A chip testing temperature control method and system

Abstract

The present invention relates to the field of chip testing, and specifically to a chip testing temperature control method and system. The present invention uses characteristic vector construction, machine learning algorithm training and other methods to establish a temperature prediction model, which can accurately predict the chip temperature change trend and improve the accuracy of the test results. After that, by setting a temperature difference threshold, the air flow temperature is adjusted in advance according to the temperature change trend to ensure that the temperature difference between any two hot spots on the chip does not exceed the set temperature difference threshold. While avoiding the chip performance degradation, failure or damage caused by the hot spot temperature exceeding the test temperature in a certain area of ​​the chip, the stability and reliability of the test are further improved.

Description

Technical Field

[0001] The present invention relates to the field of chip testing, and in particular to a chip testing temperature control method and system. Background Art

[0002] In modern electronic devices, the chip is the core component, and its performance directly affects the operation of the device. Therefore, strict temperature control of the chip is one of the key factors to ensure the stable operation of the device. However, in the actual chip testing process, due to changes in environmental conditions (such as temperature, humidity, etc.) and problems with the test equipment itself, the temperature of the chip may fluctuate, thus affecting the performance and life of the chip.

[0003] Traditional chip test temperature control systems usually rely on manual setting and adjustment of temperature. Although this method is simple and intuitive, it cannot accurately predict and respond to the impact of environmental changes. In addition, the manually adjusted temperature often cannot respond to the actual temperature changes of the chip in real time, resulting in low temperature control accuracy and possible damage to the chip due to excessive or insufficient heating or cooling.

[0004] Existing chip temperature control tests usually use high and low temperature airflows to control the chip temperature, such as patent application number: CN113484730A, patent name: Airflow temperature control method and system for chip testing. During the test, this method places the chip in a closed test chamber of a test machine. The test machine inputs high or low temperature airflows with different flow rates, causing the chip surface temperature to change dramatically, thereby simulating an extreme environment where the atmospheric temperature changes rapidly, and the chip is tested for adaptability through detection equipment.

[0005] However, the inventors have found that this type of method has the following problems during its specific implementation: when the chip is heated by high-temperature airflow, the temperature change on the chip surface is not consistent, which leads to a temperature difference inside the chip. Under the influence of the high-temperature airflow, before the overall temperature of the chip reaches the test temperature, the temperature of some areas of the chip will quickly reach and exceed the test temperature, which will cause the chip to overheat and cause chip performance degradation, failure or damage. If the airflow temperature used is too low, it will affect the chip temperature control test speed and reduce test efficiency. Summary of the invention

[0006] The technical problem of the present invention is that when the chip is heated by high-temperature airflow in the prior art, the temperature change on the chip surface is not consistent, which leads to a temperature difference inside the chip. Before the overall temperature of the chip reaches the test temperature, the temperature of some areas of the chip will quickly reach and exceed the test temperature.

[0007] The basic solution provided by the present invention is: a chip test temperature control system, including a data acquisition module, a temperature analysis module and a temperature difference control module;

[0008] The data acquisition module is used to collect chip temperature data and airflow temperature data;

[0009] The temperature analysis module is used to predict the chip temperature change trend through the chip temperature data and the airflow temperature data;

[0010] The temperature difference control module is provided with a temperature difference threshold value, which is used to control the airflow temperature according to the prediction result, and ensure that the temperature difference between any two hot spots on the chip does not exceed the set temperature difference threshold value.

[0011] Further, the data acquisition module includes a chip temperature acquisition module and an airflow temperature acquisition module;

[0012] The chip temperature acquisition module includes an infrared thermal imager and several on-chip temperature sensors. The infrared thermal imager is used to collect ambient temperature data, and the on-chip temperature sensors are used to collect hot spot temperature data of each area of ​​the chip; the airflow temperature acquisition module collects airflow temperature data through thermocouples.

[0013] Further, the temperature analysis module includes a feature vector construction module, a model training module and a temperature prediction module;

[0014] The feature vector construction module is used to construct a feature vector based on historical chip temperature data and airflow temperature data;

[0015] The model training module is used to train the feature vector through a machine learning algorithm to obtain a temperature prediction model;

[0016] The temperature prediction module is used to input the airflow temperature data and the chip temperature data into the temperature prediction model, and analyze the chip temperature change trend according to the output result of the temperature prediction model.

[0017] Further, the temperature difference control module includes a temperature difference analysis module and an airflow temperature adjustment module;

[0018] The temperature difference analysis module is provided with a temperature difference threshold, and analyzes the temperature difference of each hot spot according to the chip temperature change trend; when the temperature difference between any two hot spots reaches the temperature difference threshold before the overall temperature of the chip reaches the test temperature, the air flow temperature is adjusted by the air flow temperature adjustment module to ensure that the temperature difference between any two hot spots on the chip does not exceed the set temperature difference threshold.

[0019] Furthermore, the air flow temperature adjustment rule is:

[0020] During high-temperature testing, lower the airflow temperature to ensure that the temperature difference between any two hot spots on the chip does not exceed the set temperature difference threshold before the overall chip temperature reaches the test temperature;

[0021] During low-temperature testing, increase the air flow temperature to ensure that the temperature difference between any two hot spots on the chip does not exceed the set temperature difference threshold before the overall chip temperature reaches the test temperature.

[0022] A chip testing temperature control method comprises the following steps:

[0023] S1: Collect chip temperature data and airflow temperature data;

[0024] S2: predict the chip temperature change trend based on the airflow temperature data and chip temperature data;

[0025] S3: Control the air flow temperature according to the prediction result.

[0026] Further, the chip temperature data in S1 includes hotspot temperature data and ambient temperature data, and the airflow temperature data is the generated airflow temperature;

[0027] The hot spot temperature data is the temperature data of each area on the chip surface, and the ambient temperature data is the air temperature around the chip.

[0028] Further, the S2 comprises the steps of:

[0029] S201: constructing a feature vector according to historical chip temperature data and airflow temperature data;

[0030] S202: training the feature vector by a machine learning algorithm to obtain a temperature prediction model;

[0031] S203: inputting the airflow temperature data and the chip temperature data into a temperature prediction model;

[0032] S204: Analyze the chip temperature variation trend according to the output result of the temperature prediction model.

[0033] Further, the S3 comprises the steps of:

[0034] S301: Setting a temperature difference threshold;

[0035] S302: Analyze the temperature difference of each hot spot according to the chip temperature change trend;

[0036] S303: When the temperature difference between any two hot spots reaches the temperature difference threshold before the overall temperature of the chip reaches the test temperature, the air flow temperature is adjusted to ensure that the temperature difference between any two hot spots on the chip does not exceed the set temperature difference threshold.

[0037] Furthermore, the air flow temperature adjustment rule in S303 is:

[0038] During high-temperature testing, lower the airflow temperature to ensure that the temperature difference between any two hot spots on the chip does not exceed the set temperature difference threshold before the overall chip temperature reaches the test temperature;

[0039] During low-temperature testing, increase the air flow temperature to ensure that the temperature difference between any two hot spots on the chip does not exceed the set temperature difference threshold before the overall chip temperature reaches the test temperature.

[0040] The principles and advantages of the present invention are:

[0041] The present invention uses methods such as feature vector construction and machine learning algorithm training to establish a temperature prediction model, which can accurately predict the chip temperature change trend and improve the accuracy of the test results. After that, by setting the temperature difference threshold, the airflow temperature is adjusted in advance according to the temperature change trend, ensuring that the temperature difference between any two hot spots on the chip does not exceed the set temperature difference threshold, while avoiding the chip performance degradation, failure or damage caused by the hot spot temperature exceeding the test temperature in a certain area of ​​the chip, and further improving the stability and reliability of the test. Compared with the measures taken by conventional technical means to reduce the airflow temperature when the chip temperature reaches the test temperature, this solution adjusts the airflow temperature in advance through the temperature change trend, which can reduce the temperature difference change speed while also reducing the time required to reach the test temperature as much as possible. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 This is a logic block diagram of a first embodiment of a chip testing temperature control system of the present invention.

[0043] Figure 2 This is a flow chart of the steps of Embodiment 2 of a chip testing temperature control method of the present invention. DETAILED DESCRIPTION

[0044] The following is further described in detail through specific implementation methods:

[0045] The specific implementation process is as follows:

[0046] Embodiment 1

[0047] Embodiment 1 is basically as shown in the attached Figure 1 As shown, a chip test temperature control system includes a data acquisition module, a temperature analysis module, a high and low temperature heat flow meter and a temperature difference control module. The data acquisition module is used to collect chip temperature data and air flow temperature data; the temperature analysis module is used to predict the chip temperature change trend through the chip temperature data and air flow temperature data; the temperature difference control module is provided with a temperature difference threshold, which is used to control the air flow temperature of the high and low temperature heat flow meter according to the prediction result, and ensure that the temperature difference between any two hot spots on the chip does not exceed the set temperature difference threshold.

[0048] In an embodiment of the present invention, the test chip adopts the SA8155P chip, and the high and low temperature airflow is generated by the high and low temperature heat flow meter. The data acquisition module includes a chip temperature acquisition module and an airflow temperature acquisition module. The chip temperature acquisition module includes an infrared thermal imager and 28 on-chip temperature sensors. The infrared thermal imager is used to collect the ambient temperature data of the chip, while the on-chip temperature sensor is used to collect the hot spot temperature data of each area of ​​the chip. The airflow temperature acquisition module collects the airflow temperature of the high and low temperature heat flow meter through a thermocouple. Through the collaborative work of these sensors, comprehensive monitoring and recording of the chip temperature (the ambient temperature data of the chip and the hot spot temperature data of each area of ​​the chip) and the airflow temperature can be achieved.

[0049] Next, the temperature analysis module of the present invention includes a feature vector construction module, a model training module and a temperature prediction module. The feature vector construction module is used to construct feature vectors based on historical chip temperature data and airflow temperature data. The model training module uses a machine learning algorithm to train these feature vectors to obtain an accurate temperature prediction model. Finally, the temperature prediction module inputs the airflow temperature data and chip temperature data into the model and analyzes the chip temperature change trend based on the results output by the model.

[0050] In an embodiment of the present invention, the feature vector construction module uses the principal component analysis (PCA) method to reduce the high-dimensional feature vectors in the historical data set into low-dimensional feature vectors. Then, the model training module uses the support vector regression (SVR) algorithm to train the low-dimensional feature vectors to obtain an accurate temperature prediction model.

[0051] In addition, the temperature difference control module of the present invention is also provided with a temperature difference analysis module and an airflow temperature adjustment module. The temperature difference analysis module analyzes the temperature difference of each hot spot according to the chip temperature change trend. When it is found that the temperature difference between any two hot spots reaches the set temperature difference threshold, the airflow temperature adjustment module will adjust the airflow temperature in time to reduce the temperature difference and maintain the stability of the chip. Specifically, during low-temperature testing, increasing the airflow temperature can speed up the chip heating rate; while during high-temperature testing, lowering the airflow temperature can slow down the chip heating rate, thereby avoiding performance degradation or damage caused by overheating.

[0052] Embodiment 2

[0053] Embodiment 2 is basically as shown in the attached Figure 2 As shown, a chip testing temperature control method includes the steps of:

[0054] S1: Collect chip temperature data and airflow temperature data;

[0055] S2: predict the chip temperature change trend based on the airflow temperature data and chip temperature data;

[0056] S3: Control the air flow temperature according to the prediction result.

[0057] Specifically, this solution collects chip temperature data and airflow temperature data, predicts the chip temperature change trend based on the airflow temperature data and the chip temperature data, and controls the airflow temperature based on the prediction result.

[0058] In this embodiment, the chip temperature data and airflow temperature data are first collected. The chip temperature data includes hotspot temperature data and ambient temperature data. Among them, this embodiment collects the hotspot temperature data of each area of ​​the chip through 28 on-chip temperature sensors; and collects the air temperature around the chip, that is, the ambient temperature data, through an infrared thermal imager. The high and low temperature airflow in this scheme is generated by a high and low temperature heat flow meter, so this scheme uses a thermocouple to collect the airflow temperature data of the high and low temperature airflow generated by the high and low temperature heat flow meter.

[0059] Further, S2 comprises the steps of:

[0060] S201: constructing a feature vector according to historical chip temperature data and airflow temperature data;

[0061] S202: training the feature vector by a machine learning algorithm to obtain a temperature prediction model;

[0062] S203: inputting the airflow temperature data and the chip temperature data into a temperature prediction model;

[0063] S204: Analyze the chip temperature variation trend according to the output result of the temperature prediction model.

[0064] Specifically, this solution constructs a feature vector for training the machine learning model based on the collected chip temperature data and airflow temperature data. In this embodiment, the principal component analysis (PCA) method is used to construct the feature vector, and the feature vector of this embodiment includes the temperature data of each area on the chip surface, the ambient temperature data, and the temperature data of the airflow.

[0065] Then, the feature vector is trained using a machine learning algorithm to obtain a temperature prediction model. In this embodiment, a support vector regression (SVR) algorithm is selected as the machine learning algorithm for training. By inputting the feature vector into the trained support vector regression model, a more accurate prediction result of the chip temperature change trend can be obtained.

[0066] Next, the collected airflow temperature data and the predicted chip temperature change trend are input into the temperature prediction model for analysis. By analyzing the prediction results, the temperature change trend of each hot spot can be determined.

[0067] Finally, according to the chip temperature change trend in the prediction results, the temperature difference threshold is set. When the temperature difference between any two hot spots reaches the set temperature difference threshold before the overall chip temperature reaches the test temperature, the air flow temperature is adjusted to ensure that the temperature difference between any two hot spots on the chip does not exceed the set threshold.

[0068] In this embodiment, different airflow temperature adjustment rules are proposed for high temperature test and low temperature test. In high temperature test, the temperature difference of the hot spot area on the chip surface is reduced by lowering the airflow temperature; and in low temperature test, the temperature difference of the hot spot area on the chip surface is reduced by increasing the airflow temperature.

[0069] The embodiment of the present invention provides a method for accurately and in real time adjusting the air flow temperature, which can effectively control the temperature variation range of the chip and improve the stability and reliability of the chip. In addition, the method also has the characteristics of self-adaptation and can automatically adjust according to the actual environment and conditions, further improving the efficiency and accuracy of the test.

[0070] In summary, the chip test temperature control system and method provided in this embodiment can accurately predict the chip temperature change trend and control the airflow temperature according to the prediction results. The temperature prediction model established by adopting methods such as feature vector construction and machine learning algorithm training, as well as the airflow temperature adjustment based on the temperature difference analysis results, can improve the accuracy and reliability of the test. Compared with the measures taken by conventional technical means to reduce the airflow temperature when the chip temperature reaches the test temperature, this embodiment adjusts the airflow temperature in advance according to the temperature change trend, which can reduce the temperature difference change rate while minimizing the time required to reach the test temperature.

[0071] The above are only embodiments of the present invention. Common knowledge such as the known specific structures and characteristics in the scheme is not described in detail here. Ordinary technicians in the relevant field know all the common technical knowledge in the technical field to which the invention belongs before the application date or priority date, can obtain all the existing technologies in the field, and have the ability to apply conventional experimental means before that date. Ordinary technicians in the relevant field can improve and implement this scheme in combination with their own abilities under the enlightenment given by this application. Some typical known structures or known methods should not become obstacles for ordinary technicians in the relevant field to implement this application. It should be pointed out that for those skilled in the art, without departing from the structure of the present invention, several deformations and improvements can be made, which should also be regarded as the scope of protection of the present invention, and these will not affect the effect of the implementation of the present invention and the practicality of the patent. The scope of protection required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. A chip testing temperature control system, characterized in that: It includes data acquisition module, temperature analysis module and temperature difference control module; The data acquisition module is used to collect chip temperature data and airflow temperature data; The data acquisition module includes a chip temperature acquisition module and an airflow temperature acquisition module, and the chip temperature data includes ambient temperature data and hot spot temperature data; The chip temperature acquisition module includes an infrared thermal imager and a plurality of on-chip temperature sensors. The infrared thermal imager is used to collect ambient temperature data, and the on-chip temperature sensors are used to collect hot spot temperature data of each area of ​​the chip. The airflow temperature acquisition module collects airflow temperature data through thermocouples. The temperature analysis module is used to predict the chip temperature change trend through the chip temperature data and the airflow temperature data; The temperature difference control module includes a temperature difference analysis module and an airflow temperature adjustment module; The temperature difference analysis module is provided with a temperature difference threshold, and analyzes the temperature difference of each hot spot according to the chip temperature change trend; when the temperature difference between any two hot spots reaches the temperature difference threshold before the overall temperature of the chip reaches the test temperature, the air flow temperature is adjusted by the air flow temperature adjustment module to ensure that the temperature difference between any two hot spots on the chip does not exceed the set temperature difference threshold.

2. A chip testing temperature control system according to claim 1, characterized in that: The temperature analysis module includes a feature vector construction module, a model training module and a temperature prediction module; The feature vector construction module is used to construct a feature vector based on historical chip temperature data and airflow temperature data; The model training module is used to train the feature vector through a machine learning algorithm to obtain a temperature prediction model; The temperature prediction module is used to input the airflow temperature data and the chip temperature data into the temperature prediction model, and analyze the chip temperature change trend according to the output result of the temperature prediction model.

3. A chip testing temperature control system according to claim 1, characterized in that: The air flow temperature regulation rules are: During high-temperature testing, lower the airflow temperature to ensure that the temperature difference between any two hot spots on the chip does not exceed the set temperature difference threshold before the overall chip temperature reaches the test temperature; During low-temperature testing, increase the air flow temperature to ensure that the temperature difference between any two hot spots on the chip does not exceed the set temperature difference threshold before the overall chip temperature reaches the test temperature.

4. A chip testing temperature control method, characterized in that: Includes steps: S1: Collect chip temperature data and airflow temperature data; S2: predict the chip temperature change trend based on the airflow temperature data and chip temperature data; S3: Control the air flow temperature according to the prediction results; The chip temperature data in S1 includes hotspot temperature data and ambient temperature data, and the airflow temperature data is the generated airflow temperature; The hot spot temperature data is the temperature data of each area on the chip surface, and the ambient temperature data is the air temperature around the chip; The S3 comprises the steps of: S301: Setting a temperature difference threshold; S302: Analyze the temperature difference of each hot spot according to the chip temperature change trend; S303: When the temperature difference between any two hot spots reaches the temperature difference threshold before the overall temperature of the chip reaches the test temperature, the air flow temperature is adjusted to ensure that the temperature difference between any two hot spots on the chip does not exceed the set temperature difference threshold.

5. A chip testing temperature control method according to claim 4, characterized in that: The S2 comprises the steps of: S201: constructing a feature vector according to historical chip temperature data and airflow temperature data; S202: training the feature vector by a machine learning algorithm to obtain a temperature prediction model; S203: inputting the airflow temperature data and the chip temperature data into a temperature prediction model; S204: Analyze the chip temperature variation trend according to the output result of the temperature prediction model.

6. A chip testing temperature control method according to claim 4, characterized in that: The air flow temperature adjustment rule in S303 is: During high-temperature testing, lower the airflow temperature to ensure that the temperature difference between any two hot spots on the chip does not exceed the set temperature difference threshold before the overall chip temperature reaches the test temperature; During low-temperature testing, increase the air flow temperature to ensure that the temperature difference between any two hot spots on the chip does not exceed the set temperature difference threshold before the overall chip temperature reaches the test temperature.

Images