CN-121998989-A - Correction method for particle size parameters of particles obtained by section static image analysis method

CN121998989ACN 121998989 ACN121998989 ACN 121998989ACN-121998989-A

Abstract

The application relates to the technical field of test analysis and discloses a correction method for particle size parameters obtained by a section static image analysis method, which comprises the steps of S10, randomly generating a plurality of three-dimensional samples, calculating actual particle size parameters of the three-dimensional samples, S20, randomly cutting the three-dimensional samples, calculating section particle size parameters of the three-dimensional samples, S30, establishing a mathematical relation model of the actual particle size parameters and the section particle size parameters, and S40, correcting the particle size parameters obtained by the section static image analysis method by utilizing the mathematical relation model. In this way, the granularity parameter that gets closer to its true three-dimensional distribution can be corrected. The accuracy of the particle size analysis of the cross-section image of the compact material can be effectively improved.

Inventors

ZHANG XIAODONG
XU SHUMEI

Assignees

中国海洋大学

Dates

Publication Date: 20260508
Application Date: 20260410

Claims (10)

1. A method for correcting a particle size parameter obtained by a sectional still image analysis method, comprising: s10, randomly generating a plurality of three-dimensional samples, and calculating actual granularity parameters of the three-dimensional samples; S20, randomly cutting each three-dimensional sample, and calculating the section granularity parameter of each three-dimensional sample; S30, establishing a mathematical relationship model of the actual granularity parameter and the section granularity parameter; and S40, correcting the particle size parameters obtained by a section static image analysis method by using the mathematical relationship model.
2. The method of calibrating a particle size parameter for cross-sectional still image analysis according to claim 1, wherein in S10, the randomly generating a plurality of three-dimensional samples comprises: randomly generating one or more groups of average particle sizes and sorting coefficients according to preset probability, and further determining normal frequency distribution corresponding to each group and the total number of particles in different particle size intervals; the corresponding three-dimensional sample is composed of particles that are distributed at a single frequency that corresponds to a set of average particle sizes and sorting coefficients, or is composed of a mixture of particles that are distributed at a single frequency that corresponds to a plurality of sets of average particle sizes and sorting coefficients, and the average particle sizes and sorting coefficients of the population of particles are recalculated.
3. The method for correcting a particle size parameter obtained by a sectional still image analysis according to claim 2, When a plurality of groups of particles conforming to single frequency distribution are mixed, the mixing coefficients are generated in a random mode, the values between 0 and 1 are taken, and the sum of all the mixing coefficients is 1.
4. The method of calibrating particle size parameters for cross-sectional still image analysis according to claim 1, wherein in S20, the randomly slicing each of the three-dimensional samples comprises: Randomly selecting a cutting position to cut each three-dimensional sample for a plurality of times; generating a two-dimensional cutting section for each cutting; Wherein the probability of each particle being cut is positively correlated with its particle size.
5. The method according to claim 1, wherein in S20, the calculating the cross-sectional particle size parameter of each of the three-dimensional samples includes: Determining the cutting position of the particles to be cut of each three-dimensional sample; determining a cutting section according to the cutting position; And calculating the section granularity parameter of each three-dimensional sample on the cut section.
6. The method of calibrating particle size parameters obtained by cross-sectional still image analysis according to claim 5, wherein said calculating cross-sectional particle size parameters of each of said three-dimensional samples on said cut cross-section comprises: Calculating the diameter and the area of each particle on the cutting section, and the volume of a sphere corresponding to the cutting diameter; counting the number, total area and total volume of particles on a cutting section according to the preset diameter group; calculating the section granularity frequency distribution according to the number, the total area and the total volume of the cut sections; And calculating the number average particle size, the number sorting coefficient, the area average particle size, the area sorting coefficient, the volume average particle size and the volume sorting coefficient of the particles on the cutting section according to the section particle size frequency distribution.
7. The method for correcting a particle size parameter obtained by a sectional still image analysis according to claim 1, wherein S30 comprises: Respectively establishing mathematical relationship models of the actual granularity parameter and the section granularity parameter by adopting a regression analysis method; And analyzing the fitting degree between the average particle size calculated by the mathematical relation model and the actual average particle size sequence, and calculating a decision coefficient.
8. The method of calibrating particle size parameters for cross-sectional still image analysis according to claim 7, wherein the mathematical relationship model comprises: a first mathematical relationship model Dyc = a 1 ×Djc+b 1 ×Fjc+c 1 ; A second mathematical relationship model Fyc = a 2 ×Djc+b 2 ×Fjc+c 2 ; A third mathematical relationship model Dyv = a 3 ×Djv+b 3 ×Fjv+c 3 ; a fourth mathematical relationship model Fyv = a 4 ×Djv+b 4 ×Fjv+c 4 ; A fifth mathematical relationship model Dyv = a 5 ×Djs+b 5 ×Fjs+c 5 ; a sixth mathematical relationship model Fyv = a 6 ×Djs+b 6 ×Fjs+c 6 ; Wherein Dyc-actual number average particle diameter, fyc-actual number sorting coefficient, dyv-actual volume average particle diameter, fyv-actual volume sorting coefficient, djc-section number average particle diameter, fjc-section number sorting coefficient, djs-section area average particle diameter, fjs-section area sorting coefficient, djv-section volume average particle diameter, fjv-section volume sorting coefficient, a 1 to a 6 、b 1 to b 6 , and c 1 to c 6 are known coefficients.
9. The method of correcting a particle size parameter obtained by cross-sectional still image analysis according to claim 8, wherein S40 comprises: Correcting the particle size parameters obtained by a section static image analysis method by using the first mathematical relationship model and the second mathematical relationship model when the quantity parameters are taken as statistics, wherein Djc and Fjc are obtained by using the section static image analysis method; When the volume parameter is taken as statistics, correcting the particle size parameter obtained by a section static image analysis method by using one of the third mathematical relationship model and the fifth mathematical relationship model and one of the fourth mathematical relationship model and the sixth mathematical relationship model, wherein Djs and Fjs are obtained by using a section static image analysis method, and Djv and Fjv are obtained by using a section static image analysis method.
10. A correction device for particle size parameters obtained by a sectional still image analysis method, comprising: the sample generation module is configured to randomly generate a plurality of three-dimensional samples and calculate the actual granularity parameter of each three-dimensional sample; The sample cutting module is configured to randomly cut each three-dimensional sample and calculate the section granularity parameter of each three-dimensional sample; the relation establishing module is configured to establish a mathematical relation model of the actual granularity parameter and the section granularity parameter; and the parameter correction module is configured to correct the particle size parameter obtained by the section static image analysis method by using the mathematical relationship model.

Description

Correction method for particle size parameters of particles obtained by section static image analysis method Technical Field The application relates to the technical field of test analysis, in particular to a correction method for particle size parameters obtained by a section static image analysis method. Background Particle size analysis is a very critical test analysis technique in many fields such as scientific research and industrial production. The method is mainly used for accurately measuring the particle size distribution of particle groups and covering important information such as particle size and distribution thereof. There are various conventional particle size analysis methods, such as laser particle size analysis, sedimentation, sieving, still image analysis, and dynamic image analysis. The static image analysis method is a granularity analysis method based on an image processing technology. The core principle is as follows. First, still images of particles are acquired, which images can be taken by imaging devices such as cameras, microscopes, and scanning electron microscopes. During imaging, the particles lie in a particular plane, forming a sharp two-dimensional cross-sectional image. Then, the images are processed and analyzed by utilizing image analysis software, the particle contours in the images are identified, the geometric parameters of each particle such as the area, the perimeter, the equivalent circle diameter and the like are calculated, and then the particle size distribution condition of the whole particle group is obtained by adopting a statistical method. The static image analysis method obtains the particle size distribution of particles in a two-dimensional image space, and the particle size distribution is different from the actual particle size distribution of particles in a three-dimensional space to a certain extent. There is a difference in probability that particles of different particle diameters appear on the cross section. It is apparent that large particles are more easily cut or ground to be exposed on the cross section. Meanwhile, the particles are not cut or polished exactly at their maximum diameter, but are often cut or polished at a non-maximum diameter, resulting in a smaller diameter of the particles in cross section. The extent to which the cross-sectional diameter of the particles becomes smaller depends on the cutting location or the extent of sanding. The two effects finally lead to a difference between the particle size distribution of the particles on the cross section and the actual distribution, namely that the particle size parameters obtained by adopting a cross section static image analysis method are not consistent with the actual parameters. The frequency statistics of the particles generally include two types, namely the number frequency and the volume frequency. The number frequency means the ratio of the number of particles with different particle diameters to the total number of particles within a certain range. The volume frequency refers to the proportion of the volume of particles with different particle diameters to the total volume of the particles within a certain range. The average particle diameter of the particles obtained by the quantitative method is referred to as a number average particle diameter, and the average particle diameter of the particles obtained by the volumetric method is referred to as a volume average particle diameter. The number distribution and the volume distribution of the sample particles are generally different due to the different volumes of the particles with different particle diameters, and finally, the number average particle diameter and the volume average particle diameter of the sample particles are different. In addition to the average particle size, the classification factor is another key parameter for measuring the particle size, and there is also a quantitative classification factor obtained by a quantitative method and a volumetric classification factor obtained by a volumetric method, and there is also a common difference therebetween. In the process of implementing the embodiments of the present disclosure, it is found that at least the following problems exist in the related art: The existing static image analysis method, particle frequency statistics method and sorting coefficient method have differences between the obtained particle size parameters and actual parameters, so that the particle size analysis result is inaccurate. It should be noted that the information disclosed in the above background section is only for enhancing understanding of the background of the application and thus may include information that does not form the prior art that is already known to those of ordinary skill in the art. Disclosure of Invention The following presents a simplified summary in order to provide a basic understanding of some aspects of the disclosed embodiments. This summary is not an exte