Transient thermal simulation method for gallium nitride high electron mobility transistor

Abstract

The invention provides a transient thermal simulation method for a gallium nitride high-electron-mobility transistor, and the method comprises the steps: decomposing a whole object into a variable region and an invariable region according to the temperature change characteristic of a material of the gallium nitride high-electron-mobility transistor; carrying out transient heat transfer characteristic extraction on the invariant region, and establishing a corresponding equivalent thermal conductivity network model; converting the equivalent thermal conductivity network model into an equivalent thermal boundary condition of a change area, and calculating temperature distribution of the change area; and according to the temperature distribution of the change area, constructing an equivalent thermal boundary condition of the invariable area, and calculating the temperature distribution of the invariable area. According to the method, on the premise of ensuring the original calculation precision, the nonlinear iterative operation caused by the temperature change characteristic of the material is limited in a change area, the simulation efficiency is remarkably improved, and the consumption of calculation resources is reduced.

Description

Technical Field

[0001] The present invention relates to the field of numerical simulation computing technology, and in particular to a transient thermal simulation method for a gallium nitride high electron mobility transistor. Background Art

[0002] With the rapid increase in demand for high-power and high-frequency devices in RF applications, GaN high-electron-mobility transistors have become a popular device due to their excellent material properties. However, the self-heating effect caused by high power density may lead to device performance degradation and reliability issues. Therefore, in order to fully tap the full potential of GaN high-electron-mobility transistors and mitigate the impact of thermal failure, thermal analysis and thermal management need to be fully considered in the design process.

[0003] Numerical methods have been widely used to simulate the transient thermal behavior of GaN high electron mobility transistors. However, GaN devices have significant material temperature variation characteristics. Existing technical solutions usually need to use nonlinear iterations to obtain the temperature distribution of the entire device structure at each moment during transient thermal simulation, resulting in low computational efficiency and high consumption of computing resources. Summary of the invention

[0004] In order to solve the defects of the prior art, the purpose of the present invention is to provide a transient thermal simulation method for a gallium nitride high electron mobility transistor. Under the premise of ensuring the original calculation accuracy, the present invention limits the nonlinear iterative calculation caused by the temperature change characteristics of the material to the change area, which significantly improves the simulation efficiency and reduces the consumption of computing resources.

[0005] To solve the above problems, the technical solution of the present invention is:

[0006] A transient thermal simulation method for a gallium nitride high electron mobility transistor comprises the following steps:

[0007] According to the material temperature variation characteristics of GaN high electron mobility transistor, the whole object is decomposed into a variable area and an unchanged area;

[0008] Extract transient heat transfer characteristics of the unchanged area and establish the corresponding equivalent thermal conductivity network model;

[0009] The equivalent thermal conductivity network model is converted into the equivalent thermal boundary conditions of the changing area, and the temperature distribution of the changing area is calculated;

[0010] According to the temperature distribution in the changing area, the equivalent thermal boundary conditions in the constant area are constructed, and the temperature distribution in the constant area is calculated.

[0011] Preferably, in the step of decomposing the overall object into a variable region and an invariant region according to the material temperature variation characteristics of the GaN high electron mobility transistor, the variable region is a GaN device region whose material parameters vary with temperature, and the invariant region is a GaN device region whose material parameters do not vary with temperature.

[0012] Preferably, in the step of extracting transient heat transfer characteristics of the invariant region and establishing a corresponding equivalent thermal conductivity network model, the step of extracting the equivalent thermal conductivity network model includes:

[0013] In the first step, the matrix equation of transient heat conduction problem in the invariant region is established by using the finite volume method and the first-order backward difference formula;

[0014] The second step is to set the temperature of the interface between the changing region and the unchanged region at the current moment to zero, and set the body temperature distribution of the unchanged region to zero. According to the matrix equation obtained in the first step, the body temperature distribution of the unchanged region at the next moment is solved, and the heat flux density on the interface is further calculated, which is recorded as the reference heat flux density of the interface.

[0015] The third step is to apply uniform temperature rise excitation to each surface unit of the interface in turn, while keeping the temperature of other surface units at zero. According to the matrix equation obtained in the first step, the heat flux density at the next moment is solved, and the heat flux density matrix composed of the selected temperature rise excitation size and the corresponding interface heat flux density is recorded;

[0016] Step 4: Calculate the equivalent thermal conductivity matrix based on the interface reference heat flux obtained in step 2 and the temperature rise excitation magnitude and heat flux matrix recorded in step 3.

[0017] The fifth step is to generate an equivalent thermal conductivity network model based on the equivalent thermal conductivity matrix obtained in the fourth step.

[0018] Preferably, in the step of converting the equivalent thermal conductivity network model into the equivalent thermal boundary conditions of the changing region and calculating the temperature distribution of the changing region, the steps of constructing the equivalent thermal boundary conditions of the changing region and calculating the temperature distribution in the changing region include:

[0019] The first step is to set the current temperature of the interface to zero, and set the source term of the boundary condition and the heat source in the constant region to zero, and calculate the heat flux density of the interface in the constant region at the next moment;

[0020] The second step is to construct the equivalent thermal boundary conditions on the interface of the changing area according to the interface heat flux density and equivalent thermal conductivity network model obtained in the first step;

[0021] In the third step, based on the equivalent thermal boundary conditions constructed in the second step, a nonlinear iterative algorithm is used to solve the temperature distribution of the changing area at the next moment.

[0022] Preferably, in the step of constructing the equivalent thermal boundary conditions of the constant region according to the temperature distribution of the changing region and calculating the temperature distribution of the constant region, the steps of constructing the equivalent thermal boundary conditions of the constant region and calculating the temperature distribution in the constant region include:

[0023] In the first step, according to the body temperature distribution in the change area, the temperature value of the center node of the body unit adjacent to the interface is screened out and recorded as the adjacent body center temperature;

[0024] The second step is to construct the relationship between the interface temperature and the adjacent body center temperature based on the equivalent thermal boundary conditions of the interface of the change region and the interpolation method in the finite volume method.

[0025] The third step is to calculate the interface temperature based on the relationship between the adjacent body center temperature obtained in the first step and the interface temperature constructed in the second step and the adjacent body center temperature, and use it as the constant temperature boundary condition of the interface in the invariant region to solve the temperature distribution in the invariant region at the next moment.

[0026] Compared with the prior art, the present invention has the following beneficial effects:

[0027] 1. The present invention decomposes the GaN high electron mobility transistor into a variable region and an unchanged region, extracts the equivalent thermal conductivity network model of the unchanged region as the equivalent boundary condition of the variable region, and limits the nonlinear iterative operation caused by the temperature variation characteristics of the material to the variable region, thereby significantly reducing the dimension of the nonlinear equation group, improving the simulation efficiency, and reducing the consumption of computing resources;

[0028] 2. When the invariant region of the GaN-HEMT contains multiple independent sub-regions, its equivalent thermal conductivity network model can be extracted in parallel and its transient response can be solved in parallel, thereby further improving the simulation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0030] Figure 1 It is a flowchart of a transient thermal simulation method for a gallium nitride high electron mobility transistor according to the present invention;

[0031] Figure 2 This is a schematic diagram of the equivalent thermal conductivity network model proposed by the present invention;

[0032] Figure 3 A schematic diagram of the geometric structure of a gallium nitride high electron mobility transistor in a specific embodiment;

[0033] Figure 4is a transient temperature response curve diagram of an observation point in a specific embodiment;

[0034] Figure 5 It is a temperature distribution curve diagram on the vertical center line of the entire structure in a specific embodiment. DETAILED DESCRIPTION

[0035] The present invention is described in detail below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those of ordinary skill in the art, several changes and improvements can also be made without departing from the concept of the present invention. These all belong to the protection scope of the present invention.

[0036] Specifically, the present invention provides a transient thermal simulation method for a gallium nitride high electron mobility transistor, such as Figure 1 As shown, the method comprises the following steps:

[0037] S1: According to the material temperature variation characteristics of GaN high electron mobility transistor, the whole object is decomposed into a variable area and a constant area;

[0038] Specifically, when performing thermal simulation of GaN high electron mobility transistors, the overall object is decomposed into a variable region and an invariant region according to the simulation accuracy requirements and material properties. The variable region is the GaN device region where material parameters change with temperature, and the invariant region is the GaN device region where material parameters do not change with temperature.

[0039] S2: Extract the transient heat transfer characteristics of the unchanged area and establish the corresponding equivalent thermal conductivity network model;

[0040] Specifically, in step S2, the steps of extracting the equivalent thermal conductivity network model are as follows:

[0041] Step S21: The governing equation for the transient heat conduction problem in the constant region is:

[0042]

[0043] Where ρ is the mass density, c p is the heat capacity, T is the temperature, κ is the thermal conductivity, and Q is the heat source.

[0044] Furthermore, the spatial discretization is performed based on the finite volume method, and the time discretization is performed using the first-order backward difference formula (time interval is Δt), and the matrix equation of the transient heat conduction problem in the invariant region is established:

[0045]

[0046] Where K is the system matrix, M is the mass matrix, is the temperature distribution at the current time ((n-1)Δt) in the constant region, is the temperature distribution at the next moment (nΔt) in the constant region, Q n Represents the contribution of heat sources and boundaries to the matrix equation.

[0047] Step S22: Set the temperature of the interface between the changing region and the unchanged region at time (n-1)Δt to zero, and set the body temperature distribution of the unchanged region to Set to zero, and solve the body temperature distribution at the constant region nΔt according to the matrix equation (2) obtained in step S21. Based on the obtained body temperature distribution, combined with the zero temperature distribution on the interface, the heat flux density on the interface is further calculated and recorded as the reference heat flux density q 0 .

[0048] Step S23: Assuming that there are N surface units on the interface after spatial discretization, uniform temperature rise excitation is applied to the N surface units of the interface in turn, while the temperature of other surface units is kept at zero. According to the matrix equation (2) obtained in step S21, the heat flux density at time nΔt is solved, and the heat flux density matrix composed of the selected temperature rise excitation magnitude and the corresponding interface heat flux density is recorded. The temperature rise magnitude applied to the jth surface unit is recorded as ΔT j , the finite volume method is used to solve the vector formed by the heat flux density of the central nodes of N surface units on the interface, denoted as q j .

[0049] Step S24: Calculate the equivalent thermal conductivity matrix according to the interface reference heat flux distribution obtained in step S22 and the temperature rise excitation magnitude and heat flux matrix recorded in step S23; that is, calculate the equivalent thermal conductivity between the jth surface unit and the ith surface unit:

[0050]

[0051] Repeat the above process to construct an N×N dimensional equivalent thermal conductivity matrix G, and then obtain the temperature distribution T on the interface in the transient response of the invariant region with zero temperature at time (n-1)Δt. n Heat flux distribution The relationship between is shown in formula (4):

[0052]

[0053] Step S25: Establish an equivalent thermal conductivity network model based on the equivalent thermal conductivity matrix obtained in step S24. The local circuit structure of the equivalent thermal conductivity network model is as follows: Figure 2 shown.

[0054] S3: convert the equivalent thermal conductivity network model into equivalent thermal boundary conditions of the change area and calculate the temperature distribution of the change area;

[0055] Specifically, the steps of constructing the equivalent thermal boundary conditions of the changing region and calculating the temperature distribution in the changing region in step S3 are as follows:

[0056] Step S31: Set the temperature of the interface at time (n-1)Δt to zero, and set the source terms of the boundary conditions and the heat source in the constant region to zero, such as setting the constant temperature boundary condition to 0K and the heat flux density boundary condition to 0W / m 2 , the ambient temperature of the convection boundary condition is set to 0K, and the heat flux density at the interface of the invariant region at time nΔt is calculated

[0057] Step S32: The interface heat flux density obtained in step S31 Combined with the equivalent thermal conductivity network model and the superposition theorem, the invariant linear relationship between the heat flux density and the temperature at the interface of the changing region is constructed, that is, the equivalent thermal boundary condition, as shown in formula (5):

[0058]

[0059] Step S33: Based on the equivalent thermal boundary condition constructed in step S32, it is regarded as a convection boundary condition on the surface of the change area, and a nonlinear iterative algorithm is used to solve the temperature distribution of the change area at time nΔt.

[0060] S4: According to the temperature distribution of the changing area, construct the equivalent thermal boundary conditions of the unchanged area and calculate the temperature distribution of the unchanged area.

[0061] Specifically, step S4 includes the following steps:

[0062] Step S41: According to the body temperature distribution in the change area obtained in step S3, the center node temperature values ​​of N individual units adjacent to the interface are screened out and recorded as the adjacent body center temperature

[0063] Step S42: Assuming that there is a linear temperature distribution between the body center node and the unit center node on the interface, the heat flux density on the interface can be calculated by formula (6):

[0064]

[0065] where κ i is the thermal conductivity of the ith surface unit on the interface, L i is the distance from the center of the ith face element to the center of the adjacent volume element.

[0066] Combining formula (5) and formula (6), the interface temperature distribution T is constructed n Temperature distribution of adjacent body core The relationship between is shown in formula (7):

[0067]

[0068] Where D is a diagonal matrix, as shown in formula (8):

[0069]

[0070] Step S43: Adjacent body core temperature distribution obtained in step S41 The relationship between the interface temperature distribution constructed in step S42 and the adjacent body core temperature distribution is used to calculate the interface temperature distribution T n , which is used as the constant temperature boundary condition of the interface of the invariant region, and the finite volume method is used to solve the temperature distribution of the invariant region at the time nΔt

[0071] According to the simulation method of the present invention described above, a specific embodiment is calculated.

[0072] Below is a diagram of the Figure 3 The implementation of the present invention is further described in detail by taking the transient thermal simulation of a single-finger GaN high electron mobility transistor as an example. Obviously, the described embodiment is a part of the 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.

[0073] This embodiment is a typical single-finger GaN high electron mobility transistor, whose geometric structure size and material configuration are as follows: Figure 3 As shown. Where the gate width W g =2μm, length L g =150μm. In the thermal simulation of this embodiment, only the temperature variation characteristics of the gallium nitride material are considered, and other material parameters are set to constants that do not change with temperature. According to the above material characteristics, the entire solution domain is divided into a variable region and an invariant region: the variable region is the gallium nitride device layer; the invariant region includes the substrate and the bonding layer. Based on the equivalent thermal conductivity network model, this embodiment limits the nonlinear iterative operation to the variable region to achieve efficient and accurate transient thermal simulation.

[0074] In this embodiment, the gate surface is subjected to 5000W / mm 2 The base surface is kept at a constant temperature boundary of 300K to represent the role of heat sink, and the other outer surface boundaries are set as adiabatic boundaries.

[0075] To verify the effectiveness of the method of the present invention, a time step of 0.1 ns was selected and 1000 time steps were performed. The transient thermal simulation was performed using the method of the present invention and the standard finite volume method. The center of the GaN device layer was set as the observation point, and its temperature transient response was as follows: Figure 4 As shown. It can be found that the calculation results of the method of the present invention are very consistent with those of the traditional finite volume method, which fully verifies the effectiveness of the method of the present invention. The temperature distribution curve of the vertical center line of the structure at the time of 100ns is shown in Figure 5 It can be clearly seen that the calculation results of the two methods fit well, which further reflects the accuracy of the method of the present invention. In order to reflect the efficiency of the method of the present invention, the total time of transient thermal simulation was counted. The method of the present invention took 6m, while the traditional finite volume method took 1h 23m 42s, achieving a 14-fold acceleration effect.

[0076] The above describes the specific embodiments of the present invention. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which does not affect the essence of the present invention. In the absence of conflict, the embodiments of the present application and the features in the embodiments can be combined with each other arbitrarily.

Claims

1. A transient thermal simulation method for a gallium nitride high electron mobility transistor, characterized in that: The method comprises the following steps: According to the material temperature variation characteristics of GaN high electron mobility transistor, the whole object is decomposed into a variable area and an unchanged area; Extract transient heat transfer characteristics of the unchanged area and establish the corresponding equivalent thermal conductivity network model; The equivalent thermal conductivity network model is converted into the equivalent thermal boundary conditions of the changing area, and the temperature distribution of the changing area is calculated; According to the temperature distribution in the changing area, the equivalent thermal boundary conditions in the constant area are constructed, and the temperature distribution in the constant area is calculated.

2. The transient thermal simulation method for a gallium nitride high electron mobility transistor according to claim 1, characterized in that: In the step of decomposing the overall object into a variable region and an invariant region according to the material temperature variation characteristics of the GaN high electron mobility transistor, the variable region is a GaN device region where material parameters vary with temperature, and the invariant region is a GaN device region where material parameters do not vary with temperature.

3. The transient thermal simulation method for a gallium nitride high electron mobility transistor according to claim 1, characterized in that: In the step of extracting transient heat transfer characteristics of the unchanged area and establishing a corresponding equivalent thermal conductivity network model, the step of extracting the equivalent thermal conductivity network model includes: In the first step, the matrix equation of transient heat conduction problem in the invariant region is established by using the finite volume method and the first-order backward difference formula; The second step is to set the temperature of the interface between the changing region and the unchanged region at the current moment to zero, and set the body temperature distribution of the unchanged region to zero. According to the matrix equation obtained in the first step, the body temperature distribution of the unchanged region at the next moment is solved, and the heat flux density on the interface is further calculated, which is recorded as the reference heat flux density of the interface. The third step is to apply uniform temperature rise excitation to each surface unit of the interface in turn, while keeping the temperature of other surface units at zero. According to the matrix equation obtained in the first step, the heat flux density at the next moment is solved, and the heat flux density matrix composed of the selected temperature rise excitation size and the corresponding interface heat flux density is recorded; The fourth step is to calculate the equivalent thermal conductivity matrix based on the interface reference heat flux obtained in the second step and the temperature rise excitation magnitude and heat flux matrix recorded in the third step; The fifth step is to generate an equivalent thermal conductivity network model based on the equivalent thermal conductivity matrix obtained in the fourth step.

4. The transient thermal simulation method for a gallium nitride high electron mobility transistor according to claim 1, characterized in that: In the step of converting the equivalent thermal conductivity network model into the equivalent thermal boundary conditions of the changing region and calculating the temperature distribution of the changing region, the steps of constructing the equivalent thermal boundary conditions of the changing region and calculating the temperature distribution in the changing region include: The first step is to set the current temperature of the interface to zero, and set the source term of the boundary condition and the heat source in the constant region to zero, and calculate the heat flux density of the interface in the constant region at the next moment; The second step is to construct the equivalent thermal boundary conditions on the interface of the changing area according to the interface heat flux density and equivalent thermal conductivity network model obtained in the first step; In the third step, based on the equivalent thermal boundary conditions constructed in the second step, a nonlinear iterative algorithm is used to solve the temperature distribution of the changing area at the next moment.

5. The transient thermal simulation method for a gallium nitride high electron mobility transistor according to claim 1, characterized in that: In the step of constructing the equivalent thermal boundary conditions of the constant region according to the temperature distribution of the changing region and calculating the temperature distribution of the constant region, the steps of constructing the equivalent thermal boundary conditions of the constant region and calculating the temperature distribution in the constant region include: In the first step, according to the body temperature distribution in the change area, the temperature value of the center node of the body unit adjacent to the interface is screened out and recorded as the adjacent body center temperature; The second step is to construct the relationship between the interface temperature and the adjacent body center temperature based on the equivalent thermal boundary conditions of the interface of the change region and the interpolation method in the finite volume method. The third step is to calculate the interface temperature based on the relationship between the adjacent body center temperature obtained in the first step and the interface temperature constructed in the second step and the adjacent body center temperature, and use it as the constant temperature boundary condition of the interface in the invariant region to solve the temperature distribution in the invariant region at the next moment.

Images