A temperature control method for an intelligent power module based on a gallium nitride power chip

Abstract

This invention discloses a temperature control method for an intelligent power module based on a gallium nitride power chip. This method relates to the field of module control technology and includes collecting temperature data from the intelligent power module through distributed temperature sensors for preprocessing, dynamically adjusting the power path of the intelligent power module based on the temperature data, using an intelligent algorithm to predict future temperatures based on the adjusted power path of the intelligent power module, and making adjustments based on the predicted results, optimizing the intelligent power module cooling system based on the temperature data and the predicted results, and displaying the temperature data and power path in real time and storing them in a database. This invention ensures the safe operation of the intelligent power module and achieves precise power adjustment. Furthermore, by predicting future temperatures, it verifies the power adjustment effect and optimizes the cooling system, further improving the accuracy and flexibility of power adjustment.

Description

Technical Field

[0001] The present invention relates to the field of module control technology, and in particular to a temperature control method for an intelligent power module based on a gallium nitride power chip. Background Art

[0002] With the rapid development of power electronics technology, power modules, as one of the core components of modern electronic systems, play a vital role in efficient energy conversion, power management, and device operational stability. Gallium nitride (GaN) power chips, with their superior properties of high electron mobility, low switching losses, and high breakdown electric field, are gradually replacing traditional silicon-based power chips and becoming the mainstream technology in high-power density and high-frequency applications. However, in high-power density environments, thermal management issues within power module devices are becoming increasingly prominent. Excessively high operating temperatures not only significantly reduce device efficiency but can also cause performance degradation or even permanent damage. Current technologies are focusing on multi-point temperature acquisition and power distribution optimization, which have gradually become research hotspots. However, most existing technologies focus on improvements in a single dimension and have significant limitations in highly dynamic and high-frequency operating environments. They cannot balance accurate temperature management with energy efficiency optimization. Furthermore, the temperature control system suffers from delayed response, limited adjustment methods, and insufficient path selection flexibility, seriously affecting the stability and safety of system operation. Summary of the Invention

[0003] In view of the problems existing in the above-mentioned existing temperature control method of the intelligent power module based on the gallium nitride power chip, the present invention is proposed.

[0004] Therefore, the problem to be solved by the present invention is that the existing technology cannot take into account both the accuracy of temperature management and energy efficiency optimization, and the temperature control system has a delayed response, a single adjustment method, and insufficient flexibility in path selection.

[0005] To solve the above technical problems, the present invention provides the following technical solutions: a temperature control method for an intelligent power module based on a gallium nitride power chip, comprising: collecting temperature data of the intelligent power module through distributed temperature sensors for preprocessing, and dynamically adjusting the power path of the intelligent power module according to the temperature data; using an intelligent algorithm to predict future temperatures based on the adjusted power path of the intelligent power module, and making adjustments based on the prediction results; optimizing the cooling system of the intelligent power module based on the temperature data and the prediction results; and displaying the temperature data and power path in real time and storing them in a database.

[0006] As a preferred solution of the temperature control method of the intelligent power module based on the gallium nitride power chip described in the present invention, the preprocessing of the temperature data of the intelligent power module collected by distributed temperature sensors refers to deploying thermocouple sensors on the surface and heat dissipation port of the gallium nitride power chip in the intelligent power module, connecting all thermocouple sensors using the CAN bus protocol to form a sensor network, and connecting the sensor network to the CPU to transmit temperature data. The CPU cleans and filters the temperature data collected by the thermocouple sensors, calculates the temperature gradient, uses a bilinear interpolation algorithm to calculate the temperature distribution of the intelligent power module to form a temperature distribution map, and the CPU synchronously collects design parameters and operating parameters of the intelligent power module.

[0007] As a preferred solution of the temperature control method of the intelligent power module based on the gallium nitride power chip of the present invention, wherein: the dynamic power path adjustment of the intelligent power module according to the temperature data refers to identifying all power paths according to the design parameters of the intelligent power module, and defining the number of power paths in the intelligent power module as N;

[0008] Use finite element analysis tools to establish a thermal model based on the design parameters of the intelligent power module. Through simulation experiments, a known power is applied to power path j in the thermal model, the temperature change of power path i is recorded, and the ratio of the temperature change of power path i to the known power is calculated as the thermal impedance between power paths i and j. , traverse and calculate the thermal impedance between all power paths to form the thermal impedance matrix Z of the intelligent power module;

[0009] Calculate the real-time temperature change of power path k based on the operating parameters of the intelligent power module and the thermal impedance matrix :

[0010]

[0011] in is the thermal impedance between power path k and power path j, is the real-time power of power path j, N is the total number of power paths;

[0012] Calculate the DC current of power path k based on the real-time power and operating voltage of power path k and calculate the RMS current of power path k by the DC current :

[0013]

[0014] Where M is the calculation period and t is the time;

[0015] Calculate the skin depth of the power path k based on the design parameters of the power path , and obtain the DC resistance of the power path k , through the DC resistance and skin depth Calculate the AC resistance of the power path k ;

[0016] According to the DC resistance of the power path k , AC resistance , RMS current and DC current Calculate the equivalent resistance of the power path k :

[0017]

[0018] Get the equivalent resistance of the power path k After that, through the equivalent resistance and RMS current Calculate the power loss of power path k ;

[0019] The real-time power of all power paths of the intelligent power module is randomly initialized, and the power allocation schemes of multiple groups of initial power paths are randomly formed as the initial population. The sum of the power of all power paths in each initial population is equal to the total power of the intelligent power module;

[0020] Define the optimization objective function F for:

[0021]

[0022] in Design the maximum temperature for the intelligent power module, is the real-time power of power path k, is the temperature influence weight, is the loss weight;

[0023] Synchronously calculate the power path zero crossing frequency based on the equivalent inductance and capacitance of the power path k , define the optimization objective function F The constraints are:

[0024]

[0025] in is the current frequency of power path k, is the path frequency deviation threshold;

[0026] Calculate the optimization objective function F for each initial population Value, define the fitness value of the initial population as the optimization objective function F The roulette wheel method is used to select the initial population individuals, and the single point crossover method and mutation are used to generate a new population for initial population iteration until the objective function F is optimized. After convergence, the iteration is stopped, the fitness of the population after the iteration is calculated, and the population with the highest fitness is selected as the optimal population. The power allocation of each power path in the optimal population is extracted and applied to the intelligent power module.

[0027] As a preferred solution of the temperature control method of the intelligent power module based on the gallium nitride power chip described in the present invention, wherein: the use of an intelligent algorithm to predict the future temperature based on the power path adjusted by the intelligent power module refers to building a prediction model based on an artificial neural network (ANN) using a multi-layer perceptron architecture, defining the prediction model network model structure to include an input layer, a hidden layer, and an output layer, the input format of the input layer is the power distribution and temperature data of the power path of the intelligent power module, the output content of the output layer is the predicted temperature, historical data is used as a training set to train the prediction model, and mean square error is used as a loss function, and the Adam optimizer is used to iteratively optimize the prediction model parameters;

[0028] The power path power distribution and temperature data of the adjusted and optimized intelligent power module are input into the prediction model to obtain the future temperature data of the intelligent power module.

[0029] As a preferred solution of the temperature control method of the intelligent power module based on the gallium nitride power chip of the present invention, wherein: the temperature control adjustment of the intelligent power module based on the predicted future temperature data includes:

[0030] If the future temperature data contains temperature data that exceeds the maximum design temperature of the intelligent power module, the CPU determines that the power distribution adjustment of the intelligent power module path is unqualified and readjusts the power path of the intelligent power module;

[0031] If the temperature data in the future exceeds the intelligent power module temperature threshold, the cooling system is activated to dissipate heat for the intelligent power module;

[0032] If all future temperature data are lower than the intelligent power module temperature threshold, the power allocation of the intelligent power module power path is recorded and the operation is maintained.

[0033] As a preferred solution of the temperature control method of the intelligent power module based on the gallium nitride power chip of the present invention, wherein: the optimization of the intelligent power module cooling system by combining the temperature data and the prediction results refers to identifying the type of the intelligent power module cooling system and performing corresponding optimization according to the type of the cooling system, including:

[0034] If the cooling system is air-cooled, adjust the fan speed according to the power path temperature and ambient temperature. :

[0035]

[0036] in is the temperature of power path k, is the ambient temperature, is the temperature change rate, and is the adjustment coefficient;

[0037] If the cooling system is liquid cooling, adjust the coolant flow rate according to the GaN chip power and coolant temperature. :

[0038]

[0039] in is the GaN chip power, is the specific heat capacity of the coolant, and are the coolant outlet temperature and the coolant inlet temperature, respectively.

[0040] As a preferred solution of the temperature control method of the intelligent power module based on the gallium nitride power chip described in the present invention, the real-time display of temperature data and power path refers to the real-time display of the collected temperature data of the intelligent power module and the power distribution of the adjusted power path, and the visualization of the temperature distribution diagram of the future temperature predicted by the prediction model.

[0041] As a preferred solution of the temperature control method of the intelligent power module based on the gallium nitride power chip described in the present invention, the storage in the database refers to the CPU collecting temperature data and the power distribution of the adjusted power path and storing them in the database, and the database forms records according to timestamps for classified storage and sorting according to timestamps.

[0042] A computer device comprises: a memory and a processor; the memory stores a computer program, and the processor implements the steps of the temperature control method of the intelligent power module based on the gallium nitride power chip when executing the computer program.

[0043] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the temperature control method for an intelligent power module based on a gallium nitride power chip.

[0044] The beneficial effects of the present invention are as follows: by collecting temperature data and design operating parameters related to the gallium nitride chip in the intelligent power module, the present invention optimizes and adjusts the power path of the intelligent power module by calculating the impact of temperature changes and power loss, thereby ensuring the safe operation of the intelligent power module and achieving precise power adjustment. At the same time, by predicting future temperatures, the power adjustment effect and cooling system optimization are verified, further improving the accuracy and flexibility of power adjustment. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0046] Figure 1 The figure is a flow chart of a temperature control method for an intelligent power module based on a gallium nitride power chip. DETAILED DESCRIPTION

[0047] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0048] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0049] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it constitute a separate or selective embodiment that is mutually exclusive with other embodiments.

[0050] Example 1

[0051] Reference Figure 1 , which is the first embodiment of the present invention, provides a temperature control method for an intelligent power module based on a gallium nitride power chip. The temperature control method for an intelligent power module based on a gallium nitride power chip includes:

[0052] S1. Collect temperature data of the intelligent power module through distributed temperature sensors for preprocessing, and dynamically adjust the power path of the intelligent power module based on the temperature data;

[0053] Specifically, preprocessing the intelligent power module temperature data collected through distributed temperature sensors involves deploying thermocouple sensors on the surface and heat dissipation ports of the gallium nitride power chip in the intelligent power module. High-precision thin-film thermocouple sensors are suitable for high-temperature scenarios, have a quick response, and are deployed across key temperature gradient change areas, ensuring the comprehensiveness and accuracy of the measured data. The CAN bus protocol is used to connect all thermocouple sensors to form a sensor network, which is then connected to the CPU to transmit temperature data. The CAN bus is widely used in industrial environments, has high reliability and anti-interference capabilities, and can achieve stable data transmission in complex temperature control systems. The CPU cleans and filters the temperature data collected by the thermocouple sensors and calculates the temperature gradient. A bilinear interpolation algorithm is used to calculate the temperature distribution of the intelligent power module to form a temperature distribution map. The bilinear interpolation algorithm is simple and efficient, suitable for real-time processing, and can generate high-precision temperature field distributions when sensor density is limited. The CPU simultaneously collects the design and operating parameters of the intelligent power module, including resistance, rated power, structural parameters, operating current, operating voltage, and operating power.

[0054] By analyzing the temperature gradient, the high-temperature and cold areas within the module are identified, guiding the dynamic adjustment of the power path. The generated temperature distribution map provides input data for the prediction model of the intelligent algorithm, enhancing the accuracy of temperature control prediction. The module's operating current, voltage and other parameters are collected to provide comprehensive real-time information for dynamic power path adjustment. Combined with temperature data, it can achieve precise adjustment of the power path in a dynamic thermal environment. The introduction of design parameters such as rated power and structural parameters provides an important reference for the optimization of the cooling system.

[0055] Furthermore, dynamically adjusting the power path of the intelligent power module according to the temperature data refers to identifying all power paths according to design parameters of the intelligent power module, and defining the number of power paths in the intelligent power module as N;

[0056] Use finite element analysis tools to establish a thermal model based on the design parameters of the intelligent power module. Through simulation experiments, a known power is applied to power path j in the thermal model, the temperature change of power path i is recorded, and the ratio of the temperature change of power path i to the known power is calculated as the thermal impedance between power paths i and j. , traverse and calculate the thermal impedance between all power paths (including the power paths themselves) to form the thermal impedance matrix Z of the intelligent power module;

[0057] The thermal impedance matrix is ​​a mathematical tool that describes the heat flow transfer relationship between power paths. The elements in the matrix represent the impact of the power input of power path j on the temperature rise of power path i. The thermal impedance matrix can be used to quantify the degree of thermal coupling between power paths, providing theoretical support for dynamic temperature control. Finite element analysis is used to establish the thermal impedance matrix, quantifying the thermal coupling relationship between power paths, making temperature control strategies more precise. The thermal impedance matrix can intuitively reflect the contribution of power input to temperature rise, providing a scientific basis for dynamic adjustment of power paths.

[0058] Calculate the real-time temperature change of power path k based on the operating parameters of the intelligent power module and the thermal impedance matrix :

[0059]

[0060] in is the thermal impedance between power path k and power path j, extracted from the thermal impedance matrix Z, is the real-time power of power path j, extracted from the operating parameters, where j can be equal to k and N is the total number of power paths;

[0061] Calculating the real-time temperature changes of the power path based on the thermal impedance matrix and operating parameters enables dynamic monitoring of path temperature rise and uses the real-time path temperature to adjust the power allocation plan, effectively avoiding local overheating problems.

[0062] Calculate the DC current of power path k based on the real-time power and operating voltage of power path k and calculate the RMS current of power path k by the DC current :

[0063]

[0064] Where M is the calculation period and t is the time. The current characteristics of the power path are accurately evaluated by calculating the DC current and the RMS current.

[0065] Calculate the skin depth of the power path k based on the design parameters of the power path :

[0066]

[0067] in is the power path material resistivity, is the magnetic permeability of the power path conductor, is the power path current frequency. Skin depth is a measure of the uneven distribution of AC current in a conductor, reflecting the degree to which the current is concentrated on the conductor surface. Skin depth calculation and AC resistance analysis comprehensively consider the electrical loss of the path under high-frequency operation.

[0068] And get the DC resistance of the power path k , through the DC resistance and skin depth Calculate the AC resistance of the power path k :

[0069]

[0070] in is the wire length of the power path k, is the conductor cross-sectional width of power path k;

[0071] According to the DC resistance of the power path k , AC resistance , RMS current and DC current Calculate the equivalent resistance of the power path k :

[0072]

[0073] Get the equivalent resistance of the power path k After that, through the equivalent resistance and RMS current Calculate the power loss of power path k ;

[0074] The real-time power of all power paths of the intelligent power module is randomly initialized, and the power allocation schemes of multiple groups of initial power paths are randomly formed as the initial population. The sum of the power of all power paths in each initial population is equal to the total power of the intelligent power module;

[0075] Define the optimization objective function F for:

[0076]

[0077] in Design the maximum temperature for the intelligent power module and obtain it through design parameters. is the real-time power of power path k, is the temperature influence weight, is the loss weight, set through experience, and optimize the objective function F Taking the temperature effect and power loss into consideration, the optimal balance between temperature equilibrium and power loss can be achieved by optimizing the objective function;

[0078] Synchronously calculate the power path zero crossing frequency based on the equivalent inductance and capacitance of the power path k :

[0079]

[0080] in is the equivalent inductance of the power path k, is the equivalent capacitance of power path k. By calculating the zero-crossing frequency of the power path, high-frequency oscillation caused by path frequency mismatch is avoided. The constraint of the zero-crossing frequency ensures the stability and applicability of the optimization results, enhances the operational stability of the power path, and avoids performance degradation caused by high-frequency interference.

[0081] Define the optimization objective function F The constraints are:

[0082]

[0083] in is the current frequency of power path k, is the path frequency deviation threshold;

[0084] Calculate the optimization objective function F for each initial population Value, define the fitness value of the initial population as the optimization objective function F The roulette wheel method is used to select the initial population individuals, and the single point crossover method and mutation are used to generate a new population for initial population iteration until the objective function F is optimized. After convergence, the iteration is stopped, the fitness of the population after the iteration is calculated, and the population with the highest fitness is selected as the optimal population. The power allocation of each power path in the optimal population is extracted and applied to the intelligent power module.

[0085] By constructing a thermal impedance matrix, dynamically calculating the power path temperature and electrical parameters, and introducing an optimization objective function to achieve precise adjustment of the power path, the technical implementation combines thermal management and electrical characteristics analysis to significantly improve temperature control accuracy and system operation efficiency. In particular, in the iterative solution of the optimization objective function, temperature, loss and frequency constraints are comprehensively considered, making the power path allocation plan more scientific and reasonable.

[0086] S2. Using an intelligent algorithm to predict the future temperature based on the power path adjusted by the intelligent power module, and adjusting the temperature control of the intelligent power module according to the predicted future temperature data;

[0087] Specifically, using an intelligent algorithm to predict the future temperature based on the power path adjusted by the intelligent power module refers to building a prediction model based on an artificial neural network (ANN) using a multi-layer perceptron architecture, defining a prediction model network model structure including an input layer, a hidden layer, and an output layer. The input layer input format is the power distribution and temperature data of the power path of the intelligent power module, and the output layer outputs the predicted temperature. The prediction model is trained using historical data as a training set, and the mean square error is used as a loss function. The Adam optimizer is used to iteratively optimize the prediction model parameters.

[0088] The power path power distribution and temperature data of the adjusted and optimized intelligent power module are input into the prediction model to obtain the future temperature data of the intelligent power module.

[0089] The ANN-based multi-layer perceptron architecture can extract key features from complex power path distribution and temperature data to build a nonlinear prediction model. The input and output layers are clearly designed and can directly receive power path and temperature data and output future temperature prediction results to ensure smooth data flow. The optimized power path power distribution and real-time temperature data are input into the model, and the future temperature distribution is obtained through feedforward calculation to ensure the forward-looking nature of the power path distribution. The model can quickly respond to the adjusted power distribution plan, provide early warning of future temperature rise risks, and effectively reduce module losses caused by overheating. The predicted future temperature data can further adjust the power path distribution strategy to reduce the occurrence of local overheating.

[0090] Furthermore, the temperature control of the intelligent power module is adjusted according to the predicted future temperature data, including:

[0091] If future temperature data contains temperature data that exceeds the maximum design temperature of the intelligent power module, the CPU determines that the power distribution adjustment of the intelligent power module path is unqualified and readjusts the power path of the intelligent power module to redistribute the power path to ensure temperature balance and extend the service life of the module's internal components.

[0092] If the future temperature data exceeds the intelligent power module temperature threshold, the cooling system is activated to dissipate heat from the intelligent power module. By linking with future temperature data, the cooling system can intervene at the optimal time to reduce energy consumption.

[0093] If all future temperature data are lower than the intelligent power module temperature threshold, the power distribution of the intelligent power module power path is recorded and kept running. The recorded power path distribution data provides a reference for subsequent optimization and facilitates analysis of the optimal power distribution mode.

[0094] In the present invention, the temperature threshold is set to 80% of the designed maximum temperature of the intelligent power module. Under normal circumstances, the electronic components will be designed to have a maximum temperature during operation. When the maximum temperature is triggered, the system will shut down to protect the safety of the electronic components. A cooling temperature threshold will also be set. When the electronic components reach the cooling temperature threshold during operation, the cooling system will be turned on for cooling. Setting the temperature threshold to 80% of the designed maximum temperature can prevent the temperature of the intelligent power module from exceeding the upper temperature limit during operation, causing internal damage, and can also keep the cooling system inoperative when the temperature threshold is not reached to save energy.

[0095] Through future temperature prediction, it is possible to promptly discover unreasonable aspects of the power path allocation scheme, avoid thermal runaway caused by local power overload, prevent the module operating temperature from approaching the designed maximum temperature, improve the safety margin of system operation, establish a data foundation for the long-term operation of the intelligent power module, and further optimize the power path design by analyzing and recording data.

[0096] S3, optimize the intelligent power module cooling system by combining temperature data and prediction results;

[0097] Specifically, optimizing the intelligent power module cooling system by combining temperature data and prediction results means identifying the type of intelligent power module cooling system. The type of cooling system can be identified based on the design information and operating status of the intelligent power module, and corresponding optimization can be performed based on the type of cooling system, including:

[0098] If the cooling system is air-cooled, adjust the fan speed according to the power path temperature and ambient temperature. :

[0099]

[0100] in is the temperature of power path k, is the ambient temperature, is the temperature change rate, and Dynamically adjusts the fan speed to adjust the coefficient, quickly responding to changes in power path temperature to avoid local overheating. When the temperature rise rate is low or the ambient temperature is low, energy-saving operation is achieved by reducing the fan speed, extending the service life of the air cooling system. The dual-weighted design of temperature difference and change rate ensures the accuracy and efficiency of the cooling system response.

[0101] If the cooling system is liquid cooling, adjust the coolant flow rate according to the GaN chip power and coolant temperature. :

[0102]

[0103] in is the GaN chip power, is the specific heat capacity of the coolant, and The coolant outlet temperature and coolant inlet temperature are respectively used to dynamically adjust the coolant flow rate, optimize the heat dissipation path according to the actual heat generation of the power load, and adjust the flow rate according to the real-time heat demand to avoid system efficiency loss caused by overcooling or overheating. The high specific heat capacity characteristics of the coolant can efficiently remove a large amount of heat, ensuring the reliable operation of the gallium nitride chip under high power density.

[0104] Linked optimization significantly improves the overall efficiency of the cooling system and meets the thermal management requirements under complex working conditions. Real-time monitoring of the CPU ensures rapid response and precise implementation of cooling system adjustments, avoids waste of cooling resources, and achieves energy conservation and consumption reduction through dynamic optimization.

[0105] S4. Display the temperature data and power path in real time and store them in the database;

[0106] Specifically, real-time display of temperature data and power paths refers to real-time display of collected temperature data of intelligent power modules and power distribution of adjusted power paths, and visual display of future temperatures predicted by the prediction model into a temperature distribution graph.

[0107] Furthermore, storing in a database means that the CPU collects temperature data and the power distribution of the adjusted power path and stores them in the database. The database forms records according to timestamps, stores them in categories, and sorts them according to timestamps.

[0108] Example 2

[0109] This embodiment is different from the previous embodiment. In order to verify the effect of the method of the present invention, experimental data is presented for demonstration. The specific process is as follows:

[0110] The experiment involved a certain brand of gallium nitride chip power module with a designed maximum temperature of 125°C, four power paths, an air cooling system, a rated voltage of 48V, a total module power of 100W, and an ambient temperature of 20°C. The real-time temperature and power of the four power paths were collected using thermocouple sensors, as shown in the following table:

[0111]

[0112] The thermal impedance matrix Z is obtained by simulation calculation through finite element analysis:

[0113]

[0114] The temperature change of each power path can be calculated using the thermal impedance matrix and real-time power using the following formula:

[0115]

[0116]

[0117]

[0118]

[0119] The rated currents of the power paths can be calculated from the power and voltage of the power paths:

[0120]

[0121] Set the sampling time to 10 and calculate the RMS current of the power path:

[0122]

[0123] The material resistivity of each power path is 1.7*10 -8 , the magnetic permeability of the wire is 4π*10 -7 , the current frequency is 10000, and the skin depth of the power path is calculated as follows:

[0124]

[0125] Calculate the AC resistance of each power path using skin depth:

[0126]

[0127]

[0128]

[0129]

[0130] Calculate the equivalent resistance from the AC resistance:

[0131]

[0132] The power loss is calculated by equivalent resistance:

[0133]

[0134] This gives the power loss table for each power path:

[0135]

[0136] Calculate the optimization objective function value of each power path:

[0137]

[0138] The initial population individuals before iteration are set as follows:

[0139]

[0140] The population and fitness after iteration are as follows:

[0141]

[0142] From the above calculation results, it can be obtained that population 4 has the highest fitness value and is the optimal population. The power allocation of each power path in the population is applied to the intelligent power module.

[0143] The present invention innovatively combines distributed temperature sensing, dynamic path optimization, intelligent algorithm prediction and cooling system collaboration to achieve precise temperature control of intelligent power modules. Experimental verification shows that the technical solution significantly reduces the operating temperature of the module and improves the operating efficiency. It has broad industrial application prospects. The application of this invention in the field of high-power density chips has important technical value and economic significance, and provides a new solution for intelligent thermal management technology.

[0144] If the functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage media include various media capable of storing program code, such as USB flash drives, mobile hard drives, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0145] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0146] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, and then editing, interpreting, or processing in another suitable manner as necessary, and then storing it in a computer memory.

[0147] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used: a discrete logic circuit having logic gate circuits for implementing logic functions on data signals, an application-specific integrated circuit having suitable combinational logic gate circuits, a programmable gate array (PGA), a field-programmable gate array (FPGA), etc.

[0148] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A temperature control method for an intelligent power module based on a gallium nitride power chip, characterized by: include, Distributed temperature sensors are used to collect temperature data from intelligent power modules for preprocessing, and dynamic power path adjustment is performed on the intelligent power modules based on the temperature data. Using intelligent algorithms to predict future temperatures based on the adjusted power path of the intelligent power module, and adjusting the temperature control of the intelligent power module based on the predicted future temperature data; Optimize the intelligent power module cooling system by combining temperature data and prediction results; Display temperature data and power paths in real time and store them in the database; Dynamically adjusting the power path of the intelligent power module according to the temperature data refers to identifying all power paths according to the design parameters of the intelligent power module and defining the number of power paths in the intelligent power module as N; Use finite element analysis tools to establish a thermal model based on the design parameters of the intelligent power module. Through simulation experiments, a known power is applied to power path j in the thermal model, the temperature change of power path i is recorded, and the ratio of the temperature change of power path i to the known power is calculated as the thermal impedance between power paths i and j. , traverse and calculate the thermal impedance between all power paths to form the thermal impedance matrix Z of the intelligent power module; Calculate the real-time temperature change of power path k based on the operating parameters of the intelligent power module and the thermal impedance matrix : in is the thermal impedance between power path k and power path j, is the real-time power of power path j, N is the total number of power paths; Calculate the DC current of power path k based on the real-time power and operating voltage of power path k and calculate the RMS current of power path k by the DC current : Where M is the calculation period and t is the time; Calculate the skin depth of the power path k based on the design parameters of the power path , and obtain the DC resistance of the power path k , through the DC resistance and skin depth Calculate the AC resistance of the power path k ; According to the DC resistance of the power path k , AC resistance , RMS current and DC current Calculate the equivalent resistance of the power path k : Get the equivalent resistance of the power path k After that, through the equivalent resistance and RMS current Calculate the power loss of power path k ; The real-time power of all power paths of the intelligent power module is randomly initialized, and the power allocation schemes of multiple groups of initial power paths are randomly formed as the initial population. The sum of the power of all power paths in each initial population is equal to the total power of the intelligent power module; Define the optimization objective function F for: in Design the maximum temperature for the intelligent power module, is the real-time power of power path k, is the temperature influence weight, is the loss weight; Synchronously calculate the power path zero crossing frequency based on the equivalent inductance and capacitance of the power path k ; Define the optimization objective function F The constraints are: in is the current frequency of power path k, is the path frequency deviation threshold; Calculate the optimization objective function F for each initial population Value, define the fitness value of the initial population as the optimization objective function F The roulette wheel method is used to select the initial population individuals, and the single point crossover method and mutation are used to generate a new population for initial population iteration until the objective function F is optimized. After convergence, the iteration stops, the fitness of the population after the iteration is calculated, and the population with the highest fitness is selected as the optimal population. The power allocation of each power path in the optimal population is extracted and applied to the intelligent power module; Optimizing the intelligent power module cooling system by combining temperature data and prediction results refers to identifying the type of intelligent power module cooling system and performing corresponding optimization according to the type of cooling system, including: If the cooling system is air-cooled, adjust the fan speed according to the power path temperature and ambient temperature. : in is the temperature of power path k, is the ambient temperature, is the temperature change rate, and is the adjustment coefficient; If the cooling system is liquid cooling, adjust the coolant flow rate according to the GaN chip power and coolant temperature. : in is the GaN chip power, is the specific heat capacity of the coolant, and are the coolant outlet temperature and the coolant inlet temperature, respectively.

2. The temperature control method of the intelligent power module based on the gallium nitride power chip according to claim 1, characterized in that: The preprocessing of the intelligent power module temperature data collected by distributed temperature sensors refers to deploying thermocouple sensors on the surface and heat dissipation port of the gallium nitride power chip in the intelligent power module, connecting all thermocouple sensors using the CAN bus protocol to form a sensor network, and connecting the sensor network to the CPU to transmit temperature data. The CPU cleans and filters the temperature data collected by the thermocouple sensors, calculates the temperature gradient, uses a bilinear interpolation algorithm to calculate the temperature distribution of the intelligent power module to form a temperature distribution map, and the CPU simultaneously collects design parameters and operating parameters of the intelligent power module.

3. The temperature control method of the intelligent power module based on the gallium nitride power chip according to claim 2, characterized in that: The method of using an intelligent algorithm to predict the future temperature based on the power path adjusted by the intelligent power module refers to constructing a prediction model based on an artificial neural network (ANN) using a multi-layer perceptron architecture, defining a prediction model network model structure including an input layer, a hidden layer, and an output layer, wherein the input format of the input layer is the power distribution and temperature data of the power path of the intelligent power module, and the output content of the output layer is the predicted temperature. The prediction model is trained using historical data as a training set, and a mean square error is used as a loss function. The prediction model parameters are iteratively optimized using an Adam optimizer. The power path power distribution and temperature data of the adjusted and optimized intelligent power module are input into the prediction model to obtain the future temperature data of the intelligent power module.

4. The temperature control method of the intelligent power module based on the gallium nitride power chip according to claim 3, characterized in that: The temperature control and adjustment of the intelligent power module according to the predicted future temperature data includes: If the future temperature data contains temperature data that exceeds the maximum design temperature of the intelligent power module, the CPU determines that the power distribution adjustment of the intelligent power module path is unqualified and readjusts the power path of the intelligent power module; If the temperature data in the future exceeds the intelligent power module temperature threshold, the cooling system is activated to dissipate heat for the intelligent power module; If all future temperature data are lower than the intelligent power module temperature threshold, the power allocation of the intelligent power module power path is recorded and the operation is maintained.

5. The temperature control method of the intelligent power module based on the gallium nitride power chip according to claim 4, characterized in that: The real-time display of temperature data and power paths refers to real-time display of the collected temperature data of the intelligent power module and the power distribution of the adjusted power paths, and visual display of the future temperature predicted by the prediction model into a temperature distribution graph.

6. The temperature control method of the intelligent power module based on the gallium nitride power chip according to claim 5, characterized in that: The storing in the database means that the CPU collects temperature data and the power distribution of the adjusted power path and stores them in the database. The database forms records according to timestamps, stores them in categories, and sorts them according to timestamps.

7. A computer device comprising: memory and processor; The memory stores a computer program, and is characterized in that when the processor executes the computer program, the steps of the temperature control method of the intelligent power module based on the gallium nitride power chip according to any one of claims 1 to 6 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the temperature control method of the intelligent power module based on the gallium nitride power chip according to any one of claims 1 to 6 are implemented.

Images