A method, system and device for quickly sensing suspended particle structure in water

Abstract

A method, system and device for rapid perception of suspended particle structure in water, including obtaining time-series polarization data to measure the polarization characteristics of a single suspended particle in real time; constructing a classification module, using supervised or unsupervised machine learning methods to train or extract classification characteristics from time-series polarization data to achieve automatic classification of suspended particles; constructing a particle type code, arranging the classification results in chronological order to form a type code that characterizes the structural characteristics of a single particle; calculating polarization structure parameters, and obtaining the number distribution of particles with different structures in the water body through statistical analysis of the type code as a quantitative indicator of the suspended particle structure; obtaining the structural composition of the suspended particles and the distribution of different structural characteristics in the group through statistical analysis of the polarization structure parameters, and describing the overall properties of the suspended particle group in the water body. The present invention improves the accuracy and efficiency of suspended particle measurement in water, and is of great significance to environmental protection and ecological research.

Description

Technical Field

[0001] The present invention relates to the field of optical measurement, and in particular to a method, system and device for quickly sensing the structure of suspended particles in water. Background Art

[0002] Suspended particles include microalgae, silt and microplastics, which participate in the global material cycle through physical, chemical and biological processes and are an important part of the aquatic ecosystem. Detecting the types, states and components of suspended particles is of great significance for understanding the structure of water bodies, predicting changes in aquatic environments and protecting human health and environmental stability. Suspended particles are widely distributed in nature and have the characteristics of small size, fast movement speed and complex structure. Rapid and accurate detection of suspended particles has always been a difficult problem in the industry.

[0003] At present, the research methods for suspended particles are mainly divided into three categories: imaging method, spectroscopy method and scattering method. Imaging method commonly uses instruments such as microscopes and flow imagers, which can detect the number, size, color and shape of suspended particles, but requires professionals to conduct preliminary analysis of the test results, which takes a long time and is easily interfered by subjective factors. Spectroscopy can achieve specific detection and qualitative analysis of suspended particles by capturing the spectral characteristics of suspended particles and reflecting the biochemical composition of particles. Common methods include Raman spectroscopy, infrared spectroscopy, hyperspectroscopy, etc., but because they will generate a huge amount of data, they have high requirements for data processing and take a long time to measure. Scattering method commonly uses instruments such as turbidimeters and backscatter coefficient meters to measure optical scattering signals, which can achieve non-destructive and non-contact detection, but these measurement methods are group measurements and cannot provide information on the structure of suspended particles. Therefore, the current detection methods for suspended particles have certain limitations. How to achieve rapid and accurate classification and analysis of suspended particles in water and structural perception of suspended particles in water urgently needs to develop new methods.

[0004] It should be noted that the information disclosed in the above background technology section is only used for understanding the background of the present application, and therefore may include information that does not constitute prior art known to ordinary technicians in the field. Summary of the invention

[0005] The main purpose of the present invention is to overcome the defects existing in the above-mentioned background technology and provide a method and device for quickly sensing the structure of suspended particles in water.

[0006] To achieve the above object, the present invention adopts the following technical solutions:

[0007] A method for quickly sensing the structure of suspended particles in water comprises the following steps:

[0008] S1. Time-series polarization data acquisition: acquiring time-series polarization data of suspended particles when they pass through the illumination light path, wherein the polarization state of the illumination light is modulated continuously and rapidly, and only the polarization characteristics of a single suspended particle are measured at any given moment;

[0009] S2. Constructing a classification module: Based on an existing database and supervised machine learning methods, a model capable of classifying the time-series polarization data of suspended particles is obtained through training; alternatively, an unsupervised machine learning method is used to construct a classification module, and classification features are directly extracted from the time-series polarization data of suspended particles to achieve automatic classification of data;

[0010] S3. Constructing a particle type code: Based on the time-series polarization data obtained in S1 and the classification module constructed in S2, construct a particle type code for each suspended particle, the type code characterizing the structural characteristics of a single particle; wherein the classification module is used to classify the time-series polarization data of each suspended particle to determine the particle category corresponding to each data segment, and the classification results of each particle are arranged in chronological order to form a particle type code, the particle type code reflecting the structural characteristics of a single particle when passing through the illumination light path;

[0011] S4. Calculate polarization structure parameters: By statistically analyzing the particle type code constructed in S3, the number distribution of particles with different structures in the water body is obtained, and the polarization structure parameters are calculated as a quantitative indicator of the suspended particle structure in the water body;

[0012] S5. Perception of suspended particle structure: Receive a series of polarization structure parameters of suspended particles, obtain the structural composition of suspended particles and the distribution of different structural characteristics in the group through statistical analysis, describe the overall properties of the suspended particle group in the water body, and thus achieve rapid perception of the suspended particle structure in the water body.

[0013] A rapid sensing system for suspended particle structure in water, comprising:

[0014] A time-series polarization data acquisition module is used to acquire time-series polarization data of suspended particles when they pass through the illumination light path, wherein the polarization state of the illumination light is modulated continuously and rapidly, and only the polarization characteristics of a single suspended particle are measured at any given moment;

[0015] Classification module: Based on the existing database and supervised machine learning methods, a classification model is obtained through training to classify the time-series polarization data of suspended particles; alternatively, an unsupervised machine learning method is used to construct a classification module to directly extract classification features from the time-series polarization data of suspended particles to achieve automatic classification of data;

[0016] A particle type code construction module is used to construct a particle type code for each suspended particle based on the time-series polarization data and the output of the classification module, and the type code characterizes the structural characteristics of a single particle, wherein the classification module is used to classify the time-series polarization data of each suspended particle to determine the particle category corresponding to each data segment, and the classification results of each particle are arranged in chronological order to form a particle type code, and the particle type code reflects the structural characteristics of a single particle when passing through the illumination light path;

[0017] Polarization structure parameter acquisition module: used to perform statistical analysis on particle type codes, obtain the quantity distribution of particles with different structures in the water body, and thus calculate the polarization structure parameters as a quantitative indicator of the suspended particle structure in the water body;

[0018] Suspended particle structure perception module: used to receive a series of polarization structure parameters of suspended particles, obtain the structural composition of suspended particles and the distribution of different structural characteristics in the group through statistical analysis, describe the overall properties of the suspended particle group in the water body, and thus achieve rapid perception of the suspended particle structure of the water body.

[0019] A device for quickly sensing suspended particle structures in water used in the system for quickly sensing suspended particle structures in water, comprising:

[0020] A laser light source unit, used for emitting illumination light to irradiate the suspended particles;

[0021] A polarization adjustment module, used to convert the illumination light emitted by the laser light source into polarized light and to continuously and rapidly modulate the polarization state;

[0022] A focusing module is used to focus the modulated laser light into a focal point and irradiate the suspended particles;

[0023] The polarization state analyzer is used to measure the polarization state of the scattered light excited when the suspended particles pass through the illumination light path, and obtain time-series polarization data including the time dimension.

[0024] The present invention has the following beneficial effects:

[0025] The present invention proposes a method, system and device for rapid perception of the structure of suspended particles in water. It innovatively measures the time-series polarization data of suspended particles in water, uses the time-series polarization data of suspended particles as key features to construct a particle type code, thereby characterizing the structural characteristics of individual particles, and further obtains structural parameters based on the statistics of these type codes, thereby achieving rapid perception of the structure of suspended particles in water. The present invention breaks the limitations of traditional measurement methods and no longer relies on the subjective analysis of professionals or the time-consuming process of processing large amounts of data. It can use the polarization characteristics of suspended particles in water to quickly perceive their structure and achieve fine identification and classification. At the same time, by counting a certain number of individual particles, the present invention can obtain the quantitative distribution of particles with different structures in the water body, providing an effective technical means for applications such as early warning of algal bloom outbreaks, marine particle carbon detection, and microplastic toxicity research. This method not only improves the accuracy and efficiency of measurement, but also provides important scientific data for understanding the interaction between the complex structural composition of suspended particles and environmental media, and has important practical significance for environmental protection and ecological research.

[0026] Other beneficial effects of the embodiments of the present invention will be further described below. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 A diagram of a device for quickly sensing suspended particle structures in water according to an embodiment of the present invention

[0028] Figure 2 Flow chart of a method for rapid perception of suspended particle structure in water according to an embodiment of the present invention

[0029] Figure 3 A diagram of an implementation method for acquiring time-series polarization data according to an embodiment of the present invention

[0030] Figure 4 The present invention is a flowchart for realizing suspended particle structure perception according to an embodiment of the present invention. DETAILED DESCRIPTION

[0031] The following is a detailed description of the embodiments of the present invention. It should be emphasized that the following description is only exemplary and is not intended to limit the scope and application of the present invention.

[0032] It should be noted that when an element is referred to as being "fixed to" or "disposed on" another element, it can be directly on the other element or indirectly on the other element. When an element is referred to as being "connected to" another element, it can be directly connected to the other element or indirectly connected to the other element. In addition, connection can be used for fixing as well as for coupling or communication.

[0033] It should be understood that the orientation or position relationship indicated by terms such as "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside" and "outside" are based on the orientation or position relationship shown in the accompanying drawings, and are only for the convenience of describing the embodiments of the present invention and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0034] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0035] See also Figures 1 to 4 The embodiment of the present invention provides a method for quickly sensing the structure of suspended particles in water, comprising the following steps:

[0036] S1. Time-series polarization data acquisition: acquiring time-series polarization data of suspended particles when they pass through the illumination light path, wherein the polarization state of the illumination light is modulated continuously and rapidly, and only the polarization characteristics of a single suspended particle are measured at any given moment;

[0037] S2. Constructing a classification module: Based on an existing database and supervised machine learning methods, a model capable of classifying the time-series polarization data of suspended particles is obtained through training; alternatively, an unsupervised machine learning method is used to construct a classification module, and classification features are directly extracted from the time-series polarization data of suspended particles to achieve automatic classification of data;

[0038] S3. Constructing a particle type code: Based on the time-series polarization data obtained in S1 and the classification module constructed in S2, construct a particle type code for each suspended particle, the type code characterizing the structural characteristics of a single particle; wherein the classification module is used to classify the time-series polarization data of each suspended particle to determine the particle category corresponding to each data segment, and the classification results of each particle are arranged in chronological order to form a particle type code, the particle type code reflecting the structural characteristics of a single particle when passing through the illumination light path;

[0039] S4. Calculate polarization structure parameters: By statistically analyzing the particle type code constructed in S3, the number distribution of particles with different structures in the water body is obtained, and the polarization structure parameters are calculated as a quantitative indicator of the suspended particle structure in the water body;

[0040] S5. Perception of suspended particle structure: Receive a series of polarization structure parameters of suspended particles, obtain the structural composition of suspended particles and the distribution of different structural characteristics in the group through statistical analysis, describe the overall properties of the suspended particle group in the water body, and thus achieve rapid perception of the suspended particle structure in the water body.

[0041] The method for rapid perception of suspended particle structure in water of the present invention achieves characterization of the structural features of individual particles by acquiring the time-series polarization data of suspended particles and constructing a particle type code. The method uses the time-series polarization data of suspended particles as a key feature, combined with supervised or unsupervised machine learning technology, to quickly classify and identify suspended particles. By statistically analyzing the particle type code, the quantitative distribution of particles with different structures in the water body is obtained, and the polarization structure parameters are calculated as a quantitative indicator of the suspended particle structure of the water body. The present invention breaks the limitations of traditional measurements, improves the accuracy and efficiency of measurements, and provides important scientific data for water environment monitoring and ecological research.

[0042] In a preferred embodiment, in step S1, the Stokes vector is used to describe the polarization state of the illumination light and the scattered light, and the Mueller matrix M is used to describe the change in the polarization state of the light. The Stokes vector of the scattered light is associated with the Stokes vector of the illumination light to reflect the polarization optical characteristics of the suspended particles. In step S2, the structural information of the suspended particles contained in the Stokes vector and the Mueller matrix is ​​mined through a machine learning method to construct a classification module so that it can identify and classify the structural composition and state changes of the suspended particles. In step S4, based on the polarization information of the Stokes vector and the Mueller matrix, the polarization structure parameters are calculated to quantify the structural characteristics of the suspended particles in the water body.

[0043] Specifically, the Stokes vector Sout of the scattered light can be related to the Stokes vector Sin of the illumination light by the following formulas (1) and (2) to reflect the polarization optical characteristics of the suspended particles:

[0044] (1)

[0045] (2)

[0046] Where S is the Stokes vector, I is the total light intensity, Q, U, and V are the polarization component differences of 0-degree linear polarization, 45-degree linear polarization, and right-hand circular polarization, respectively; M is the Mueller matrix, Sout is the Stokes vector of the scattered light, Sin is the Stokes vector of the illumination light, and m11 to m44 are the 16 elements that make up the Mueller matrix, which contain information about how suspended particles affect the polarization state of light.

[0047] The Mueller matrix can describe the polarization optical characteristics of suspended particles and reflect the complete microscopic structural characteristics of suspended particles. Compared with the traditional optical scattering method, it can provide richer information and realize the rapid measurement of the types and morphologies of suspended particles. Applied to the method for rapid perception of suspended particle structure in water of the present invention, with the help of machine learning technology, the suspended particle structure information contained in polarization information such as Stokes vector and Mueller matrix can be deeply mined, and the characterization ability of the structural composition and state change of suspended particles can be improved, thereby realizing the structural perception and fine classification of suspended particles.

[0048] See also Figure 1 In a preferred embodiment, in step S1: using a laser light source 1 as illumination light to illuminate the suspended particles 4; continuously and rapidly modulating the polarization state of the laser light source 1 through a polarization adjustment module 2; focusing the modulated laser light into a focus using a focusing module 3 and irradiating it onto the suspended particles; collecting the scattered light excited when the suspended particles 4 pass through the illumination light path; the suspended particles 4 can pass through the illumination light path by setting a fluid delivery system (such as a flow channel, a peristaltic pump or a pressure drive system), which can pass the water sample containing the suspended particles through the illumination light path at a stable speed and direction to ensure that the suspended particles 4 pass through the action area of ​​the focused light beam in sequence; using a polarization state analyzer 5 to measure the polarization state of the scattered light, and obtain time-series polarization data containing a time dimension; and real-time acquisition and preprocessing of the collected time-series polarization data. The polarization adjustment module 2 optimizes the single polarization state of the illumination light by adjustment, or, by setting to use multiple different illumination light polarization states, preferably at least 4 different illumination light polarization states, to obtain the time-series polarization data of the particles and optimize the classification capability of the time-series polarization data.

[0049] See also Figure 1 and Figure 3 In some embodiments, the intersection of the optical paths of the illumination light and the scattered light is determined as the scattering volume, wherein the scattering volume is limited so that there is only one suspended particle in the scattering volume; when a single suspended particle passes through the scattering volume, the polarization state analyzer 5 is used to collect the timing signal of the suspended particle within the detection time length, and a series of timing polarization data containing time information is obtained; wherein, the accuracy and credibility of the polarization information are ensured by single-point multiple measurements (continuous and multiple polarization state measurements of a single suspended particle when passing through the illumination light path). The collected polarization information can be smoothed, filtered and other data processing operations, and the polarization information detected when the suspended particle passes through the scattering volume is sorted according to the time dimension to obtain the timing polarization data of each suspended particle; wherein, according to the path and method of the suspended particle passing through the scattering volume, the length of the timing polarization data of each particle is the same or different, while the result of structural perception is kept consistent.

[0050] In some embodiments, in step S2: in the process of constructing the classification module, the output of the classification module is verified, that is, the category of the particulate matter output by the classification module is compared with the actual category of the particulate matter; the accuracy of the classification result is calculated to evaluate the performance of the classification module; and a classification module with a high classification accuracy is selected as the final classification module to ensure the accuracy of the method for rapid perception of suspended particle structure.

[0051] In some embodiments, in step S3: the particle structure at each moment or time period is classified into a specific category characterized by polarization information in the time series, thereby generating differences in particle type codes; the construction result of the particle type code depends on factors such as the type, state and attachment conditions of the suspended particles, and the specific expression of the particle type code is caused by the structural differences of the suspended particles themselves.

[0052] See also Figures 2 to 4 An embodiment of the present invention also provides a system for rapid perception of suspended particle structures in water, including a time-series polarization data acquisition module, a classification module, a particle type code construction module, a polarization structure parameter acquisition module, and a suspended particle structure perception module.

[0053] The time-series polarization data acquisition module is used to acquire the time-series polarization data of suspended particles when they pass through the illumination light path, wherein the polarization state of the illumination light is modulated continuously and rapidly, and only the polarization characteristics of a single suspended particle are measured at any given moment. The classification module is based on an existing database and a supervised machine learning method, and a classification model is obtained through training, which is used to classify the time-series polarization data of suspended particles; or, an unsupervised machine learning method is used to construct a classification module, and classification features are directly extracted from the time-series polarization data of suspended particles to achieve automatic classification of data. The particle type code construction module is used to construct a particle type code for each suspended particle based on the time-series polarization data and the output of the classification module, and the type code characterizes the structural characteristics of a single particle, wherein the classification module is used to classify the time-series polarization data of each suspended particle to determine the particle category corresponding to each data segment, and the classification results of each particle are arranged in chronological order to form a particle type code, and the particle type code reflects the structural characteristics of a single particle when it passes through the illumination light path. The polarization structure parameter acquisition module is used to perform statistical analysis on the particle type code to obtain the quantity distribution of particles with different structures in the water body, thereby calculating the polarization structure parameters as a quantitative indicator of the suspended particle structure in the water body. The suspended particle structure perception module is used to receive the polarization structure parameters of a series of suspended particles, obtain the structural composition of the suspended particles and the distribution of different structural characteristics in the group through statistical analysis, describe the overall properties of the suspended particle group in the water body, and thus realize the rapid perception of the suspended particle structure in the water body.

[0054] See also Figure 1The embodiment of the present invention also provides a device for quickly sensing the structure of suspended particles in water for use in the system for quickly sensing the structure of suspended particles in water, comprising: a laser light source unit 1, for emitting illumination light to illuminate the suspended particles; a polarization adjustment module 2, for converting the illumination light emitted by the laser light source 1 into polarized light and continuously and quickly modulating the polarization state; a focusing module 3, for focusing the modulated laser light into a focus and irradiating it onto the suspended particles; a polarization state analyzer 5, for measuring the polarization state of scattered light excited when the suspended particles pass through the illumination light path, and obtaining time-series polarization data including a time dimension.

[0055] The specific embodiments of the present invention are further described below.

[0056] Figure 1 The schematic diagram of the rapid sensing device for the structure of suspended particles in water is shown in FIG. The light source is a laser light source 1, which serves as the illumination light for the suspended particles. The polarization adjustment module 2 can realize continuous and rapid modulation of the polarization state of the illumination light in a short time. The illumination light passes through the focusing module 3, converges into a focus and then irradiates the suspended particles. The scattered light excited by the suspended particles passing through the illumination light is measured by a polarization state analyzer 5 after passing through the focusing module 3. The polarization information of the suspended particles is a series of signals including the time dimension.

[0057] Figure 2 This is a schematic diagram of a method system for rapidly sensing the structure of suspended particles in water, which mainly includes: a time-series polarization data acquisition module, a classification module, a particle type code construction module, a polarization structure parameter acquisition module, and a suspended particle structure perception module.

[0058] The time-series polarization data acquisition module is composed of a rapid sensing device for suspended particle structures in water, which can quickly obtain polarization information when particles pass through the illumination light path. By acquiring the scattered light polarization state data in real time, the time-series polarization data of suspended particles is obtained according to the time sequence of particles passing through the illumination light path, and the data is pre-processed.

[0059] The classification module can be constructed based on an existing database and a specific machine learning algorithm, or it can directly extract classification features from the data itself through unsupervised learning methods such as clustering and dimensionality reduction. The flexible classification method allows the classification module to be applicable to different data environments and application scenarios, providing a more effective classification solution. When constructing a classification module, the category of the particles output by the classification module is verified against the actual category of the particles, the accuracy of the classification results is calculated, and the one with the highest classification accuracy is used as the final classification module.

[0060] The particle type code construction module uses the classification module to classify each time series data segment in the time series polarization data of the suspended particles into a corresponding particle category, and then obtains the category sequence in turn, which is the particle type code.

[0061] The polarization structure parameter acquisition module is used to calculate the quantity distribution of different particle structures in the water body. Due to the different structures of different particles in the water body, the corresponding particle type code structures are quite different. The polarization structure parameter acquisition module counts all particle type codes and outputs polarization structure parameters as a quantitative indicator of the suspended particle structure in the water body.

[0062] In this way, the suspended particle structure perception module receives a series of polarization structure parameters of suspended particles, obtains the structural composition of the suspended particles and the differences in the suspended particles due to structural changes through statistical analysis, describes the overall properties of the suspended particles, and realizes rapid perception of the structure of suspended particles in the water body.

[0063] (1) Acquisition of time-series polarization data

[0064] When suspended particles pass through the illumination light path, the particles will scatter the illumination light, and the polarization state of the scattered light contains the structural information of the particles. However, due to the large number of suspended particles and the extremely short detection time, high requirements are placed on the measurement accuracy. When building a rapid perception device for suspended particle structures in water, a single-particle scattering detection light path is designed to achieve rapid polarization data measurement of a single suspended particle. At the same time, a polarization adjustment module and a polarization state analyzer are used to quickly modulate and detect polarization information, thereby achieving the acquisition of time-series polarization data of suspended particles. The polarization state of the illumination light provided by the polarization adjustment module affects the classification ability of time-series polarization data. The best classification effect of time-series polarization data can be obtained by optimizing the polarization state of the illumination light; or multiple (at least 4) illumination light polarization states can be directly adjusted to obtain time-series polarization data of particles.

[0065] (2) Construction of PM type code

[0066] The particle type code is a one-dimensional sequence composed of the particle type information at each time point (segment) in the time series. The particle type code is constructed through a classification algorithm, and the type of the particle can be determined based on the polarization data of the particle at a certain time point (segment). The classification algorithm can be composed of a pre-prepared database and algorithm, or the data itself can be formed through unsupervised methods such as clustering. Polarization data is an important part of the construction of particle type codes. The amount of information and dimensions it contains directly affect the classification accuracy and precision, and can be selected as needed.

[0067] (3) Calculation of polarization structure parameters

[0068] By measuring individual suspended particles, we can obtain the individual information of each suspended particle in the detected water sample, namely the particle type code. Polarization structure parameters are statistical analysis of the particle type codes of a large number of individual suspended particles, and the quantity distribution of different particle structures is obtained, thereby obtaining the overall information of the water sample. By calculating the particle polarization structure parameters, we can obtain the quantity and proportion distribution of different types of suspended particles in the water body, and realize the overall understanding of the water particles.

[0069] Figure 1 A device for quickly sensing the structure of suspended particles in water is shown. The light source emits laser illumination, and the polarization adjustment module is placed in front of the light source, so that the illumination light becomes polarized light after passing through the polarization adjustment module, and the polarization state of the illumination light can be changed quickly in a very short time. The illumination light passes through the focusing module placed in front of the polarization adjustment module and converges to irradiate the suspended particles. After the suspended particles excite and scatter the illumination light, its scattered light passes through the focusing module and finally irradiates the polarization state analyzer.

[0070] Figure 3 A method for implementing time-series polarization data acquisition is shown. The intersection of the optical paths of the illumination light and the scattered light is the scattering volume. The scattering volume is limited to a very small range, ensuring that there is only one suspended particle in the scattering volume, and the scattered light is all generated by the excitation of the single suspended particle. When a single suspended particle passes through the scattering volume, the polarization state analyzer will collect the time-series signal within the detection time of the suspended particle, that is, a series of time-series polarization data containing time information. The acquisition of time-series polarization data is affected by the sampling frequency of the polarization state analyzer and the polarization adjustment module. Single-point multiple measurements ensure the accuracy and reliability of the polarization information, and are also conducive to data processing operations such as smoothing and filtering. The polarization information detected by the suspended particles passing through the scattering volume is sorted according to the time dimension to obtain the time-series polarization data of the suspended particles. Since the paths and methods of the suspended particles passing through the scattering volume are inconsistent, the length of the time-series polarization data of each particle may be different, but the results of their structural perception are consistent.

[0071] The implementation method of the particle type code construction is as follows. First, the rapid perception device of the suspended particle structure in water of the present invention realizes the accurate and rapid measurement of the time series polarization data, and ensures the credibility and integrity of the suspended particle data set. After the suspended particle data set is constructed, the classification module is used to process the data set to construct a classifier that can identify the type of particles. After the time series polarization data of the suspended particles to be detected are identified by the classification module, the corresponding particle category will be output, and the particle category will be arranged in the original time sequence to obtain the particle type code. The difference in the particle type code is generated by the classification module classifying the particle structure at each moment (time period) into a specific category characterized by polarization information in the time series. The construction result of the particle type code is determined by factors such as the type, state and attachment status of the suspended particles, and its specific expression is caused by the structural differences of the suspended particles themselves. Therefore, the construction of the particle type code can effectively realize the rapid perception of the suspended particle structure.

[0072] The implementation method of polarization structure parameter calculation is as follows. After the particle type code is constructed, the structural information of each suspended particle in the sample to be tested is obtained, the structural perception of a single suspended particle is realized, and the structural differences of the suspended particles are understood. In order to better describe the group information of suspended particles in the water body, the present invention provides a method for calculating polarization structure parameters, and implements group measurement of suspended particles in a statistical manner. The classification module can divide the time series polarization data generated by a single suspended particle into multiple categories. After the particle type code of the suspended particle is constructed, each unit in the time series polarization data has a corresponding type label. The units with the same label are counted to obtain the polarization structure parameters of different structural components. By calculating the polarization structure parameters, the differences in the content of different suspended particle components can be obtained. Using the group information, unknown water samples can be identified and classified.

[0073] Figure 4 The flow of a method for sensing the structure of suspended particles in water is shown. When the suspended particle structure sensing system is used for the first time, a database is established for the suspended particle structure of the target detection, the corresponding suspended particle time series polarization data is obtained, and the corresponding classification module is constructed. When a new suspended particle sample is detected, it is only necessary to obtain the suspended particle time series polarization data of the sample, use the established classification module, construct the particle type code, calculate the polarization structure parameters of the suspended particles, and realize the perception of the suspended particle structure.

[0074] In summary, the present invention constructs a particle type code by measuring the time-series polarization data of suspended particles as a key feature to characterize the structural characteristics of a single particle; then obtains structural parameters based on the statistics of the particle type code to achieve rapid perception of the structure of suspended particles in water.

[0075] The method and device for rapid perception of the structure of suspended particles in water of the present invention measure the time-series polarization data of suspended particles, construct particle type codes and polarization structure parameters, and realize rapid and in-situ measurement of the structure of suspended particles. The advantage of the present invention is that it breaks the limitations of traditional measurements, and by directly detecting the polarization characteristics of suspended particles in water bodies, it can rapidly perceive the structure of individual suspended particles and realize fine identification and classification of suspended particles; on this basis, by counting a certain number of individual particles, the number distribution of particles with different structures in the water body can be obtained. The present invention can be applied to aspects such as early warning of algal bloom outbreaks, detection of marine particle carbon, and research on the toxicity of microplastics. It is of great significance for understanding the interaction between the complex structural composition of suspended particles and environmental media, and provides an effective solution for suspended particle research and environmental monitoring in water.

[0076] The above content is a further detailed description of the present invention in combination with specific / preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, without departing from the concept of the present invention, it can also make several substitutions or modifications to these described embodiments, and these substitutions or modifications should be regarded as belonging to the protection scope of the present invention. In the description of this specification, the description of the reference terms "an embodiment", "some embodiments", "preferred embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in combination with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily target the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In the absence of mutual contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples. Although the embodiments of the present invention and their advantages have been described in detail, it should be understood that various changes, substitutions and modifications can be made herein without departing from the scope of protection of the patent application.

Claims

1. A method for quickly sensing the structure of suspended particles in water, characterized in that: The following steps are involved: S1. Time-series polarization data acquisition: acquiring time-series polarization data of suspended particles when they pass through the illumination light path, wherein the polarization state of the illumination light is modulated continuously and rapidly, and only the polarization characteristics of a single suspended particle are measured at any given moment; S2. Constructing a classification module: Based on an existing database and a supervised machine learning method, a model capable of classifying the time-series polarization data of suspended particles is obtained through training; alternatively, an unsupervised machine learning method is used to construct a classification module, and classification features are directly extracted from the time-series polarization data of suspended particles to achieve automatic classification of data; S3. Constructing a particle type code: Based on the time-series polarization data obtained in S1 and the classification module constructed in S2, construct a particle type code for each suspended particle, the type code characterizing the structural characteristics of a single particle; wherein the classification module is used to classify the time-series polarization data of each suspended particle to determine the particle category corresponding to each data segment, and the classification results of each particle are arranged in chronological order to form a particle type code, the particle type code reflecting the structural characteristics of a single particle when passing through the illumination light path; S4. Calculate polarization structure parameters: By statistically analyzing the particle type code constructed in S3, the number distribution of particles with different structures in the water body is obtained, and the polarization structure parameters are calculated as a quantitative indicator of the suspended particle structure in the water body; S5. Perception of suspended particle structure: Receive a series of polarization structure parameters of suspended particles, obtain the structural composition of suspended particles and the distribution of different structural characteristics in the group through statistical analysis, describe the overall properties of the suspended particle group in the water body, and thus achieve rapid perception of the suspended particle structure in the water body.

2. The method for rapid perception of suspended particle structure in water according to claim 1, characterized in that: In step S1, the polarization states of the illumination light and the scattered light are described by the Stokes vector, the change of the polarization state of the light is described by the Mueller matrix M, and the Stokes vector of the scattered light is associated with the Stokes vector of the illumination light to reflect the polarization optical characteristics of the suspended particles; In step S2, the suspended particle structure information contained in the Stokes vector and the Mueller matrix is ​​mined by a machine learning method to construct a classification module, so that it can identify and classify the structural composition and state changes of the suspended particles; In step S4, polarization structure parameters are calculated based on the polarization information of the Stokes vector and the Mueller matrix to quantify the structural characteristics of the suspended particles in the water.

3. The method for rapid perception of suspended particle structure in water according to claim 2, characterized in that: The Stokes vector Sout of the scattered light is related to the Stokes vector Sin of the illumination light by the following formulas (1) and (2) to reflect the polarization optical characteristics of the suspended particles: （1） （2） Where S is the Stokes vector, I is the total light intensity, Q, U, and V are the polarization component differences of 0-degree linear polarization, 45-degree linear polarization, and right-hand circular polarization, respectively; M is the Mueller matrix, Sout is the Stokes vector of the scattered light, Sin is the Stokes vector of the illumination light, and m11 to m44 are the 16 elements that make up the Mueller matrix, which contain information about how suspended particles affect the polarization state of light.

4. The method for quickly sensing the structure of suspended particles in water according to any one of claims 1 to 3, characterized in that: In step S1: a laser light source is used as illumination light to illuminate the suspended particles; the polarization state of the laser light source is continuously and rapidly modulated through a polarization adjustment module; the modulated laser is focused into a focus using a focusing module and irradiated onto the suspended particles; scattered light excited when the suspended particles pass through the illumination light path is collected; the polarization state of the scattered light is measured using a polarization state analyzer to obtain time-series polarization data including a time dimension; and the collected time-series polarization data is acquired and preprocessed in real time.

5. The method for rapid perception of suspended particle structure in water according to claim 4, characterized in that: The polarization adjustment module optimizes a single polarization state of the illumination light through adjustment, or is configured to use a plurality of different polarization states of the illumination light to obtain the time-series polarization data of the particles, thereby optimizing the classification capability of the time-series polarization data.

6. The method for rapid perception of suspended particle structure in water according to claim 5, characterized in that: The multiple different illumination light polarization states are at least 4 different illumination light polarization states.

7. The method for quickly sensing the structure of suspended particles in water according to any one of claims 1 to 3, characterized in that: In step S1: Determine the intersection of the optical paths of the illumination light and the scattered light as a scattering volume, wherein the scattering volume is limited so that only one suspended particle exists in the scattering volume; when a single suspended particle passes through the scattering volume, use a polarization state analyzer to collect a time sequence signal of the suspended particle within the detection time length, and obtain a series of time sequence polarization data containing time information; The collected polarization information is smoothed and filtered, and the polarization information detected when the suspended particles pass through the scattering volume is sorted according to the time dimension to obtain the time-series polarization data of each suspended particle. According to the path and method of the suspended particles passing through the scattering volume, the length of the time-series polarization data of each particle is the same or different, while the results of structural perception remain consistent.

8. The method for quickly sensing the structure of suspended particles in water according to any one of claims 1 to 3, characterized in that: In step S2: in the process of constructing the classification module, the output of the classification module is verified, that is, the category of the particles output by the classification module is compared with the actual category of the particles; the accuracy of the classification result is calculated to evaluate the performance of the classification module; and the classification module with high classification accuracy is selected as the final classification module.

9. The method for quickly sensing the structure of suspended particles in water according to any one of claims 1 to 3, characterized in that: In step S3: the particle structure at each moment or time period is classified into categories characterized by polarization information in the time series, thereby generating differences in particle type codes; the construction result of the particle type code depends on the type, state and attachment conditions of the suspended particles, and the specific expression of the particle type code is caused by the structural differences of the suspended particles themselves.

10. A rapid sensing system for suspended particle structure in water, characterized in that: include: A time-series polarization data acquisition module is used to acquire time-series polarization data of suspended particles when they pass through the illumination light path, wherein the polarization state of the illumination light is modulated continuously and rapidly, and only the polarization characteristics of a single suspended particle are measured at any given moment; Classification module: Based on the existing database and supervised machine learning methods, a classification model is obtained through training to classify the time-series polarization data of suspended particles; alternatively, an unsupervised machine learning method is used to construct a classification module to directly extract classification features from the time-series polarization data of suspended particles to achieve automatic classification of data; A particle type code construction module is used to construct a particle type code for each suspended particle based on the time-series polarization data and the output of the classification module. The type code characterizes the structural characteristics of a single particle. The classification module is used to classify the time-series polarization data of each suspended particle to determine the particle category corresponding to each data segment. The classification results of each particle are arranged in chronological order to form a particle type code. The particle type code reflects the structural characteristics of a single particle when it passes through the illumination light path. Polarization structure parameter acquisition module: used to perform statistical analysis on particle type codes, obtain the quantity distribution of particles with different structures in the water body, and thus calculate the polarization structure parameters as a quantitative indicator of the suspended particle structure in the water body; Suspended particle structure perception module: used to receive a series of polarization structure parameters of suspended particles, obtain the structural composition of suspended particles and the distribution of different structural characteristics in the group through statistical analysis, describe the overall properties of the suspended particle group in the water body, and thus achieve rapid perception of the suspended particle structure of the water body.

11. A device for quickly sensing suspended particle structures in water used in the system for quickly sensing suspended particle structures in water as claimed in claim 10, characterized in that: include: A laser light source unit, used for emitting illumination light to irradiate the suspended particles; A polarization adjustment module, used to convert the illumination light emitted by the laser light source into polarized light and to continuously and rapidly modulate the polarization state; A focusing module is used to focus the modulated laser light into a focal point and irradiate the suspended particles; The polarization state analyzer is used to measure the polarization state of the scattered light excited when the suspended particles pass through the illumination light path, and obtain time-series polarization data including the time dimension.

Images