Particle detection device based on dual-wavelength illumination and online identification method thereof

Abstract

The invention discloses a particle detection device based on dual-wavelength illumination and an online identification method thereof, and relates to the field of particle optical detection, and the particle detection device is characterized by comprising a collimation particle flow generation module, a multi-wavelength combined light illumination module, a light splitting and sampling module and a differential processing module. According to the particle detection device and the online recognition method thereof, different characteristics of particles can be detected, meanwhile, after the differential processing module carries out differential processing on the electric pulse signals, the recognition influence of shapes on particle components can be reduced, and the detection accuracy and the measurement precision are improved.

Description

Technical Field

[0001] The present invention relates to the field of particle optical detection, and in particular to a particle detection device based on dual-wavelength illumination and an online identification method thereof. Background Art

[0002] Accurate particulate matter detection is of great significance. In the environmental sector, it enables precise tracing of pollution sources, refined air quality management, and accurate assessment of ecological impacts. In industrial applications, it helps improve product quality, optimize production processes, and control costs. In medicine and health, it enables more precise disease diagnosis and optimizes drug delivery and treatment. In short, particulate matter classification and detection plays a key role in multiple fields, providing important support for environmental protection, industrial development, and human health.

[0003] Laser diffraction is currently a key method for particle detection. It can quickly measure particle size distribution, making it suitable for rapid, real-time particle size monitoring on production lines. The sample can be a dry powder, liquid, suspension, or aerosol. By selecting different measurement configurations and optical settings, a wide range of particle sizes can be covered, typically from tens of nanometers to several millimeters. This method can generally measure particle diameter and distribution.

[0004] However, due to the need for refined detection, it is often necessary to identify the composition of single particles. By using laser scattering technology and polarization analysis of the scattered signals, researchers have found that the composition of different particles can be classified.

[0005] However, the light source used in laser scattering technology is usually a single-wavelength laser beam, which is easily affected by the external environment and may cause problems such as wavelength drift, affecting output stability and resulting in reduced measurement accuracy.

[0006] While the above classification generally has a high accuracy rate for round particles, when particle shapes vary significantly, the difference in polarized light may not be sufficient to effectively distinguish different types of particles. In practical applications, particles often exhibit different shapes, with common ones being flakes, needles, blocks, and fibers. Due to their irregular shapes, these particles interact differently with the laser, affecting the distribution pattern of scattered light and, in turn, the accuracy of particle diameter detection, making it difficult to effectively distinguish particle composition through analysis of polarization signals. Summary of the Invention

[0007] The technical problem solved by the present invention is that conventional laser scattering technology uses single-wavelength laser illumination, resulting in low measurement accuracy. At the same time, the detection of particles is easily affected by the particle shape, which affects the accuracy of particle diameter detection and makes it difficult to identify particle components.

[0008] In a first aspect, the present invention provides a particle detection device based on dual-wavelength illumination, characterized in that it includes a collimated particle flow generation module, a multi-wavelength combined light illumination module, a spectroscopic and sampling module, and a differential processing module;

[0009] A collimated particle flow generation module, used for forming a collimated particle flow to be measured;

[0010] A multi-wavelength combined light illumination module, configured to provide combined light sources of different wavelengths to illuminate the particle flow, comprising a laser group and a front dichroic mirror. The laser group comprises two or more lasers of different wavelengths. The light beams of different wavelengths emitted by the laser group are combined by the front dichroic mirror and then illuminate the particle flow in a parallel light manner at the same angle.

[0011] A light splitting and sampling module is used to split the scattered light within the same angular range of the particle flow into scattered light of different wavelengths. The scattered light of different wavelengths is collected by corresponding photoelectric detectors and converted into corresponding electrical pulse signals;

[0012] The differential data processing module is used to perform differential data processing on the electrical pulse signals of scattered light of different wavelengths to reduce the influence of shape on the identification and classification of the particle flow.

[0013] The collimated particle flow generation module includes an air pump and a nozzle. The nozzle is connected to the air pump. Different types of micro-nano particles are pumped by the air pump and form collimated particle flows through the nozzle. The particle flows are in an aerosol state.

[0014] A polarizer and a quarter-wave plate are arranged in parallel between the front dichroic mirror and the particle flow. The polarizer is arranged close to the front dichroic mirror. The combined light beam passing through the front dichroic mirror passes through the polarizer and the quarter-wave plate in sequence to form a circularly polarized light beam that shines onto the particle flow.

[0015] The laser group of the multi-wavelength combined light illumination module includes a red laser and a green laser. The red laser and the green laser are arranged perpendicular to each other according to the beam transmission direction. The light beams emitted by the red laser and the green laser are combined by the front dichroic mirror and then illuminate the particle flow.

[0016] The spectrometry and sampling module includes two detector arrays and two narrowband filters. The two detector arrays are arranged at 90° and -90° relative to the transmission direction of the combined light beam. The two narrowband filters are respectively arranged between the two detector arrays and the particle flow. The wavelengths of the two narrowband filters correspond to the wavelengths of the red laser and the green laser, respectively.

[0017] Or the spectroscopic and sampling module includes two detectors, a rear dichroic mirror, and a parabolic mirror. The parabolic mirror is arranged between the quarter-wave plate and the particle flow, the particle flow is located at the focal position of the parabolic mirror, and the rear dichroic mirror is arranged above the parabolic mirror. The rear dichroic mirror splits the scattered light of the particle flow into lights of different wavelengths, which are then collected by the two detectors and converted into electrical pulse signals of different wavelengths.

[0018] Or the spectroscopic and sampling module includes two groups of sampling units with the same structure, and the two groups of sampling units are arranged at 90° and -90° relative to the transmission direction of the combined light beam. The sampling units include focusing lenses, beam splitters, linear polarizers, filters, and detectors. The scattered light directed toward the particle flow is directed toward the beam splitter through the focusing lens of one of the sampling units and is split into two light beams of different wavelengths. The two light beams of different wavelengths pass through the linear polarizer and the filter in sequence and are collected by the two detectors and converted into electrical pulse signals of different wavelengths. The scattered light directed toward the particle flow is directed toward the beam splitter through the focusing lens of the other sampling unit and is split into another two light beams of different wavelengths. The two light beams of different wavelengths pass through the linear polarizer and the filter in sequence and are collected by the other two detectors and converted into electrical pulse signals of different wavelengths.

[0019] In a second aspect, the present invention provides an online particle identification method based on dual-wavelength illumination, which is applied to the aforementioned particle detection device based on dual-wavelength illumination, and the method comprises:

[0020] The collimated particle flow generation module generates a collimated particle flow from powdered micro-nano particles, and the collimated particle flow is emitted to the multi-wavelength combined light illumination module;

[0021] The multi-wavelength beam combining illumination module includes a laser group and a front dichroic mirror. The laser group includes two or more lasers with different wavelengths. The laser group respectively emits two or more laser beams with different wavelengths, which are combined by the front dichroic mirror and then illuminate the particle flow in a parallel light manner.

[0022] The light splitting and sampling module splits the scattered light of the particle flow into scattered light of different wavelengths, and converts the collected scattered light of different wavelengths into electrical pulse signals of different wavelengths;

[0023] The differential data processing module performs differential processing on the electrical pulse signals of different wavelengths, and judges the similarity of the signals based on the differential processing results, thereby reducing the influence of the shape on the identification and classification of the particle flow.

[0024] The differential data processing module includes processing the electrical pulse signals of different wavelengths using mean square error (MSE), and / or processing the electrical pulse signals of different wavelengths using mean absolute error (MAE).

[0025] The differential data processing module further includes classifying particles based on Mie scattering intensity or Rayleigh scattering intensity, and setting the ratio of particle size to wavelength to x: Where r is the particle radius, λ is the wavelength, when x<1, Rayleigh scattering intensity processing is used; when x≥1, Mie scattering intensity processing is used; and then N-order derivative processing is performed after processing the Mie scattering intensity and Rayleigh scattering intensity respectively, where N is an integer between 2 and 6.

[0026] The technical effects of the present invention are as follows:

[0027] The present invention provides a particle detection device based on dual-wavelength illumination and an online identification method thereof. The detection device provides two or more illumination light beams of different wavelengths in a multi-wavelength combined light illumination module. The spectroscopic and sampling module splits the scattered light on the particle flow into different wavelength regions within the same angle range and converts them into electrical pulse signals, thereby being able to detect different characteristics of the particles. At the same time, the differential processing module differentially processes the electrical pulse signals of different wavelengths, thereby reducing the influence of the shape on the identification of the particle components and improving the detection accuracy and measurement precision. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 This is a schematic diagram of the structure of the connection between the air pump outlet and the tapered nozzle in the present invention;

[0029] Figure 2 This is a schematic diagram of the optical design principles of the multi-wavelength combined light illumination module and the light splitting and sampling module (scattered light in the upward and downward directions) of the first embodiment of the present invention;

[0030] Figure 3 This is the electrical pulse diagram of the same particle with different shapes under a single wavelength laser source in the prior art;

[0031] Figure 4 This is the electrical pulse diagram of the same particle with different shapes under the dual-wavelength laser source of the present invention;

[0032] Figure 5 The present invention makes a difference map of scattered light signals of different shapes of the same particles irradiated with different wavelengths under a dual-wavelength laser source;

[0033] Figure 6 The present invention makes a diagram of scattered light signals of different shapes of the same particles irradiated with different wavelengths under a dual-wavelength laser source;

[0034] Figure 7 This is the electrical pulse diagram of different particles under a single wavelength laser source in the prior art;

[0035] Figure 8 This is an electric pulse diagram after differential processing of scattered light signals from different particles under the dual-wavelength laser source of the present invention;

[0036] Figure 9 The electrical pulse diagrams of different particle diameters and shapes in the prior art;

[0037] Figure 10 Particle size distribution diagrams of different particles used in the present invention;

[0038] Figure 11 This is a diagram showing the classification effect of different particles after multi-order derivative difference processing in the present invention;

[0039] Figure 12 This is a schematic diagram of the optical design principles of the multi-wavelength beam combining illumination module, the light splitting and sampling module (backscattered light) in the second embodiment of the present invention;

[0040] Figure 13 This is a schematic diagram of the optical design principles of the multi-wavelength beam combining illumination module, the light splitting and sampling module (upward and downward scattered light) of the third embodiment of the present invention;

[0041] Figure 14 This is a schematic diagram of the optical design principles of the multi-wavelength combined light illumination module, the light splitting and sampling module (scattered light in the same angle range) according to the fourth embodiment of the present invention.

[0042] Description of Reference Numerals

[0043] 1. Red laser; 2. Green laser; 3. Front dichroic mirror; 4. Particle stream; 5. Polarizer; 6. Quarter-wave plate; 7. Detector array; Detectors (9, 10); 11. Rear dichroic mirror; 12. Parabolic mirror; 13. Focusing lens; 14. Beam splitter; 15. Linear polarizer; 16. Filter; Detectors (17, 18). DETAILED DESCRIPTION

[0044] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0045] Among them, the same parts are represented by the same figure marks. It should be noted that the words "front", "rear", "left", "right", "up" and "down" used in the following description refer to the directions in the drawings of the present application specification, and the words "bottom" and "top", "inside" and "outside" refer to directions toward or away from specific parts, respectively. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present application specification, the meaning of "multiple" is two or more.

[0046] The present application will be further described below with reference to the accompanying drawings and embodiments.

[0047] Example 1

[0048] A particle detection device based on dual-wavelength illumination comprises a collimated particle flow generation module, a multi-wavelength combined light illumination module, a light splitting and sampling module and a differential processing module.

[0049] The collimated particle flow generation module is used to generate the collimated particle flow to be measured. Specifically, a sample air pump (AT-772) is used to quickly and uniformly excite different types of micro- and nano-particles into a stable aerosol-like particle flow. A sheath flow air pump (NMP 830KNDC) is also equipped to focus and stabilize the particle flow by pumping sheath air at four times the speed of the sample air pump. A recovery air pump (NMP 830KNDC) is also provided to recover the particle flow.

[0050] like Figure 1 As shown in the figure, the coaxial tapered nozzle design optimizes the fluid dynamics and significantly reduces the losses caused by particle collisions, diffusion, or turbulence in traditional nozzles. In addition, the focusing of the particles reduces the area where the beam is required, which in turn reduces power consumption and allows the use of low-power light sources.

[0051] The multi-wavelength combined beam illumination module is used to provide two or more combined beam light sources with different wavelengths to illuminate the particle flow. Figure 2As shown, this embodiment provides two lasers. Specifically, the module includes a red laser 1, a green laser 2, and a front dichroic mirror 3. The red laser 1 and the green laser 2 are arranged perpendicular to each other according to the beam transmission direction. In the figure, the beam transmission direction is the X direction, and the direction perpendicular to it is the Y direction. The beams emitted by the red laser 1 and the green laser 2 are combined by the front dichroic mirror 3 and then directed toward the particle stream 4. The two lasers can be two semiconductor laser sources HL6544FM and L520P50, which can generate a red laser beam with an intensity of 50mW and a wavelength of 660nm and a green laser beam of 520nm, respectively. The front dichroic mirror 3 uses a DMSP650 to combine laser beams of different wavelengths.

[0052] A polarizer 5 and a quarter-wave plate 6 are also provided between the front dichroic mirror 3 and the particle beam 4. The multi-wavelength linear laser beams combined by the front dichroic mirror 3 are converted into circularly polarized light after passing through the polarizer 5 and the quarter-wave plate 6 in sequence and then irradiated onto the particle flow.

[0053] The spectrometer and sampling module is used to split the backscattered light of the particle flow into red scattered light and green scattered light, and convert the red scattered light and green scattered light into red photoelectric pulse signals and green photoelectric pulse signals. The spectrometer and sampling module includes two detector arrays 7 and two narrowband filters 8. The two detector arrays 7 use AXUV20ELG and are arranged at 90° and -90° relative to the transmission direction of the combined light beam. The two narrowband filters 8 are respectively located between the two detector arrays 7 and the particle flow 4. The wavelengths of the two narrowband filters 8 correspond to the wavelengths of the red laser 1 and the green laser 2, respectively. The two detector arrays 7 detect scattered light of different wavelengths within the same angular range, which includes scattered light within the ±5° range of the detection area directly opposite the detector arrays, and convert it into red photoelectric pulse signals and green photoelectric pulse signals. These electrical pulse signals are used to measure different characteristics of the particle flow.

[0054] In order to demonstrate the influence of irregular shapes on the accuracy of particle component identification, we present the scattered photoelectric signals collected by the above detector array in different directions for the same particle in three different shapes and merge them. Figure 3 As can be seen from the electrical pulse diagrams of the same particle with different shapes using a single-wavelength laser source, when the particle cross-section is circular (shape 1), its scattering cross-section exhibits a certain regular shape. When shapes 2 and 3 are irregular, the corresponding scattering cross-sections also exhibit irregular shapes. In real-world scenarios, particles often deviate from a spherical shape and appear in a variety of shapes. Therefore, it is necessary to consider the influence of shape in scattering characterization and measurement. When the particle deviates from a spherical shape, the Mie scattering theory requires the introduction of a shape correction factor F(q) to modify the scattering intensity distribution.

[0055] According to the Mie scattering efficiency Q, this parameter characterizes the ratio of the particle's ability to scatter light to its geometric cross-sectional area. Its theoretical calculation formula is:

[0056]

[0057] in is the size parameter, λ is the wavelength of light in the medium, d is the equivalent diameter of the particle, n is the refractive index, a n , b n is the Mie coefficient related to m,x and the complex refractive index of the particle m=n+ik, where the real part n is the refractive index and the imaginary part k is the absorption coefficient, which is related to the type of material. It can be calculated by recursion or special functions (such as Bessel function and Hankel function):

[0058]

[0059] The shape factor F(q) characterizes the modulation of the scattering amplitude by the particle shape, where q is the scattering vector where θ is the scattering angle.

[0060] The corrected scattering intensity of non-spherical particles I(θ) can be expressed as: I(θ) = I Mie (θ)|F(q)| 2

[0061] The following are some examples of shape factors for non-spherical particles:

[0062] (1) Ellipsoid: in a and b are the major axis and minor axis respectively, and φ is the angle between q and the major axis.

[0063] (2) Cylinder: in q || =qcosψ, ψ is the angle between the scattering angle and the cylinder axis.

[0064] (3) Thin disc: Where R is the radius of the disk.

[0065] The above analysis shows that particles of different shapes, or parts of particles of different shapes, significantly influence the light scattering cross section. Irregular shapes negatively impact particle size detection and classification. In this application, a differential data processing module is used to perform differential data processing on the red and green photoelectric pulse signals, effectively avoiding or reducing shape-related errors in analyzing particle scattering signals.

[0066] Specifically, the differential data processing module includes processing the red photoelectric pulse signal and the green photoelectric pulse signal using the mean square error (MSE), and / or processing the red photoelectric pulse signal and the green photoelectric pulse signal using the mean absolute error (MAE).

[0067] From the above analysis, it is clear that the shape factor is actually independent of the particle's illumination wavelength. Therefore, the influence of the shape factor on the particle's scattering cross section can be effectively eliminated by performing differential processing on the scattered signal, such as a division operation. To achieve this, at least two wavelengths of light sources are generally required to combine the beams.

[0068] In this embodiment, by using dual-wavelength illumination, scattered light of different wavelengths is collected in two directions that are symmetrical and perpendicular to the incident light beam, and the influence of irregular shaped scattered signals on our detection accuracy is reduced through the differential data processing module. Figure 4 The figure shows the scattering cross-section signals of three different shapes under dual wavelengths. It can be seen that due to different wavelengths, the scattering signal intensities are different, but the shape of the scattering cross-section is basically unchanged when the same angle of illumination and the same angle of collection are used. Therefore, the red photoelectric pulse signal and the green photoelectric pulse signal are processed by using the mean square error MSE and the mean absolute error MAE. Figure 5 and Figure 6 As shown, after differential data processing, the influence of shape scattering signal analysis and particle composition classification can be basically offset.

[0069] The uncertainty of the scattering cross-section signal under single wavelength and dual wavelength is analyzed below. The similarity of the signals can be judged by the mean square error (MSE) and mean absolute error (MAE). The mathematical calculation formulas for MSE and MAE are as follows:

[0070]

[0071] Where y1i is the value of the first curve, y2i is the value of the second curve, and N is the sample value.

[0072] The following table shows the average values of MSE and MAE calculated for each of the three different shapes of single-wavelength signals (electrical pulse signals collected using a single wavelength laser illumination in the prior art) and dual-wavelength signals (electrical pulse signals collected using two wavelength laser illumination in the present invention):

[0073] Single wavelength signal Dual wavelength signal difference Dual wavelength signal division MSE mean 0.0592 0.0204 0.000325 MAE mean 0.1879 0.1251 0.013423

[0074] Generally speaking, when the MSE is less than 0.01 and the MAE is less than 0.1, the differences between the arrays can be considered small. The table above shows that after subtracting the dual-wavelength signals, while the mean MSE and MAE values are not exactly less than 0.01 and 0.1, they are within an acceptable range. Notably, after dividing the dual-wavelength signals, the MSE and MAE values are both significantly less than 0.01 and 0.1, respectively. This difference is considered a significant improvement in normalized signals.

[0075] The data processing module also includes classifying particles based on Mie scattering intensity or Rayleigh scattering intensity, and setting the ratio of particle size to wavelength to x: Where r is the particle radius, λ is the wavelength, when x < 1, Rayleigh scattering intensity processing is used; when x ≥ 1, Mie scattering intensity processing is used; and then the Mie scattering intensity and Rayleigh scattering intensity processing are respectively subjected to N-order derivative processing, where N is an integer between 2 and 6.

[0076] After using differential processing MSE and MAE to effectively reduce the detection error caused by particle shape factors, we classify three different particles based on the processed high-similarity data in combination with Mie scattering or Rayleigh scattering. As mentioned above, the Mie scattering efficiency Q is closely related to the material type of the particles. Therefore, the different scattering efficiencies of different types of particles will lead to different electric pulse signal intensities detected by the detector. If the electric pulse signal obtained by a single-wavelength light source in the existing technology is directly used as the classification basis, the following will be obtained: Figure 7 The classification of .

[0077] Although the average electric pulse signal intensity of various types of particles is obviously different, e.g. Figure 9 However, when particles of similar size overlap due to the complexity of their shapes, the electrical pulse signals from different types of particles significantly reduce classification accuracy. However, the dual-wavelength signal differential processing proposed in this invention averages the amplitudes of the scattered signals from different particles, thereby amplifying the differences in electrical pulse signal amplitudes caused by different particle materials and using this as a basis for classification, significantly enhancing classification accuracy.

[0078] like Figure 8 As shown in the figure, the particles are classified using the differentially processed signal. The classification results show that the above method can effectively amplify the pulse amplitude characteristics of different types of particles and distinguish the characteristic signals of particles of different materials.

[0079] like Figure 10 The particle size distributions of the different particles shown are different. Mie scattering occurs for larger particles, while Rayleigh scattering occurs for smaller particles. The scattering intensity I of Rayleigh scattering is related to the following factors:

[0080]

[0081] Where I0 is the incident light intensity, λ is the incident light wavelength, is the ratio of the refractive index of the particle to the refractive index of the medium, d is the particle diameter, and θ is the scattering angle.

[0082] The above formula shows that the scattering cross section of Rayleigh scattering is proportional to the sixth power of the particle diameter, and the diameter has a great influence on its component classification. In order to reduce the influence of particle diameter, the Nth-order derivative is used to process it. After Rayleigh scattering, the sixth-order derivative of the above formula is taken to become the following form:

[0083]

[0084] By doing this, when particles are classified by scattering intensity, the effect of their size is significantly reduced. Figure 11 As shown in the figure, for the same kind of particles, the influence of different shapes and diameters is significantly reduced, while for different particles, the curve distribution differences are more obvious.

[0085] This approach has two advantages:

[0086] 1. The principle of using the high-order derivatives of light intensity with respect to wavelength to amplify the differences in the spectral characteristics of materials is particularly suitable for materials with similar absorption peaks.

[0087] 2. This method can reduce the influence of particle size on scattering measurement and particle classification. After multiple-order derivatives, it can be found that the influence decreases in sequence.

[0088] Through mathematical differentiation, the mechanism is based on high-order derivatives of the scattered signal with respect to wavelength, which strips the intrinsic absorption characteristics of the material from the scattering background. Specifically, when the original spectrum is transformed into an N-order derivative (N is an integer between 2 and 6), the broad scattering components caused by particle size and shape decay exponentially as the derivative order increases, while the narrowband absorption peaks related to the material are retained by retaining their differential extremes, thus extracting the characteristics of the particle material.

[0089] In addition, we give the calculation formula of Mie scattered light intensity I(θ) and scattering angle θ:

[0090]

[0091] Where I0 is the incident light intensity, r is the distance from the observation point to the scatterer, φ is the angle between the polarization direction and the scattering surface, S1(θ) and S2(θ) are the Mie scattering amplitude functions, corresponding to the polarization components perpendicular and parallel to the scattering surface, respectively. And:

[0092]

[0093] In the above formula, a n and b n is the Mie coefficient, the mathematical formula of which has been given above and is related to the particle size parameter x and the complex refractive index m; π n and τ n is the angular distribution function, and the Legendre polynomial Related:

[0094]

[0095] The above definition is the size parameter. Combined with the above-mentioned formula for calculating the meter coefficient, we can know that the meter coefficient a n , b n It is proportional to the size parameter x, that is, the particle diameter d, while the Mie scattering amplitude function S1(θ) and S2(θ) are proportional to the Mie coefficient a n , b n The Mie scattering intensity I(θ) is proportional to the square of the Mie scattering amplitude function S1(θ) and S2(θ). Therefore, for particles that meet the Mie scattering conditions, the Mie scattering intensity is positively correlated with the square of the particle diameter. We can take the second derivative of the Mie scattering intensity function to reduce the detection error caused by particles of different diameters.

[0096] Example 2

[0097] The difference between the second embodiment and the first embodiment is that the light splitting and sampling modules used are different. Specifically, Figure 12 As shown, the light splitting and sampling module includes two detectors (9, 10), a rear dichroic mirror 11, and a parabolic mirror 12. The parabolic mirror 12 is arranged between the quarter-wave plate 6 and the particle flow 4. The particle flow 4 is located at the focal position of the parabolic mirror 12. The rear dichroic mirror 11 is arranged above the parabolic mirror 12. The rear dichroic mirror 11 splits the scattered light of the particle flow 4 into different wavelengths, which are then collected by the two detectors (9, 10) and converted into red photoelectric pulse signals and green photoelectric pulse signals. The combined light beam formed by the two lasers passes through a 3mm circular hole reserved in the parabolic mirror 12 and shines on the particle flow 4. Since the particle flow 4 is located at the focal position of the parabolic mirror 12, its scattered light is collimated by the parabolic mirror 12 and then split by the rear dichroic mirror 11 before being received by the two detectors (9, 10).

[0098] Example 3

[0099] In this embodiment, it is still possible to add a polarizer and a wave plate to the multi-wavelength combined light illumination module of this embodiment. The polarizer and wave plate are arranged between the front dichroic mirror and the particle flow. In the diagram of this embodiment three, these components are omitted. The biggest difference between this embodiment three and embodiment one is still that the light splitting and sampling modules used are different. Specifically, Figure 13 As shown, the light splitting and sampling module includes two groups of sampling units with the same structure. The two groups of sampling units are arranged in the 90° direction and -90° direction relative to the transmission direction of the combined light beam. Taking one group of sampling units as an example, one group of sampling units includes a focusing lens 13, a beam splitter 14, a linear polarizer 15, a filter 16, and a detector (17, 18). The scattered light directed to the particle flow 4 passes through the focusing lens 13 of one sampling unit and then is directed to the beam splitter 14 and is split into two light beams of different wavelengths. The two light beams of different wavelengths pass through the focusing lens 13 of one sampling unit and are respectively split into two light beams of different wavelengths. The linear polarizer 15 and the filter 16 are collected by two detectors (17, 18) and converted into a first red photoelectric pulse signal and a first green photoelectric pulse signal. Then, another set of sampling units is as follows: the scattered light directed toward the particle flow 4 is directed toward the beam splitter 14 through the focusing lens 13 of another sampling unit and is split into two other light beams of different wavelengths. The two light beams of different wavelengths are sequentially collected by the linear polarizer 15 and the filter 16 by the two detectors (17, 18) and converted into a second red photoelectric pulse signal and a second green photoelectric pulse signal. In this embodiment, laser 1 can emit a continuous laser source at a wavelength of 660nm, and laser 2 can emit a continuous laser source at a wavelength of 520nm. The lasers emitted by the two laser sources are respectively transmitted and reflected by the front dichroic mirror 3 to form a combined beam of light that is irradiated toward the particle 4. The scattered light in the 90° and -90° directions passes through the focusing lens 13 and then is irradiated toward the beam splitter 9 and is split into two light beams with wavelengths of 660nm and 520nm respectively. The two light beams pass through the linear polarizer 15 and the filter 16 in turn and then are irradiated toward the detectors (17, 18) and converted into electrical pulse signals.

[0100] Example 4

[0101] In this embodiment, Figure 14 As shown, the biggest difference between it and other embodiments is that the multi-wavelength combined light illumination module of this embodiment includes four lasers of different wavelengths. The light source emitted by each laser is directed to the front dichroic mirror for combining after passing through a collimating lens. The combined parallel light beam is directed to the particle flow. The spectroscopic and sampling module includes four photodetectors. The scattered light area of the particle flow is arranged with four photodetectors through filters placed in a 2×2 array. Since the photodetectors use micro-detectors with a size of 3mm×3mm, it can be approximately considered that the four photodetectors detect scattered light within the same angle range, thereby realizing the collection of more electrical pulse signals. Through more sampling data, the detection accuracy and measurement accuracy are further guaranteed.

[0102] The present invention provides an online particle identification method based on dual-wavelength illumination, which is applied to the above-mentioned particle detection device based on dual-wavelength illumination. The method includes:

[0103] The collimated particle flow generation module generates a collimated particle flow from powdered micro-nano particles, and the collimated particle flow is emitted to the multi-wavelength combined light illumination module;

[0104] The multi-wavelength beam combining illumination module includes a laser group and a front dichroic mirror, wherein the laser group includes two or more lasers of different wavelengths, and the laser group respectively emits two or more laser beams of different wavelengths, which are combined by the front dichroic mirror and then illuminate the particle flow in a parallel light manner;

[0105] The spectrometer and sampling module splits the scattered light of the particle flow into scattered light of different wavelengths, and converts the collected scattered light of different wavelengths into electrical pulse signals of different wavelengths;

[0106] The differential data processing module performs differential processing on the electric pulse signals of different wavelengths, and judges the similarity of the signals based on the differential processing results, thereby reducing the influence of the shape on the identification and classification of the particle flow.

[0107] Specifically, the differential data processing module includes processing the electrical pulse signals of different wavelengths using the mean square error (MSE), and / or processing the electrical pulse signals of different wavelengths using the mean absolute error (MAE).

[0108] Specifically, the differential data processing module further includes classifying particles based on Mie scattering intensity or Rayleigh scattering intensity, and setting the ratio of particle size to wavelength as x: Where r is the particle radius, λ is the wavelength,

[0109] When x<1, Rayleigh scattering intensity processing is used; when x≥1, Mie scattering intensity processing is used; and then N-order derivative processing is performed on the Mie scattering intensity and Rayleigh scattering intensity after processing, where N is an integer between 2 and 6.

Claims

1. A particle detection device based on dual-wavelength illumination, characterized in that: It includes a collimated particle flow generation module, a multi-wavelength beam combining light illumination module, a light splitting and sampling module and a differential processing module; A collimated particle flow generation module, used for forming a collimated particle flow to be measured; A multi-wavelength combined light illumination module, configured to provide combined light sources of different wavelengths to illuminate the particle flow, comprising a laser group and a front dichroic mirror. The laser group comprises two or more lasers of different wavelengths. The light beams of different wavelengths emitted by the laser group are combined by the front dichroic mirror and then illuminate the particle flow in a parallel light manner at the same angle. A light splitting and sampling module is used to split the scattered light within the same angular range of the particle flow into scattered light of different wavelengths. The scattered light of different wavelengths is collected by corresponding photoelectric detectors and converted into corresponding electrical pulse signals; The differential data processing module is used to perform differential data processing on the electrical pulse signals of scattered light of different wavelengths to reduce the influence of shape on the identification and classification of the particle flow.

2. A particle detection device based on dual-wavelength illumination according to claim 1, characterized in that: The collimated particle flow generation module includes an air pump and a nozzle. The nozzle is connected to the air pump. Different types of micro-nano particles are pumped by the air pump and form collimated particle flows through the nozzle. The particle flows are in an aerosol state.

3. The particle detection device based on dual-wavelength illumination according to claim 1, characterized in that: A polarizer and a quarter-wave plate are arranged in parallel between the front dichroic mirror and the particle flow. The polarizer is arranged close to the front dichroic mirror. The combined light beam passing through the front dichroic mirror passes through the polarizer and the quarter-wave plate in sequence to form a circularly polarized light beam that shines onto the particle flow.

4. The particle detection device based on dual-wavelength illumination according to claim 3, characterized in that: The laser group of the multi-wavelength combined light illumination module includes a red laser and a green laser. The red laser and the green laser are arranged perpendicular to each other according to the beam transmission direction. The light beams emitted by the red laser and the green laser are combined by the front dichroic mirror and then illuminate the particle flow.

5. The particle detection device based on dual-wavelength illumination according to claim 4, characterized in that: The spectrometry and sampling module includes two detector arrays and two narrowband filters. The two detector arrays are arranged at 90° and -90° relative to the transmission direction of the combined light beam. The two narrowband filters are respectively arranged between the two detector arrays and the particle flow. The wavelengths of the two narrowband filters correspond to the wavelengths of the red laser and the green laser, respectively.

6. The particle detection device based on dual-wavelength illumination according to claim 4, characterized in that: The spectroscopic and sampling module includes two detectors, a rear dichroic mirror, and a parabolic mirror. The parabolic mirror is arranged between the quarter-wave plate and the particle flow. The particle flow is located at the focal position of the parabolic mirror. The rear dichroic mirror is arranged above the parabolic mirror. The rear dichroic mirror splits the scattered light of the particle flow into light of different wavelengths, which are then collected by the two detectors and converted into red photoelectric pulse signals and green photoelectric pulse signals.

7. The particle detection device based on dual-wavelength illumination according to claim 4, characterized in that: The spectroscopic and sampling module includes two groups of sampling units with identical structures, which are arranged at 90° and -90° relative to the transmission direction of the combined light beam. The sampling units include focusing lenses, beam splitters, linear polarizers, filters, and detectors. The scattered light directed toward the particle flow is directed toward the beam splitter through the focusing lens of one sampling unit and is split into two light beams of different wavelengths. The two light beams of different wavelengths pass through the linear polarizer and the filter in sequence and are collected by the two detectors and converted into a first red photoelectric pulse signal and a first green photoelectric pulse signal. The scattered light directed toward the particle flow is directed toward the beam splitter through the focusing lens of the other sampling unit and is split into two light beams of different wavelengths. The two light beams of different wavelengths pass through the linear polarizer and the filter in sequence and are collected by the other two detectors and converted into a second red photoelectric pulse signal and a second green photoelectric pulse signal.

8. A method for online particle identification based on dual-wavelength illumination, characterized in that: Applied to the particle detection device based on dual-wavelength illumination, the method includes: The collimated particle flow generation module generates a collimated particle flow from powdered micro-nano particles, and emits the collimated particle flow to the multi-wavelength combined light illumination module; The multi-wavelength beam combining illumination module includes a laser group and a front dichroic mirror. The laser group includes two or more lasers with different wavelengths. The laser group respectively emits two or more laser beams with different wavelengths, which are combined by the front dichroic mirror and then illuminate the particle flow in a parallel light manner. The light splitting and sampling module splits the scattered light of the particle flow into scattered light of different wavelengths, and converts the collected scattered light of different wavelengths into electrical pulse signals of different wavelengths; The differential data processing module performs differential processing on the electrical pulse signals of different wavelengths, and judges the similarity of the signals based on the differential processing results, thereby reducing the influence of the shape on the identification and classification of the particle flow.

9. The method for online particle identification based on dual-wavelength illumination according to claim 8, characterized in that: The differential data processing module includes processing the red photoelectric pulse signal and the green photoelectric pulse signal using mean square error (MSE), and / or processing the red photoelectric pulse signal and the green photoelectric pulse signal using mean absolute error (MAE).

10. The method for online particle identification based on dual-wavelength illumination according to claim 9, characterized in that: The differential data processing module further includes classifying particles based on Mie scattering intensity or Rayleigh scattering intensity, and setting the ratio of particle size to wavelength to x: Where r is the particle radius, λ is the wavelength, and when x < 1, Rayleigh scattering intensity treatment is used; When x≥1, Mie scattering intensity processing is adopted; and then N-order derivative processing is performed on the Mie scattering intensity and Rayleigh scattering intensity respectively, where N is an integer between 2 and 6.

Images