A quality control detection method for mullite / quartz composite ceramic hollow microspheres

Abstract

This invention discloses a quality control and testing method for mullite / quartz composite ceramic hollow microspheres, comprising the following steps: particle size detection: using a laser particle size analyzer to detect the particle size distribution of the microspheres; sphericity and surface defect detection: calculating the sphericity and the type and number of surface defects through AI image analysis; true density detection: placing the microspheres in a true density analyzer to determine the true density; dielectric constant detection: detecting the dielectric constant and loss of 5G millimeter waves; single-particle compressive strength detection: testing the compressive strength of a single particle; and quality grading: classifying the ceramic microspheres into different quality grades. The advantage of this invention is that it integrates six steps: particle size detection, sphericity and surface defect detection, true density detection, dielectric constant detection, single-particle compressive strength detection, and quality grading, forming a complete testing chain from microscopic morphology to macroscopic performance, and from single-particle characteristics to batch quality. This fills the gap in the current market for specialized quality control methods for mullite / quartz composite ceramic hollow microspheres.

Description

Technical Field

[0001] This invention belongs to the field of materials testing and analysis technology, specifically relating to a quality control testing method for mullite / quartz composite ceramic hollow microspheres. Background Technology

[0002] Hollow microspheres, due to their excellent properties such as lightweight, high strength, and low dielectric constant, have significant application prospects in aerospace, 5G communication, and deep-sea exploration. Mullite / quartz composite ceramic hollow microspheres are a novel type of high-performance hollow microsphere, but there are currently no mature commercial products available in the domestic and international markets.

[0003] Due to the lack of mature products, existing testing methods are mainly designed for hollow glass microspheres and cannot be directly applied to the quality control of mullite / quartz composite ceramic hollow microspheres. The main problems are as follows: First, in terms of sphericity and surface defect detection, existing methods mostly rely on manual visual inspection or simple image analysis, which is inefficient and makes it difficult to identify defects such as microcracks and pits unique to composite ceramics. Furthermore, there is a lack of a judgment mechanism to "stop the detection when a fatal defect is found," resulting in a waste of subsequent detection resources.

[0004] Secondly, regarding dielectric performance testing, the hollow microspheres easily trap trace amounts of moisture inside, and these impurities significantly affect the dielectric loss of 5G millimeter waves. Existing testing methods only use conventional drying (such as 50-80℃), which cannot effectively remove the moisture trapped inside, leading to distorted test results.

[0005] Third, in terms of compressive strength testing, existing methods are mostly batch tests, which can only obtain average strength and cannot reflect the strength differences between individual microspheres. Furthermore, due to the complex preparation process, mullite / quartz composite ceramic microspheres exhibit significant dispersion in individual particle strength, and a small number of low-strength microspheres may lead to premature failure of the entire material.

[0006] Fourth, in terms of quality grading, existing products are simply graded by particle size or density, which cannot meet the differentiated requirements of different scenarios such as 5G communication, deep-sea buoyancy, and well cementing for dielectric, compressive strength, and density properties.

[0007] Therefore, there is an urgent need to develop a quality control and testing method specifically for mullite / quartz composite ceramic hollow microspheres, so as to achieve systematic testing and quality grading of particle size, sphericity, surface defects, true density, dielectric properties and single-particle compressive strength. Summary of the Invention

[0008] To address the aforementioned problems, the main objective of this invention is to design a quality control and testing method for mullite / quartz composite ceramic hollow microspheres.

[0009] To achieve the above objectives, the present invention adopts the following technical solution: A quality control and testing method for mullite / quartz composite ceramic hollow microspheres includes the following steps: S1: Particle size detection: The particle size distribution of microspheres is detected by a laser particle size analyzer to obtain particle size parameters; S2: Sphericity and Surface Defect Detection: The microspheres to be tested are immersed in anhydrous ethanol for dispersion. After dispersion, the geometric morphology image and surface defect image of the microspheres are acquired simultaneously through a multimodal imaging system. The sphericity and the type and number of surface defects are calculated by AI image analysis. If the sphericity is lower than the preset threshold or there is a preset fatal defect type, the sample is directly judged as unqualified and all subsequent testing procedures are terminated. S3: True density test: Place the sample that has passed the sphericity and surface defect test into the true density meter and use the gas displacement method to determine the true density of the microspheres; S4: Dielectric detection: The microspheres were placed in a vacuum oven and dried in a vacuum constant temperature environment of 112℃ for 90 minutes to remove the water vapor sealed in the inner cavity. Then, the dielectric constant and loss of 5G millimeter wave were detected by the resonance method using a vector network analyzer. S5: Single-particle compressive strength test: The compressive strength of a single particle is tested by applying pressure at a constant low speed of 0.2μm / s using a microsphere compressive strength tester. S6: Quality Classification: Ceramic microspheres are classified into different quality grades based on differences in particle size, sphericity, surface defects, true density, dielectric loss, and compressive strength.

[0010] As a further description of the present invention, in step S2, the multimodal imaging system includes a visible light camera and a hyperspectral or infrared camera. The visible light channel is used to acquire geometric shape images to calculate the sphericity, and the hyperspectral or infrared camera is used to acquire surface defect images to detect microcracks, pits and pores. The detection channels of the visible light camera and the hyperspectral or infrared camera are associated through an image fusion algorithm to achieve synchronous determination of sphericity and surface defects.

[0011] As a further description of the present invention, in step S2, the temperature of the low-temperature low-power dispersion is 30-35°C, the ultrasonic power is 20W, and the ultrasonic duration is 2min.

[0012] As a further description of the present invention, in step S2, the preset fatal defects include penetrating microcracks, pits with a diameter greater than 10% of the microsphere particle size, and perforations connecting the inside and outside.

[0013] As a further description of the present invention, the sphericity threshold in step S2 can be dynamically adjusted. When used as filler for 5G millimeter-wave radomes, the sphericity threshold is 90%–92%; When used in lightweight composite materials for aerospace applications, the sphericity threshold is 93%–95%. When used in deep-sea buoyancy materials, the sphericity threshold is 92% to 94%.

[0014] As a further description of the present invention, in step S2, the sphericity threshold can also be dynamically adjusted according to the particle size D50 of the hollow microspheres. When D50 ≤ 50 μm, the sphericity threshold is reduced by 1% to 2%; When D50 ≥ 200 μm, the sphericity threshold is increased by 1% to 2%; When 50μm < D50 < 200μm, the sphericity is based on the benchmark threshold.

[0015] As a further description of the present invention, it also includes dynamic feedback. In step S2, when the sphericity and surface defect detection failure rate of three consecutive batches exceed the set ratio threshold, an early warning signal is issued and a process parameter adjustment suggestion is pushed.

[0016] As a further description of the present invention, in step S2, the AI ​​recognition accuracy of sphericity is ±0.1%, and the minimum identifiable defect size for surface defect detection is no greater than 10% of the ceramic microsphere particle size.

[0017] As a further description of the present invention, in step S3, the true density meter has a measuring range of 0.1–1.0 g / cm³. 3 In step S4, the vacuum degree of vacuum drying is 0.08 MPa; in step S5, the accuracy of the compressive strength test is ±1 MPa.

[0018] Compared with the prior art, the technical advantages of the present invention are as follows: 1. This invention integrates six steps: particle size detection, sphericity and surface defect detection, true density detection, dielectric detection, single particle compressive strength detection, and quality grading. It forms a complete testing chain from microscopic morphology to macroscopic performance, and from single particle characteristics to batch quality, filling the gap in the current market for specialized quality control methods for mullite / quartz composite ceramic hollow microspheres.

[0019] 2. This invention achieves simultaneous acquisition and correlation analysis of sphericity and surface defects by immersing the microspheres under test in anhydrous ethanol for low-temperature, low-power dispersion, combined with a multimodal imaging system (visible light camera + hyperspectral / infrared camera) and image fusion algorithm. Compared with traditional manual visual inspection or single optical inspection, this invention offers faster detection speed, higher accuracy, and can identify surface defects unique to composite ceramics such as microcracks, pits, and pores, significantly improving the precision and automation of the inspection.

[0020] 3. In the sphericity and surface defect detection process, this invention defines penetrating microcracks, large-sized pits, and connecting internal and external perforations as fatal defects. Once any of these defects are detected or the sphericity falls below a preset threshold, all subsequent detection processes are immediately terminated. This avoids wasting detection resources on obviously unqualified samples, reduces detection costs, and shortens the detection cycle.

[0021] 4. This invention sets differentiated sphericity thresholds for different application scenarios, and these thresholds can be dynamically adjusted based on the microsphere size. This flexible threshold setting method ensures the rigor of quality control while avoiding overly stringent requirements that could lead to misjudgments of qualified batches, demonstrating the adaptability of the detection method to different application needs.

[0022] 5. This invention addresses the problem of moisture trapping within the inner cavity of mullite / quartz composite ceramic hollow microspheres by employing a vacuum oven at a higher temperature for drying. Compared to conventional low-temperature drying methods, this method more thoroughly removes trace amounts of moisture and volatile residues trapped within the cavity. Based on this, the dielectric constant and loss of millimeter waves are measured, and the results accurately reflect the intrinsic dielectric properties of the microspheres, avoiding the problems of artificially high dielectric losses or unstable test results caused by moisture residue.

[0023] 6. This invention employs a microsphere compressive strength tester to test the compressive strength of individual microspheres using a uniform, low-speed pressurization method. Compared to traditional batch testing methods that can only obtain average strength, this invention can reflect the strength distribution differences between individual microspheres, effectively identifying a small number of low-strength microspheres, and providing more accurate data support for reliability assessment under extreme conditions such as deep-sea buoyancy and deep-well cementing.

[0024] 7. This invention classifies ceramic microspheres into different quality grades based on differences in multiple dimensions, including particle size, sphericity, surface defects, true density, dielectric loss, and compressive strength. Users can select the appropriate grade of product according to their application scenarios, achieving a refined and serialized supply of microsphere products.

[0025] 8. This invention also includes a dynamic feedback mechanism: when the sphericity and surface defect detection failure rates of multiple consecutive batches exceed a set threshold, the system automatically issues an early warning signal and pushes process parameter adjustment suggestions. This design links quality inspection with the production process, helping to promptly identify systemic problems in the preparation process, promoting continuous optimization of the production process, and improving the pass rate of microsphere products from the source. Attached Figure Description Detailed Implementation

[0027] The present invention will now be described in detail: This application provides a quality control and testing method for mullite / quartz composite ceramic hollow microspheres. It mainly includes the following steps, S1: Particle size detection: The particle size distribution of microspheres is detected by a laser particle size analyzer to obtain particle size parameters; S2: Sphericity and Surface Defect Detection: The microspheres to be tested are immersed in anhydrous ethanol for dispersion. After dispersion, the geometric morphology image and surface defect image of the microspheres are acquired simultaneously through a multimodal imaging system. The sphericity and the type and number of surface defects are calculated by AI image analysis. If the sphericity is lower than the preset threshold or there is a preset fatal defect type, the sample is directly judged as unqualified and all subsequent testing procedures are terminated. S3: True density test: Place the sample that has passed the sphericity and surface defect test into the true density meter and use the gas displacement method to determine the true density of the microspheres; S4: Dielectric detection: The microspheres were placed in a vacuum oven and dried in a vacuum constant temperature environment of 112℃ for 90 minutes to remove the water vapor sealed in the inner cavity. Then, the dielectric constant and loss of 5G millimeter wave were detected by the resonance method using a vector network analyzer. S5: Single-particle compressive strength test: The compressive strength of a single particle is tested by applying pressure at a constant low speed of 0.2μm / s using a microsphere compressive strength tester. S6: Quality Classification: Ceramic microspheres are classified into different quality grades based on differences in particle size, sphericity, surface defects, true density, dielectric loss, and compressive strength.

[0028] In step S2, the multimodal imaging system includes a visible light camera and a hyperspectral or infrared camera. The visible light channel is used to acquire geometric shape images and calculate sphericity. The hyperspectral or infrared camera is used to acquire surface defect images and detect microcracks, pits, and pores. The detection channels of the visible light camera and the hyperspectral or infrared camera are associated through an image fusion algorithm to achieve simultaneous determination of sphericity and surface defects.

[0029] In step S2, the temperature for low-temperature, low-power dispersion is 30–35°C, the ultrasonic power is 20W, and the ultrasonic duration is 2 minutes.

[0030] In step S2, the preset fatal defects include penetrating microcracks, pits with a diameter greater than 10% of the microsphere diameter, and perforations that connect the inside and outside.

[0031] The sphericity threshold in step S2 can be dynamically adjusted. When used as filler for 5G millimeter-wave radomes, the sphericity threshold is 90%–92%; When used in lightweight composite materials for aerospace applications, the sphericity threshold is 93%–95%. When used in deep-sea buoyancy materials, the sphericity threshold is 92% to 94%.

[0032] In step S2, the sphericity threshold can also be dynamically adjusted according to the particle size D50 of the hollow microspheres. When D50 ≤ 50 μm, the sphericity threshold is reduced by 1% to 2%; When D50 ≥ 200 μm, the sphericity threshold is increased by 1% to 2%; When 50μm < D50 < 200μm, the sphericity is based on the benchmark threshold.

[0033] In step S2, if the sphericity and surface defect detection failure rate of three consecutive batches exceed the set threshold, an early warning signal is issued and a process parameter adjustment suggestion is pushed.

[0034] In step S2, the AI ​​recognition accuracy of sphericity is ±0.1%, and the minimum identifiable defect size for surface defect detection is no greater than 10% of the ceramic microsphere particle size.

[0035] In step S3, the true density meter has a measuring range of 0.1–1.0 g / cm³. 3 In step S4, the vacuum degree of vacuum drying is 0.08 MPa; in step S5, the accuracy of the compressive strength test is ±1 MPa. Example

[0036] This application discloses a quality control and testing method for mullite / quartz composite ceramic hollow microspheres. The actual testing process includes the following steps: S1: Particle size detection: The particle size distribution of microspheres is detected by a laser particle size analyzer to obtain particle size parameters; S2: Sphericity and Surface Defect Detection: The microspheres to be tested were immersed in anhydrous ethanol and dispersed at low temperature and low power using an ultrasonic power of 20W and an ultrasonic duration of 2min at 30-35℃. After dispersion, geometric morphology images and surface defect images of microspheres are simultaneously acquired through a multimodal imaging system. The multimodal imaging system includes a visible light camera and a hyperspectral or infrared camera. The visible light channel is used to acquire geometric morphology images and calculate the sphericity, while the hyperspectral or infrared camera is used to acquire surface defect images to detect microcracks, pits, and pores. The detection channels of the visible light camera and the hyperspectral or infrared camera are associated through an image fusion algorithm to achieve simultaneous determination of sphericity and surface defects. The sphericity and the type and number of surface defects are calculated through AI image analysis. The AI ​​recognition accuracy of sphericity is ±0.1%, and the minimum identifiable defect size for surface defect detection is no greater than 10% of the ceramic microsphere particle size. If the sphericity is lower than the preset threshold or there is a preset fatal defect type (through microcracks, pits with a diameter greater than 10% of the microsphere diameter, and perforations connecting the inside and outside), the sample is directly judged as unqualified and all subsequent testing procedures are terminated. The sphericity thresholds are as follows: When used as filler for 5G millimeter-wave radomes, the sphericity threshold is 90%–92%; When used in lightweight composite materials for aerospace applications, the sphericity threshold is 93%–95%. When used in deep-sea buoyancy materials, the sphericity threshold is 92% to 94%.

[0037] at the same time, When D50 ≤ 50 μm, the sphericity threshold is reduced by 1% to 2%; When D50 ≥ 200 μm, the sphericity threshold is increased by 1% to 2%; When 50μm < D50 < 200μm, the sphericity is based on the benchmark threshold.

[0038] Furthermore, if the sphericity and surface defect detection failure rate of three consecutive batches exceed the set threshold, an early warning signal will be issued, and suggestions for adjusting process parameters will be pushed.

[0039] The sphericity detection method is compared with traditional methods as follows. Microsphere breakage rate (%) 16.8% 2.1% Sphericity test value (%) 81.5% 94.2% Relative error of sphericity (repeated 5 times, ±%) ±7.8% ±1.5% In traditional methods, about 17 out of every 100 balls break; in this patented method, only about 2 balls break, reducing the breakage rate by 87.5%. Traditional methods misjudge the sphericity as unacceptable due to breakage; this patented method reduces the error by 80.8%.

[0040] S3: True density detection: Samples that passed the tests for sphericity and surface defects were placed in a true density meter, and the true density of the microspheres was determined using the gas displacement method. The true density meter had a measurement range of 0.1–1.0 g / cm³. 3 .

[0041] S4: Dielectric detection: The microspheres were placed in a vacuum oven and dried in a vacuum constant temperature environment of 112℃ for 90 minutes to remove the water vapor sealed in the inner cavity. The vacuum degree of vacuum drying was 0.08Mpa. Then, the dielectric constant and loss of 5G millimeter wave were detected by the resonance method using a vector network analyzer. The dielectric property testing method is compared with traditional methods as follows: Average value of dielectric constant ε' (5 GHz) 2.95 2.93 Standard deviation of dielectric constant ε' (n=10) 0.132 0.034 Average value of dielectric loss tanδ 0.0048 0.0041 Dielectric loss tanδ fluctuation range (n=10) 0.0040 ~ 0.0056 (range 0.0016) 0.0040 ~ 0.0042 (range 0.0002) tanδ Relative Standard Deviation (RSD) 12.3% 2.4% Atmospheric pressure drying cannot remove moisture trapped in the hollow inner cavity, which affects dielectric loss and causes significant fluctuations during testing; vacuum drying thoroughly dehumidifies, resulting in stable and reliable dielectric data.

[0042] S5: Single-particle compressive strength test: The compressive strength of a single particle is tested by applying pressure at a constant low speed of 0.2μm / s using a microsphere compressive strength tester. The accuracy of the compressive strength test is ±1MPa. The following is a comparison between single-particle compressive strength testing and traditional methods. Traditional rapid loading causes thin-walled spheres to brittlely burst before they can deform, resulting in low and inconsistent strength values. This patented slow loading allows the shell to deform slowly, yielding accurate and highly consistent compressive strength measurements.

[0043] S6: Quality Classification: Ceramic microspheres are classified into different quality grades based on differences in particle size, sphericity, surface defects, true density, dielectric loss, and compressive strength.

[0044] In summary, the quality control and testing method for mullite / quartz composite ceramic hollow microspheres of this application, compared with other particulate quality control and testing methods, yields the following results: Therefore, compared with the prior art, the technical effects of the present invention are as follows: 1. This invention integrates six steps: particle size detection, sphericity and surface defect detection, true density detection, dielectric detection, single particle compressive strength detection, and quality grading. It forms a complete testing chain from microscopic morphology to macroscopic performance, and from single particle characteristics to batch quality, filling the gap in the current market for specialized quality control methods for mullite / quartz composite ceramic hollow microspheres.

[0045] 2. This invention achieves simultaneous acquisition and correlation analysis of sphericity and surface defects by immersing the microspheres under test in anhydrous ethanol for low-temperature, low-power dispersion, combined with a multimodal imaging system (visible light camera + hyperspectral / infrared camera) and image fusion algorithm. Compared with traditional manual visual inspection or single optical inspection, this invention offers faster detection speed, higher accuracy, and can identify surface defects unique to composite ceramics such as microcracks, pits, and pores, significantly improving the precision and automation of the inspection.

[0046] 3. In the sphericity and surface defect detection process, this invention defines penetrating microcracks, large-sized pits, and connecting internal and external perforations as fatal defects. Once any of these defects are detected or the sphericity falls below a preset threshold, all subsequent detection processes are immediately terminated. This avoids wasting detection resources on obviously unqualified samples, reduces detection costs, and shortens the detection cycle.

[0047] 4. This invention sets differentiated sphericity thresholds for different application scenarios, and these thresholds can be dynamically adjusted based on the microsphere size. This flexible threshold setting method ensures the rigor of quality control while avoiding overly stringent requirements that could lead to misjudgments of qualified batches, demonstrating the adaptability of the detection method to different application needs.

[0048] 5. This invention addresses the problem of moisture trapping within the inner cavity of mullite / quartz composite ceramic hollow microspheres by employing a vacuum oven at a higher temperature for drying. Compared to conventional low-temperature drying methods, this method more thoroughly removes trace amounts of moisture and volatile residues trapped within the cavity. Based on this, the dielectric constant and loss of millimeter waves are measured, and the results accurately reflect the intrinsic dielectric properties of the microspheres, avoiding the problems of artificially high dielectric losses or unstable test results caused by moisture residue.

[0049] 6. This invention employs a microsphere compressive strength tester to test the compressive strength of individual microspheres using a uniform, low-speed pressurization method. Compared to traditional batch testing methods that can only obtain average strength, this invention can reflect the strength distribution differences between individual microspheres, effectively identifying a small number of low-strength microspheres, and providing more accurate data support for reliability assessment under extreme conditions such as deep-sea buoyancy and deep-well cementing.

[0050] 7. This invention classifies ceramic microspheres into different quality grades based on differences in multiple dimensions, including particle size, sphericity, surface defects, true density, dielectric loss, and compressive strength. Users can select the appropriate grade of product according to their application scenarios, achieving a refined and serialized supply of microsphere products.

[0051] 8. This invention also includes a dynamic feedback mechanism: when the sphericity and surface defect detection failure rates of multiple consecutive batches exceed a set threshold, the system automatically issues an early warning signal and pushes process parameter adjustment suggestions. This design links quality inspection with the production process, helping to promptly identify systemic problems in the preparation process, promoting continuous optimization of the production process, and improving the pass rate of microsphere products from the source.

[0052] The above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Any other modifications or equivalent substitutions made by those skilled in the art to the technical solutions of the present invention, as long as they do not depart from the direction and scope of the technical solutions of the present invention, should be covered within the scope of the claims of the present invention.

Claims

1. A quality control and testing method for mullite / quartz composite ceramic hollow microspheres, characterized in that: Includes the following steps, S1: Particle size detection: The particle size distribution of microspheres is detected by a laser particle size analyzer to obtain particle size parameters; S2: Sphericity and Surface Defect Detection: The microspheres to be tested are immersed in anhydrous ethanol for dispersion. After dispersion, the geometric morphology image and surface defect image of the microspheres are acquired simultaneously through a multimodal imaging system. The sphericity and the type and number of surface defects are calculated by AI image analysis. If the sphericity is lower than the preset threshold or there is a preset fatal defect type, the sample is directly judged as unqualified and all subsequent testing procedures are terminated. S3: True density test: Place the sample that has passed the sphericity and surface defect test into the true density meter and use the gas displacement method to determine the true density of the microspheres; S4: Dielectric detection: The microspheres were placed in a vacuum oven and dried in a vacuum constant temperature environment of 112℃ for 90 minutes to remove the water vapor sealed in the inner cavity. Then, the dielectric constant and loss of 5G millimeter wave were detected by the resonance method using a vector network analyzer. S5: Single-particle compressive strength test: The compressive strength of a single particle is tested by applying pressure at a constant low speed of 0.2μm / s using a microsphere compressive strength tester. S6: Quality Classification: Ceramic microspheres are classified into different quality grades based on differences in particle size, sphericity, surface defects, true density, dielectric loss, and compressive strength.

2. The quality control and testing method for mullite / quartz composite ceramic hollow microspheres according to claim 1, characterized in that: In step S2, the multimodal imaging system includes a visible light camera and a hyperspectral or infrared camera. The visible light channel is used to acquire geometric shape images and calculate sphericity. The hyperspectral or infrared camera is used to acquire surface defect images and detect microcracks, pits, and pores. The detection channels of the visible light camera and the hyperspectral or infrared camera are associated through an image fusion algorithm to achieve simultaneous determination of sphericity and surface defects.

3. The quality control and testing method for mullite / quartz composite ceramic hollow microspheres according to claim 1, characterized in that: In step S2, the temperature of the low-temperature, low-power dispersion is 30-35°C, the ultrasonic power is 20W, and the ultrasonic duration is 2min.

4. The quality control and testing method for mullite / quartz composite ceramic hollow microspheres according to claim 1, characterized in that: In step S2, the preset fatal defects include penetrating microcracks, pits with a diameter greater than 10% of the microsphere particle size, and perforations connecting the inside and outside.

5. The quality control and testing method for mullite / quartz composite ceramic hollow microspheres according to claim 1, characterized in that: The sphericity threshold in step S2 can be dynamically adjusted. When used as filler for 5G millimeter-wave radomes, the sphericity threshold is 90%–92%; When used in lightweight composite materials for aerospace applications, the sphericity threshold is 93%–95%. When used in deep-sea buoyancy materials, the sphericity threshold is 92% to 94%.

6. The quality control and testing method for mullite / quartz composite ceramic hollow microspheres according to claim 1, characterized in that: In step S2, the sphericity threshold can also be dynamically adjusted according to the particle size D50 of the hollow microspheres. When D50 ≤ 50 μm, the sphericity threshold is reduced by 1% to 2%; When D50 ≥ 200 μm, the sphericity threshold is increased by 1% to 2%; When 50μm < D50 < 200μm, the sphericity is based on the benchmark threshold.

7. The quality control and testing method for mullite / quartz composite ceramic hollow microspheres according to claim 1, characterized in that: It also includes dynamic feedback. In step S2, when the sphericity and surface defect detection failure rate of three consecutive batches exceed the set ratio threshold, an early warning signal is issued and a process parameter adjustment suggestion is pushed.

8. The quality control and testing method for mullite / quartz composite ceramic hollow microspheres according to claim 1, characterized in that: In step S2, the AI ​​recognition accuracy of sphericity is ±0.1%, and the minimum identifiable defect size for surface defect detection is no greater than 10% of the ceramic microsphere particle size.

9. The quality control and testing method for mullite / quartz composite ceramic hollow microspheres according to claim 1, characterized in that: In step S3, the true density meter has a measuring range of 0.1–1.0 g / cm³. 3 In step S4, the vacuum degree of vacuum drying is 0.08 MPa; in step S5, the accuracy of the compressive strength test is ±1 MPa.