A non-contact measurement method for dust distribution

Abstract

The present invention provides a non-contact measurement method for dust distribution, including: obtaining an image of the surface of an item to be measured; sampling the image of the surface of the item to be measured and obtaining the gray-scale values of each sampling point; comparing the gray-scale values of the sampling points with the gray-scale standard values in the dust accumulation thickness standard database, and determining the dust distribution on the surface of the item to be measured according to the comparison result; wherein, the dust accumulation thickness standard database is a pre-established database that stores gray-scale standard values corresponding to different thicknesses of dust. The non-contact measurement method of the present invention can obtain the dust distribution on the surface of the chip to be measured, provide data support for timely cleaning the surface of the chip, and ensure the stable and efficient operation of the data center.

Description

Technical Field

[0001] The invention relates to the field of energy science, and in particular to a non-contact measurement method for dust distribution. Background Art

[0002] Since the beginning of the 21st century, the application of computer technology has become increasingly widespread. A host of emerging industries, such as big data, cloud computing, artificial intelligence, and 5G, are booming, driving a growing demand for data storage. Against this backdrop, numerous data centers have emerged. By the end of 2019, my country had 2.658 million data center racks in operation, a year-on-year increase of 28.7%. For the foreseeable future, data centers will continue to develop towards high speed and high quality.

[0003] The high heat generation of electronic chips in data centers has long been a key issue hindering their development. As data centers become increasingly integrated and electronic chips shrink in size, the heat generation rate per unit volume of electronic components continues to rise. Electronic chips like CPUs and graphics cards are primarily made of silicon. When temperatures are too high, the performance of these chips can drop significantly, even affecting their lifespan. Therefore, addressing the heat dissipation issue is urgent.

[0004] Dust accumulation on chip surfaces is one of the main issues that restricts chip heat dissipation. In computer rooms, a large number of devices are densely stacked. To address this heat dissipation issue, air cooling is typically used. This involves using air as a heat transfer medium, with fans driving the air to remove excess heat from the chips, thereby achieving the desired heat dissipation effect. However, during the cooling process, dust is carried by the air into the cabinet and adheres to the chip surface. Failure to promptly remove dust accumulation increases the thermal resistance of the chip surface, impacting overall heat dissipation. To avoid unnecessary labor costs, it is necessary to accurately measure the dust thickness on the chip surface and promptly remove it based on the thickness. Due to the complex structure and high precision of chip surfaces, traditional contact measurement methods cannot effectively meet these requirements. Therefore, it is necessary to develop non-contact measurement methods for dust distribution on chip surfaces. Summary of the Invention

[0005] In order to solve the problems of the above-mentioned prior art, the present invention provides a non-contact measurement method for dust distribution. The non-contact measurement method of the present invention can obtain the dust distribution on the surface of the chip to be tested, provide data support for timely cleaning of the chip surface, and ensure the smooth and efficient operation of the data center.

[0006] The present invention is achieved through the following technical solutions:

[0007] A non-contact measurement method for dust distribution, comprising:

[0008] Acquire the surface image of the object to be tested;

[0009] Sampling the surface image of the object to be tested and obtaining the grayscale value of each sampling point;

[0010] Compare the grayscale value of the sampling point with the grayscale standard value in the dust accumulation thickness standard database, and determine the dust distribution on the surface of the object to be tested based on the comparison result;

[0011] The dust accumulation thickness standard database is a pre-established database storing grayscale standard values corresponding to dust of different thicknesses.

[0012] Preferably, the object is a chip.

[0013] Preferably, the acquiring of the surface image of the object to be tested is specifically: photographing the surface of the object to be tested with a sampling optical camera; wherein, a red light is used as the only light source during the photographing process.

[0014] Preferably, the process of sampling the surface image of the object to be tested and obtaining the grayscale value of each sampling point includes:

[0015] Use MATLAB software to read the surface image of the object to be tested, crop the surface image of the object to be tested, and only retain the area of the object to be tested to obtain the area image of the object to be tested;

[0016] Use MATLAB software to evenly select multiple sampling points in the image of the object area to be tested, obtain the grayscale value of each pixel in the sampling point and calculate the average value to obtain the grayscale value of each sampling point.

[0017] Preferably, the grayscale value of the sampling point is compared with the grayscale standard value in the dust accumulation thickness standard database, and the dust distribution on the surface of the object to be tested is determined according to the comparison result. The specific process includes:

[0018] Compare the grayscale value of the sampling point with the grayscale standard value in the dust accumulation thickness standard database, and select the dust accumulation thickness corresponding to the grayscale standard value closest to the grayscale value of the sampling point as the dust accumulation thickness of the sampling point;

[0019] Assign different colors to sampling points with different dust accumulation thicknesses, draw images based on the colors, and obtain the dust distribution on the surface of the object to be tested.

[0020] Preferably, the method for establishing the dust accumulation thickness standard database includes:

[0021] A plurality of standard plates with dust of different thicknesses evenly distributed on their surfaces are set, and an image of the surface of each standard plate is obtained;

[0022] The grayscale value of each standard plate surface image is obtained to obtain the grayscale standard values corresponding to dust of different thicknesses. All grayscale standard values together constitute a dust accumulation thickness standard database.

[0023] Further, a series of standard plates with different thicknesses of dust evenly distributed on the set surface are provided, and surface images of each standard plate are obtained, which specifically includes:

[0024] Select multiple plates of the same size, with the same structure and size as the item to be measured;

[0025] Coat the dust on multiple plates evenly according to different thickness standards respectively to obtain multiple standard plates;

[0026] Use an optical camera to take surface images of the standard plates.

[0027] Further, when using an optical camera to take surface images of the standard plates, use a red light lamp as the only light source.

[0028] Further, the gray values of the surface images of each standard plate are obtained to obtain the gray standard values corresponding to different thicknesses of dust, which specifically includes:

[0029] Use MATLAB software to read the surface images of the standard plates, and crop the surface images of the standard plates, only retaining the standard plate area to obtain the standard plate area image;

[0030] In MATLAB software, use a matrix form to evenly select multiple sampling points in the standard plate area image, obtain the gray value of each pixel point in the sampling points, calculate the average value of the gray values of all pixel points in a sampling point to obtain the sampling point gray value; calculate the average value of the gray values of all sampling points as the gray standard value corresponding to the dust accumulation thickness on the standard plate;

[0031] Repeat the above operations to obtain the gray standard values corresponding to different dust accumulation thicknesses.

[0032] Preferably, the resolution of the surface image of the item to be measured and / or the surface image of the standard plate is not less than 600 DPI.

[0033] Compared with the prior art, the present invention has the following beneficial effects:

[0034] A non-contact measurement method for dust distribution of the present invention obtains the dust accumulation situation on the surface of the item to be measured by collecting the surface image of the item to be measured and obtaining the gray value of the sampling point in the image. According to the relationship between the dust accumulation thickness and the gray value, the method of the present invention does not need to contact the surface of the item, so it can well realize the measurement of dust distribution for items with complex surface shapes or components. For example, it can obtain the dust distribution situation on the surface of the chip to be measured, providing data support for timely cleaning the surface of the chip and ensuring the stable and efficient operation of the data center.

[0035] Furthermore, the ambient light source should be turned off during the shooting process, and red light should be used as the only light source, because red light has a longer wavelength and its reflectivity varies significantly on surfaces with different dust thicknesses.

[0036] Furthermore, a plurality of standard plates with a series of dusts of different thicknesses evenly distributed on the surface are used to construct a standard database of dust thickness. The obtained grayscale standard values are highly accurate, thereby ensuring the accuracy of the measurement results. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 A flow chart of the non-contact measurement method for dust distribution of the present invention;

[0038] Figure 2 This is a specific step diagram of step S2 of the present invention;

[0039] Figure 3 This is a specific step diagram of step S4 of the present invention;

[0040] Figure 4 Schematic diagram of the method of the present invention. DETAILED DESCRIPTION

[0041] In order to further understand the present invention, the present invention is described below in conjunction with embodiments. These descriptions are only for further explaining the features and advantages of the present invention and are not intended to limit the claims of the present invention.

[0042] Combine Figure 1 The present invention provides a non-contact measurement method for dust distribution, which specifically includes the following steps:

[0043] S1: Set a standard plate with dust of a specific thickness evenly distributed on its surface and use an optical camera to obtain an image of the standard plate surface; the specific steps are as follows:

[0044] S101: Select multiple plates of the same size, whose structure and size should be the same as the chip to be tested;

[0045] S102: Common local dust is evenly applied to a plurality of boards according to different thickness standards to obtain standard boards; during the implementation process, the dust mass distributed per square centimeter can be used to represent the dust accumulation thickness;

[0046] S102: Use an optical camera to capture images of the standard plate surface. Note the following when capturing: the image resolution must be no less than 600 DPI; turn off ambient light during the capture process, and use red light as the sole light source. This is because red light has a longer wavelength, and its reflectivity varies significantly on the surfaces of standard plates with varying dust accumulation thicknesses; ensure that the surface of the standard plate is completely within the image, and try to keep the edge of the standard plate parallel to the edge of the image to facilitate post-processing.

[0047] S2: Read the surface image information of the standard board through MATLAB software, and then obtain the gray-scale values corresponding to different thicknesses of dust, which are used as gray-scale standard values (i.e., establish a standard database for dust accumulation thickness); combined with Figure 2 , the specific steps are as follows:

[0048] S201: Use MATLAB software to read the surface image of the standard board. That is, in MATLAB software, use the "imread" function to read the surface image of the standard board into the MATLAB environment, use the "imshow" function to display the read surface image of the standard board, and use the "imcrop" function to crop the image so that it only retains the standard board area in the image, obtaining the standard board area image;

[0049] S202: In MATLAB software, uniformly select multiple sampling points in the standard board area image in matrix form, use the "impixel" function to obtain the gray-scale value of each pixel point in the sampling points, and calculate the average value of the gray-scale values of all pixel points within a sampling point, that is, obtain the gray-scale value of this sampling point. Calculate the average value of the gray-scale values of all sampling points according to formula 1, and use it as the gray-scale standard value corresponding to the dust accumulation thickness on this standard board:

[0050]

[0051] In the above formula, n is the number of sampling points, GL n is the gray-scale value of different sampling points, is the gray-scale standard value corresponding to this dust accumulation thickness;

[0052] S203: Repeat the above steps S201 and S202 to obtain the gray-scale standard values corresponding to different dust accumulation thicknesses (taking I, II, III, and IV as examples), and all gray-scale standard values together constitute the standard database for dust accumulation thickness.

[0053] S3: Obtain the surface image of the chip to be tested. The specific shooting steps and requirements are the same as those in step S1;

[0054] S4: Obtain the gray-scale distribution of the surface image of the chip to be tested through MATLAB software, and compare it with the gray-scale standard values in the standard database for dust accumulation thickness, and finally determine the dust distribution on the surface of the chip to be tested. Combined with Figure 3 , the specific steps are as follows:

[0055] S401: Use MATLAB software to read the surface image of the chip to be tested. That is, in MATLAB software, use the "imread" function to read the surface image of the chip to be tested into the MATLAB environment, use the "imshow" function to display the read surface image of the chip to be tested, and use the "imcrop" function to crop the image so that it only retains the area of the chip to be tested in the image, obtaining the area image of the chip to be tested;

[0056] S402: Define the three-dimensional array "GLM[X,Y,Z]", where X represents the horizontal coordinate of the sampling point, Y represents the vertical coordinate of the sampling point, and Z represents the grayscale value of the sampling point. Use the "impixel" function to obtain the grayscale value of a single pixel within each sampling point in the image of the chip area to be tested. Calculate the average grayscale value of all pixels within the sampling point as the grayscale value of the sampling point, and store this grayscale value in the three-dimensional array "GLM[X,Y,Z]". Compare the grayscale value of each sampling point with the grayscale standard value. Select the dust accumulation thickness corresponding to the grayscale standard value closest to the grayscale value of the sampling point as the dust accumulation thickness of the sampling point. Assign different dust accumulation thicknesses different colors, and draw an image based on the colors to ultimately obtain the dust distribution on the surface of the chip to be tested.

[0057] The uniform selection of sampling points mentioned in the above method specifically refers to:

[0058] (1) The sampling point is circular;

[0059] (2) In general, a single sampling point should contain no less than 10 pixels and no more than 20 pixels;

[0060] (3) The sampling points should meet the following requirements: the difference between the horizontal and vertical coordinates should be the same; in general, the difference between the horizontal and vertical coordinates should not be greater than 0.5 inches.

[0061] The schematic diagram of the method of the present invention is as follows Figure 4 shown.

[0062] To address the problem of dust accumulation on the surface of electronic chips in data centers, the present invention adopts a non-contact optical camera method. By collecting grayscale information on the surface of standard plates, a standard database of dust accumulation thickness is established, and the dust distribution on the surface of the chip to be tested is obtained. This provides data support for timely cleaning of the chip surface and ensures the smooth and efficient operation of the data center.

[0063] It should be noted that the method of the present invention has a wide range of applications and is not only applicable to the measurement of dust accumulation on the surface of chips, but can also be extended to the measurement of dust accumulation on the surfaces of other components.

Claims

1. A non-contact measurement method for dust distribution, characterized in that Including: Obtain the surface image of the item to be measured; the item is a chip; Sample the surface image of the item to be measured and obtain the gray values of each sampling point; Compare the gray values of the sampling points with the gray standard values in the dust accumulation thickness standard database, select the dust accumulation thickness corresponding to the gray standard value closest to the gray value of the sampling point as the dust accumulation thickness of the sampling point; assign different colors to the sampling points with different dust accumulation thicknesses, and draw an image according to the colors to obtain the dust distribution on the surface of the item to be measured; Among them, the dust accumulation thickness standard database is a pre-established database that stores the gray standard values corresponding to different thicknesses of dust; The method for establishing the dust accumulation thickness standard database includes: Set multiple standard plates with different thicknesses of dust, and the dust on the surface of each standard plate is evenly distributed, and obtain the surface images of each standard plate; Obtain the gray values of the surface images of each standard plate to obtain the gray standard values corresponding to different thicknesses of dust, and all the gray standard values together form the dust accumulation thickness standard database.

2. The non-contact measurement method of dust distribution according to claim 1, characterized in that, The obtaining of the surface image of the item to be measured is specifically: sample the surface of the item to be measured with an optical camera; among them, a red light lamp is used as the only light source during the shooting process.

3. The non-contact measurement method of dust distribution according to claim 1, characterized in that, The sampling of the surface image of the item to be measured and the obtaining of the gray values of each sampling point specifically include the process of: Use MATLAB software to read the surface image of the item to be measured, crop the surface image of the item to be measured, and only retain the area of the item to be measured to obtain the area image of the item to be measured; Use MATLAB software to evenly select multiple sampling points in the area image of the item to be measured, obtain the gray value of each pixel point in the sampling point and calculate the average value to obtain the gray values of each sampling point.

4. The non-contact measurement method for dust distribution according to claim 1, characterized in that, The setting of multiple standard plates with different thicknesses of dust, and the dust on the surface of each standard plate is evenly distributed, and the obtaining of the surface images of each standard plate specifically includes: Select multiple plates of the same size, and the structure and size of the plates are the same as those of the item to be measured; Coat the dust evenly on multiple plates according to different thickness standards to obtain multiple standard plates; Use an optical camera to take the surface image of the standard plate.

5. The non-contact measurement method for dust distribution according to claim 4, characterized in that, When using an optical camera to take the surface image of the standard plate, use a red light lamp as the only light source.

6. The non-contact measurement method for dust distribution according to claim 1, characterized in that, The obtaining of the gray values of the surface images of each standard plate to obtain the gray standard values corresponding to different thicknesses of dust specifically includes: Use MATLAB software to read the surface image of the standard plate, and crop the surface image of the standard plate to only retain the area of the standard plate to obtain the area image of the standard plate; In MATLAB software, use a matrix form to evenly select multiple sampling points in the area image of the standard plate, obtain the gray value of each pixel point in the sampling point, calculate the average value of the gray values of all pixel points in a sampling point to obtain the gray value of the sampling point; calculate the average value of the gray values of all sampling points as the gray standard value corresponding to the dust accumulation thickness on the standard plate. Repeat the above operations to obtain the gray standard values corresponding to different dust accumulation thicknesses.

7. The non-contact measurement method of dust distribution according to claim 1, characterized in that, The resolution of the surface image of the item to be measured and / or the surface image of the standard plate is not less than 600 DPI.

Images