Power component performance testing method and system based on thermal feedback topology

Abstract

The present invention discloses a power component performance testing method and system based on thermal feedback topology, which relates to the field of electrical performance testing. The method includes: obtaining the access circuit of the target component, performing circuit scenario simulation, determining the simulation circuit, and replacing the non-target components of the access circuit with a black box structure; for the target component, determining the abstract surface under the layer-by-layer packaging of the structure with the chip as the center, determining the multi-axis with multiple performance parameters, temperature and time, constructing a thermal feedback map and performing dynamic drive training; by performing periodic simulation of the simulation circuit, dynamically updating the thermal feedback map, intervening in the target thermal feedback state, performing decomposition call and directional evaluation based on the thermal feedback map, and determining the performance test results. The method solves the technical problem of insufficient accuracy and comprehensiveness of performance evaluation in existing power component performance testing, and achieves the technical effect of accurately and comprehensively testing the performance of power components.

Description

Technical Field

[0001] The present application relates to the field of electrical performance testing, and in particular to a method and system for testing the performance of power components based on a thermal feedback topology. Background Art

[0002] As the core components of electronic devices, the stability and reliability of the performance of power components are crucial to the normal operation of the entire electronic system. Accurately testing the performance of power components is a key link in ensuring the high-quality operation of electronic equipment. At present, the main method to solve the problem of power component performance testing is usually to conduct actual circuit access tests on the components and evaluate their performance by collecting various data during actual operation. Due to the complex and changeable circuit environment and the interference of a large number of non-target components in the actual circuit access test, the current method is difficult to accurately focus on the performance evaluation of the target components. At the same time, for performance tests under some extreme or specific working conditions, the actual test cost is high, the cycle is long, and it is difficult to achieve comprehensive test coverage, which may lead to deviations in the test results and cannot fully and accurately reflect the true performance of the power components.

[0003] In the current related technologies, the performance testing of power components has the technical problem of insufficient accuracy and comprehensiveness of performance evaluation. Summary of the Invention

[0004] The present application provides a power component performance testing method and system based on thermal feedback topology, adopts technical means based on thermal feedback topology, obtains the access circuit of the target component and performs circuit scenario simulation, replaces non-target components with a black box structure to reduce interference, determines the abstract surface of the structure under layer-by-layer packaging with the chip as the center, constructs a multi-axial thermal feedback map containing multiple performance parameters, temperature and time, and performs dynamic drive training, dynamically updates the thermal feedback map using periodic simulation of the analog circuit, and performs decomposition call and directional evaluation after intervening in the target thermal feedback state, thereby solving the technical problems of insufficient accuracy and comprehensiveness of performance evaluation in existing power component performance tests, and achieving the technical effect of accurately and comprehensively testing the performance of power components.

[0005] The present application provides a power component performance testing method based on a thermal feedback topology, comprising: obtaining an access circuit of a target component, performing circuit scenario simulation, and determining an analog circuit, wherein non-target components of the access circuit are replaced with a black box structure; for the target component, determining an abstract surface under layer-by-layer packaging of the structure with the chip as the center, determining multi-dimensional axes with multiple performance parameters, temperature, and time, constructing a thermal feedback map, and performing dynamic drive training, wherein the thermal feedback map is built into a test system and establishes temporary communication with the analog circuit; dynamically updating the thermal feedback map by performing periodic simulation of the analog circuit, intervening in a target thermal feedback state, performing decomposition calling and directional evaluation based on the thermal feedback map, and determining performance test results, wherein the dynamic update includes direct update based on analog data and indirect update based on non-analog data.

[0006] In a possible implementation, the following processing is performed: the access circuit includes at least one target component; for the access circuit, the circuit access position of the non-target component is located, wherein the input-output direction of each non-target component at the circuit access position is determined according to the circuit flow direction; with input-output as the demand orientation, the linear state relationship of each non-target component is mined; based on the circuit access position and the linear state relationship, the black box structure of the non-target component is replaced on the access circuit.

[0007] In a possible implementation, the abstract surface of the structure under layer-by-layer packaging is determined with the chip as the center, and the following processing is performed: with the chip as the center and the outermost packaging layer as the boundary, the component structure from the center to the boundary is abstractly displayed to determine the multi-layer structure surface; for the multi-layer structure surface, the multi-layer packaging characteristics are located, the multi-layer structure surface is marked, and the abstract surface is determined.

[0008] In a possible implementation, a thermal feedback map is constructed and the following processing is performed: determining multiple performance parameters, wherein the multiple performance parameters define all performance factors of the target component; determining multiple performance axes for the multiple performance parameters; and coupling the abstract surface, temperature axis, time axis and the multiple performance axes to determine the thermal feedback map.

[0009] In a possible implementation, the abstract surface, temperature axis, time axis and the multiple performance axes are coupled, and the following processing is performed: performance-related variables are determined for the multiple performance parameters; and correlation coupling between each performance axis and the abstract surface, temperature axis and time axis is established using the performance-related variables.

[0010] In a possible implementation, a thermal feedback map is constructed and dynamically driven training is performed, and the following processing is performed: the thermal feedback map is dynamically driven training with sample test tasks as constraints, sample circuit data as input, and the update of the thermal feedback map as output; if the convergence conditions are met, the trained thermal feedback map is embedded and deployed in the test system, and temporary communication between the thermal feedback map and the analog circuit is established.

[0011] In a possible implementation, the following processing is performed: with the switching of the target component to be tested or the switching of the access circuit, determining to switch the simulation circuit; interrupting the temporary communication between the thermal feedback map and the simulation circuit, and establishing temporary communication between the switching simulation circuit and the thermal feedback map.

[0012] In a possible implementation, the target thermal feedback state is intervened, and a decomposition call and a directional evaluation based on the thermal feedback map are performed, and the following processing is performed: the thermal feedback state is set, wherein the thermal feedback state at least includes temperature cycling, constant high temperature, thermal stress accumulation under dynamic temperature, thermal stability, mechanical strength and packaging reliability, and the thermal stability is based on stability over a wide temperature range; the thermal feedback map is updated in real time through the analog circuit, which includes direct update of simulation data and indirect update of performance parameters; the target thermal feedback state is determined, and a directional evaluation is performed on the thermal feedback map updated in real time.

[0013] In a possible implementation, the performance test result is determined by performing the following processing: determining a directional evaluation result based on the target thermal feedback state; and determining the performance test result based on the directional evaluation result and the thermal feedback map updated in real time.

[0014] The present application also provides a power component performance testing system based on a thermal feedback topology, including: an analog circuit determination module, used to obtain the access circuit of the target component, perform circuit scenario simulation, and determine the analog circuit, wherein the non-target components of the access circuit are replaced with a black box structure; a thermal feedback map construction module, used to determine the abstract surface under the layer-by-layer packaging of the structure with the chip as the center for the target component, determine the multi-axis with multiple performance parameters, temperature and time, construct a thermal feedback map and perform dynamic drive training, wherein the thermal feedback map is built into the test system and establishes temporary communication with the analog circuit; a performance test result determination module, used to dynamically update the thermal feedback map by performing periodic simulation of the analog circuit, intervene in the target thermal feedback state, perform decomposition call and directional evaluation based on the thermal feedback map, and determine the performance test result, wherein the dynamic update includes direct update based on analog data and indirect update based on non-analog data.

[0015] The power component performance testing method and system based on thermal feedback topology proposed in this application first obtains the access circuit of the target component, performs circuit scenario simulation, and determines the simulation circuit, wherein the non-target components of the access circuit are replaced with a black box structure. Then, for the target component, the abstract surface under the layer-by-layer packaging of the structure is determined with the chip as the center, and the multi-axis is determined by multiple performance parameters, temperature and time. A thermal feedback map is constructed and dynamic drive training is performed. The thermal feedback map is built into the test system and establishes temporary communication with the simulation circuit. Finally, by performing periodic simulation of the simulation circuit, the thermal feedback map is dynamically updated, the target thermal feedback state is intervened, and decomposition call and directional evaluation based on the thermal feedback map are performed to determine the performance test results. The dynamic update includes direct update based on simulation data and indirect update based on non-simulation data. The technical effect of accurately and comprehensively testing the performance of power components is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings of the embodiments of the present invention are briefly introduced below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in precise order. Instead, various steps may be processed in reverse order or simultaneously as needed. Furthermore, other operations may be added to these processes, or one or more operations may be removed from these processes.

[0017] Figure 1 A schematic flow chart of a method for testing the performance of a power component based on a thermal feedback topology according to an embodiment of the present application.

[0018] Figure 2 A schematic structural diagram of a power component performance testing system based on a thermal feedback topology provided in an embodiment of the present application.

[0019] Description of the reference numerals: analog circuit determination module 10 , thermal feedback map construction module 20 , performance test result determination module 30 . DETAILED DESCRIPTION

[0020] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0021] In order to make the purpose, technical solutions and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limiting this application. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0022] In the following description, reference is made to “some embodiments” which describe a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict. The terms “including” and “having” and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application belongs. The terms used herein are for the purpose of describing the embodiments of this application only.

[0023] The present application provides a method for testing the performance of power components based on a thermal feedback topology. Figure 1 As shown, the method includes:

[0024] Step S100 , obtaining an access circuit of a target component, performing circuit scenario simulation, and determining a simulated circuit, wherein non-target components of the access circuit are replaced with a black box structure.

[0025] Specifically, the target components are power components requiring performance testing, such as power MOSFETs and power capacitors. Circuit design software (such as Cadence or Altium Designer) is used to read the target component's circuit diagram and extract its connection information, including pin connections, power supply, and ground connections. Other components in the circuit are identified and marked as non-target components. These non-target components are then blackboxed, retaining only their input and output interfaces while ignoring their internal structure. For example, for a resistor, only the voltage and current interfaces at its ends are retained, ignoring the specific calculation of its resistance value. The extracted circuit is simulated using circuit simulation software (such as LTspice or MATLAB / Simulink). Simulation conditions, such as power supply voltage, signal frequency, and load conditions, are set. Using the simulation software's interface functions, the blackboxed non-target components are connected to the target component to form a complete simulation circuit.

[0026] For example, suppose the target component is a power MOSFET, and its circuit diagram also includes components such as capacitors, resistors, and inductors. Cadence software is used to read the circuit diagram, extract the MOSFET's pin connection information, and mark the capacitors, resistors, and inductors as non-target components. The capacitors, resistors, and inductors are simplified as black boxes, retaining only their input and output interfaces. LTspice software is used to simulate this simplified circuit, setting the power supply voltage to 12V, the signal frequency to 100kHz, and the load condition to a 1Ω resistor. Using LTspice's interface function, the non-target components are connected to the MOSFET in the black box to form a complete simulation circuit.

[0027] In one possible implementation, step S100 further includes step S110, where the circuit includes at least one target component. For the circuit, the circuit connection locations of non-target components are located. The input-output direction of each non-target component at the circuit connection location is determined based on the circuit power flow direction. Specifically, the complete circuit diagram of the circuit is read using circuit design software (such as Cadence or Altium Designer), all components in the circuit are identified, and the connection location of each component, including the pin connections and connection sequence, is determined. For example, for a circuit containing multiple components, the circuit diagram is parsed to determine the pin connections and connection sequence of each component. Based on the power and load distribution of the circuit, the current and voltage flow directions of the circuit are analyzed to determine the input and output directions of each non-target component in the circuit. For example, for a circuit containing an amplifier, the direction of current flow from the power supply to the amplifier and then from the amplifier to the load is analyzed.

[0028] For example, suppose the access circuit contains a power MOSFET (target component), an amplifier, and a resistor. Use Cadence software to read the circuit diagram and identify the connection locations of each component. For example, the amplifier's input is connected to the signal source, and its output is connected to the gate of the power MOSFET. Analyzing the current flow in the circuit, we determine that the amplifier's input is directed toward the signal source, and its output is directed toward the gate of the power MOSFET. Therefore, the amplifier's input-output direction is from the signal source to the power MOSFET.

[0029] Step S120: Using input-output requirements as a guide, linear state relationships are mined for each non-target component. Specifically, the input-output state relationship analysis is performed for each non-target component. Circuit simulation software (such as LTspice, MATLAB / Simulink, etc.) is used to simulate the non-target component, record its input and output states, and analyze the linear or approximately linear relationship between the input and output states. For example, for an amplifier, the relationship between the amplitude and phase of the input signal and the amplitude and phase of the output signal is analyzed. Data fitting methods (such as the least squares method) are used to fit the input-output data and establish a linear model. For example, for an amplifier, a linear model is established between the input signal amplitude and the output signal amplitude.

[0030] For example, suppose the non-target component is an amplifier, and it is simulated using LTspice software. The input signal amplitude is recorded as 1V, and the output signal amplitude is 10V. Analyzing the linear relationship between the input and output signals reveals that the output signal amplitude is 10 times the input signal amplitude. Using the least squares method, a linear model is constructed by fitting the input-output data. The model formula is: output amplitude = 10 × input amplitude.

[0031] Step S130: Based on the relationship between the circuit access location and the linear state, the non-target components of the access circuit are replaced with a black box structure. Specifically, based on the linear state relationship, a black box structure is defined for the non-target components. The black box structure only retains the input and output interfaces, ignoring the internal structure. For example, for an amplifier, the black box structure is defined such that the input terminal receives the signal and the output terminal outputs the amplified signal. Using circuit design software (such as Cadence or Altium Designer), the non-target components are replaced with a black box structure, and the circuit is reconstructed to ensure that the circuit's connectivity and functionality remain unchanged. For example, replacing an amplifier with a black box structure retains the input and output interfaces.

[0032] This approach simplifies the circuit model and reduces computational complexity by replacing non-target components with black-box structures. It also improves test efficiency by focusing solely on the performance testing of target components, avoiding detailed analysis of non-target components.

[0033] Step S200, for the target component, determine the abstract surface of the structure under layer-by-layer packaging with the chip as the center, determine the multi-axis with multiple performance parameters, temperature and time, construct a thermal feedback map and perform dynamic drive training, wherein the thermal feedback map is built into the test system and establishes temporary communication with the analog circuit.

[0034] Specifically, package design software (such as Ansys Icepak and Siemens Xpedition) is used to analyze the package structure of the target component and determine the chip's layer-by-layer packaging information, including the chip, package substrate, heat sink, etc. For example, for a packaged power chip, the structure and material properties of the chip, package substrate, and heat sink are analyzed.

[0035] Use thermal analysis software (such as Ansys Fluent and COMSOL Multiphysics) to perform thermal analysis on the target component. A thermal feedback map is constructed, centered around the chip. This map includes multiple axes for performance parameters (such as voltage, current, and power), temperature, and time. This map records the temperature changes of the target component under different operating conditions. For example, the map can record the temperature changes of the chip under different currents (0A-10A), temperatures (25°C-100°C), and time (0s-1000s).

[0036] Dynamically train the thermal feedback map using machine learning algorithms (such as neural networks and support vector machines). The trained model can predict the temperature changes of target components under different operating conditions. For example, a neural network can be used to train the thermal feedback map with inputs of current, temperature, and time, and the output is the temperature change of the target component.

[0037] For example, suppose the target component is a packaged power chip. Ansys Icepak software is used to analyze its package structure and determine the structural and material properties of the chip, package substrate, and heat sink. The chip measures 10 mm x 10 mm, the package substrate is made of ceramic, and the heat sink is made of copper. Ansys Fluent software is used to perform thermal analysis on the target component and construct a thermal feedback map. The map includes multi-axis curves for current (0 A–10 A), temperature (25°C–100°C), and time (0–1000 seconds). The map records the temperature changes of the chip at different currents, temperatures, and times, as shown in Table 1.

[0038] Table 1: Example of thermal feedback map data

[0039]

[0040] A neural network is used to train the thermal feedback map, with inputs of current, temperature, and time, and outputs the temperature change of the target component. Once trained, the model can predict the temperature change of the target component under different operating conditions.

[0041] In one possible implementation, the abstract surface of the structure under layer-by-layer packaging is determined with the chip as the center. Step S200 further includes step S210, which abstractly displays the component structure from the center to the boundary with the chip as the center and the outermost packaging layer as the boundary to determine the multi-layer structure surface. Specifically, the packaging structure of the target component is read using packaging design software (such as Ansys Icepak, Siemens Xpedition, etc.), starting from the center of the chip, and gradually analyzing each layer of the packaging structure outward until the outermost packaging layer. For example, for a packaged power chip, the structure of each layer of the chip, packaging substrate, heat sink, etc. is analyzed. Each layer of the packaging structure is abstracted as a region, and the specific structural features are not displayed, but only defined by the region. For example, the chip is abstracted as a central region, the packaging substrate is abstracted as a middle region, and the heat sink is abstracted as an outer region. Colors or marks are used to distinguish different regions to facilitate subsequent analysis of temperature distribution.

[0042] For example, suppose the target component is a power chip, and its package structure is analyzed using Ansys Icepak software. Starting from the center of the chip, each layer of the package structure is analyzed outward, including the chip, package substrate, and heat sink. The chip is abstracted as a central region (labeled as Region 1), the package substrate as a middle region (labeled as Region 2), and the heat sink as an outer region (labeled as Region 3). Different colors or markers are used to distinguish these regions, as shown in Table 2.

[0043] Table 2: Examples of multi-layered structures

[0044]

[0045] Step S220, locate the multi-layer packaging characteristics for the multi-layer structural surface, mark the multi-layer structural surface, and determine the abstract surface. Specifically, analyze the characteristics of each layer of the packaging structure, such as material properties (thermal conductivity, thermal expansion coefficient, etc.), thickness, etc. For example, the thermal conductivity of the chip is relatively high, the thermal conductivity of the packaging substrate is relatively low, and the thermal conductivity of the heat sink is relatively high. Use thermal analysis software (such as Ansys Fluent, COMSOL Multiphysics, etc.) to analyze the thermal characteristics of each layer of the packaging structure. Mark each layer of the structural surface according to the packaging characteristics. For example, mark the chip area as a "high thermal conductivity area", the packaging substrate area as a "low thermal conductivity area", and the heat sink area as a "high thermal conductivity area". Determine the abstract surface, that is, the boundary of each layer of the structure, for subsequent temperature distribution analysis.

[0046] For example, Ansys Fluent software was used to analyze the thermal characteristics of each package layer. The analysis results are as follows: the chip region has high thermal conductivity and low thermal expansion coefficient; the package substrate region has low thermal conductivity and high thermal expansion coefficient; and the heat sink region has high thermal conductivity and low thermal expansion coefficient. Based on the analysis results, each structural surface is labeled: the chip region is labeled "high thermal conductivity region" (marked A); the package substrate region is labeled "low thermal conductivity region" (marked B); and the heat sink region is labeled "high thermal conductivity region" (marked C). Abstract surfaces, or the boundaries of each structural layer, are determined, as shown in Table 3.

[0047] Table 3: Examples of abstract faces

[0048]

[0049] This implementation abstracts each layer of the package structure into a region and uses these regions to more intuitively display the temperature distribution at different internal structural locations. By marking the packaging characteristics of each layer, the impact of temperature distribution on the package state can be more accurately analyzed, thereby improving the accuracy of temperature distribution analysis. By analyzing the impact of temperature distribution on the package state, the mechanical properties of the package structure can be more accurately evaluated. By marking the multi-layer structure surface, the behavior of each layer under temperature changes can be more clearly understood, providing more reliable data support for mechanical performance evaluation, thereby enhancing the reliability of mechanical performance evaluation.

[0050] In one possible implementation, constructing a thermal feedback map, step S200 further includes step S230, determining multiple performance parameters, wherein these multiple performance parameters define all performance factors of the target component. Specifically, key performance parameters are extracted based on the specifications and application scenarios of the target component. These parameters cover all important characteristics of the target component in actual operation. For example, for a power MOSFET, performance parameters include: voltage (drain-source voltage (Vds) and gate-source voltage (Vgs); current (drain current (Id); power (power dissipation (Pd); efficiency (conversion efficiency); and switching speed (turn-on and turn-off times). To ensure that the extracted performance parameters fully reflect the performance of the target component, these parameters should include electrical, thermal, and mechanical properties. For example, in addition to electrical performance parameters (such as voltage, current, and power), thermal performance parameters (such as temperature and thermal resistance) and mechanical performance parameters (such as package strength and thermal expansion coefficient) should also be included.

[0051] For example, assuming the target component is a power MOSFET, its key performance parameters include: electrical performance parameters: drain-source voltage (Vds), gate-source voltage (Vgs), drain current (Id), power consumption (Pd); thermal performance parameters: chip temperature (Tj), package temperature (Tc), thermal resistance (Rth); mechanical performance parameters: package strength, thermal expansion coefficient.

[0052] Step S240: Determine multiple performance axes for the multiple performance parameters. Specifically, define an independent axis for each performance parameter to represent the range of variation of the parameter in the graph. For example, define a voltage axis for the drain-source voltage (Vds) ranging from 0V to 100V; define a current axis for the drain current (Id) ranging from 0A to 10A. Similarly, define a temperature axis for the temperature (Tj) ranging from 25°C to 150°C; and define a time axis for the time (t) ranging from 0s to 1000s. Construct a multi-axis system to integrate the axes of all performance parameters to form a multidimensional space. For example, construct a four-dimensional space containing a voltage axis, a current axis, a temperature axis, and a time axis.

[0053] Step S250 couples the abstract surface, temperature axis, time axis, and multiple performance axes to determine the thermal feedback map. Specifically, the abstract surface (e.g., chip area, package substrate area, heat sink area) is coupled with the temperature axis, time axis, and multiple performance axes. For example, for the chip area, the changes in drain-source voltage (voltage axis) and drain current (current axis) at different times (time axis) and different temperatures (temperature axis) are recorded. Data processing software (e.g., MATLAB, Python) is used to process and visualize these multidimensional data. A thermal feedback map is constructed, which includes all coupled data points. For example, the map can show the changes in performance parameters such as voltage, current, and power consumption of the target component at different times and temperatures. Thermal analysis software (e.g., Ansys Fluent, COMSOL Multiphysics) is used to dynamically update and analyze the map.

[0054] This implementation enables a comprehensive assessment of target component performance by coupling abstract surfaces, temperature axes, time axes, and multiple performance axes. This multidimensional map reflects the electrical, thermal, and mechanical performance of the target component under different operating conditions. The thermal feedback map dynamically updates, reflecting performance changes of the target component in real time, providing real-time data support for performance optimization and enhancing the comprehensiveness of performance evaluation. By defining multiple performance parameters and integrating them into the map, the performance of the target component can be more accurately assessed. This multidimensional analysis approach captures more details and improves test accuracy.

[0055] In one possible implementation, the abstract surface, temperature axis, time axis, and the multiple performance axes are coupled, and step S250 further includes step S251, determining performance-related variables for the multiple performance parameters. Specifically, each performance parameter is analyzed to determine variables related to it. These variables can be direct physical quantities (such as temperature, current, and voltage) or derived performance indicators (such as power consumption, efficiency, and thermal resistance). For example, for voltage (Vds) and current (Id), the related variable can be power consumption Pd = Vds × Id; for temperature (Tj), the related variable can be thermal resistance Rth = (Tj-Tc) / Pd.

[0056] Select relevant variables based on the specific test requirements. For example, if the test requirement is to evaluate thermal failure, relevant variables may include temperature limit locations, structural failure states at package locations, etc. If the test requirement is to evaluate thermal stability, relevant variables may include temperature fluctuation partitions based on time trends.

[0057] For example, assuming the target component is a power MOSFET, its performance parameters include drain-source voltage (Vds), drain current (Id), die temperature (Tj), package temperature (Tc), and power dissipation (Pd). The relevant variables are as follows: power dissipation (Pd): Pd = Vds × Id; thermal resistance (Rth): Rth = (Tj - Tc) / Pd; temperature change rate (dT / dt): used to evaluate temperature fluctuations.

[0058] If the test requirement is to evaluate thermal failure, the relevant variables include: temperature limit location: Tj>150°C; structural failure state at the package location: abnormal increase in thermal resistance.

[0059] If the test requirement is to evaluate thermal stability, relevant variables include: Temperature fluctuation partition based on time trend: dT / dt>1°C / s.

[0060] Step S252 uses the performance-related variables to establish correlation coupling between each performance axis and the abstract surface, temperature axis, and time axis. Specifically, data processing software (such as MATLAB or Python) is used to analyze the performance-related variables and determine their correlations. For example, the relationship between temperature (Tj) and power consumption (Pd) or the relationship between temperature change rate (dT / dt) and time (t) can be analyzed. Based on the results of the correlation analysis, a coupling relationship is established between each performance axis and the abstract surface, temperature axis, and time axis.

[0061] For example, establish a coupling relationship between the temperature axis (Tj) and the power consumption axis (Pd), or establish a coupling relationship between the temperature change rate axis (dT / dt) and the time axis (t). Use thermal analysis software (such as Ansys Fluent and COMSOL Multiphysics) to dynamically simulate and visualize the coupling relationship.

[0062] This implementation method establishes correlation coupling to quickly locate data relevant to test requirements. This correlation coupling enables efficient performance evaluation of target components. By analyzing the relationships between performance-related variables, the performance of target components can be more accurately assessed. This precise analysis method captures more details and improves test accuracy.

[0063] In one possible implementation, a thermal feedback map is constructed and dynamic drive training is performed. Step S200 further includes step S260, where dynamic drive training is performed on the thermal feedback map using sample test tasks as constraints, sample circuit data as input, and an update of the thermal feedback map as output. Specifically, a set of sample test tasks is defined that cover various operating scenarios of the target component in actual applications. For example, for a power MOSFET, sample test tasks include: switching operation under different load conditions; continuous operation at different ambient temperatures; and performance evaluation under different input voltages.

[0064] Collect circuit data corresponding to the sample test task, including voltage, current, and temperature. This data is obtained through circuit simulation. Use machine learning algorithms (such as neural networks and support vector machines) to train the thermal feedback map. Dynamically drive training is performed using the sample test task as a constraint, sample circuit data as input, and an update to the thermal feedback map as output. For example, a neural network can be used to train the thermal feedback map, with sample circuit data (voltage, current, and temperature) as input and an update to the thermal feedback map as output.

[0065] Step S270, if the convergence condition is met, the trained thermal feedback map is embedded and deployed in the test system, and temporary communication between the thermal feedback map and the analog circuit is established. Specifically, during the training process, convergence conditions are set, such as the error is less than a certain threshold (such as 0.01) or the number of training times reaches a certain upper limit (such as 1000 times). The trained thermal feedback map is embedded in the test system, and the test system can be a hardware test platform or a software simulation platform for real-time evaluation of the performance of the target components. Temporary communication between the thermal feedback map and the analog circuit is established so that data can be updated and fed back in real time during the test process. For example, a data interface (such as USB, Ethernet) is used to establish communication between the thermal feedback map and the analog circuit.

[0066] This implementation uses dynamic drive training to update the thermal feedback map in real time based on actual circuit data, improving test accuracy. By establishing temporary communication between the thermal feedback map and the simulated circuit, data can be fed back in real time, improving the real-time and dynamic nature of the test.

[0067] In one possible implementation, step S270 further includes step S271, determining whether to switch the analog circuit as the target component to be tested is switched or the access circuit is switched. Specifically, in the test system, real-time monitoring is performed to determine whether the target component to be tested has been replaced through a hardware interface (such as GPIO, I2C, etc.) or a software interface (such as an API call). For example, when the test system detects that the target component has been switched from a power MOSFET to a power capacitor, the switching process is triggered. Similarly, real-time monitoring is performed to determine whether the access circuit has changed by detecting the connection status in the circuit (such as relay status, switch status) or through the interface of the circuit simulation software. For example, when the access circuit switches from a simple load circuit to a complex multi-load circuit, the switching process is triggered. Based on the new target component or access circuit, the analog circuit is redefined by reloading the circuit model or reconfiguring the simulation parameters. For example, if the target component is switched from a power MOSFET to a power capacitor, the circuit model of the power capacitor is reloaded, and the simulation parameters are updated to match the new circuit configuration.

[0068] Step S272, interrupting the temporary communication between the thermal feedback map and the analog circuit, and establishing temporary communication between the switched analog circuit and the thermal feedback map. Specifically, when switching target components or accessing circuits, the temporary communication between the current thermal feedback map and the analog circuit is interrupted by closing the communication interface (such as USB, Ethernet) or pausing data transmission. For example, the USB communication interface between the current analog circuit and the thermal feedback map is closed. Based on the new analog circuit, the temporary communication between the thermal feedback map and the switched analog circuit is re-established by reconfiguring the communication interface or restarting data transmission. For example, the USB communication interface is reconfigured so that it is connected to the new analog circuit and data transmission is resumed.

[0069] This approach monitors changes in target components and access circuits in real time, enabling rapid switching of analog circuits to accommodate varying test requirements. By interrupting and re-establishing temporary communication, it ensures seamless integration between the thermal feedback map and the new analog circuit, minimizing data loss and communication delays during the switching process and improving the test system's flexibility and adaptability.

[0070] Step S300, by performing periodic simulation of the analog circuit, dynamically updating the thermal feedback map, intervening in the target thermal feedback state, performing decomposition call and directional evaluation based on the thermal feedback map, and determining the performance test results, wherein the dynamic update includes direct update based on analog data and indirect update based on non-analog data.

[0071] Specifically, use circuit simulation software (such as LTspice, MATLAB / Simulink, etc.) to periodically simulate the analog circuit. Set the simulation period, for example, every 100ms, and record the electrical parameters (such as voltage, current, and power) and temperature changes of the target components. For example, use LTspice software to set the simulation period to every 100ms and record the voltage, current, and temperature changes of the target components.

[0072] Use data processing software (such as Python and MATLAB) to process the simulation data and directly update the thermal feedback map based on the simulation data, such as updating the temperature change of the target component under different operating conditions. For non-simulation data, use empirical formulas or historical data for indirect updates.

[0073] Use evaluation software (such as Excel or MATLAB) to decompose and access the thermal feedback map. Based on the target thermal feedback state and the data in the thermal feedback map, conduct a targeted performance evaluation of the target component. For example, evaluate performance indicators such as power loss and efficiency at different temperatures. Once the evaluation is complete, determine the performance test results of the target component.

[0074] In one possible implementation, a target thermal feedback state is involved, and decomposition, invocation, and targeted evaluation based on the thermal feedback map are performed. Step S300 further includes step S310, where thermal feedback states are set. These thermal feedback states include at least temperature cycling, constant high temperature, thermal stress accumulation under dynamic temperature, thermal stability, mechanical strength, and package reliability. Thermal stability is conditional on stability over a wide temperature range. Specifically, multiple thermal feedback states are set to comprehensively evaluate the performance of the target component under different thermal conditions. These states include temperature cycling (simulating a component cycling between different temperatures), constant high temperature (simulating long-term operation of a component in a high-temperature environment), thermal stress accumulation under dynamic temperature (simulating thermal stress accumulation under dynamic temperature changes), thermal stability (assessing component stability over a wide temperature range), mechanical strength (assessing the mechanical strength of the package structure under thermal stress), and package reliability (assessing the reliability of the package structure under thermal stress). Conditions for thermal stability are defined, such as maintaining stable performance over a wide temperature range (e.g., -40°C to 150°C). For example, the thermal stability condition is set as follows: within the range of -40°C to 150°C, the performance parameters of the components (such as voltage, current, and power consumption) do not change by more than 5%.

[0075] Step S320: Update the thermal feedback map in real time through the analog circuit, including direct updates of analog data and indirect updates of performance parameters. Specifically, the thermal feedback map is directly updated using data generated by the analog circuit, which includes real-time voltage, current, temperature, etc. For example, through LTspice or MATLAB / Simulink analog circuits, the voltage, current, and temperature data of the components are obtained in real time, and the thermal feedback map is updated. The performance parameters in the thermal feedback map are indirectly updated based on existing performance models or empirical formulas. For example, the temperature data in the thermal feedback map is indirectly updated based on the relationship between the temperature change rate (dT / dt) and time (t).

[0076] Step S330, determine the target thermal feedback state, and perform a targeted evaluation on the thermal feedback map updated in real time. Specifically, determine the target thermal feedback state according to the set thermal feedback state. For example, determine the thermal stress accumulation, thermal stability, mechanical strength and packaging reliability of the target component under temperature cycling, constant high temperature and dynamic temperature. For example, determine the thermal stability of the target component in the range of -40°C to 150°C. According to the target thermal feedback state, perform a targeted evaluation on the thermal feedback map updated in real time. For example, evaluate the performance changes of the target component under temperature cycling, or the stability under constant high temperature. Use data processing software (such as MATLAB, Python) to analyze the thermal feedback map and generate an evaluation report.

[0077] For example, assuming the target component is a power MOSFET, the following target thermal feedback conditions are determined: thermal stability within the -40°C to 150°C range; constant high-temperature performance at 150°C for 1000 hours; and thermal stress accumulation under dynamic temperature with hourly temperature changes between 0°C and 100°C. Based on the target thermal feedback conditions, a targeted evaluation is performed on the thermal feedback map, which is updated in real time. For example, the performance changes of the target component under temperature cycling are evaluated, as shown in Table 4.

[0078] Table 4: Example of thermal stability evaluation under temperature cycling

[0079]

[0080] This implementation comprehensively evaluates the performance of target components under varying thermal conditions by setting multiple thermal feedback states. Real-time updates to the thermal feedback map capture changes in the target component's performance under dynamic conditions, enhancing the comprehensiveness of the performance evaluation. Direct updates to simulation data and indirect updates to performance parameters ensure the accuracy and completeness of the thermal feedback map. Targeted evaluation based on the target thermal feedback state enables more precise problem identification and improves test reliability.

[0081] In one possible implementation, the performance test results are determined, and step S300 further includes step S340, which determines a directional evaluation result based on the target thermal feedback state. Specifically, according to the target thermal feedback state (such as temperature cycling, constant high temperature, thermal stress accumulation at dynamic temperature, thermal stability, mechanical strength and packaging reliability), relevant evaluation results are extracted from the real-time updated thermal feedback map. For example, for thermal stability at constant high temperature, the performance parameter changes after running at 150°C for 1000 hours are extracted. The extracted evaluation results are statistically analyzed to determine the change trend and stability of the performance parameters. For example, the statistical quantities such as the mean, standard deviation, maximum value, minimum value, etc. of the performance parameters are calculated to evaluate the change range and fluctuation of the performance parameters.

[0082] Step S350, based on the directional evaluation result and the real-time updated thermal feedback map, determine the performance test result. Specifically, the performance of the target component is comprehensively evaluated in combination with the directional evaluation result and the real-time updated thermal feedback map. For example, based on the statistical analysis results, evaluate whether the thermal stability of the target component at a constant high temperature meets the design requirements. Perform trend analysis on the real-time updated thermal feedback map to determine the changing trend of the performance parameters in the full test cycle. For example, analyze the changing trend of power consumption over time to determine whether there is performance degradation or abnormality. Determine the final performance test result based on the comprehensive evaluation and trend analysis results. For example, if the average performance change is within the design allowable range (such as less than 1%), it is considered that the thermal stability of the target component at a constant high temperature is good.

[0083] This approach combines targeted evaluation results with real-time updated thermal feedback maps to more comprehensively assess the performance of target components. Statistical analysis allows for more accurate assessment of performance parameter trends and stability, improving test reliability.

[0084] The embodiment of the present application adopts a technical means based on thermal feedback topology. By obtaining the access circuit of the target component and performing circuit scenario simulation, the non-target components are replaced with a black box structure to reduce interference, and the abstract surface of the structure under layer-by-layer packaging is determined with the chip as the center. A multi-axial thermal feedback map containing multiple performance parameters, temperature and time is constructed and dynamic drive training is performed. The thermal feedback map is dynamically updated using periodic simulation of the analog circuit, and decomposition call and directional evaluation are performed after intervening in the target thermal feedback state. The technical problem of insufficient accuracy and comprehensiveness of performance evaluation in existing power component performance tests is solved, and the technical effect of accurately and comprehensively testing the performance of power components is achieved.

[0085] In the above, refer to Figure 1 The power component performance testing method based on thermal feedback topology according to an embodiment of the present invention is described in detail. Figure 2 A power component performance testing system based on a thermal feedback topology according to an embodiment of the present invention is described.

[0086] A power component performance testing system based on a thermal feedback topology according to an embodiment of the present invention is designed to address the technical issues of insufficient accuracy and comprehensiveness in performance evaluation in existing power component performance testing, thereby achieving the technical effect of accurately and comprehensively testing power component performance. The power component performance testing system based on a thermal feedback topology includes: an analog circuit determination module 10, a thermal feedback map construction module 20, and a performance test result determination module 30.

[0087] The analog circuit determination module 10 is used to obtain the access circuit of the target component, perform circuit scenario simulation, and determine the analog circuit, wherein the non-target components of the access circuit are replaced with a black box structure; the thermal feedback map construction module 20 is used to determine the abstract surface under the layer-by-layer packaging of the structure with the chip as the center for the target component, determine the multi-axis with multiple performance parameters, temperature and time, construct a thermal feedback map and perform dynamic drive training, wherein the thermal feedback map is built into the test system and establishes temporary communication with the analog circuit; the performance test result determination module 30 is used to dynamically update the thermal feedback map by performing periodic simulation of the analog circuit, intervene in the target thermal feedback state, perform decomposition call and directional evaluation based on the thermal feedback map, and determine the performance test result, wherein the dynamic update includes direct update based on simulation data and indirect update based on non-simulation data.

[0088] The specific configuration of the analog circuit determination module 10 will be described in detail below. As described above, the analog circuit determination module 10 may further include: a non-target component circuit access position positioning unit for locating the circuit access position of the non-target component in the access circuit including at least one target component, wherein the input-output direction of each non-target component at the circuit access position is determined according to the circuit flow direction; a linear state relationship mining unit for mining the linear state relationship of each non-target component based on input-output requirements; and a black box structure replacement unit for performing black box structure replacement of the non-target component on the access circuit according to the circuit access position and the linear state relationship.

[0089] The specific configuration of the thermal feedback map construction module 20 will be described in detail below. As described above, the abstract surface of the structure packaged layer by layer is determined with the chip as the center. The thermal feedback map construction module 20 can further include: a multi-layer structure surface determination unit for abstractly displaying the component structure from the center to the boundary, with the chip as the center and the outermost package layer as the boundary, to determine the multi-layer structure surface; and a multi-layer package characteristic location unit for locating the multi-layer package characteristics with respect to the multi-layer structure surface, marking the multi-layer structure surface, and determining the abstract surface.

[0090] Among them, a thermal feedback map is constructed, and the thermal feedback map construction module 20 may further include: multiple performance parameter determination units for determining multiple performance parameters, wherein the multiple performance parameters define all performance factors of the target component; multiple performance axis determination units for determining multiple performance axes for the multiple performance parameters; a thermal feedback map determination unit for coupling the abstract surface, temperature axis, time axis and the multiple performance axes to determine the thermal feedback map.

[0091] Among them, the abstract surface, temperature axis, time axis and the multiple performance axes are coupled, and the thermal feedback map determination unit may further include: a performance-related variable determination subunit for determining performance-related variables for the multiple performance parameters; a correlation coupling establishment subunit for establishing correlation coupling between each performance axis and the abstract surface, temperature axis and time axis based on the performance-related variables.

[0092] Among them, a thermal feedback map is constructed and dynamic drive training is performed. The thermal feedback map construction module 20 may further include: a dynamic drive training unit is used to perform dynamic drive training on the thermal feedback map with sample test tasks as constraints, sample circuit data as input, and the update of the thermal feedback map as output; a thermal feedback map embedding unit is used to embed the trained thermal feedback map into the test system if the convergence conditions are met, and establish temporary communication between the thermal feedback map and the analog circuit.

[0093] Among them, the thermal feedback map construction module 20 can further include: a switching analog circuit determination unit for determining the switching analog circuit as the target component to be tested is switched or the access circuit is switched; a switching analog circuit temporary communication establishment unit for interrupting the temporary communication between the thermal feedback map and the analog circuit, and establishing temporary communication between the switching analog circuit and the thermal feedback map.

[0094] The specific configuration of the performance test result determination module 30 will be described in detail below. As described above, the target thermal feedback state is intervened, and decomposition call and directional evaluation based on the thermal feedback map are performed. The performance test result determination module 30 may further include: a thermal feedback state setting unit for setting the thermal feedback state, wherein the thermal feedback state at least includes temperature cycling, constant high temperature, thermal stress accumulation under dynamic temperature, thermal stability, mechanical strength and packaging reliability, and the thermal stability is based on stability over a wide temperature range; a thermal feedback map updating unit for updating the thermal feedback map in real time through the analog circuit, wherein the updating includes direct updating of simulation data and indirect updating of performance parameters; and a directional evaluation unit for determining the target thermal feedback state and performing directional evaluation on the thermal feedback map updated in real time.

[0095] Among them, to determine the performance test result, the performance test result determination module 30 may further include: a directional evaluation result determination unit for determining the directional evaluation result based on the target thermal feedback state; a performance test result determination unit for determining the performance test result based on the directional evaluation result and the thermal feedback map updated in real time.

[0096] The power component performance testing system based on thermal feedback topology provided by the embodiment of the present invention can execute the power component performance testing method based on thermal feedback topology provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0097] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other and are not used to limit the scope of protection of the present invention.

[0098] The above specific embodiments do not constitute a limitation to the scope of protection of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements and improvements made within the spirit and principles of this application should be included in the scope of protection of this application. In some cases, the actions or steps recorded in this application can be performed in an order different from that in the embodiments and can still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A power component performance testing method based on thermal feedback topology, characterized in that: The method comprises: Obtaining the access circuit of the target component, performing circuit scenario simulation, and determining the simulated circuit, wherein non-target components of the access circuit are replaced with a black box structure; For the target components, the abstract surface of the structure under layer-by-layer packaging is determined with the chip as the center, and the multi-axis is determined based on multiple performance parameters, temperature and time. A thermal feedback map is constructed and dynamic drive training is performed, wherein the thermal feedback map is built into the test system and establishes temporary communication with the simulation circuit; Dynamically updating the thermal feedback map by performing periodic simulation of the analog circuit, intervening in a target thermal feedback state, performing decomposition call and directional evaluation based on the thermal feedback map, and determining performance test results, wherein the dynamic update includes direct update based on analog data and indirect update based on non-analog data; Wherein, the access circuit includes at least one target component; For the access circuit, locating the circuit access position of the non-target components, wherein the input-output direction of each non-target component at the circuit access position is determined according to the circuit power flow direction; Guided by input-output requirements, the linear state relationship of non-target components is mined; performing a black box structure replacement of non-target components on the access circuit according to the relationship between the circuit access position and the linear state; The abstract aspects of the chip-centric structure and layer-by-layer packaging include: With the chip as the center and the outermost package layer as the boundary, the component structure from the center to the boundary is abstractly displayed to determine the multi-layer structure surface; Locating multi-layer packaging characteristics for the multi-layer structural surface, marking the multi-layer structural surface, and determining the abstract surface; Among them, constructing a thermal feedback map includes: Determining a plurality of performance parameters, wherein the plurality of performance parameters define all performance factors of the target component; determining a plurality of performance axes for the plurality of performance parameters; The abstract surface, the temperature axis, the time axis and the plurality of performance axes are coupled to determine the thermal feedback map.

2. The power component performance testing method based on thermal feedback topology according to claim 1, characterized in that: Coupling the abstract surface, the temperature axis, the time axis, and the multiple performance axes includes: determining performance-related variables for the plurality of performance parameters; The performance-related variables are used to establish correlation coupling between each performance axis and the abstract surface, temperature axis, and time axis.

3. The power component performance testing method based on thermal feedback topology according to claim 1, characterized in that: Build a thermal feedback map and conduct dynamic driving training, including: Dynamically driving training the thermal feedback map is performed using a sample test task as a constraint, sample circuit data as input, and an update of the thermal feedback map as output; If the convergence condition is met, the trained thermal feedback map is embedded and deployed in the test system, and temporary communication between the thermal feedback map and the analog circuit is established.

4. The power component performance testing method based on thermal feedback topology according to claim 1, characterized in that: Determining a switching simulation circuit along with the switching of the target component to be tested or the switching of the access circuit; Temporary communication between the thermal feedback map and the analog circuit is interrupted, and temporary communication between the switching analog circuit and the thermal feedback map is established.

5. The power component performance testing method based on thermal feedback topology according to claim 1, characterized in that: Intervening in the target thermal feedback state, performing decomposition call and directional evaluation based on the thermal feedback map, including: Setting a thermal feedback state, wherein the thermal feedback state at least includes temperature cycling, constant high temperature, thermal stress accumulation under dynamic temperature, thermal stability, mechanical strength and packaging reliability, wherein the thermal stability is based on stability over a wide temperature range; updating the thermal feedback map in real time through the analog circuit, including direct updating of analog data and indirect updating of performance parameters; The target thermal feedback state is determined, and a directional evaluation is performed on the thermal feedback map updated in real time.

6. The power component performance testing method based on thermal feedback topology according to claim 5, characterized in that: Determine performance test results, including: determining a directional assessment result based on the target thermal feedback state; The performance test result is determined based on the directional evaluation result and the thermal feedback map updated in real time.

7. A power component performance test system based on thermal feedback topology, characterized in that: The system is used to implement the power component performance testing method based on thermal feedback topology according to any one of claims 1 to 6, and the system includes: An analog circuit determination module is used to obtain the access circuit of the target component, perform circuit scenario simulation, and determine the analog circuit, wherein non-target components of the access circuit are replaced with a black box structure; A thermal feedback map construction module is used to determine the abstract surface of the structure package layer by layer with the chip as the center for the target component, determine the multi-axis with multiple performance parameters, temperature and time, construct a thermal feedback map and perform dynamic drive training, wherein the thermal feedback map is built into the test system and establishes temporary communication with the simulation circuit; The performance test result determination module is used to dynamically update the thermal feedback map by performing periodic simulation of the analog circuit, intervene in the target thermal feedback state, perform decomposition call and directional evaluation based on the thermal feedback map, and determine the performance test result, wherein the dynamic update includes direct update based on simulation data and indirect update based on non-simulation data.

Images