A method for measuring the projected area of soot accumulation particles

Abstract

The application discloses a kind of measurement methods of soot accumulation particle projection area, comprising the following steps: S10, collection soot accumulation particle;S20, obtain the gray scale image of soot accumulation particle, and calculate the real size represented by single pixel in gray scale image;S30, the gray scale image of soot accumulation particle is handled;S40, calculate soot accumulation particle projection area.The measurement method of soot accumulation particle projection area of the hydrocarbon fuel coaxial diffusion flame of the present application changes all by computer program automatic analysis processing by artificial operation, automatically divides out the projection area of soot accumulation particle in image and image background, and filters the noise point interference in image, compared with manual operation, this method can realize batch image analysis, greatly saves the time of artificial processing, while guaranteeing measurement accuracy, avoids subjective error generated by artificial operation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of soot particle microstructure analysis, and particularly relates to a measurement method of soot accumulation particle projection area of a hydrocarbon fuel coaxial diffusion flame. BACKGROUND

[0002] The research on the morphology of soot particles is a hotspot in the field of combustion science, and the measurement of the projection area of soot accumulation particles is an important way in the research process. Transmission electron microscopy (TEM) and image analysis software (ImageJ) are commonly used equipment and software for obtaining and measuring soot accumulation particle images. Soot is the main component of diesel engine emission particles, and the amount of soot emission accounts for 50% to 80% of the total amount of diesel engine emission particles. Research on the physical and chemical properties of soot particles is of great significance for guiding diesel engines to reduce the number of nano-particle emissions.

[0003] The measurement of the projection area of soot accumulation particles is an important way to observe the morphological changes of the particles. The current measurement method of the projection area of soot accumulation particles is mainly to manually outline the outline of the soot accumulation particles, and measure the area surrounded by the outline. However, this method has obvious drawbacks. First, due to the irregular and complex morphology of soot accumulation particles, manual outlining has obvious boundary demarcation errors and unmeasurable uncertainties. Second, there are excessive areas on the projection background and the particle projection boundary in the picture, so there are subjective factors in the selection of the projection color depth threshold, which will lead to inconsistent measurement standards and further expand the error degree.

[0004] Due to the influence of human error and uncertainty factors in the above method, a lot of time and effort of researchers are consumed in the characterization analysis process, which is not conducive to the efficient development of related research work, and the uncertainty error generated in the measurement process will further reduce the reliability of the research results. SUMMARY

[0005] The purpose of the present application is to reduce the workload of soot accumulation particle projection area image analysis, while ensuring the accuracy of the measurement results and the consistency of the measurement standards, and based on the current technical means, a measurement method of soot accumulation particle projection area of a hydrocarbon fuel coaxial diffusion flame is provided.

[0006] The technical problem solved by the present application adopts the following technical solution:

[0007] A measurement method of soot accumulation particle projection area, comprising the following steps:

[0008] S10, collecting soot accumulation particles;

[0009] S20, obtain a gray scale image of the soot accumulation particles, and calculate the real size represented by a single pixel in the gray scale image;

[0010] S30, process the gray scale image of the soot accumulation particles;

[0011] S40, calculate the projection area of the soot accumulation particles;

[0012] In step S30, the processing of the gray scale image of the soot accumulation particles comprises:

[0013] S301, perform histogram equalization processing on the gray scale image;

[0014] S302, perform Gaussian smoothing processing on the image obtained in step S301;

[0015] S303, use a watershed algorithm to process the edges of the image obtained in step S302;

[0016] S304, binarize the image obtained in step S303, and calculate the area of each soot particle projection region;

[0017] S305, filter the soot accumulation particle projection region in step S304.

[0018] Further, in step S10, the soot accumulation particles are collected by a soot accumulation particle collection device, which comprises an air compressor, an air flow meter, an air heating furnace, a fuel storage bottle, a liquid piston pump, a fuel driving gas source, a nitrogen flow meter, a fuel heating furnace, a burner, and a sampling probe;

[0019] The air compressor, air flow meter, and air heating furnace are connected in sequence to form an air pipeline and are connected with the burner; the fuel driving gas source and the nitrogen flow meter are connected to form a driving gas pipeline and are connected with the fuel heating furnace; the fuel storage bottle and the liquid piston pump are connected to form a fuel pipeline and are connected with the driving gas pipeline; the fuel heating furnace is connected with the burner; the sampling probe is fixed on the shaft of a reciprocating linear motor cylinder and is used to collect soot accumulation particles.

[0020] Further, in step S20, the gray scale image of the soot accumulation particles is obtained by the following method: placing the obtained soot accumulation particle detection sample into a transmission electron microscope, setting the magnification to 30000 times, adjusting the focusing knob until the edges of the soot accumulation particles in the field of view appear white, and obtaining the gray scale image.

[0021] Further, in step S303, the gradient intensity of any pixel point is compared with the gradient intensity of two pixel points along the positive and negative gradient directions, if the gradient intensity of the pixel point is greater than the gradient intensity of the other two pixel points, the pixel point is reserved as an edge point, otherwise the pixel point is suppressed (the gray value is set to 0).

[0022] Further, in step S304, each pixel number contained in each region is counted by performing line-by-line scanning in the formed soot accumulation particle projection region, and the area of each soot accumulation particle projection region can be obtained by counting the number of pixels and converting according to the scale.

[0023] Further, in step S305, the areas of the calculated soot accumulation particle projection regions are sorted in ascending order, a curve is fitted according to the sorted values, then the average value of all connected region area values is obtained, the curve is divided into two curves from the average value, and the midpoints of the two curves are taken as tangent points to draw tangent lines, and the intersection point of the two tangent lines is taken as a threshold value, the regions with area values greater than and equal to the threshold value in the connected region are particle agglomerate regions, and the regions with area values less than the threshold value are smaller particles or image noise points, and the smaller connected region areas are deleted from the results.

[0024] The present application has the following advantages: the measurement method of the soot accumulation particle projection area of the coaxial diffusion flame of the hydrocarbon fuel in the present application automatically analyzes and processes by a computer program, automatically divides the soot accumulation particle projection region and the image background in the image, and filters the noise interference in the image, compared with manual operation, the method can realize batch image analysis, greatly saves the time of manual processing, ensures the measurement accuracy, and avoids subjective errors caused by manual operation. BRIEF DESCRIPTION OF DRAWINGS

[0025] Figure 1 The flowchart of the measurement method of the present application is shown in the figure;

[0026] Figure 2 The structure diagram of the soot accumulation particle collecting device in the present application is shown in the figure;

[0027] Figure 3 The gray scale image of the soot accumulation particle in the present application is shown in the figure;

[0028] Figure 4 The gray scale image of the soot accumulation particle after histogram equalization in the present application is shown in the figure;

[0029] Figure 5 The gray scale image of the soot accumulation particle after Gaussian smoothing processing in the present application is shown in the figure;

[0030] Figure 6A carbon smoke accumulation particle projection area image detected by using a watershed algorithm in the present application;

[0031] Figure 7 A carbon smoke accumulation particle projection area image binarized in the present application.

[0032] The figure is marked as: 1-air compressor; 2-air flow meter; 3-nitrogen flow meter; 4-air heating furnace; 5-fuel driving gas source; 6-liquid piston pump; 7-fuel heating furnace; 8-burner; 9-sampling probe; 10-control computer. DETAILED DESCRIPTION

[0033] The technical solutions of the present application are further described below in combination with examples and drawings.

[0034] Example 1

[0035] A method for measuring the projection area of carbon smoke accumulation particles, comprising the following steps:

[0036] S10, collecting carbon smoke accumulation particles.

[0037] In this embodiment, carbon smoke accumulation particles are collected by a carbon smoke accumulation particle collecting device, as shown in the figure. Figure 2 The carbon smoke accumulation particle collecting device comprises an air compressor 1, an air flow meter 2, an air heating furnace 4, a fuel storage bottle 11, a liquid piston pump 6, a fuel driving gas source 5, a nitrogen flow meter 3, a fuel heating furnace 7, a burner 8 and a sampling probe 9.

[0038] Specifically, the air compressor 1, the air flow meter 2 and the air heating furnace 4 are connected in sequence to form an air pipeline and are connected with the burner 8, for providing hot air with large flow, and at the same time, can block the interference of the external environment to the flame.

[0039] The fuel driving gas source 5 is connected with the nitrogen flow meter 3 to form a driving gas pipeline, and is connected with the fuel heating furnace 7; the fuel storage bottle 11 is connected with the liquid piston pump 6 to form a fuel pipeline, and is connected with the driving gas pipeline. Among them, the fuel is pumped from the fuel storage bottle 11 into the driving gas pipeline by the liquid piston pump 6, and then is carried into the fuel heating furnace 7 by the high-pressure nitrogen gas provided by the fuel driving gas source 5, and the fuel evaporates into fuel vapor inside the fuel heating furnace 7.

[0040] The fuel heating furnace 7 is connected with the burner 8, and the fuel vapor inside the fuel heating furnace 7 enters the burner 8 through the pipeline and mixes with the hot air, and burns in the burner 8.

[0041] The sampling probe 9 is a high-precision self-locking tweezers fixed on the shaft of a reciprocating linear motor, which is used to hold the sampling carrier film to collect the soot accumulation particles. The reciprocating linear motor can accurately control the position and time of the sampling probe through the control of the computer 10, and the time accuracy can be controlled within 30 ms. The sampling carrier film is a 230-mesh (0.063 mm) carbon support film with a diameter of 3 mm and a thickness of 30 μm. When sampling, the sampling probe 9 holds the sampling carrier film and quickly inserts it into the flame center of the burner 8. The copper mesh of the sampling carrier film rapidly cools the hot gas around it. At this time, the soot accumulation particles reach the surface of the copper mesh under the action of thermal force and are adsorbed by the carbon film on the copper mesh.

[0042] S20, obtaining a gray-scale image of the soot accumulation particles and calculating the real size represented by a single pixel in the gray-scale image.

[0043] In this embodiment, the obtained soot accumulation particle detection sample is placed in a LaB6 transmission electron microscope (TEM), and the magnification is set to 30000 times. The focusing knob is adjusted until the edges of the soot accumulation particles in the field of view appear white, a gray-scale image is obtained, and saved to the computer. Preferably, the soot accumulation particle detection sample is derived from a n-heptane coaxial diffusion flame.

[0044] According to the scale generated by the software (marked in the lower right corner of the TEM image) and the screen resolution of the display screen, the real size represented by a single pixel in the calculated gray-scale image is calculated. The specific method is as follows: draw a line segment with the same length as the scale line, and calculate the number of pixels contained in the length of the line segment through the screen resolution, so that the real size represented by a single pixel in the image can be calculated.

[0045] S30, processing the gray-scale image of the soot accumulation particles.

[0046] In this embodiment, the gray-scale image of the soot accumulation particles is processed by the following method:

[0047] S301, histogram equalization processing is performed on the gray-scale image.

[0048] Specifically, let r (i,j) and s (i,j) represent the original image gray scale and the histogram equalized image gray scale of the pixel coordinate (i, j) in the gray-scale image after normalization, respectively, r (i,j) ∈ [0, 1], s (i,j) ∈ [0, 1], when r = s = 0, it represents black; when r = s = 1, it represents white; when r, s ∈ (0, 1), it represents that the pixel gray scale changes between black and white. In the interval [0, 1], for any r (i,j) value, there is a corresponding s (r) value.(i,j) :

[0049] s = T(r) (1)

[0050] In the formula, the transform function T (r) satisfies the following two conditions: T (r) is a monotonically increasing function in 0≤r (i,j) ≤1, ensuring that the order of gray levels from black to white is unchanged; in 0≤r (i,j) ≤1, 0≤s (i,j) ≤1, ensuring that the pixel gray level after the mapping transformation is within the allowed range.

[0051] According to probability theory, given the probability density of a random variable r (i,j) is p r (r), and the random variable s (i,j) is a function of r (i,j) , then the probability density p (i,j) (s) of s s can be obtained from p r (r).

[0052] Let F s (s) represent the distribution function of the random variable s (i,j) , according to the definition of the distribution function, we have:

[0053]

[0054] Since the probability density function is the derivative of the distribution function, taking the derivative of both sides of equation (7) with respect to s gives:

[0055]

[0056] In the formula, is the inverse operation of equation (6).

[0057] To achieve an equal distribution of image gray levels, P s (s) should be a uniform probability density function on [0,1], so P s (s) = 1, and from equation (7) we have:

[0058]

[0059] Therefore, when the transform function T(r) is the cumulative distribution probability of the original image histogram, the original image can be converted into a histogram-equalized gray image.

[0060] On the basis of the histogram, further define the normalized histogram as the relative frequency P r (r k ) of each gray level, i.e.:

[0061]

[0062] wherein r k is the gray value of the kth gray level, n k is the number of pixels in the image with the gray value r k , and N is the total number of pixels in the image.

[0063] By replacing the probability with the frequency, the discrete form of the transform function T (r) may be expressed as:

[0064]

[0065] wherein k = 1, 2,..., L, and L is the current gray level.

[0066] In this step, the histogram of the original image is transformed into an equidistributed form, which can increase the dynamic range of the difference between the gray values of the pixels, thereby enhancing the global contrast of the image.

[0067] S302, performing Gaussian smoothing on the image obtained in step S301.

[0068] Specifically, the specific operation process of Gaussian smoothing is as follows:

[0069] Taking a 5x5 convolution kernel as an example:

[0070]

[0071] Suppose the gray matrix:

[0072] F ij = (b rs ) 5×5 (8)

[0073] wherein b rs is the gray value of the pixel coordinate (r, s), i-2≤r≤i+2, j-2≤s≤j+2.

[0074] Then the pixel gray value of the pixel coordinate (i, j) after Gaussian smoothing is:

[0075] H(i, j) = F ij *G5 (9)

[0076] wherein * is a convolution operator. For the boundary region with insufficient pixel points, the existing points are copied to the corresponding positions on the other side to complete the matrix.

[0077] In this step, by performing Gaussian smoothing, the image noise generated by histogram equalization and the detail level can be reduced to obtain a better image edge.

[0078] S303, using the watershed algorithm to process the edges of the image obtained in step S302.

[0079] Specifically, two sobel operators are provided, respectively:

[0080]

[0081] The pixel gradient matrix of the image in the x, y direction is respectively:

[0082] g x = S x *I,g y = S y *I (11)

[0083] Where I is the gray matrix at pixel coordinates (i, j).

[0084] The gray gradient at pixel coordinates (i, j) can be calculated:

[0085]

[0086] The gradient strength of any pixel point is compared with the gradient strength of two pixel points along the positive and negative gradient directions. If the gradient strength of this pixel point is greater than that of the other two pixel points, the pixel point is retained as an edge point, otherwise the pixel point is removed. Usually, in order to calculate more accurately, linear interpolation is used between the two adjacent pixels across the gradient direction to obtain the pixel gradient to be compared.

[0087] After this step, the pseudo boundary points located at the edge of the soot accumulation particle region and having boundary point characteristics are removed, and the soot accumulation particle projection region and the background are effectively separated.

[0088] S304, binarizing the image obtained in step S303, and calculating the area of each soot accumulation particle projection region.

[0089] Specifically, the overall gray average value of the image processed in step S303 is selected as the threshold value for binarization segmentation, that is:

[0090]

[0091] The pixel gray value less than the threshold value is set to 0, and all values greater than or equal to the gray value are set to 1, that is:

[0092]

[0093] The number of pixels contained in each area is counted by performing line-by-line scanning in the formed soot accumulation particle projection area (black closed area).

[0094] S305, filtering the soot accumulation particle projection area in step S304.

[0095] Specifically, the calculated area of each soot accumulation particle projection area is sorted in order from small to large, and a curve is fitted according to the sorted values, then the average value of all connected area values is calculated, the curve is divided into two curves from the average value, and tangent lines are drawn with the midpoints of the two curves as tangent points, and the intersection point of the two tangent lines is the threshold value for distinguishing particle agglomerates, the area with an area value greater than and equal to the threshold value in the divided connected area is the particle agglomerate area, and the area with an area value less than the threshold value is smaller particles or image noise, and these smaller connected area areas are deleted from the results.

[0096] S40, calculating the soot accumulation particle projection area.

[0097] In this embodiment, the area of the region in the soot accumulation particle projection area with an area value less than the threshold value is deleted, the area of the remaining soot accumulation particle projection area is counted, and the final soot accumulation particle projection area is obtained.

[0098] The TEM image of the same soot particle is measured multiple times by the measurement method of the present application, and the measurement results are shown in Table 1, and the fluctuation of the data values of multiple measurements is within 1.6%.

[0099]

[0100] Table 1

[0101] The measurement method of the soot accumulation particle projection area of the hydrocarbon fuel coaxial diffusion flame of the present application changes all manual operations to automatic analysis and processing by a computer program, automatically divides the soot accumulation particle projection area and the image background in the image, and filters the noise interference in the image, compared with manual operation, this method can realize batch image analysis, greatly saves the time of manual processing, ensures the measurement accuracy, and avoids subjective errors caused by manual operation.

[0102] The above sequence of embodiments is only for description, and does not represent the advantages and disadvantages of the embodiments.

[0103] It should be pointed out finally that the above embodiments are only used to illustrate the technical solutions of the present application, but not to limit the same; and although the present application has been described in detail with reference to the foregoing embodiments, it should be appreciated by those skilled in the art that the technical solutions recorded in the foregoing embodiments can be modified, or some technical features thereof can be replaced equivalently; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method of measuring the projected area of soot accumulation particles, characterized by, The method comprises the following steps: S10, collecting carbon soot accumulation particles; S20, obtaining a gray scale image of the carbon soot accumulation particles and calculating the real size represented by a single pixel in the gray scale image; S30, processing the gray scale image of the carbon soot accumulation particles; S40, calculating the projection area of the carbon soot accumulation particles; In step S30, the processing of the gray scale image of the carbon soot accumulation particles comprises: S301, histogram equalization processing of the gray scale image; S302, Gaussian smoothing processing of the image obtained in step S301; S303, using a watershed algorithm to process the edges of the image obtained in step S302: comparing the gradient intensity of any pixel point with the gradient intensities of two pixel points along the positive and negative gradient directions, if the gradient intensity of the pixel point is greater than the gradient intensities of the other two pixel points, the pixel point is retained as an edge point, otherwise the pixel point is suppressed; S304, binarizing the image obtained in step S303 and calculating the area of each carbon soot particle projection region: performing line-by-line scanning in the formed carbon soot accumulation particle projection region to count the number of pixels contained in each region; the area of each carbon soot accumulation particle projection region can be obtained by counting the number of pixels and converting according to the scale; S305, filtering the carbon soot accumulation particle projection region in step S304: sorting the calculated areas of each carbon soot accumulation particle projection region in ascending order, fitting a curve according to the sorted values, then obtaining the average value of all connected region area values, dividing the curve into two curves from the average value, and making tangent lines with the midpoints of the two curves as tangent points, and taking the intersection point of the two tangent lines as a threshold, the regions with area values greater than and equal to the threshold in the connected region are particle agglomerate regions, and the regions with area values less than the threshold are smaller particles or image noise, and the smaller connected region areas are deleted from the results.

2. The method of measuring the projected area of soot accumulation particles according to claim 1, wherein In step S10, the carbon soot accumulation particles are collected by a carbon soot accumulation particle collecting device, which comprises an air compressor, an air flowmeter, an air heating furnace, a fuel storage bottle, a liquid piston pump, a fuel driving gas source, a nitrogen flowmeter, a fuel heating furnace, a burner and a sampling probe. The air compressor, the air flowmeter and the air heating furnace are connected in sequence to form an air pipeline and are connected with the burner; the fuel driving gas source and the nitrogen flowmeter are connected to form a driving gas pipeline and are connected with the fuel heating furnace; the fuel storage bottle and the liquid piston pump are connected to form a fuel pipeline and are connected with the driving gas pipeline; the fuel heating furnace is connected with the burner; the sampling probe is fixed on the shaft of a reciprocating linear motor cylinder and is used for collecting carbon soot accumulation particles.

3. The method of measuring the projected area of soot accumulation particles according to claim 1, wherein In step S20, the gray scale image of the carbon soot accumulation particles is obtained by the following method: placing the obtained carbon soot accumulation particle detection sample into a transmission electron microscope, setting the magnification to 30000 times, adjusting the focusing knob until the edges of the carbon soot accumulation particles in the field of view appear white, and obtaining the gray scale image.

Images