaN HEMT DC IV reverse conduction characteristic model further includes:

[0030] According to the optimized DC IV forward conduction characteristic model, the initial DC IV reverse conduction characteristic model is determined, and the initial model is:

[0031]

[0032] Among them, V sd =-V ds , I sd =-I ds , V gd =V gs -V ds ; K 21 is the saturation current modulation coefficient, K 22 V gs The related saturation current modulation factor, V th2 is the threshold voltage, c2 is the modulation coefficient related to transconductance, a 20 is the knee voltage modulation coefficient, α2 is the parameter related to the intermodulation between the transfer characteristic and the output characteristic, β2 is the trap effect modulation coefficient, λ 21 is the second-order channel length modulation coefficient, λ 22 is the second-order channel modulation parameter that characterizes the output characteristics.

[0033] As a preferred technical solution, the step of establishing the initial GaN HEMT DC IV reverse conduction characteristic model further includes:

[0034] According to the reverse conduction characteristics of GaN HEMT, the second-order channel modulation parameter λ characterizing the output characteristics in the initial DC IV reverse conduction characteristic model is removed. 22 , the parameter α2 related to the intermodulation between the transfer characteristic and the output characteristic, determines the simplified DC IV reverse conduction characteristic model, which is:

[0035]

[0036] As a preferred technical solution, in the measurement of the reverse conduction characteristics of the GaN HEMT, the drain-source current I ds , drain-source voltage V ds and gate-source voltage V gsThe step of comparing the value of the GaN HEMT DC IV reverse conduction characteristic model with the initial GaN HEMT DC IV reverse conduction characteristic model to optimize the initial GaN HEMT DC IV reverse conduction characteristic model also includes:

[0037] By comparing the simplified DC IV reverse conduction characteristic model with the reverse conduction characteristic of GaN HEMT, a second-order saturation current modulation compensation coefficient K is introduced. 23 and the trap effect modulation compensation coefficient β3, the simplified DC IV reverse conduction characteristic model is optimized, and the optimization model is:

[0038]

[0039] As a preferred technical solution, in the step of obtaining a non-segmented GaN HEMT DC IV model according to the forward conduction characteristic model and the reverse conduction characteristic model, the DC IV model of the non-segmented GaN HEMT is:

[0040]

[0041] As a preferred technical solution, in the step of determining the optimal value of each parameter in the DC IV model by nonlinear fitting, it also includes:

[0042] By using the global optimization algorithm in the nonlinear curve fitting and comprehensive optimization analysis and calculation software platform 1stOpt, the measured data are nonlinearly fitted with the forward conduction equation and the reverse conduction equation in the DC IV model of the non-segmented GaN HEMT to obtain the initial values ​​of each parameter;

[0043] Based on the initial values ​​of the parameters, the surface fitting tool in Matlab is used to perform nonlinear fitting on the measured data with the forward conduction equation and the reverse conduction equation in the DC IV model of the non-segmented GaN HEMT, respectively, to obtain the optimal values ​​of the parameters.

[0044] The technical solution adopted by the present invention can achieve the following beneficial effects:

[0045] The invention provides a method for establishing an IV model of a GaN HEMT power device. The method is based on a MET model and an Angelov model, and starting from the meaning of each parameter of the model itself, a direct current IV forward conduction characteristic model and a reverse conduction characteristic model of the GaN HEMT are improved and established respectively, and each parameter in the model has its meaning; each parameter of the model can be directly fitted and extracted by using 1stOpt and matlab and measured IV data, without extracting the initial value of the parameter at a specific voltage, so that the final GaN HEMT direct current IV model is more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments, which constitute a part of the present invention. The exemplary embodiments of the present invention and their descriptions explain the present invention and do not constitute improper limitations on the present invention. In the drawings:

[0047] Figure 1 A flowchart of a method for establishing an IV model of a GaN HEMT power device in a preferred embodiment of the present invention;

[0048] Figure 2 A flow chart of establishing a DC IV forward conduction characteristic model of GaN HEMT in a preferred embodiment of the present invention;

[0049] Figure 3 The parameters in formula (2) of the present invention are related to the drain-source current I ds The modulation effect of

[0050] Figure 4 The magnitude of each parameter in formula (3) of the present invention is related to the drain-source current I ds Specific modulation effects;

[0051] Figure 5 A comparison diagram of the model established by formula (4) of the present invention and the forward conduction characteristics of EPC 2010;

[0052] Figure 6 A comparison diagram of the model established by formula (5) of the present invention and the forward conduction characteristics of EPC 2010;

[0053] Figure 7 A comparison diagram of the model established by formula (6) of the present invention and the forward conduction characteristics of EPC 2010;

[0054] Figure 8 Flow chart of establishing a DC IV reverse conduction characteristic model of GaN HEMT in a preferred embodiment of the present invention

[0055] Fig. 9 A comparison diagram of the reverse conduction characteristics of the model established by formula (8) of the present invention and EPC 2010;

[0056] Fig.10 is the parameter K in formula (8) of the present invention 22 Schematic diagram of the effect of β2 on reverse IV characteristics;

[0057] Fig.11 This is a comparison diagram of the reverse conduction characteristics of the formula (9) of the present invention and EPC 2010. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the specific embodiments of the present invention and the corresponding drawings. In the description of the present invention, it should be noted that the term "or" is usually used in the sense of including "and / or", unless the content clearly indicates otherwise.

[0059] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected" and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be internal communication between two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance.

[0060] Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0061] The specific implementation steps and methods of the invention are described below in conjunction with the accompanying drawings.

[0062] refer to Figure 1 , Figure 1 FIG. 1 is a flow chart of a method for establishing an IV model of a GaN HEMT power device in a preferred embodiment of the present invention. Figure 1 As shown, the method comprises the following steps:

[0063] Step S100: establishing a DC IV forward conduction characteristic model of GaN HEMT based on the MET model and the Angelov model;

[0064] Step S200: establishing a DC IV reverse conduction characteristic model of GaN HEMT;

[0065] Step S300: obtaining a DC IV model of a non-segmented GaN HEMT according to a forward conduction characteristic model and a reverse conduction characteristic model;

[0066] Step S400: determining the optimal value of each parameter in the DC IV model by nonlinear fitting.

[0067] Specifically, before step S100, it is first necessary to select a suitable DC IV model. In a preferred embodiment, taking the EPC2010 (200V / 12A) GaN HEMT power device as an example, a non-segmented DC IV model is established:

[0068] I ds =f transfer (V gs )f output (V ds ); (1)

[0069] Due to the asymmetry of GaN HEMT, when V ds ≥0 and V ds When <0, its IV also shows asymmetry, so the DC IV forward conduction characteristics (V ds ≥0) and reverse conduction characteristics (V ds <0) are modeled separately to improve the overall accuracy of the model.

[0070] Considering that the DC IV characteristics of GaN HEMT mainly include transfer characteristics and output characteristics, preferably, step S100 further includes steps S101 to S104, such as Figure 2 :

[0071] Step S101: Selecting transfer characteristic and output characteristic equations;

[0072] Step S102: multiplying the transfer characteristic and the output characteristic equations to construct a continuous and unified initial DC IV model;

[0073] Step S103: Determine the effect of various parameters in the initial DC IV model on the drain-source current I ds The degree of influence;

[0074] Step S104: Optimizing the initial DC IV model.

[0075] Specifically, in step S101, equations are selected whose parameters have relatively clear meanings and can characterize the direct current IV transfer characteristics and output characteristics of the GaN HEMT.

[0076] The transfer characteristics are based on the equation modified by MET:

[0077]

[0078] The output characteristics are determined by the equation in the Angelov IV model that can characterize the output characteristics:

[0079] I output =f output (V ds)=tanh(aV ds )(1+λV ds ). (3)

[0080] Specifically, in step S102, the transfer characteristic equation and the output characteristic equation selected in step S101 are multiplied to construct an initial DC IV model:

[0081]

[0082] Preferably, in order to facilitate the construction of a unified forward and reverse IV model and the subsequent study of the significance of each parameter, some adjustments are made to the letter identifiers of the parameters from formula (2) to formula (4).

[0083] Specifically, in step S103: the role of the comprehensive formula (2) in the entire IV DC model can determine the meaning of each parameter in the equation. Figure 3 As shown in the figure, the parameters in formula (2) are related to the drain-source current I ds The figure shows the modulation effect. It can be seen from the figure that K1 is the saturation current coefficient, b1 is the threshold voltage and c1 is the modulation parameter related to transconductance. This formula can explicitly show the threshold voltage in the device transfer characteristics, which is crucial for extracting parameters in the model based on the transfer characteristic curve.

[0084] Similarly, the role of the comprehensive formula (3) in the entire IV DC model can determine the significance of each parameter in the equation. The specific magnitude of each parameter has an impact on the drain-source current I ds Specific modulation effects such as Figure 4 As shown, it can be seen that a can characterize the slope of the IV curve in the knee voltage region, and λ is the channel length modulation parameter, which characterizes the slope of the IV curve in the saturation region.

[0085] Specifically, in step S104, refer to Figure 5 , which shows the comparison results between the model established using formula (4) and the EPC 2010 forward static characteristics. The directly constructed IV DC equation can only achieve the accuracy of the GaN HEMT DC IV model constructed based on the Shichman-Hodges model. It is obviously insufficient in characterizing the nonlinear behavior of the transfer characteristics and output characteristics, and further optimization is needed.

[0086] from Figure 5 From the results in (b), the model can hardly accurately characterize the characteristics of each output characteristic curve in the saturation region. From the meaning of each parameter in formula (4), it can be seen that the parameters that affect and modulate the accuracy of the saturation region are mainly K 11 and 11 , so consider introducing the gate-source voltage V gs The related saturation current modulation coefficient K12 and the second-order channel length modulation parameter λ 12 , so formula (4) can be changed to:

[0087]

[0088] The comparison results with the EPC 2010 forward static characteristics are as follows: Figure 6 As shown in the figure, the model's ability to characterize the nonlinear behavior of EPCI-V is significantly improved. However, the accuracy of the output characteristic curve near the knee voltage at low gate-source voltage still needs to be improved. This is due to the leakage hysteresis effect (surface trap) and gate hysteresis effect (substrate trap) caused by the trap effect, which specifically manifests as the response V ds The increase in I ds The combined effect of these two trap effects is to cause the V gs and V ds Therefore, the modulation parameters α1 and β1 related to the trap effect are introduced into the modulation parameters related to the knee voltage in the output characteristic equation, so formula (5) can be further changed to:

[0089]

[0090] refer to Figure 7 After steps S101 to S104, the model finally established, that is, formula (6), can basically correctly characterize the forward IV characteristics of EPC 2010.

[0091] Preferably, in step S200, further steps S201 and S202 are included, referring to Figure 8 :

[0092] Step S201: determining an initial reverse DC IV model;

[0093] Specifically, when V ds When <0, GaN HEMT exhibits reverse conduction characteristics, that is, the drain-source current flows from the source to the drain, which is exactly the opposite of forward conduction. At this time, the conduction channel of GaN HEMT is caused by the gate-drain voltage V gd Therefore, the reverse characteristic equation can be obtained from the forward characteristic equation:

[0094]

[0095] Among them, V sd =-V ds , I sd =-I ds , V gd =V gs -Vds Since GaN HEMT always works in the linear region when reverse conducting, the second-order channel modulation parameter λ that characterizes the output characteristics can be considered 22 And the parameter α2 related to the intermodulation between the transfer characteristic and the output characteristic is removed to simplify the model. In this way, the reverse characteristic equation can be simplified to:

[0096]

[0097] Step S202: Optimizing the initial reverse DC IV model;

[0098] The reverse conduction characteristics of EPC 2010 can be directly characterized using formula (8). Fig. 9 As shown, from Fig. 9 It can be seen that as V gs and V sd The increase in model (8) I sd There is a clear contrast between the reverse conduction characteristics of EPC 2010. After analysis, it is found that as the parameter K 22 As well as the increase in the absolute value of β2, it will lead to I sd There is a sudden drop in the phenomenon, such as Fig.10 As shown, this is because when characterizing the reverse conductivity of the tube, the gate-source voltage V gs The related saturation current modulation coefficient K 22 And the trap effect modulation parameter β2 is negative, when V gs and V sd Increase to a certain extent will lead to I sd In order to avoid this error, the second-order saturation current modulation coefficient K is introduced. 23 And the trap effect modulation coefficient β3. In this way, the reverse conduction characteristics can be improved as follows:

[0099]

[0100] The results of using the modified model to characterize the reverse conduction characteristics of EPC 2010 are shown in Fig.11 shown.

[0101] Specifically, in step S300: according to the forward conduction characteristic model and the reverse conduction characteristic model, the DC IV model of the non-segmented GaN HEMT is obtained as follows:

[0102]

[0103] The meaning of each parameter is shown in Table 1.

[0104] Table 1. Meaning of parameters in formula (10)

[0105]

[0106]

[0107] Specifically, in step S400, the optimal value of each parameter in the DC IV model is extracted by a two-step method:

[0108] Step S401, using the global optimization algorithm in the nonlinear curve fitting and comprehensive optimization analysis and calculation software platform 1stOpt, the measurement data are nonlinearly fitted with the forward conduction equation and the reverse conduction equation in formula (10) to obtain the initial values ​​of each parameter;

[0109] Step S402, using the initial values ​​obtained in step S401 and the surface fitting tool in MATLAB, the measured data are again nonlinearly fitted with the forward conduction equation and the reverse conduction equation in formula (10) to obtain the optimal values ​​of each parameter.

[0110] The measured data used to extract the parameters of the forward conduction characteristic equation is Figure 7 The dotted part in the figure; for the extraction of reverse conduction characteristic equation parameters, if V gs =V0 (constant), according to the voltage relationship between the three ports of GaN HEMT gate, drain and source:

[0111] V gd =V gs +V sd =V0+V sd , (11)

[0112] When V sd <V th -V0, V can be obtained from formula (11) gd <V th , at this time the tube is in the cut-off state; similarly, when V sd >V th -V0, we can get V gd >V th , at this time the tube is in the on state. Therefore, the reverse-conducting GaN HEMT can be regarded as a GaN HEMT whose threshold voltage is affected by the gate-source voltage V gs The diode controlled by the threshold voltage is V th -V gs The diode has a terminal voltage of V sd Therefore, the IV model fitting data during reverse conduction can select the relationship between source-drain current and source-drain voltage under different gate-source voltage conditions. sd -V sd ,like Fig.11The dotted part in the figure. The model parameter values ​​extracted according to the “two-step method” are shown in Tables 2 and 3. The results between the model and the data are shown in Figure 7 and Fig.11 shown.

[0113] Table 2. Forward conduction characteristic parameters

[0114]

[0115] Table 3. Reverse conduction characteristic parameters

[0116]

[0117] Preferably, in order to measure the accuracy of the model before and after improvement, the mean square error (RMSE) and correlation coefficient (R) between formulas (4)-(6) and the EPC 2010 forward conduction characteristic data are calculated. 2 ) and the RMSE and R between formulas (7)-(9) and the EPC 2010 reverse conduction characteristic data respectively. 2 The results are shown in Table 4 and Table 5. From Table 4, it can be seen that by introducing the gate-source voltage V gs The related saturation current modulation coefficient K 12 and the second-order channel length modulation parameter λ 12 Extending the forward conduction characteristic equation to formula (5) can greatly improve the model's ability to characterize the forward conduction characteristics of EPC 2010, and it is effective to introduce modulation parameters α1 and β1 related to characterizing the trap effect on the basis of formula (5). Similarly, it can be seen from Table 5 that the second-order saturation current modulation coefficient K 23 Introducing the trap effect modulation coefficient β3 into the reverse conduction characteristic equation can effectively solve the problem caused by directly using the forward conduction characteristic equation to characterize the reverse conduction characteristic. In summary, the DC IV model establishment and optimization method proposed in the present invention can establish an accurate GaN HEMT DC IV model.

[0118] Table 4. RMSE and R between formulas (4)-(6) and EPC 2010 forward conduction characteristic data 2 Comparison results

[0119]

[0120] Table 5. RMSE and R between formulas (7)-(9) and EPC 2010 reverse conduction characteristic data 2 Comparison results

[0121]

[0122] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation modes, which are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the present invention and the claims, all of which are within the protection of the present invention.

Claims

1. A method for establishing an IV model of a GaN HEMT power device, characterized in that: include: Based on the MET model and Angelov model, the initial GaN HEMT DC IV forward conduction characteristic model is established; Obtain the drain-source current I in the initial GaN HEMT DC IV forward conduction characteristic model under different parameters ds The value of each parameter determines the effect of the initial GaN HEMT DC IV forward conduction characteristic model on the drain-source current I ds The weight of the impact; Measuring the Drain-Source Current I in the Forward Conducting Characteristics of GaN HEMT ds , drain-source voltage V ds and gate-source voltage V gs and compare it with the initial GaN HEMT DC IV forward conduction characteristic model to optimize the initial GaN HEMT DC IV forward conduction characteristic model; Establish an initial GaN HEMT DC IV reverse conduction characteristic model; Measuring the drain-source current I in the reverse conduction characteristics of GaN HEMT ds , drain-source voltage V ds and gate-source voltage V gs and compare it with the initial GaN HEMT DC IV reverse conduction characteristic model to optimize the initial GaN HEMT DC IV reverse conduction characteristic model; According to the forward conduction characteristic model and the reverse conduction characteristic model, a non-segmented GaN HEMT DC IV model is obtained; Determine the optimal values ​​of each parameter in the DC IV model through nonlinear fitting; Specifically, the step of establishing an initial GaN HEMT DC IV forward conduction characteristic model based on the MET model and the Angelov model also includes: The transfer characteristic equation is established based on the MET model: The output characteristic equation is established based on the equation characterizing the output characteristic in the Angelov IV model: I output =f output (V ds )=tanh(aV ds )(1+λV ds ); Wherein, K1 is the saturation current coefficient, b1 is the threshold voltage, c1 is the modulation parameter related to transconductance, a represents the slope of the IV curve in the knee voltage region; λ is the channel length modulation parameter, representing the slope of the IV curve in the saturation region; The transfer characteristic equation is multiplied by the output characteristic equation to construct the initial GaN HEMT DC IV forward conduction characteristic model: Among them, K 11 is the saturation current modulation coefficient, V th1 is the threshold voltage, a 10 is the knee voltage region modulation coefficient, λ 11 is the channel length modulation coefficient; Specifically, in the measurement of the forward conduction characteristics of the GaN HEMT, the drain-source current I ds , drain-source voltage V ds and gate-source voltage V gs The step of comparing the value of , and optimizing the initial GaN HEMT DC IV forward conduction characteristic model with the initial GaN HEMT DC IV forward conduction characteristic model also includes: Comparing the initial DC IV forward conduction characteristic model with the forward conduction characteristic of GaN HEMT, the gate-source voltage V gs The related saturation current modulation coefficient K 12 , trap effect modulation coefficients α1 and β1, second-order channel length modulation coefficient λ 12 , the initial DC IV forward conduction characteristic model is optimized, and the optimized model is:

2. The method for establishing an IV model of a GaN HEMT power device according to claim 1, characterized in that: The step of establishing the initial GaN HEMT DC IV reverse conduction characteristic model also includes: According to the optimized DC IV forward conduction characteristic model, the initial DC IV reverse conduction characteristic model is determined, and the initial model is: Among them, V sd =-V ds , I sd =-I ds , V gd =V gs -V ds ; K 21 is the saturation current modulation coefficient, K 22 V gs The related saturation current modulation factor, V th2 is the threshold voltage, c2 is the modulation coefficient related to transconductance, a 20 is the knee voltage modulation coefficient, α2 is the parameter related to the intermodulation between the transfer characteristic and the output characteristic, β2 is the trap effect modulation coefficient, λ 21 is the second-order channel length modulation coefficient, λ 22 is the second-order channel modulation parameter that characterizes the output characteristics.

3. The method for establishing an IV model of a GaN HEMT power device according to claim 2, characterized in that: The step of establishing the initial GaN HEMT DC IV reverse conduction characteristic model also includes: According to the reverse conduction characteristics of GaN HEMT, the second-order channel modulation parameter λ characterizing the output characteristics in the initial DC IV reverse conduction characteristic model is removed. 22 , the parameter α2 related to the intermodulation between the transfer characteristic and the output characteristic, determines the simplified DC IV reverse conduction characteristic model, the simplified model is:

4. The method for establishing an IV model of a GaN HEMT power device according to claim 3, characterized in that: In the measurement of GaN HEMT reverse conduction characteristics, the drain-source current I ds , drain-source voltage V ds and gate-source voltage V gs The step of comparing the value of the GaN HEMT DC IV reverse conduction characteristic model with the initial GaN HEMT DC IV reverse conduction characteristic model to optimize the initial GaN HEMT DC IV reverse conduction characteristic model also includes: By comparing the simplified DC IV reverse conduction characteristic model with the reverse conduction characteristic of GaN HEMT, a second-order saturation current modulation compensation coefficient K is introduced. 23 and the trap effect modulation compensation coefficient β3, the simplified DC IV reverse conduction characteristic model is optimized, and the optimized model is:

5. The method for establishing an IV model of a GaN HEMT power device according to claim 4, characterized in that: In the step of obtaining a non-segmented GaN HEMT DC IV model according to the forward conduction characteristic model and the reverse conduction characteristic model, the DC IV model of the non-segmented GaN HEMT is:

6. The method for establishing an IV model of a GaN HEMT power device according to claim 3, characterized in that: In the step of determining the optimal value of each parameter in the DC IV model by nonlinear fitting, it also includes: Through the global optimization algorithm in the nonlinear curve fitting and comprehensive optimization analysis and calculation software platform 1stOpt, the measured data are nonlinearly fitted with the forward conduction equation and the reverse conduction equation in the DC IV model of the non-segmented GaN HEMT to obtain the initial values ​​of each parameter; Based on the initial values ​​of the parameters, the surface fitting tool in Matlab is used to perform nonlinear fitting on the measured data with the forward conduction equation and the reverse conduction equation in the DC IV model of the non-segmented GaN HEMT, respectively, to obtain the optimal values ​​of the parameters.

Images