Laser granularity inversion method based on Mie scattering theory

Abstract

The invention relates to the technical field of optical detection, and discloses a Mie scattering theory-based laser particle size inversion method, which comprises the following steps of: constructing a laser scattering measurement system which comprises a laser light source, a sample cell, an optical collimation assembly, a scattered light collection assembly and a photoelectric detection assembly, monochromatic laser emitted by the laser light source is processed by the optical collimation assembly to form parallel light beams which are incident to a particle sample in the sample cell, and the light beams and the particle sample interact to generate scattered light. Scattered light collection assemblies distributed in an annular array are adopted to cover forward, lateral and backward multiple scattering angles, and a Vondrak smoothing algorithm is combined to carry out noise filtering and baseline correction preprocessing, so that the integrity and purity of scattered light signal collection are improved, and the interference of ambient light, circuit noise and background light on signals is effectively reduced. And accurate data support is provided for subsequent correlation model construction, so that the accuracy of a particle size inversion result is improved.

Description

Technical Field

[0001] This invention relates to the field of optical detection technology, specifically to a laser particle size inversion method based on Mie scattering theory. Background Technology

[0002] In industrial production, scientific research, and environmental monitoring, particle size distribution is a core parameter determining product performance, experimental accuracy, and monitoring indicators. Laser particle size inversion technology, as a non-contact, rapid, and efficient detection method, has replaced traditional sieving and sedimentation methods as the mainstream approach, widely used in industries such as pharmaceuticals, chemicals, and environmental protection. Among these, methods based on Mie scattering theory are one of the most widely used particle size inversion techniques due to their adaptability to a wide particle size range from nanometers to micrometers and their ability to accurately describe the scattering characteristics of particles with different morphologies.

[0003] The performance of laser particle size inversion technology depends on the applicability of the scattering theory and the rationality of the signal processing flow. Existing core scattering theories are divided into three categories. Rayleigh scattering theory is only applicable to extremely fine particles with a diameter much smaller than the laser wavelength. When the particle size is close to or exceeds the laser wavelength, the error increases dramatically, which cannot meet the requirements for wide particle size detection.

[0004] Fraunhofer diffraction theory simplifies the scattering process to achieve rapid calculations and is suitable for detecting large-diameter particles, but it has obvious limitations: it is only suitable for large-diameter particles, and the resolution error for small-diameter particles exceeds 15%; it ignores internal scattering, multiple scattering and non-spherical characteristics, resulting in large detection deviations for complex samples; and it has strict requirements on detection distance, which limits the miniaturization of equipment.

[0005] Mie scattering theory overcomes the aforementioned limitations, comprehensively considering the influence of multiple parameters and adapting to particle size ranges from 1 nm to 1000 μm. It is the only theory that can simultaneously detect wide particle sizes and complex particle morphologies. Its core process includes scattering signal acquisition, preprocessing, model construction, algorithm solving, and result output. The design of each step directly determines the inversion accuracy and stability. However, existing schemes based on Mie scattering theory still have many shortcomings. In the signal acquisition stage, most schemes use 6-12 limited-angle acquisition structures, which cannot capture the characteristic scattering signals of small-diameter particles at lateral angles of 30°-150° and rearward angles of 150°-180°, resulting in insufficient data integrity and high inversion errors for small-diameter particles. The angle interval is set empirically, making it difficult to balance detection accuracy and speed. In the signal preprocessing stage, fixed threshold filtering cannot adapt to dynamic changes in the signal, easily leading to signal distortion or noise residue. Baseline correction uses preset fixed reference values, which cannot track background light fluctuations, leaving a 1%-3% deviation, which cannot meet the requirements of high-precision detection. In the algorithm solution stage, the initial values ​​for iteration rely on empirical settings, which can easily lead to local optima and result distortion in multi-peak sample detection. Furthermore, the lack of a dynamic fitting residual judgment mechanism and the inability to balance accuracy and efficiency with a fixed number of iterations make it difficult to meet the rapid response requirements of industrial online detection. In addition, existing methods lack result verification and error correction mechanisms; directly outputting data makes it impossible to identify the source of errors, compromising reliability and limiting their application in high-end pharmaceuticals, precision electronics, and other fields.

[0006] In summary, existing solutions have inherent deficiencies in signal acquisition integrity, preprocessing adaptability, algorithm stability, and result reliability, failing to meet the high-precision detection requirements of high-end scenarios. Developing a systematically optimized Mie scattering laser particle size inversion method has become an urgent problem to be solved in this field. Summary of the Invention

[0007] To address the shortcomings of existing technologies, this invention provides a laser particle size inversion method based on Mie scattering theory, which solves the problems mentioned in the background section.

[0008] To achieve the above objectives, the present invention provides the following technical solution: a laser particle size inversion method based on Mie scattering theory, comprising the following steps: Step 1: Construct a laser scattering measurement system. The system includes a laser source, a sample cell, an optical collimation component, a scattered light collection component, and a photoelectric detection component. The monochromatic laser emitted by the laser source is processed by the optical collimation component to form a parallel beam that is incident on the particle sample in the sample cell. The beam interacts with the particle sample to generate scattered light. Step 2: The scattered light signals under multiple scattering angles are collected by the scattered light collection component. The scattered light collection component is distributed in a ring array around the sample cell, covering the forward, side and backward scattering angle range. The collected scattered light signals are converted into electrical signals by the photoelectric detection component and transmitted to the data processing module. Step 3: The data processing module preprocesses the electrical signal. The preprocessing process includes noise filtering, signal amplitude calibration, and baseline correction. Noise filtering uses the Vondrak smoothing algorithm to reduce the interference of ambient light and circuit noise on the signal. Signal amplitude calibration is completed based on the scattered light intensity reference value of the standard particle sample. Step 4: Construct a correlation model between scattered light intensity and particle size based on Mie scattering theory. The model input parameters include laser wavelength, medium refractive index, particle refractive index, and scattered light intensity data after preprocessing in Step 3. The model establishes a mapping relationship between different particle size ranges and corresponding scattered light intensities. Step 5: The projection iteration algorithm is used to solve the correlation model. During the iteration process, a fitting residual judgment mechanism is introduced. The fitting residual is calculated by the deviation between the preprocessed scattered light intensity data and the model predicted light intensity data. When the fitting residual is greater than the preset threshold, the initial value of the particle size distribution is optimized and the iteration is repeated. Step 6: When the fitting residual is less than the preset threshold, stop the iteration and output the particle size distribution data. The data processing module simultaneously calculates the characteristic parameters of the particle size distribution, including the average particle size, median particle size, and particle size distribution span.

[0009] Preferably, in step 1, the laser source is a continuously output monochromatic laser with the output laser wavelength fluctuation range controlled within ±0.5nm. The optical collimation component is composed of a convex lens group, used to collimate the laser beam into a parallel beam with a parallelism better than 0.01mrad.

[0010] Preferably, the ring array of the scattered light collection component in step 2 includes at least 16 optical collection units, each optical collection unit corresponding to a fixed scattering angle, the scattering angle interval being 2° to 15°, the forward scattering angle covering 0° to 30°, the side scattering angle covering 30° to 150°, and the back scattering angle covering 150° to 180°.

[0011] Preferably, in step 3, baseline correction is achieved by acquiring the background light signal when the sample is free of particles. The acquired background light signal is used as a reference value, and the reference value is subtracted from the scattered light signal of the sample containing particles to eliminate background interference.

[0012] Preferably, in step 4, the refractive index of the medium is the refractive index of the dispersion medium in the sample cell, and the refractive index of the particles is the inherent refractive index of the particles being tested. Both are pre-determined using a standard refractive index measuring device, and the measurement accuracy is controlled within ±0.001.

[0013] Preferably, in step 5, the initial value of particle size distribution is obtained in advance by laser diffraction. The particle size range of the initial value of particle size distribution covers 1 nm to 1000 μm, and the proportion of particles in each particle size range is used as the initial iteration parameter.

[0014] Preferably, in step 5, the preset threshold is set according to the measurement accuracy requirements, ranging from 0.001 to 0.01, and the upper limit of the number of iterations is set to 50. When the number of iterations reaches the upper limit but the fitting residual is still greater than the preset threshold, a data anomaly alarm is triggered.

[0015] Preferably, in step 6, the average particle size is calculated using a volume-weighted average algorithm, the median particle size is the particle size value corresponding to the cumulative proportion of particle size distribution reaching 50%, and the particle size distribution span is calculated by (D90-D10) / D50, where D10 is the particle size corresponding to the cumulative proportion of 10%, and D90 is the particle size corresponding to the cumulative proportion of 90%.

[0016] Preferably, the method further includes step 7: verifying the output particle size distribution data by comparing it with a standard particle sample. If the verification error is less than 2%, the particle size distribution data is confirmed to be valid. If the verification error is greater than or equal to 2%, the method returns to step 3 to re-perform signal preprocessing and subsequent steps.

[0017] Preferably, in step 7, the particle size of the standard particle sample is known and the particle size distribution span is less than 0.2. The standard particle sample and the particle sample to be tested use the same dispersion medium, and the dispersion concentration is controlled within the same range to ensure consistency of measurement conditions.

[0018] This invention provides a laser grain size inversion method based on Mie scattering theory. It has the following advantages: 1. By employing a ring array of scattered light collection components to cover forward, lateral, and backward scattering angles, and combining this with the Vondrak smoothing algorithm for noise filtering and baseline correction preprocessing, the integrity and purity of the scattered light signal acquisition are improved. This effectively reduces the interference of ambient light, circuit noise, and background light on the signal, providing accurate data support for subsequent correlation model construction, and thus improving the accuracy of particle size inversion results.

[0019] 2. By constructing a correlation model between scattered light intensity and particle size based on Mie scattering theory, and combining it with laser diffraction to obtain initial values ​​of particle size distribution in advance and projection iterative algorithm for solution, a fitting residual judgment mechanism and iterative optimization strategy are introduced to achieve targeted and efficient particle size inversion process, avoid blind iteration, shorten inversion time while ensuring consistency between particle size distribution data and actual particle characteristics, and adapt to measurement needs in a wide particle size range from 1nm to 1000μm.

[0020] 3. By adding a standard particle sample comparison and verification step, the dispersion medium and dispersion concentration of the standard sample and the test sample are kept consistent. Combined with a 2% verification error threshold and a data anomaly feedback mechanism, the validity of the inversion results can be verified and corrected in real time. When the verification error does not meet the standard, the preprocessing step is returned to be re-executed, ensuring the reliability of the output particle size distribution data and characteristic parameters, and improving the applicability and credibility of the method in actual detection scenarios. Attached Figure Description

[0021] Figure 1 This is the overall flowchart of the present invention; Figure 2 This is a flowchart of the signal preprocessing process of the present invention; Figure 3 This is a flowchart of the iterative solution and verification process of the present invention. Detailed Implementation

[0022] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] Please see the appendix Figure 1 - Appendix Figure 3 This invention provides a laser particle size inversion method based on Mie scattering theory, comprising the following steps: Step 1: Construct a laser scattering measurement system. The system includes a laser source, a sample cell, an optical collimation component, a scattered light collection component, and a photoelectric detection component. The monochromatic laser emitted by the laser source is processed by the optical collimation component to form a parallel beam that is incident on the particle sample in the sample cell. The beam interacts with the particle sample to generate scattered light. The laser source is a continuously output monochromatic laser with the output laser wavelength fluctuation range controlled within ±0.5nm. The optical collimation component is composed of a convex lens group and is used to collimate the laser beam into a parallel beam with a parallelism better than 0.01mrad.

[0024] Specifically, the core components of the system include a laser source, a sample cell, an optical collimation component, a scattered light collection component, and a photoelectric detection component. Each component is selected and integrated based on the characteristics of the incident and scattered light according to the Mie scattering theory. Among them, the laser source, as the excitation source of the scattered signal, directly affects the consistency and repeatability of the scattered light intensity due to the stability of its output light. In existing technologies, some light sources have large wavelength fluctuations, leading to instability in the scattered light intensity reference value and introducing additional detection errors. Therefore, this system selects a continuously output monochromatic laser, strictly controlling the output laser wavelength fluctuation range within ±0.5nm to ensure the monochromaticity and stability of the incident laser, providing a stable light source foundation for the accurate correlation between scattered light intensity and particle size.

[0025] The optical collimation component, constructed from a convex lens group, functions primarily to transform the diverging laser light emitted from the laser source into a highly parallel beam. In Mie scattering theory, the calculation of scattered light intensity is based on the assumption of parallel incident light interacting with particles. Insufficient parallelism of the incident beam leads to uneven illumination range and intensity distribution within the sample cell, resulting in differences in the scattered signals from particles at different locations and affecting the uniformity of signal acquisition. Therefore, this component employs an optical adaptation design with a convex lens group to collimate the laser beam into a parallel beam with a parallelism better than 0.01 mrad. This ensures that the incident beam vertically and uniformly illuminates the particle sample within the sample cell, guaranteeing that the scattered light generated by the interaction between particles and the beam conforms to the calculation premises of Mie scattering theory, thereby reducing detection errors introduced by beam characteristic deviations at the source.

[0026] The sample cell, serving as the carrier for the particle sample, provides a closed and stable environment for the interaction between the laser and the particles, preventing interference from external airflow and impurities on the sample distribution and scattering process, thus ensuring the authenticity of the scattered light signal. When a parallel light beam is incident on the particle sample in the sample cell, the beam interacts with particles of different sizes and shapes, generating omnidirectional scattered light based on Mie scattering theory. This scattered light contains core information about the particle size distribution. Subsequently, the scattered light collection component and photoelectric detection component complete the signal acquisition and conversion, providing raw data for subsequent preprocessing and inversion calculations.

[0027] Overall, this laser scattering measurement system, through precise selection and coordinated design of its components, solves the problems of scattering signal distortion caused by insufficient light source stability and poor incident beam parallelism in existing technologies. It lays a solid hardware foundation for subsequent complete capture of multi-angle scattering signals and improvement of inversion accuracy. At the same time, it conforms to the application conditions of Mie scattering theory, ensuring the scientificity and reliability of the entire granularity inversion method.

[0028] Step 2: Collect scattered light signals from multiple scattering angles using a scattered light collection component. The scattered light collection component is arranged in a ring array around the sample cell, covering the forward, lateral, and backward scattering angle ranges. The collected scattered light signals are converted into electrical signals by a photoelectric detection component and transmitted to the data processing module. The ring array of the scattered light collection component contains at least 16 optical collection units, each corresponding to a fixed scattering angle. The scattering angle interval ranges from 2° to 15°, with the forward scattering angle covering 0° to 30°, the lateral scattering angle covering 30° to 150°, and the backward scattering angle covering 150° to 180°.

[0029] Specifically, the optical collimation component is composed of a convex lens group, whose core function is to process the diverging laser emitted from the laser source into a highly parallel beam. In Mie scattering theory, the calculation of scattered light intensity is based on the assumption of parallel incident light interacting with particles. If the parallelism of the incident beam is insufficient, it will lead to uneven illumination range and intensity distribution of the beam within the sample cell, resulting in differences in the scattered signals from particles at different locations and affecting the uniformity of signal acquisition. Therefore, this component, through the optical adaptation design of the convex lens group, collimates the laser beam into a parallel beam with a parallelism better than 0.01 mrad, ensuring that the incident beam vertically and uniformly illuminates the particle sample within the sample cell. This ensures that the scattered light generated by the interaction between particles and the beam conforms to the calculation premise of Mie scattering theory, reducing the detection error introduced by beam characteristic deviations from the source.

[0030] The sample cell, serving as the carrier for the particle sample, provides a closed and stable environment for the interaction between the laser and the particles, preventing interference from external airflow and impurities on the sample distribution and scattering process, thus ensuring the authenticity of the scattered light signal. When a parallel light beam is incident on the particle sample in the sample cell, the beam interacts with particles of different sizes and shapes, generating omnidirectional scattered light based on Mie scattering theory. This scattered light contains core information about the particle size distribution. Subsequently, the scattered light collection component and photoelectric detection component complete the signal acquisition and conversion, providing raw data for subsequent preprocessing and inversion calculations.

[0031] The coordinated design of the scattered light collection component and the photoelectric detection component is key to accurately capturing the spatial distribution characteristics of scattered light and overcoming the shortcomings of existing technologies in signal acquisition completeness. Based on Mie scattering characteristics, the intensity of scattered light from particles of different sizes varies significantly across different angles. In particular, the intensity of scattered light from small-diameter particles exhibits characteristic peaks in the lateral and backward scattering regions, making it impossible to fully extract particle size correlation information using only limited-angle acquisition. Therefore, in this system, the scattered light collection components are arranged in a ring array around the sample cell. This ring layout achieves full coverage of forward, lateral, and backward scattering angles, ensuring comprehensive capture of scattered light signals without blind spots.

[0032] This ring array comprises at least 16 optical collection units, each corresponding to a fixed scattering angle. The angle intervals are controlled within the range of 2° to 15°. This design avoids the problem of failing to capture details of abrupt changes in scattered light intensity due to excessively large intervals, while also preventing signal redundancy and reduced processing efficiency due to excessively small intervals. Specifically, the forward 0° to 30° interval covers the strong scattering signal dominated by large-diameter particles, the lateral 30° to 150° interval covers the scattering signal of medium-diameter particles, and the backward 150° to 180° interval covers the characteristic scattering signal of small-diameter particles. Through precise acquisition in these intervals, comprehensive capture of scattering information from particles of different sizes is achieved. The acquired scattered light signals are converted into electrical signals by a photoelectric detection component, completing the conversion from optical signals to quantifiable and processable electrical signals. These signals are then transmitted to the data processing module, providing complete and accurate raw signal support for subsequent model construction and inversion solutions based on Mie scattering theory. This further improves the accuracy of particle size inversion from the signal acquisition stage.

[0033] Overall, this laser scattering measurement system, through precise selection and coordinated design of its components, solves the problems of insufficient light source stability, poor incident beam parallelism, and scattering signal distortion caused by incomplete scattering signal acquisition in existing technologies. This lays a solid hardware foundation for subsequent complete capture of multi-angle scattering signals and improved inversion accuracy. Simultaneously, it conforms to the application conditions of Mie scattering theory, ensuring the scientific validity and reliability of the entire grain size inversion method. Step 3: The data processing module preprocesses the electrical signal. The preprocessing process includes noise filtering, signal amplitude calibration, and baseline correction. Noise filtering uses the Vondrak smoothing algorithm to reduce the interference of ambient light and circuit noise on the signal. Signal amplitude calibration is completed based on the reference value of the scattered light intensity of the standard particle sample. Baseline correction is achieved by collecting the background light signal when there is no particle sample. The collected background light signal is used as the reference value, and the reference value is subtracted from the scattered light signal containing the particle sample to eliminate background interference.

[0034] Specifically, the data processing module preprocesses the transmitted electrical signals, a crucial step in eliminating signal interference and ensuring the accuracy of inversion calculations. It specifically addresses signal distortion issues caused by poor filter adaptability and incomplete baseline correction in existing technologies, providing high-purity signal data for subsequent model construction. The preprocessing process encompasses three core steps: noise filtering, signal amplitude calibration, and baseline correction. Each step is designed based on the characteristics of signal interference and the quantization requirements of the Mie scattering signal, forming a collaborative noise reduction and calibration system.

[0035] The noise filtering stage employs the Vondrak smoothing algorithm. Unlike existing fixed-threshold filtering methods, this algorithm can adaptively adjust the smoothing coefficient to effectively reduce the interference of ambient light fluctuations and inherent circuit noise on the signal while preserving the feature details of the scattered signal. Since the scattered signal contains both effective information related to particle size and random noise, traditional filtering easily leads to signal distortion or noise residue. The Vondrak smoothing algorithm, by balancing signal smoothness and feature preservation, accurately separates the effective signal from noise, preventing the weak scattered signal corresponding to small-diameter particles from being covered by noise, thus ensuring the integrity and authenticity of the signal.

[0036] Signal amplitude calibration is performed based on the reference value of scattered light intensity from standard particle samples. The core purpose is to eliminate the influence of individual equipment differences and environmental factors on signal amplitude and establish a unified standard for light intensity quantification. The particle size and scattering characteristics of standard particle samples are known, and the scattered light intensity they produce under the same detection conditions can serve as a reference. Calibration is achieved by comparing the electrical signal amplitude of the sample under test with the reference value, correcting amplitude deviations caused by laser source intensity drift and differences in the sensitivity of photoelectric detection components. This ensures that the correlation between the scattered light intensity signal and particle size conforms to the Mie scattering theory calculation model, thereby improving signal quantization accuracy.

[0037] Baseline correction is achieved by acquiring the background light signal when the sample is free of particles. This background light signal is used as a reference value and subtracted from the scattered light signal of the sample containing particles, thus completely eliminating background interference. Existing technologies that preset a fixed reference value cannot adapt to the dynamic fluctuations of background light during detection. This method, however, acquires the background signal in real time, accurately subtracting fixed and dynamic background interference such as sample cell reflected light and ambient stray light. It avoids baseline shift caused by the background signal being superimposed on the effective scattered signal, further improving the purity of the preprocessed signal and providing accurate data support for subsequent inversion algorithms.

[0038] Overall, this preprocessing process, through the collaborative design of algorithms and operations, systematically solves the shortcomings of incomplete signal interference elimination and insufficient quantization accuracy in existing technologies. Combined with the high-precision signal acquisition stage mentioned above, it forms a closed-loop optimization of "acquisition-preprocessing", ensuring that the signal input to the inversion model is both complete and accurate, providing double assurance for the accuracy of granular inversion results based on Mie scattering theory.

[0039] Step 4: Construct a correlation model between scattered light intensity and particle size based on Mie scattering theory. The model input parameters include laser wavelength, medium refractive index, particle refractive index, and scattered light intensity data after preprocessing in Step 3. The model establishes a mapping relationship between different particle size ranges and corresponding scattered light intensities. The medium refractive index is the refractive index of the dispersion medium in the sample cell, and the particle refractive index is the inherent refractive index of the measured particle. Both are pre-determined using a standard refractive index measuring device, with the measurement accuracy controlled within ±0.001.

[0040] Step 5: The projection iteration algorithm is used to solve the correlation model. During the iteration process, a fitting residual judgment mechanism is introduced. The fitting residual is calculated by the deviation between the preprocessed scattered light intensity data and the model predicted light intensity data. When the fitting residual is greater than the preset threshold, the initial value of the particle size distribution is optimized and the iteration is repeated. The initial value of the particle size distribution is obtained in advance by laser diffraction. The particle size range of the initial value of the particle size distribution covers 1nm to 1000μm. The proportion of particles in each particle size range is used as the initial iteration parameter.

[0041] Specifically, step 4 constructs a correlation model between scattered light intensity and particle size based on Mie scattering theory. The core of this model is to transform the physical laws of Mie scattering into a quantifiable mathematical mapping relationship, providing theoretical support for particle size inversion. Mie scattering theory clearly demonstrates a strong correlation between scattered light intensity and particle size, laser wavelength, medium refractive index, and particle refractive index. Therefore, the model input parameters accurately cover these key variables. The laser wavelength is a fixed output parameter of the laser source mentioned earlier, the preprocessed scattered light intensity data is the measured input, and the medium refractive index and particle refractive index are core physical parameters that directly determine the model's calculation accuracy.

[0042] The refractive index of the medium is an inherent property of the dispersion medium within the sample cell, while the particle refractive index is an inherent property of the measured particles. The measurement accuracy of both directly affects the accuracy of the correlation between scattered light intensity and particle size. Some existing methods use empirical refractive index values, which can easily introduce model calculation errors. Therefore, this method pre-determines both values ​​using a standard refractive index measuring device, strictly controlling the measurement accuracy within ±0.001, and minimizing the impact of parameter deviations on the model. This model establishes a one-to-one mapping relationship between different particle size ranges and corresponding scattered light intensities, achieving a theoretical transformation from light intensity signals to particle size information, and providing a reliable computational framework for subsequent algorithm solutions.

[0043] Step 5 employs a projection iterative algorithm to solve the associated model. The core principle is to gradually approximate the true particle size distribution through iterative optimization, addressing the problems of traditional single-iterative algorithms easily getting trapped in local optima and lacking sufficient convergence accuracy. The key to the iterative process lies in the coordinated design of initial value setting and the fitting residual judgment mechanism, both ensuring the efficiency and accuracy of the iteration. Specifically, the initial particle size distribution is obtained in advance using laser diffraction. This method can quickly obtain the approximate particle size distribution range, avoiding the blindness of traditional empirical initial value setting. Furthermore, the initial particle size range covers 1 nm to 1000 μm, consistent with the suitable particle size range of Mie scattering theory. The proportion of particles within each size range serves as the initial iteration parameter, providing a reasonable starting point for the iterative calculation.

[0044] The fitting residual judgment mechanism provides a convergence basis for the iterative process. The fitting residual is obtained by calculating the deviation between the preprocessed measured scattered light intensity data and the model-predicted light intensity data, quantifying the degree of agreement between the model's predicted and measured values. When the fitting residual is greater than a preset threshold, it indicates a large deviation between the initial particle size distribution value and the true value, requiring optimization of the initial particle size distribution value and repeated iterations to gradually reduce the residual. When the fitting residual is less than the preset threshold, the iteration terminates, and the particle size distribution at this point represents the optimal inversion result. This mechanism avoids the insufficient accuracy or low efficiency problems caused by the fixed number of iterations in existing technologies, achieving dynamic control of the iterative process and balancing inversion accuracy and speed.

[0045] In summary, step 4 constructs a highly reliable correlation model through precise parameter control, and step 5 optimizes the iterative process through scientific initial value setting and dynamic residual judgment. The two work together to achieve a deep integration of Mie scattering theory and engineering calculation, effectively making up for the defects of existing technologies, ensuring the accuracy and stability of particle size inversion results, and laying the foundation for subsequent result output and verification.

[0046] The preset threshold is set according to the measurement accuracy requirements, ranging from 0.001 to 0.01. The maximum number of iterations is set to 50. When the number of iterations reaches the maximum but the fitting residual is still greater than the preset threshold, a data anomaly alarm is triggered.

[0047] Step 6: When the fitting residual is less than a preset threshold, stop the iteration and output the particle size distribution data. The data processing module simultaneously calculates the characteristic parameters of the particle size distribution, including the average particle size, median particle size, and particle size distribution span. The average particle size is calculated using a volume-weighted average algorithm. The median particle size is the particle size value corresponding to a cumulative proportion of 50% in the particle size distribution. The particle size distribution span is calculated by (D90-D10) / D50, where D10 is the particle size corresponding to a cumulative proportion of 10%, and D90 is the particle size corresponding to a cumulative proportion of 90%.

[0048] Specifically, after the data processing module preprocesses the transmitted electrical signal, it needs to form a complete inversion link through model construction, algorithm solution and result output. Steps 4 to 6 sequentially realize the closed loop of "model construction - iterative solution - result output". Step 6, as the closing link of the inversion process, is to complete the particle size data output and feature parameter quantification, so as to provide intuitive and accurate indicator support for subsequent result verification and practical application. At the same time, the comparability of the results is ensured through standardized parameter calculation. Step 6 is based on the iterative convergence result of step 5. When the fitting residual is less than the preset threshold, it indicates that the model-predicted light intensity and the measured scattered light intensity have reached an ideal degree of agreement. The current particle size distribution data can truly reflect the actual characteristics of the measured particles. At this time, the iteration is stopped and the particle size distribution data is output, thus ending the core inversion calculation process. The data processing module synchronously calculates the characteristic parameters of particle size distribution. Its core purpose is to transform continuous particle size distribution data into concise and representative quantitative indicators to meet the evaluation needs of particle characteristics in different application scenarios. The characteristic parameters selected are average particle size, median particle size, and particle size distribution span. These three parameters characterize the particle size distribution from different dimensions, forming a comprehensive parameter system: the average particle size reflects the overall size level of the particle population, the median particle size reflects the central trend of the particle size distribution, and the particle size distribution span characterizes the degree of dispersion of particle size. The combination of these three parameters can fully outline the core characteristics of particle size distribution. For each characteristic parameter, standardized algorithms are employed to ensure calculation accuracy and industry compatibility. The average particle size is calculated using a volume-weighted average algorithm. Compared to the quantity-weighted average, this algorithm is more suitable for scenarios in industrial production and scientific research experiments where particle volume is the core influencing factor, and can accurately reflect the impact of the particle group on actual application performance. The median particle size is defined as the particle size value corresponding to the cumulative proportion of the particle size distribution reaching 50%. It has the advantage of resisting extreme value interference, can stably represent the central position of the particle size distribution, and avoid the misleading influence of a few large or small particles on core indicators. The particle size distribution span is calculated as (D90-D10) / D50, where D10 is the particle size corresponding to the 10% cumulative percentage of the particle size distribution (i.e., 10% of the particles are smaller than this size), D90 is the particle size corresponding to the 90% cumulative percentage (i.e., 90% of the particles are smaller than this size), and D50 is the median particle size. This calculation method is a standardized method for characterizing particle dispersion in the industry. The smaller the value, the more concentrated the particle size; the larger the value, the more dispersed the particle size distribution. The calculation results are intuitive and have a unified comparison benchmark, which can meet the evaluation needs of particle uniformity in different detection scenarios. In summary, step 6, through precise data output after convergence judgment and standardized feature parameter calculation, not only achieves the core objective of granular inversion, but also provides quantitative support for subsequent result verification and scenario application. At the same time, through the standardized design of algorithm selection and parameter definition, the universality and reliability of the inversion results are improved.

[0049] The process also includes step 7: verifying the output particle size distribution data by comparing it with a standard particle sample. If the verification error is less than 2%, the particle size distribution data is confirmed to be valid; if the verification error is greater than or equal to 2%, return to step 3 to re-process the signal and perform subsequent steps. The particle size of the standard particle sample is known and the particle size distribution span is less than 0.2. The standard particle sample and the particle sample being tested use the same dispersion medium, and the dispersion concentration is controlled within the same range to ensure consistency of measurement conditions.

[0050] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A laser particle size inversion method based on Mie scattering theory, characterized in that, Includes the following steps: Step 1: Construct a laser scattering measurement system. The system includes a laser source, a sample cell, an optical collimation component, a scattered light collection component, and a photoelectric detection component. The monochromatic laser emitted by the laser source is processed by the optical collimation component to form a parallel beam that is incident on the particle sample in the sample cell. The beam interacts with the particle sample to generate scattered light. Step 2: The scattered light signals under multiple scattering angles are collected by the scattered light collection component. The scattered light collection component is distributed in a ring array around the sample cell, covering the forward, side and backward scattering angle range. The collected scattered light signals are converted into electrical signals by the photoelectric detection component and transmitted to the data processing module. Step 3: The data processing module preprocesses the electrical signal. The preprocessing process includes noise filtering, signal amplitude calibration, and baseline correction. Noise filtering uses the Vondrak smoothing algorithm to reduce the interference of ambient light and circuit noise on the signal. Signal amplitude calibration is completed based on the scattered light intensity reference value of the standard particle sample. Step 4: Construct a correlation model between scattered light intensity and particle size based on Mie scattering theory. The model input parameters include laser wavelength, medium refractive index, particle refractive index, and scattered light intensity data after preprocessing in Step 3. The model establishes a mapping relationship between different particle size ranges and corresponding scattered light intensities. Step 5: The projection iteration algorithm is used to solve the correlation model. During the iteration process, a fitting residual judgment mechanism is introduced. The fitting residual is calculated by the deviation between the preprocessed scattered light intensity data and the model predicted light intensity data. When the fitting residual is greater than the preset threshold, the initial value of the particle size distribution is optimized and the iteration is repeated. Step 6: When the fitting residual is less than the preset threshold, stop the iteration and output the particle size distribution data. The data processing module simultaneously calculates the characteristic parameters of the particle size distribution, including the average particle size, median particle size, and particle size distribution span.

2. The laser grain size inversion method based on Mie scattering theory according to claim 1, characterized in that, In step 1, a continuously output monochromatic laser is selected as the laser source, and the output laser wavelength fluctuation range is controlled within ±0.5nm. The optical collimation component is composed of a convex lens group, which is used to collimate the laser beam into a parallel beam with a parallelism better than 0.01mrad.

3. The laser grain size inversion method based on Mie scattering theory according to claim 1, characterized in that, In step 2, the ring array of the scattered light collection component contains at least 16 optical collection units, each optical collection unit corresponding to a fixed scattering angle. The scattering angle interval ranges from 2° to 15°, the forward scattering angle covers 0° to 30°, the side scattering angle covers 30° to 150°, and the back scattering angle covers 150° to 180°.

4. The laser grain size inversion method based on Mie scattering theory according to claim 1, characterized in that, In step 3, baseline correction is achieved by acquiring the background light signal when there is no particle sample. The acquired background light signal is used as a reference value, and the reference value is subtracted from the scattered light signal of the particle sample to eliminate background interference.

5. The laser grain size inversion method based on Mie scattering theory according to claim 1, characterized in that, In step 4, the refractive index of the medium is the refractive index of the dispersion medium in the sample cell, and the refractive index of the particles is the inherent refractive index of the particles being measured. Both are pre-determined using a standard refractive index measuring device, with the measurement accuracy controlled within ±0.

001.

6. The laser grain size inversion method based on Mie scattering theory according to claim 1, characterized in that, In step 5, the initial value of particle size distribution is obtained in advance by laser diffraction. The particle size range of the initial value of particle size distribution covers 1 nm to 1000 μm, and the proportion of particles in each particle size range is used as the initial iteration parameter.

7. The laser grain size inversion method based on Mie scattering theory according to claim 1, characterized in that, In step 5, the preset threshold is set according to the measurement accuracy requirements, ranging from 0.001 to 0.01, and the upper limit of the number of iterations is set to 50. When the number of iterations reaches the upper limit but the fitting residual is still greater than the preset threshold, a data anomaly alarm is triggered.

8. The laser grain size inversion method based on Mie scattering theory according to claim 1, characterized in that, In step 6, the average particle size is calculated using a volume-weighted average algorithm. The median particle size is the particle size value corresponding to a cumulative proportion of 50% in the particle size distribution. The particle size distribution span is calculated by (D90-D10) / D50, where D10 is the particle size corresponding to a cumulative proportion of 10% and D90 is the particle size corresponding to a cumulative proportion of 90%.

9. The laser grain size inversion method based on Mie scattering theory according to claim 1, characterized in that, It also includes step 7: verifying the output particle size distribution data by comparing it with standard particle samples. When the verification error is less than 2%, the particle size distribution data is confirmed to be valid; when the verification error is greater than or equal to 2%, return to step 3 to re-perform signal preprocessing and subsequent steps.

10. The laser grain size inversion method based on Mie scattering theory according to claim 9, characterized in that, In step 7, the particle size of the standard particle sample is known and the particle size distribution span is less than 0.

2. The standard particle sample and the particle sample to be tested use the same dispersion medium, and the dispersion concentration is controlled within the same range to ensure the consistency of measurement conditions.

Images