A thermal circuit simulation method with non-uniform time step and related equipment

Abstract

The present invention proposes a simulation method and related equipment with non-uniform time steps, which can quickly realize the convolution calculation of the temperature T(t) of the observation point, greatly improving the simulation rate. A thermal circuit simulation method with non-uniform time steps proposed by the present invention includes: step S1, importing a simulation model, setting thermal simulation parameters, performing thermal circuit extraction, and obtaining a thermal circuit response function H(s). Wherein, setting the thermal simulation parameters includes setting the observation point. Step S2, setting the heat source power curve P(t), the heat source power curve P(t) is a power curve that defines the change of the heat source over time, and the heat source power curve P(t) is set according to the use scenario of the heat source. Step S3, setting the output time series of the non-uniform step. Step S4, performing thermal transient simulation according to the thermal circuit response function H(s), the heat source power curve P(t) and the output time series of the non-uniform step. Step S5, based on the thermal transient simulation results, checking the temperature response curve of the observation point.

Description

Technical Field

[0001] The invention relates to the field of thermal simulation, and in particular to a thermal circuit simulation method with a non-uniform time step and related equipment. Background Art

[0002] Thermal circuits are a set of system functions with circuit response characteristics extracted through simulation models, which reflect the direct mapping relationship between observation points and heat sources. Through the extraction of thermal circuits, the function of one-time extraction and multiple simulations can be realized.

[0003] Thermal circuit simulation is a technology that uses computer simulation to quickly calculate the temperature response of an observation point through a model's thermal circuit and input heat source. This technology is widely used in high-end thermal design due to its extremely short calculation time and fast temperature response of the observation point. It plays a key role in optimizing the heat dissipation design of electronic products and shortening the product design cycle. For example, the prior art WO2016053470A1 discloses a method for thermal loop simulation using convolution and iteration methods. The patent elaborates on the technical details and implementation steps of thermal loop simulation using convolution and iteration methods, providing an effective solution for the field of system thermal management.

[0004] In traditional thermal circuit simulation, the temperature T(t) of the observation point is obtained by convolving the thermal circuit response function H(t) with the uniform time step of the heat source function P(t). However, with the development of thermal design, the heat source function P(t) has become more and more complex, the required time step has become smaller and smaller, the number of simulation steps has become larger and larger, and the simulation efficiency has also decreased. Summary of the invention

[0005] In order to solve the above problems, the present invention proposes a simulation method with non-uniform time step, which can quickly realize the convolution calculation of the temperature T(t) of the observation point, greatly improving the simulation rate.

[0006] The present invention proposes a thermal circuit simulation method with non-uniform time steps, comprising:

[0007] Step S1, importing a simulation model, setting thermal simulation parameters, performing thermal circuit extraction, and obtaining a thermal circuit response function H(s), wherein setting the thermal simulation parameters includes setting observation points;

[0008] Step S2, setting a heat source power curve P(t), wherein the heat source power curve P(t) is a power curve defining the heat source changing over time, and the heat source power curve P(t) is set according to a usage scenario of the heat source;

[0009] Step S3, setting an output time series with a non-uniform step size;

[0010] Step S4, performing thermal transient simulation according to the thermal circuit response function H(s), the heat source power curve P(t) and the output time series of the non-uniform step size;

[0011] Step S5: checking the temperature response curve of the observation point based on the thermal transient simulation result.

[0012] In one embodiment, in step S1, setting the thermal simulation parameters further includes:

[0013] Set material and convective heat transfer boundary conditions;

[0014] Set the heat source.

[0015] In one embodiment, in step S3, an output time series is set according to the moment of interest in the usage scenario, and the time step length in the output time series is non-uniform.

[0016] In one embodiment, step S4 includes:

[0017] Step S41, discretize the heat source power curve P(t) according to the output time series, and store it in a linked list L1, each node in the linked list L1 represents a time period, and the linked list L1 stores the initial time t of the node m , time step dt and average power p m , the time period from t m Start, to t m +dt ends, the time step dt is non-uniform, the average power p m It represents the average power of the heat source in a corresponding period of time;

[0018] Step S42, performing an inverse Laplace transform on the thermal circuit response function H(s) to obtain h(t), ;

[0019] Step S43, discretize h(t) according to the output time series and store it in a linked list L2, where each node in the linked list L2 represents a time period, and the linked list L2 stores the initial time t of the node. m , time step dt and average response value h m , the time period also starts from t m Start, to t m +dt ends, the time step dt is non-uniform;

[0020] Step S44, based on the linked lists L1 and L2, performing linked list convolution calculation to obtain the temperature at time t;

[0021] Step S45, obtaining the temperature of the observation point at each time point in the output time series with non-uniform step size.

[0022] In one embodiment, in step S44, when performing linked list convolution calculation, the temperature at time t can be obtained by the reverse linked list L1 of linked list L1 T Multiply by linked list L2 and integrate to get.

[0023] In one embodiment, in step S44, in performing linked list convolution calculation, the formula for convolution calculation is: ;

[0024] The temperature at time t can be obtained by the reverse link list L1 of link list L1 T Multiplying with the linked list L2 and integrating it gives: .

[0025] In one embodiment, step S45 includes:

[0026] Step S451: Obtain the power chain table L corresponding to n moments according to the output time series 1i , the power chain table L 1i It is the sub-link list of L1, from t0 to t i The time nodes are composed of;

[0027] Step S452: n power linked lists L 1i Reverse the linked list separately to get L 1i T ;

[0028] Step S453: n power linked lists L 1i T Perform linked list convolution calculations with L2 respectively to obtain the temperature T of the observation point at n moments i .

[0029] In one embodiment, in step S5, checking the temperature response curve of the observation point is: according to the time t corresponding to n moments i and temperature T i , draw the temperature response curve.

[0030] The present invention also provides a thermal circuit simulation device with non-uniform time steps, comprising a memory and a processor;

[0031] The memory is used to store computer programs;

[0032] The processor is used to implement the thermal circuit simulation method with non-uniform time step as described above when executing the computer program.

[0033] The present invention further proposes a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the thermal circuit simulation method with non-uniform time steps as described above is implemented.

[0034] The non-uniform time step thermal circuit simulation method and related equipment of the present invention have the following beneficial effects:

[0035] 1. The present invention solves the problem of low simulation rate caused by using complex heat source function P(t) for thermal circuit simulation, and can quickly realize the convolution calculation of observation point temperature T(t), greatly improving the simulation rate.

[0036] 2. The simulation method with non-uniform time step proposed in the present invention can discretize the heat source function P(t) and the response function h(t) to different degrees during simulation calculation, thereby improving the discrete accuracy of the response function h(t) and thus improving the simulation accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 Schematic diagram of a flow chart of a thermal circuit simulation method with non-uniform time steps according to an embodiment of the present invention.

[0038] Figure 2 FIG. 4 is a schematic diagram of a heat source power curve P(t) according to an embodiment of the present invention.

[0039] Figure 3 FIG. 4 is a schematic diagram of an output time series according to an embodiment of the present invention.

[0040] Figure 4 FIG. 4 is a schematic diagram of a linked list L1 according to an embodiment of the present invention.

[0041] Figure 5 FIG. 4 is a schematic diagram of a linked list L2 according to an embodiment of the present invention.

[0042] Figure 6 Schematic diagram of linked list convolution calculation in one embodiment of the present invention.

[0043] Figure 7 The linked lists L1 and L2 in one embodiment of the present invention are 1i T Schematic diagram of .

[0044] Figure 8 Schematic diagram of a temperature response curve of an observation point according to an embodiment of the present invention. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the invention and are not used to limit the invention.

[0046] like Figure 1 As shown, the present invention proposes a thermal circuit simulation method with a non-uniform time step, comprising the following steps:

[0047] Step S1, importing the simulation model, setting thermal simulation parameters, performing thermal circuit extraction, and obtaining the thermal circuit response function H(s).

[0048] Among them, setting thermal simulation parameters includes setting observation points. Observation points are points used to monitor and analyze temperature changes or other thermal physical quantities at specific locations of the model during the simulation process. In the software, you can usually directly specify the location of the observation point on the model and set the physical quantity to be monitored.

[0049] In this embodiment, a thermal circuit model is extracted from the simulation model, which generally involves analyzing the thermal behavior in the model, abstracting it into components in the circuit (such as resistors, capacitors, inductors, etc.), and determining the connection method between these components. In electronic system design, this may mean converting thermal behavior into an equivalent circuit model for more convenient analysis and calculation. Based on the extracted thermal circuit model, the thermal circuit response function H(s) can be further obtained. This function describes the response characteristics of the system to thermal input and is an important tool for analyzing the thermal performance of the system. In frequency domain analysis, H(s) is generally expressed as a transfer function between the system output and input, where s is a complex frequency variable.

[0050] In a specific embodiment, setting the thermal simulation parameters further includes: setting the material and convection heat transfer boundary conditions, and setting the heat source.

[0051] Among them, material properties have an important impact on thermal simulation results, because different materials have different thermal conductivity, specific heat capacity, density and other thermophysical properties. In the software, it is usually necessary to assign corresponding material properties to each part or material in the model. This is usually done by selecting a material from a predefined material library or manually entering the thermophysical parameters of the material.

[0052] Convective heat transfer is the process of heat transfer through the contact between a fluid and a solid surface. In thermal simulation, these boundary conditions need to be set to simulate this heat transfer. Convective heat transfer boundary conditions usually include parameters such as surface temperature, fluid temperature, and convection heat transfer coefficient. These parameters can be obtained through experimental measurement or estimated through theoretical calculations or empirical formulas. In the software, these parameters can be set on the boundaries of the model.

[0053] A heat source is a component or area that generates heat. In thermal simulation, you need to set parameters such as the location, shape, size, and heat generated by the heat source. These parameters may be based on experimental data, design specifications, or theoretical calculations. In the software, heat sources can appear in many forms, such as point heat sources, surface heat sources, or volume heat sources, depending on the complexity of the model and the simulation requirements.

[0054] Step S2, setting the heat source power curve P(t). Figure 2 As shown, in the heat source power curve P(t), the heat source is the GPU, the horizontal axis is time (time), the unit is second (s), and the vertical axis is power (power), the unit is watt (W).

[0055] The heat source power curve P(t) is a power curve that defines the change of the heat source over time. The heat source power curve P(t) is set according to the use scenario of the heat source. This curve describes the power output of the heat source at different time points and is one of the important inputs in thermal transient simulation. The heat source power curve is a user input curve and is set according to the actual simulation application scenario. For example, the power curve of the CPU (i.e., the heat source) is determined by the actual power consumption of the CPU and is set according to the CPU use scenario.

[0056] Step S3, setting the output time series t0, t1, …, t n , see Figure 3 The output time series starts at t0 and ends at t n , the time interval dt between each two time points is not required to be equal, that is, t m -t m-1 But not equal to t m+1 -t m .

[0057] Specifically, according to the moment of interest in the usage scenario, an output time series is set, and the time step in the output time series is non-uniform. This step directly sets the time points of interest. The intervals between these time points are not fixed, but are determined according to actual needs. Generally, it is set according to the set heat source power curve P(t) and the usage scenario. For example, more dense time points can be set near the 0 moment and where the power changes greatly.

[0058] In this embodiment, t0=0, t n = 60s, see Figure 2 .

[0059] Step S4, performing thermal transient simulation based on the thermal circuit response function H(s), the heat source power curve P(t) and the output time series with non-uniform step size. In this step, the temperature response at each time point in the output time series is calculated based on these inputs.

[0060] Specifically, step S4 includes the following steps:

[0061] Step S41, discretize the heat source power curve P(t) according to the output time series and store it in the linked list L1, see Figure 4 Each node in the linked list L1 represents a time period and stores the initial time t of the current node. m , time step dt and average power p m Here, a time period starts from t m Start, to t m +dt ends, the time step dt is non-uniform. The average power p stored in the node m It represents the average power of the heat source during this period of time.

[0062] Step S42, performing an inverse Laplace transform on the thermal circuit response function H(s) to obtain h(t), This step performs an inverse Laplace transform on the function H(s) in the complex frequency domain to obtain the response function h(t) in the time domain.

[0063] Step S43, discretize h(t) according to the output time series and store it in linked list L2, see Figure 5 Each node in the linked list L2 represents a time period and stores the initial time t of the current node. m , time step dt and average response value h m , the time step dt is non-uniform. Here, a time period is also from t m Start, to t m +dt ends.

[0064] Step S44, based on the linked lists L1 and L2, perform linked list convolution calculation to obtain the temperature at time t. The purpose of this step is to calculate the temperature at time t by combining the discrete values ​​of P(t) and h(t) through linked list convolution calculation. The size of t is determined by the final moment of the heat source power curve P(t) (this moment may not exist in the output time series set by the user). For example, if the heat source power curve P(t) is set to 0-3000s, then t=3000s. This step demonstrates in detail how to calculate the temperature at a specific moment based on linked list convolution.

[0065] Specifically, in step S44, in the linked list convolution calculation, the formula for the convolution calculation is: .

[0066] The temperature at time t can be obtained from the reverse list L1 of list L1 T Multiplying with linked list L2 and integrating gives: .

[0067] like Figure 6As shown, reverse list L1 T Multiply it with linked list L2 to get linked list T i ,in:

[0068] T1=P3*h1*dt1;

[0069] T2=P2*h2*dt2;

[0070] T3=P2*h3*dt3;

[0071] T4=P1*h3*dt4;

[0072] T5=P1*h4*dt5;

[0073] T6=P1*h5*dt6.

[0074] Among them, dt i T i The duration of i Corresponding time step. P1, P2, P3 are the discrete values ​​of P(t) stored in the linked list L1, h1, h2, h3, h4, h5, h6 are the discrete values ​​of h(t) stored in the linked list L2, and the temperature at the observation point T = T1 + T2 + T3 + T4 + T5 + T6 can be obtained by integration.

[0075] Step S45, obtaining the temperature of the observation point at each time point in the output time series of non-uniform step size. Specifically, step S45 includes:

[0076] Step S451: Obtain the power chain table L corresponding to n moments according to the output time series 1i , power chain table L 1i It is the sub-link list of L1, from t0 to t i time nodes.

[0077] Step S452: n power linked lists L 1i Reverse the linked list separately to get L 1i T , see Figure 7 .

[0078] Step S453: n power linked lists L 1i T Perform linked list convolution calculations with L2 respectively to obtain the temperature T of the observation point at n moments i .

[0079] The above step S45 is to obtain the sub-list L1 of the list L1 according to the time point i set by the user. i Then use L1 iThe temperature at time i is obtained by performing a linked list convolution calculation with L2 (refer to step S44), thereby obtaining the temperature at all times corresponding to the output time series. In order to explain step S44 and step S45 more clearly, the following example is given: In one embodiment, the heat source power curve P(t) is a CPU power curve of 1-300s, and the time points of interest to the user are 1s, 10s, and 100s. In this case, step S44: the temperature at 300s is calculated to be 60C°. Step S45: Perform linked list convolution calculations according to the output time series, and obtain the temperatures at 1s, 10s, and 100s as 30C°, 40C°, and 50C°, respectively.

[0080] Step S5: Based on the thermal transient simulation results obtained in step S4, check the temperature response curve of the observation point. i and temperature T i , draw the temperature response curve. Figure 8 As shown, the horizontal axis of the temperature response curve of the observation point is time (time), the unit is seconds (s), the initial time is 0s, and the end time is 60s. The vertical axis is temperature (T), the unit is degrees Celsius (℃)

[0081] The present invention also provides a thermal circuit simulation device with non-uniform time steps, comprising a memory and a processor. The memory is used to store a computer program, and the processor is used to implement the thermal circuit simulation method with non-uniform time steps as described above when executing the computer program.

[0082] The present invention further proposes a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the thermal circuit simulation method with non-uniform time steps as described above is implemented.

[0083] The steps of the method or algorithm described in conjunction with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. The software module may reside in a RAM memory, a flash memory, a ROM memory, an EPROM memory, an EEPROM memory, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. An exemplary storage medium is coupled to a processor so that the processor can read and write information from / to the storage medium. In an alternative, a storage medium may be integrated into a processor. The processor and the storage medium may reside in an ASIC. The ASIC may reside in a user terminal. In an alternative, the processor and the storage medium may reside in a user terminal as discrete components.

[0084] In one or more exemplary embodiments, the functions described may be implemented in hardware, software, firmware, or any combination thereof. If implemented as a computer program product in software, each function may be stored on or transmitted by a computer-readable medium as one or more instructions or codes. Computer-readable media include both computer storage media and communication media, including any media that facilitates the transfer of a computer program from one place to another. Storage media may be any available media that can be accessed by a computer. As an example and not limitation, such a computer-readable medium may include RAM, ROM, EEPROM, CD-ROM or other optical disk storage, disk storage or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of an instruction or data structure and can be accessed by a computer. Any connection is also properly referred to as a computer-readable medium. For example, if the software is transmitted from a website, server, or other remote source using a coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwaves, the coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwaves are included in the definition of the medium. Disk and disc as used herein include compact disc (CD), laser disc, optical disc, digital versatile disc (DVD), floppy disk and Blu-ray disc, where disks usually reproduce data magnetically, while discs reproduce data optically with lasers. Combinations of the above should also be included within the scope of computer-readable media.

[0085] The non-uniform time step thermal circuit simulation method and related equipment of the present invention have the following beneficial effects:

[0086] 1. The present invention solves the problem of low simulation rate caused by using complex heat source function P(t) for thermal circuit simulation, and can quickly realize the convolution calculation of observation point temperature T(t), greatly improving the simulation rate.

[0087] 2. The simulation method with non-uniform time step proposed in the present invention can discretize the heat source function P(t) and the response function h(t) to different degrees during simulation calculation, thereby improving the discrete accuracy of the response function h(t) and thus improving the simulation accuracy.

[0088] The above-described embodiments are only further explanations of the present invention, and are not intended to limit the present invention in other forms. The present invention may also have other various embodiments. Without departing from the spirit and essence of the present invention, those skilled in the art may make various corresponding modifications and changes according to the present invention, but these corresponding modifications and changes should all fall within the protection scope of the present invention.

Claims

1. A thermal circuit simulation method with non-uniform time steps, characterized in that: include: Step S1, importing the simulation model, setting thermal simulation parameters, performing thermal circuit extraction, and obtaining the thermal circuit response function H(s), wherein H(s) represents the transfer function between the system output and input, s is a complex frequency variable, and setting the thermal simulation parameters includes setting the observation point, setting the material and convection heat transfer boundary conditions, and setting the heat source; Step S2, setting a heat source power curve P(t), wherein the heat source power curve P(t) is a power curve defining the heat source changing over time, and the heat source power curve P(t) is set according to a usage scenario of the heat source; Step S3, setting an output time series with a non-uniform step size; Step S4, performing thermal transient simulation according to the thermal circuit response function H(s), the heat source power curve P(t) and the output time series of the non-uniform step size; Step S5: checking the temperature response curve of the observation point based on the thermal transient simulation result.

2. The thermal circuit simulation method with non-uniform time step according to claim 1, characterized in that: In the step S3, an output time series is set according to the moment of interest in the usage scenario, and the time step length in the output time series is non-uniform.

3. The thermal circuit simulation method with non-uniform time step according to claim 1, characterized in that: The step S4 comprises: Step S41, discretize the heat source power curve P(t) according to the output time series, and store it in a linked list L1, each node in the linked list L1 represents a time period, and the linked list L1 stores the initial time t of the node m , time step dt and average power p m , the time period from t m Start, to t m +dt ends, the time step dt is non-uniform, the average power p m It represents the average power of the heat source in a corresponding period of time; Step S42, performing an inverse Laplace transform on the thermal circuit response function H(s) to obtain h(t), ; Step S43, discretize h(t) according to the output time series and store it in a linked list L2, where each node in the linked list L2 represents a time period, and the linked list L2 stores the initial time t of the node. m , time step dt and average response value h m , the time period also starts from t m Start, to t m +dt ends, the time step dt is non-uniform; Step S44, based on the linked lists L1 and L2, performing linked list convolution calculation to obtain the temperature at time t; Step S45, obtaining the temperature of the observation point at each time point in the output time series with non-uniform step size.

4. The thermal circuit simulation method with non-uniform time step according to claim 3, characterized in that: In the step S44, when performing the linked list convolution calculation, the temperature at time t can be obtained by the reverse linked list L1 of the linked list L1 T Multiply by linked list L2 and integrate to get.

5. The thermal circuit simulation method with non-uniform time step according to claim 4, characterized in that: In step S44, the linked list convolution calculation is performed, and the convolution calculation formula is: ; The temperature at time t can be obtained by the reverse link list L1 of link list L1 T Multiplying with the linked list L2 and integrating it gives: .

6. The thermal circuit simulation method with non-uniform time step according to claim 3, characterized in that: The step S45 comprises: Step S451: Obtain the power chain table L corresponding to n moments according to the output time series 1i , the power chain table L 1i It is the sub-link list of L1, from t0 to t i The time nodes are composed of; Step S452: n power linked lists L 1i Reverse the linked list separately to get L 1i T ; Step S453: n power linked lists L 1i T Perform linked list convolution calculations with L2 respectively to obtain the temperature T of the observation point at n moments i .

7. The thermal circuit simulation method with non-uniform time step according to claim 1, characterized in that: In step S5, the temperature response curve of the observation point is checked as follows: according to the time t corresponding to n moments i and temperature T i , draw the temperature response curve.

8. A thermal circuit simulation device with non-uniform time steps, characterized in that: including memory and processor; The memory is used to store computer programs; The processor is used to implement the thermal circuit simulation method with non-uniform time steps as claimed in any one of claims 1 to 7 when executing the computer program.

9. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the thermal circuit simulation method with a non-uniform time step as claimed in any one of claims 1 to 7 is implemented.

Images