A surface temperature detection method for a chip-level GaAs power device

Abstract

The application discloses a kind of surface temperature detection methods of chip-level GaAs power device, belong to semiconductor device temperature detection technical field.The method includes the following steps: obtaining the device surface temperature image under working condition, according to surface temperature image identification several virtual heat source points of device;Several thermocouples are arranged on the surface of the device;The temperature image of the device surface is obtained under different ambient temperature, and the first test temperature of the virtual heat source point is calculated according to the temperature image, while the second test temperature of the virtual heat source point is obtained by thermocouple;According to the ambient surface temperature difference and the test temperature difference, the temperature compensation model of each virtual heat source point is fitted;The working temperature data of ambient temperature and virtual heat source point are acquired in real time, and the working temperature data is corrected according to ambient temperature and temperature compensation model.The working temperature data collected by thermocouple is compensated in real time by ambient temperature and temperature compensation model, and high-precision device surface temperature detection is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of semiconductor device temperature detection, in particular to a surface temperature detection method for a chip-level GaAs power device. BACKGROUND

[0002] Power devices are widely used in various electronic equipment, and their performance (such as on-resistance, current capacity, switching speed, etc.) and operating environment (such as working temperature, voltage, maximum current, etc.) are important factors to determine their final function and efficiency. Since semiconductor power devices generate a large amount of heat during operation, good thermal design and thermal management are needed to ensure normal and stable operation and improve their service life.

[0003] Among them, GaAs semiconductor has high electron mobility and saturated electron flow speed, so that the chip-level GaAs power device is superior to many other types of semiconductor power devices, such as silicon semiconductor power devices, in terms of operating speed and frequency. Although the working performance of GaAs power devices is excellent, the demand for energy conversion will cause more heat generation, so accurate temperature detection is needed to provide data support for the thermal management of power devices.

[0004] On the one hand, using a thermocouple to measure the surface temperature of a semiconductor power device is one of the commonly used temperature measurement methods. However, since the thermocouple is a contact type detection method, this method is prone to cause large detection errors. At the same time, the thermocouple measures the single-point temperature of the device surface, and its fixed method and position as well as the environmental temperature will all affect the measurement accuracy, so the use of a thermocouple to measure the temperature of a semiconductor power device needs to be calibrated with high precision to ensure the accuracy of the measurement results.

[0005] On the other hand, an infrared thermal imager can determine the surface temperature of an object by capturing infrared photons emitted or reflected by a power device without the need for contact. The advantage of the thermal imager is that it can accurately measure the surface temperature of the power device and obtain the overall temperature distribution of the power device. This has high value for understanding the thermal behavior of the device, determining possible thermal problem areas, and evaluating heat dissipation designs. However, the infrared thermal imager is expensive, making this method not widely applicable to practical testing scenarios that require continuous detection of a large number of power devices. SUMMARY

[0006] To solve the problems in the background art, the present application adopts the following technical solutions:

[0007] The present application provides a surface temperature detection method for a chip-level GaAs power device, comprising the steps of:

[0008] acquire a device surface temperature image in a working state, and identify a plurality of virtual heat source points of the device according to the surface temperature image;

[0009] a plurality of thermocouples are arranged on the device surface, and the thermocouples are used to collect temperatures of the virtual heat source points; the number of the thermocouples is the same as that of the virtual heat source points;

[0010] acquire a device surface temperature image in a working state, and identify a plurality of virtual heat source points of the device according to the surface temperature image;

[0011] calculate a test temperature difference value according to the first test temperature and the second test temperature, calculate an ambient surface temperature difference value according to the second test temperature and the ambient temperature, and fit a temperature compensation model of each virtual heat source point according to the ambient surface temperature difference value and the test temperature difference value;

[0012] acquire real-time ambient temperature and working temperature data of the virtual heat source points, and correct the working temperature data according to the ambient temperature and the temperature compensation model.

[0013] As a preferred scheme of the present application, the identification of the plurality of virtual heat source points of the device according to the surface temperature image specifically includes the following steps:

[0014] convert the surface temperature image from a color image to a gray-scale image;

[0015] set a reference temperature, and draw a plurality of closed isothermals in the gray-scale image according to a gray-scale value corresponding to the reference temperature;

[0016] record a lowest point of the gray-scale value in the closed isotheral as a virtual heat source point.

[0017] As a preferred scheme of the present application, the reference temperature is expressed as:

[0018] T ref = k (T j,max -P × R jc ),

[0019] wherein T j,max is a standard maximum junction temperature, R jc is a standard thermal resistance, P is a working power of the device, and k is a proportional adjustment coefficient.

[0020] As a preferred scheme of the present application, the identification of the plurality of virtual heat source points of the device according to the surface temperature image further includes the following steps:

[0021] According to the temperature distribution gradient characteristics, a real heat source point in each closed isotheral line is identified; when the virtual heat source point does not coincide with the real heat source point and the virtual heat source point is on the line connecting the first real heat source point and the second real heat source point, the set virtual heat source point is deleted and the first real heat source point and the second real heat source point are respectively set as the virtual heat source point; wherein the first real heat source point is the lowest point of the real heat source point in the closed isotheral line in terms of gray value; and the second real heat source point is the second lowest point of the real heat source point in the closed isotheral line in terms of gray value.

[0022] As a preferred scheme of the present application, the temperature compensation model of each virtual heat source point is fitted according to the environment surface temperature difference and the test temperature difference, and specifically includes the following steps:

[0023] The environment surface temperature difference and the test temperature difference data are divided into a training set and a test set;

[0024] The training set data is fitted by a least square method to obtain a temperature compensation model; the temperature compensation model is an electrical parameter analysis model;

[0025] The test set data is used to test the temperature compensation model, and a mean square error index is used to evaluate the prediction accuracy of the model;

[0026] An error curve of the temperature compensation model is drawn, overfitting or underfitting phenomena in the training process of the temperature compensation model are identified through the error curve, and the temperature compensation model is optimized by increasing or decreasing the number of polynomials or the regularization method according to the phenomena; the error curve is used to represent the distribution of training errors and validation errors with the change of iteration times.

[0027] As a preferred scheme of the present application, the following steps are further included:

[0028] A heat-sensitive electrical parameter of the power device is obtained;

[0029] A plurality of electrical parameter analysis models are constructed, each of which corresponds to a virtual heat source point;

[0030] The heat-sensitive electrical parameter, the working temperature data and the heat-sensitive electrical parameter are used to train the electrical parameter analysis model;

[0031] The real-time obtained heat-sensitive electrical parameter and working temperature data are input into the trained electrical parameter analysis model to predict the temperature of the virtual heat source point, and when the ratio of the prediction result to the real-time obtained working temperature data of the virtual heat source point is greater than a preset threshold, it is determined that the power device has a fault.

[0032] As a preferred scheme of the present application, the electrical parameter analysis model is a polynomial regression model; the independent variables of the electrical parameter analysis model include the heat-sensitive electrical parameter and the ambient temperature; and the dependent variable of the electrical parameter analysis model is the working temperature data of the virtual heat source point.

[0033] As a preferred scheme of the present application, the temperature compensation value is expressed as:

[0034] T i,com = T i,1 - T i,2 ,

[0035] wherein i is a positive integer, representing the serial number of the virtual heat source point; T i,com is the temperature compensation value of the i th virtual heat source point, T i,1 is the first test temperature of the i th virtual heat source point, T i,2 is the second test temperature of the i th virtual heat source point.

[0036] The ambient surface temperature difference value is expressed as:

[0037] T i,def = T i,2 - T s ,

[0038] wherein T i,def represents the ambient surface temperature difference value of the i th virtual heat source point, and T s represents the ambient temperature.

[0039] As a preferred scheme of the present application, before converting the surface temperature image from a color image to a grayscale image, the method further comprises the step of: performing histogram equalization processing on the surface temperature image to obtain a surface temperature image with optimized brightness and contrast.

[0040] As a preferred scheme of the present application, before converting the surface temperature image from a color image to a grayscale image, the method further comprises the step of:

[0041] selecting the size of the filter window, sequentially traversing each pixel in the image, sorting all pixel values in the filter window, and taking the median value as the new value of the pixel point; the new value of the pixel point is expressed as:

[0042] P(x, y) = median{Q(x + i, y + j) | i, j ∈ W},

[0043] Wherein, P(x, y) represents a new value of a pixel point, x and y represent horizontal and vertical coordinates of the pixel point respectively, Q(x+i, y+j) represents a value of a corresponding pixel neighborhood around the pixel in the original image, W represents a filter window, i and j are parameters circulating in the filter window, and median{} represents a median operation of pixel values in the neighborhood.

[0044] Compared with the prior art, the present application has the following beneficial effects:

[0045] The present application detects the temperature of each virtual heat source point by means of a thermocouple to obtain a high-precision second test temperature, and especially for equipment that needs to work continuously for a long time, can provide a data basis for simulating and predicting the thermal behavior of the chip under varying working conditions, so as to discover possible faults early and take necessary heat dissipation or protection measures. Further, the present application acquires a first temperature image by means of a thermal imager, calculates a test temperature difference value according to the first test temperature and the second test temperature, and combines the environmental temperature factor to be used for fitting a temperature compensation model, so as to correct the second test temperature corresponding to each virtual heat source point collected by the thermocouple with high precision; after the correction is completed, the working temperature data collected by the thermocouple can be compensated in real time through the real-time acquired environmental temperature and the temperature compensation model, and further, higher-precision device surface temperature detection is realized.

[0046] The present application embodiment corrects the virtual heat source point located on the connecting line of the first real heat source point and the second real heat source point by identifying the real heat source point, so that the obtained virtual heat source point can accurately correspond to the first real heat source point and the second real heat source point in the closed isotherm, thereby improving the accuracy of temperature calculation, and only needs to calculate according to the obtained working temperature data, without understanding the specific physical properties or structure of the device, so the application cost is low and the range is wide.

[0047] The present application embodiment acquires real-time heat-sensitive electrical parameters and environmental and working temperature data, constructs an electrical parameter analysis model based on the data, and predicts the temperature of the virtual heat source point through the model. When the deviation between the predicted temperature and the actually acquired temperature exceeds a preset threshold, it can be judged in time that the power device has a fault, thereby improving the monitoring accuracy of the power device and the timeliness of fault detection, so as to achieve the purposes of improving the efficiency of the equipment and prolonging the service life of the equipment. BRIEF DESCRIPTION OF DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creating any inventive labor.

[0049] Figure 1 A flowchart of a surface temperature detection method of a chip-level GaAs power device according to an embodiment of the present application;

[0050] Figure 2 A flowchart of identifying a plurality of virtual heat source points of a device according to a surface temperature image according to an embodiment of the present application;

[0051] Figure 3 A flowchart of fitting a temperature compensation model of each virtual heat source point according to an environmental surface temperature difference and a test temperature difference according to another embodiment of the present application;

[0052] Figure 4 A flowchart of step S6 according to an embodiment of the present application. DETAILED DESCRIPTION

[0053] The technical solutions of the present application will be described clearly and completely below in conjunction with embodiments. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application. It should be noted that the relationship terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or sequence between the entities or operations.

[0054] Referring to Figure 1 The present application provides a surface temperature detection method of a chip-level GaAs power device, comprising the steps of:

[0055] S1, acquiring a device surface temperature image in a working state, and identifying a plurality of virtual heat source points of the device according to the surface temperature image; specifically, the surface temperature image is acquired by an infrared thermal imager.

[0056] S2, arranging a plurality of thermocouples on the device surface, the thermocouples being used to collect temperatures of the virtual heat source points; the number of the thermocouples is the same as the number of the virtual heat source points.

[0057] S3, acquiring temperature images of the device surface under different environmental temperatures, calculating first test temperatures of the virtual heat source points according to the temperature images, and simultaneously acquiring second test temperatures of the virtual heat source points by the thermocouples; wherein the different environmental temperatures are specifically a plurality of arithmetic sequences in a standard working environmental temperature range; the environmental temperatures are acquired by a temperature sensor integrated with the power device.

[0058] S4, calculating a test temperature difference value according to the first test temperature and the second test temperature, calculating an ambient surface temperature difference value according to the second test temperature and the ambient temperature, and fitting a temperature compensation model of each virtual heat source point according to the ambient surface temperature difference value and the test temperature difference value;

[0059] S5, acquiring the ambient temperature and the working temperature data of the virtual heat source point in real time, and correcting the working temperature data according to the ambient temperature and the temperature compensation model.

[0060] Specifically, the temperature compensation value is expressed as:

[0061] T i,com = T i,1 -T i,2 ,

[0062] wherein i is a positive integer, representing the serial number of the virtual heat source point; T i,com is the temperature compensation value of the i-th virtual heat source point, T i,1 is the first test temperature of the i-th virtual heat source point, and T i,2 is the second test temperature of the i-th virtual heat source point.

[0063] Specifically, the ambient surface temperature difference value is expressed as:

[0064] T i,def = T i,2 -T s ,

[0065] wherein T i,def represents the ambient surface temperature difference value of the i-th virtual heat source point, and T s represents the ambient temperature. It can be understood that the test temperature difference value between the surface temperature of the power device and its surrounding environment directly affects the heat dissipation speed of the object, therefore, by fitting the temperature compensation model according to the ambient surface temperature difference value and the test temperature difference value, the heat dissipation condition of the heat source can be accurately reflected, so that the actual temperature of the virtual heat source point of the device can be more accurately predicted through the temperature compensation model.

[0066] When a power device is in operation, it may contain heat sources of varying degrees. Existing technologies typically use a single thermocouple to detect the surface temperature of the device, obtaining only a single-point temperature on the device surface and failing to capture the real-time temperature of specific high-temperature points. This application uses thermocouples to detect the temperature of each virtual heat source point to obtain a high-precision second test temperature. Especially for devices requiring long-term continuous operation, this provides a data foundation for simulating and predicting the thermal behavior of chips under varying operating conditions, thereby enabling early detection of potential faults and the implementation of necessary heat dissipation or protection measures. Furthermore, this application uses a thermal imager to acquire a first temperature image. Based on the difference between the first and second test temperatures, it calculates the test temperature difference and combines it with the ambient temperature factor to fit a temperature compensation model, thereby performing high-precision correction on the second test temperature corresponding to each virtual heat source point acquired by the thermocouple. After correction, the operating temperature data acquired by the thermocouple can be compensated in real-time using the acquired ambient temperature and the temperature compensation model, further achieving higher-precision device surface temperature detection.

[0067] Further, please refer to Figure 2 The step of identifying several virtual heat source points based on surface temperature image recognition devices specifically includes the following steps:

[0068] S11. Convert the surface temperature image from a color image to a grayscale image. Grayscale images only contain information about grayscale changes, making them simpler to process than color images.

[0069] S12. Set a reference temperature and draw several closed isotherms in the grayscale image based on the grayscale value corresponding to the reference temperature.

[0070] S13. The point with the lowest gray value within the closed isotherm is recorded as the virtual heat source point; where the point with the lowest gray value is the point with the highest temperature in the region.

[0071] Specifically, the reference temperature is calculated using the standard maximum junction temperature and standard thermal resistance; the reference temperature is expressed as:

[0072] T ref =k(T) j,max -P×R jc ),

[0073] Among them, T j,max R is the standard maximum junction temperature. jc The standard thermal resistance is the highest junction temperature and standard thermal resistance provided in the product datasheet; P is the operating power of the device; k is the proportional control factor. Reference temperature T refThe reference temperature is a monitoring index set by the embodiment, and a position exceeding the reference temperature is regarded as a possible heat source. The proportional adjustment coefficient k is used to fine-tune the reference temperature in the embodiment. The setting of the proportional adjustment coefficient k affects the detection accuracy of the temperature and the search speed of the heat source, and the specific value can be set according to the prior knowledge of the technician and experimental feedback. In an embodiment, the value range is [0.5, 0.8].

[0074] It can be understood that the virtual heat source point searched in the embodiment is not necessarily the real heat source point corresponding to the working chip; for example, when two working chips are included in an isotherm, and both of the working chips are real heat source points, the virtual heat source point searched through step S13 may be one of the two real heat source points, or a point on the line connecting the two real heat source points. The specific position of the virtual heat source point is determined by the structure of the power device and the performance of the working chip. Therefore, in the embodiment, the smaller the coefficient k is, the farther the temperature corresponding to the drawn isotherm is from the maximum junction temperature, and thus the real heat source point with a lower temperature in the device can be more accurately included in the isotherm. However, a smaller value of the proportional adjustment coefficient k also increases the area for searching the virtual heat source point, which may result in the problem that more than one real heat source point is included in the isotherm, and thus the accuracy of the test result is reduced when the performance of one of the real heat source points deviates from the standard. To solve this problem, as a preferred embodiment, the identifying a plurality of virtual heat source points of the device according to the surface temperature image further includes the step of:

[0075] S14, identifying a real heat source point in each closed isotherm according to a temperature distribution gradient feature; when the virtual heat source point does not coincide with the real heat source point and the virtual heat source point is on the line connecting the first real heat source point and the second real heat source point, deleting the set virtual heat source point and setting the first real heat source point and the second real heat source point as the virtual heat source points, respectively.

[0076] The first real heat source point is the point with the lowest gray value among the real heat source points of the closed isotherm, and the second real heat source point is the point with the second lowest gray value among the real heat source points of the closed isotherm. The temperature distribution gradient represents the rate of change of the temperature in space. The temperature distribution gradient is obtained by calculating the temperature difference between each point and its surrounding points. The values of these temperature gradients are sorted, and the point with the maximum temperature gradient value (i.e., the most dramatic change in temperature) is selected as the real heat source point.

[0077] Further, before converting the surface temperature image from a color image to a grayscale image in the step S11, the method further comprises the step of: performing histogram equalization on the surface temperature image to obtain a surface temperature image with optimized brightness and contrast. The histogram is a statistical distribution of the gray levels of each pixel in the image. The histogram equalization is a method for enhancing the contrast of an image by stretching the histogram to make it more uniform, so that the brightness levels of the image are smoother.

[0078] Further, before converting the surface temperature image from a color image to a grayscale image in the step S11, the method further comprises the step of: performing median filtering on the surface temperature image. Specifically, first, the size of the filtering window is selected, and then each pixel in the image is iterated in turn, and all pixel values in the filtering window are sorted, and the median value is taken as the new value of the pixel. The new value of the pixel is denoted as:

[0079] P(x, y) = median{Q(x + i, y + j) | i, j ∈ W},

[0080] where P(x, y) represents the new value of the pixel, x and y represent the horizontal and vertical coordinates of the pixel, respectively, Q(x + i, y + j) represents the value of the neighborhood around the corresponding pixel in the original image, W represents the filtering window, i and j are parameters that cycle within the filtering window, and median{} represents the operation of taking the median of the pixel values in the neighborhood.

[0081] In this embodiment, histogram equalization and median filtering are performed on the surface temperature image before converting the surface temperature image from a color image to a grayscale image. The histogram equalization is used for image enhancement. In many cases, due to problems such as lighting and reflection, the dynamic range of the image can be limited, resulting in an image that is too dark or too bright. In this case, the histogram equalization can expand the dynamic range of the image and enhance the contrast of the image, making the detailed information in the image more obvious. At the same time, the histogram equalization can also make the brightness distribution of the image more uniform. The median filtering can effectively remove some random and isolated noise, such as salt and pepper noise. At the same time, compared with the average filter, the median filtering can better protect the edge information of the image while removing noise, avoiding excessive blurring of the image during the noise removal process.

[0082] The embodiment of the present application corrects the virtual heat source point on the line connecting the first real heat source point and the second real heat source point by identifying the real heat source point, so that the obtained virtual heat source point can accurately correspond to the first real heat source point and the second real heat source point in the closed isotherm, thereby improving the accuracy of temperature calculation, and only needs to calculate according to the obtained working temperature data, without the need to understand the specific physical properties or structure of the device, so the application cost is low and the range is wide. It should be noted that, since the multiple points in the closed isotherm have a greater influence on the virtual heat source point, the performance of one real heat source point deviates from the standard, and the influence on the test result is relatively smaller, and setting too many thermocouples in a closed isotherm will increase the real-time cost, therefore, when the virtual heat source point does not coincide with the real heat source point but the virtual heat source point is not on the line connecting the first real heat source point and the second real heat source point, the step S14 does not process the virtual heat source point for the comprehensive consideration between the detection accuracy and the cost of thermocouples.

[0083] Further, please refer to Figure 3 , the temperature compensation model of each virtual heat source point is fitted according to the environmental surface temperature difference and the test temperature difference, and specifically includes the following steps:

[0084] Divide the environmental surface temperature difference and the test temperature difference data into a training set and a test set;

[0085] Obtain the temperature compensation model by fitting the training set data by the least square method; the temperature compensation model is an electrical parameter analysis model;

[0086] Test the temperature compensation model using the test set data, and use the mean square error index to evaluate the prediction accuracy of the model;

[0087] Draw an error curve of the temperature compensation model, identify the overfitting or underfitting phenomenon in the training process of the temperature compensation model through the error curve, and optimize the temperature compensation model by increasing or decreasing the number of polynomials or the regularization method.

[0088] The error curve is used to represent the distribution of training error and validation error with the change of iteration number. If the training error and the validation error are both in the state of convergence and the gap between them is small, it means that the temperature compensation model is in a relatively optimal state; if the training error and the validation error are both in the state of convergence but the gap between them is large, it is judged that the temperature compensation model is overfitting; if the convergence degree of the training error and the validation error is low, it is judged that the temperature compensation model is underfitting.

[0089] The embodiment combines data set division, least square method modeling, model testing and evaluation, and model optimization, to construct and optimize the temperature compensation model of the device environment surface temperature and the test temperature difference. In the implementation process of the embodiment, first, the approximate relationship between the environment surface temperature difference and the test temperature difference is visually observed through the scatter plot, then the data is divided into training and test sets to effectively promote the generalization ability of the model and prevent overfitting. Next, the least square method is used to establish a preliminary model, and the mean square error is used to evaluate the prediction performance of the model. Finally, the overfitting or underfitting phenomenon of the model is identified through the training error curve, so that the temperature compensation model can accurately fit the environment surface temperature and the test temperature difference, thereby providing accurate and reliable prediction and evaluation for temperature compensation of different virtual heat source points, and further improving the efficiency and performance of the device.

[0090] Another embodiment of the present application provides a surface temperature detection method for a chip-level GaAs power device, which further comprises step S6 based on the above embodiment. Please refer to Figure 4 , the step S6 specifically comprises the following steps:

[0091] S61, while acquiring the environment temperature and the working temperature data of the virtual heat source point in real time in step S5, acquiring the heat-sensitive electrical parameters of the power device; taking an IGBT module as an example, the heat-sensitive electrical parameters at least include the threshold voltage, the saturation voltage and the forward conductance.

[0092] S62, constructing a plurality of electrical parameter analysis models, each of which corresponds to a virtual heat source point; wherein the electrical parameter analysis model is a polynomial regression model; the independent variables of the electrical parameter analysis model include the heat-sensitive electrical parameters and the environment temperature; the dependent variable of the electrical parameter analysis model is the working temperature data of the virtual heat source point.

[0093] S63, training the electrical parameter analysis model through the heat-sensitive electrical parameters, the working temperature data and the heat-sensitive electrical parameters;

[0094] S64, inputting the heat-sensitive electrical parameters and the working temperature data acquired in real time into the trained electrical parameter analysis model to predict the temperature of the virtual heat source point, and when the ratio of the prediction result to the working temperature data of the virtual heat source point acquired in real time is greater than a preset threshold, it is judged that the power device has a fault.

[0095] The embodiment obtains the thermal sensitive electrical parameters and the environment and working temperature data in real time, constructs an electrical parameter analysis model based on the data, and predicts the temperature of the virtual heat source point through the model. When the deviation between the predicted temperature and the actually obtained temperature exceeds a preset threshold, it can be determined in time that the power device has a fault, so as to improve the monitoring accuracy of the power device and the timeliness of fault detection, so as to improve the equipment efficiency and prolong the service life of the equipment.

[0096] In several embodiments provided in the present application, it should be understood that the disclosed system and method can be implemented in other manners. For example, the system embodiments described above are merely schematic. For example, the division of the modules is only a logical function division. There can be another division manner for actual implementation. For example, a plurality of modules or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections between different modules can be indirect couplings or communication connections through some interfaces, and electrical, mechanical or other forms.

[0097] The modules described as separated components can or can not be physically separated, and the components displayed as modules can or can not be physical modules, i.e., can be located in one place, or can be distributed on a plurality of network modules. Some or all of the modules can be selected according to actual needs to achieve the purposes of the embodiments.

[0098] In addition, each functional module in each embodiment of the present application can be integrated in one processing module, or each module can be physically present alone, or two or more modules can be integrated in one module. The integrated module can be realized in the form of hardware or in the form of a software functional module.

[0099] When the integrated module is realized in the form of a software functional module and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

Claims

1. A method for detecting surface temperature of a chip-level GaAs power device, characterized by: The method comprises the steps of: acquiring a device surface temperature image in a working state, and identifying a plurality of virtual heat source points of the device according to the surface temperature image; setting a plurality of thermocouples on the device surface, the thermocouples being used to collect temperatures of the virtual heat source points, and the number of the thermocouples being the same as that of the virtual heat source points; acquiring temperature images of the device surface under different environmental temperatures, calculating first test temperatures of the virtual heat source points according to the temperature images, and collecting second test temperatures of the virtual heat source points through the thermocouples; calculating a test temperature difference value according to the first test temperatures and the second test temperatures, calculating an environmental surface temperature difference value according to the second test temperatures and the environmental temperatures, and fitting a temperature compensation model of each virtual heat source point according to the environmental surface temperature difference value and the test temperature difference value; acquiring environmental temperature and working temperature data of the virtual heat source points in real time, and correcting the working temperature data according to the environmental temperature and the temperature compensation model; wherein the identification of the plurality of virtual heat source points of the device according to the surface temperature image comprises the steps of: converting the surface temperature image from a color image to a grayscale image; setting a reference temperature, and drawing a plurality of closed isothermals in the grayscale image according to a grayscale value corresponding to the reference temperature; recording a lowest point of a grayscale value in the closed isotheral as a virtual heat source point.

2. The surface temperature detecting method according to claim 1, characterized by: The reference temperature is expressed as: wherein, is the standard maximum junction temperature, is the standard thermal resistance; P is the operating power of the device; k is a proportional adjustment coefficient.

3. The surface temperature detecting method according to claim 1, characterized by: The identification of the plurality of virtual heat source points of the device according to the surface temperature image further comprises the steps of: identifying a real heat source point in each closed isothermal according to a temperature distribution gradient characteristic; when the virtual heat source point does not coincide with the real heat source point and the virtual heat source point is on a line connecting a first real heat source point and a second real heat source point, deleting the set virtual heat source point and setting the first real heat source point and the second real heat source point as the virtual heat source points respectively; wherein the first real heat source point is a lowest point of a grayscale value among the real heat source points of the closed isothermal, and the second real heat source point is a second lowest point of a grayscale value among the real heat source points of the closed isothermal.

4. The surface temperature detecting method according to claim 1, characterized by: The fitting of the temperature compensation model of each virtual heat source point according to the environmental surface temperature difference value and the test temperature difference value comprises the steps of: dividing the environmental surface temperature difference value and the test temperature difference value data into a training set and a test set; fitting the training set data by a least square method to obtain a temperature compensation model; the temperature compensation model is an electrical parameter analysis model; testing the temperature compensation model by using the test set data, and evaluating the prediction accuracy of the model by using a mean square error index; drawing an error curve of the temperature compensation model, identifying overfitting or underfitting phenomena in the training process of the temperature compensation model through the error curve, and optimizing the temperature compensation model by increasing or decreasing the number of polynomials or using a regularization method according to the phenomena; the error curve is used to represent the distribution of training errors and validation errors with the change of iteration numbers.

5. The surface temperature detecting method according to claim 1, characterized by: The method further comprises the steps of: acquiring heat-sensitive electrical parameters of the power device; constructing a plurality of electrical parameter analysis models, each of the electrical parameter analysis models corresponding to a virtual heat source point; training the electrical parameter analysis models by using the heat-sensitive electrical parameters, the working temperature data, and the heat-sensitive electrical parameters. The real-time acquired thermal sensitive electrical parameter and working temperature data are input into the trained electrical parameter analysis model to predict the temperature of the virtual heat source point, and when the ratio of the prediction result to the real-time acquired working temperature data of the virtual heat source point is greater than a preset threshold, it is determined that the power device has a fault.

6. The surface temperature detecting method according to claim 5, characterized by: The electrical parameter analysis model is a polynomial regression model; the independent variables of the electrical parameter analysis model include the thermal sensitive electrical parameter and the environmental temperature; and the dependent variable of the electrical parameter analysis model is the working temperature data of the virtual heat source point.

7. The surface temperature detecting method according to claim 1, characterized by: The test temperature difference is represented as: wherein i is a positive integer, representing the serial number of the virtual heat source point; is the test temperature difference of the i-th virtual heat source point, is the first test temperature of the i-th virtual heat source point, is the second test temperature of the i-th virtual heat source point; The ambient surface temperature difference is expressed as: wherein, represents the ambient surface temperature difference of the i-th virtual heat source point, represents the ambient temperature.

8. The surface temperature detection method according to claim 1, characterized by: Before the surface temperature image is converted from a color image to a gray scale image, the method further includes the step of: performing histogram equalization processing on the surface temperature image to obtain a surface temperature image with optimized brightness and contrast.

9. The surface temperature detecting method according to claim 1, characterized by: Before the surface temperature image is converted from a color image to a gray scale image, the method further includes the step of: The size of a filter window is selected, each pixel in the image is sequentially traversed, all pixel values in the filter window are sorted, and a middle value is taken as a new value of the pixel point; the new value of the pixel point is represented as: P(x,y) = median {Q(x+i,y+j) | i,j ∈ W}, wherein P(x,y) represents the new value of the pixel point, x and y represent the horizontal coordinate and the vertical coordinate of the pixel point respectively, Q(x+i, y+j) represents the value of a neighborhood around a corresponding pixel in the original image, W represents the filter window, i and j are parameters that are cycled in the filter window, and median {} represents a median operation on the pixel values in the neighborhood.

Images