Atmospheric single particle source identification method and device based on morphology, particle size and components

Abstract

The invention discloses an atmospheric single particle source identification method and device based on morphology, particle size and components, and the method comprises the steps: analyzing a sample of a set pollution source through a scanning electron microscope controlled by a computer, obtaining an image, component data and particle size parameters of a single particle, constructing a single particle data set, and randomly dividing the single particle data set into a first data set and a second data set; constructing a source analysis model according to the residual network model and the extreme gradient lifting model; training the source analysis model through the first data set and the second data set; and through the trained source analysis model, obtaining the probability value of the target single particle belonging to the set pollution source, and determining the source of the target single particle. The morphology, particle size and component characteristics of the particulate matter can be fully excavated, and the source of the particulate matter is identified; and the physical and chemical property research of the ambient atmospheric particulates and the atmospheric particulate source analysis research can be further carried out.

Description

Technical Field

[0001] The present invention relates to the technical field of atmospheric particle source analysis, and in particular to a method and device for identifying the source of atmospheric single particles based on morphology, particle size and composition. Background Art

[0002] Atmospheric particulate matter pollution has a significant impact on climate, ecosystems and human health, and has attracted widespread attention from all walks of life in recent years. Particulate matter source analysis is an important tool for formulating particulate matter pollution reduction policies.

[0003] The sources of atmospheric particulate matter are complex, including natural and anthropogenic sources. Accurately identifying the sources of atmospheric particulate matter can provide important technical support for formulating atmospheric particulate matter pollution reduction and protecting human health. Scanning electron microscope energy spectrum analysis (SEM-EDS) can obtain the morphology, particle size and composition information of atmospheric particulate matter, and has been widely used in the study of the physical and chemical properties of atmospheric particles.

[0004] At present, in the direct observation research on particulate matter emitted by pollution sources, it is found that the particulate matter emitted by different pollution sources has certain differences in morphology, particle size and composition. However, there is currently no method that can identify the source of single particles in the atmospheric environment based on the single particle morphology, particle size and composition information.

[0005] Therefore, how to invent a source analysis method that can identify the source of single particles in the atmospheric environment based on single particle morphology, particle size and composition information has become an urgent problem to be solved. Summary of the invention

[0006] To this end, the present invention provides a method and device for identifying the source of single atmospheric particles based on morphology, particle size, and composition. The method can fully explore the morphology, particle size, and composition characteristics of particles, identify the source of particles, and at the same time, help further study the physical and chemical properties of ambient atmospheric particles and study the source analysis of atmospheric particles.

[0007] In order to achieve the above object, the present invention provides the following technical solution: a method for identifying the source of a single atmospheric particle based on morphology, particle size and composition, comprising:

[0008] Analyzing the collected samples of the set pollution source by using a computer-controlled scanning electron microscope to obtain images, composition data and particle size parameters of single particles in the samples of the set pollution source, and constructing a single particle data set of the pollution source samples;

[0009] The single particle data set is randomly divided into a first data set and a second data set according to a set ratio; the first data set is used for training a source analysis model; and the second data set is used for verifying the source analysis model;

[0010] Construct a source analysis model based on the residual network model and the extreme gradient boosting model;

[0011] Iteratively training the source analysis model using the first data set; verifying the prediction result of the source analysis model using the second data set to obtain the trained source analysis model;

[0012] The component data and particle size parameters of the target single particle are input into the trained source analysis model for processing, the probability value of the target single particle belonging to the set pollution source is obtained, and the source of the target single particle is determined.

[0013] As a preferred scheme of the atmospheric single particle source identification method based on morphology, particle size and composition, in the process of analyzing the collected samples of the set pollution sources by applying a computer-controlled scanning electron microscope, the set pollution sources include: soil dust, road dust, construction dust, coal smoke dust, biomass combustion dust and steel smelting dust.

[0014] As a preferred solution of the atmospheric single particle source identification method based on morphology, particle size and composition, the image of the single particle includes: a high-definition microscopic photograph of the single particle;

[0015] The composition data of the single particle includes: carbon, oxygen, sodium, magnesium, aluminum, silicon, phosphorus, sulfur, chlorine, potassium, calcium, zinc, barium, vanadium, chromium, cobalt, selenium, tin, titanium, manganese, nickel, copper and lead;

[0016] The particle size parameters of the single particle include: maximum diameter, minimum diameter, average diameter, vertical diameter and equivalent circle diameter.

[0017] As a preferred solution for the atmospheric single particle source identification method based on morphology, particle size and composition, in the process of inputting the image, composition data and particle size parameters of the target single particle into the trained source analysis model for processing, and obtaining the probability value that the target single particle belongs to the set pollution source, the image features of the target single particle are extracted through the residual network model, and the source of the target single particle is preliminarily predicted based on the image features of the target single particle, so as to obtain the preliminary probability value that the target single particle belongs to the set pollution source.

[0018] As a preferred solution for the method of identifying the source of atmospheric single particles based on morphology, particle size and composition, in the process of inputting the image, composition data and particle size parameters of the target single particle into the trained source analysis model for processing to obtain the probability value of the target single particle belonging to the set pollution source, the preliminary probability value of the target single particle belonging to the set pollution source is merged with the particle size parameters and composition data of the target single particle, and input into the extreme gradient boosting model, and the probability value of the target single particle belonging to the set pollution source is obtained through prediction by the extreme gradient boosting model.

[0019] The present invention also provides an atmospheric single particle source identification device based on morphology, particle size, and composition, which adopts the above-mentioned atmospheric single particle source identification method based on morphology, particle size, and composition, including:

[0020] A single particle data set acquisition module is used to analyze the collected samples of the set pollution source by using a scanning electron microscope controlled by a computer, obtain the image, composition data and particle size parameters of the single particles in the samples of the set pollution source, and construct a single particle data set of the pollution source samples;

[0021] A single particle data set division module, used to randomly divide the single particle data set into a first data set and a second data set according to a set ratio; the first data set is used for training the source analysis model; the second data set is used for verifying the source analysis model;

[0022] The source analysis model building module is used to build a source analysis model based on the residual network model and the extreme gradient boosting model;

[0023] A source analysis model training module, configured to iteratively train the source analysis model using the first data set; verify the prediction result of the source analysis model using the second data set to obtain the trained source analysis model;

[0024] The single particle source determination module is used to input the image, component data and particle size parameters of the target single particle into the trained source analysis model for processing, obtain the probability value that the target single particle belongs to the set pollution source, and determine the source of the target single particle.

[0025] As a preferred solution of the atmospheric single particle source identification device based on morphology, particle size and composition, in the single particle data set acquisition module, in the process of analyzing the collected samples of the set pollution source by applying a computer-controlled scanning electron microscope, the set pollution source includes: soil dust, road dust, construction dust, coal smoke dust, biomass combustion dust and steel smelting dust.

[0026] As a preferred solution of the atmospheric single particle source identification device based on morphology, particle size and composition, in the single particle data set acquisition module, the image of the single particle includes: a single particle high-definition microscopic photograph;

[0027] The composition data of the single particle includes: carbon, oxygen, sodium, magnesium, aluminum, silicon, phosphorus, sulfur, chlorine, potassium, calcium, zinc, barium, vanadium, chromium, cobalt, selenium, tin, titanium, manganese, nickel, copper and lead;

[0028] The particle size parameters of the single particle include: maximum diameter, minimum diameter, average diameter, vertical diameter and equivalent circle diameter.

[0029] As a preferred solution for the atmospheric single particle source identification device based on morphology, particle size and composition, in the single particle source determination module, while inputting the image, composition data and particle size parameters of the target single particle into the trained source analysis model for processing, and obtaining the probability value that the target single particle belongs to the set pollution source, the image features of the target single particle are extracted through the residual network model, and the source of the target single particle is preliminarily predicted based on the image features of the target single particle, so as to obtain the preliminary probability value that the target single particle belongs to the set pollution source.

[0030] As a preferred solution for the atmospheric single particle source identification device based on morphology, particle size and composition, in the single particle source determination module, while inputting the image, composition data and particle size parameters of the target single particle into the trained source analysis model for processing, and obtaining the probability value that the target single particle belongs to the set pollution source, the preliminary probability value that the target single particle belongs to the set pollution source is merged with the particle size parameters and composition data of the target single particle, and input into the extreme gradient boosting model, and the probability value that the target single particle belongs to the set pollution source is obtained through prediction by the extreme gradient boosting model.

[0031] The present invention has the following advantages: by using a computer-controlled scanning electron microscope to analyze the collected samples of the set pollution source, the image, component data and particle size parameters of the single particles in the sample of the set pollution source are obtained, and the single particle data set of the pollution source sample is constructed; the single particle data set is randomly divided into a first data set and a second data set according to a set ratio; the first data set is used for training the source analysis model; the second data set is used for verifying the source analysis model; the source analysis model is constructed according to the residual network model and the extreme gradient lifting model; the source analysis model is iteratively trained through the first data set; the prediction result of the source analysis model is verified through the second data set to obtain the trained source analysis model; the image, component data and particle size parameters of the target single particle are input into the trained source analysis model for processing, the probability value of the target single particle belonging to the set pollution source is obtained, and the source of the target single particle is determined. The present invention can fully explore the morphology, particle size and component characteristics of particulate matter, identify the source of particulate matter and quantify the contribution of different pollution sources. At the same time, it is helpful to further study the physical and chemical properties of ambient atmospheric particulate matter and the source analysis of atmospheric particulate matter. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the implementation methods of the present invention or the technical solutions in the prior art, the drawings required for the implementation methods or the description of the prior art are briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other implementation drawings can be derived from the provided drawings without creative work.

[0033] The structures, proportions, sizes, etc. illustrated in this specification are only used to match the contents disclosed in the specification so as to facilitate understanding and reading by persons familiar with the technology. They are not used to limit the conditions under which the present invention can be implemented, and therefore have no substantial technical significance. Any structural modification, change in proportion or adjustment of size shall still fall within the scope of the technical contents disclosed in the present invention without affecting the effects and purposes that can be achieved by the present invention.

[0034] Figure 1 A schematic diagram of a flow chart of a method for identifying the source of a single atmospheric particle based on morphology, particle size, and composition provided in Example 1 of the present invention;

[0035] Figure 2 A schematic diagram of specific implementation steps of the method for identifying the source of a single atmospheric particle based on morphology, particle size, and composition provided in Example 1 of the present invention;

[0036] Figure 3 A schematic diagram of the accuracy of the source analysis model in predicting different pollution sources in the atmospheric single particle source identification method based on morphology, particle size and composition provided in Example 1 of the present invention;

[0037] Figure 4 A schematic diagram of the accuracy of the prediction results of the source analysis model for predicting the sources of different types of particulate matter in the atmospheric single particle source identification method based on morphology, particle size and composition provided in Example 1 of the present invention;

[0038] Figure 5 A schematic diagram of a confusion matrix for predicting different types of particulate matter using a source analysis model in a method for identifying atmospheric single particle sources based on morphology, particle size, and composition provided in Example 1 of the present invention;

[0039] Figure 6 Schematic diagram of the architecture of an atmospheric single particle source identification device based on morphology, particle size, and composition provided in Example 2 of the present invention. DETAILED DESCRIPTION

[0040] The following is a description of the implementation of the present invention by specific embodiments. People familiar with the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] Example 1

[0042] See also Figure 1 and Figure 2 Embodiment 1 of the present invention provides a method for identifying the source of a single atmospheric particle based on morphology, particle size, and composition, including:

[0043] S1. Analyze the collected samples of the set pollution source by using a scanning electron microscope controlled by a computer, obtain the image, composition data and particle size parameters of single particles in the samples of the set pollution source, and construct a single particle data set of the pollution source samples;

[0044] S2. randomly dividing the single particle data set into a first data set and a second data set according to a set ratio; the first data set is used for training a source analysis model; and the second data set is used for verifying the source analysis model;

[0045] S3. Construct a source analysis model based on the residual network model and the extreme gradient boosting model;

[0046] S4, iteratively training the source analysis model using the first data set; verifying the prediction result of the source analysis model using the second data set to obtain the trained source analysis model;

[0047] S5. Input the image, component data and particle size parameters of the target single particle into the trained source analysis model for processing, obtain the probability value that the target single particle belongs to the set pollution source, and determine the source of the target single particle.

[0048] In this embodiment, in step S1, in the process of analyzing the collected samples of the set pollution sources by applying a computer-controlled scanning electron microscope, the set pollution sources include: soil dust, road dust, construction dust, coal smoke dust, biomass combustion dust and steel smelting dust.

[0049] Specifically, in this embodiment, six types of pollution source samples were collected, including soil dust, road dust, construction dust, coal smoke dust, biomass combustion dust, and steel smelting dust. The pollution source samples were analyzed using the scanning electron microscope energy spectrum analysis (SEM-EDS) method to build a typical pollution source single particle database. The computer-controlled scanning electron microscope energy spectrum analysis (CCSEM-EDS) method was used to analyze at least 10,000 particles of each type of source sample, of which at least 2,000 particles were analyzed for each sample.

[0050] In this embodiment, in step S1, in the process of obtaining the image, component data and particle size parameters of a single particle in the sample of the set pollution source, the image, component data and particle size parameters of the single particle;

[0051] Specifically, the data collected by CCSEM-EDS include a microscopic picture of the particles, particle size parameters (such as maximum diameter, minimum diameter, average diameter, vertical diameter and equivalent circle diameter, etc.) and element data (including carbon (C), oxygen (O), sodium (Na), magnesium (Mg), aluminum (Al), silicon (Si), phosphorus (P), sulfur (S), chlorine (Cl), potassium (K), calcium (Ca), zinc (Zn), barium (Ba), vanadium (V), chromium (Cr), cobalt (Co), selenium (Se), tin (Sn), titanium (Ti), manganese (Mn), nickel (Ni), copper (Cu) and lead (Pb), etc.). The maximum, minimum and average diameters refer to the maximum, minimum and average diameters of the particles calculated by multiple rotation measurements, the vertical diameter refers to the diameter of the particles perpendicular to the maximum diameter, and the equivalent circle diameter refers to the diameter of a circle with the same area as the particle. In this embodiment, in step S2, the single particle data set is randomly divided into a first data set and a second data set according to a set ratio; the first data set is used for training the source analysis model; the second data set is used for verifying the source analysis model;

[0052] Specifically, the single-particle data sets in the source samples obtained from the analysis are divided into two sets. 80% of the single-particle data from each type of pollution source are randomly selected as the first data set, which is mainly used for training the source analysis model; the remaining 20% ​​of the single-particle data from each type of pollution source are the second data set, which is used to verify the accuracy of the source analysis results.

[0053] In this embodiment, in step S3, a source analysis model is constructed according to the residual network model and the extreme gradient boosting model;

[0054] Specifically, the residual network model is combined with the extreme gradient boosting model to form a source analysis model. The preliminary probability value of a single particle belonging to a set pollution source is obtained through the residual network model. The preliminary probability value of the single particle belonging to the set pollution source and its corresponding particle size parameters and component data are simultaneously input into the extreme gradient boosting model to predict the probability value of the single particle belonging to the set pollution source.

[0055] In this embodiment, in step S4, the source analysis model is iteratively trained using the first data set;

[0056] Specifically, the single particle data in the first data set are input into the source analysis model in batches, the pollution source of the single particle is predicted by the source analysis model, the predicted result is compared with the actual pollution source, and the source analysis model is continuously optimized.

[0057] Verifying the prediction result of the source analysis model by using the second data set to obtain the trained source analysis model;

[0058] Specifically, the trained source analysis model is used to predict the pollution source of a single particle in the second data set. The prediction accuracy of different pollution sources is as follows: Figure 3 shown.

[0059] The probability value of the single particle belonging to the set pollution source is weighted and summed up by source category according to the mass or quantity of the single particle to obtain the mass or quantity ratio of the atmospheric particulate matter emitted by the set pollution source. Figure 3 The values ​​on the upper left to lower right diagonal indicate the prediction accuracy of various pollution sources, where the values ​​in brackets indicate the number of samples. The prediction accuracy of the six pollution sources is 91.53-99.09%, and the average accuracy is 95.22%. Considering that the same pollution source may contain different types of particulate matter (having differences in morphology and chemical composition), and different pollution sources may contain similar types of particulate matter, such as road dust may contain metal particles and aluminosilicates, and aluminosilicates also exist in soil dust. In order to further explore the accuracy of the present invention in identifying different types of particulate pollution sources, the particulate matter is now classified, and the steps of particulate matter classification are as follows:

[0060] T1. Variable selection: Considering the different intrinsic meanings of shape parameter, particle size and element concentration variables, it may be difficult to interpret them through cluster analysis. Therefore, chemical element variables (including C, O, Na, Mg, Al, Si, P, S, Cl, K, Ca, Zn, Ba, V, Cr, Co, Se, Sn, Ti, Mn, Ni, Cu and Pb) were first selected from all variables for cluster analysis.

[0061] T2. Data noise processing: If any X-ray count of a particle is less than twice the square root of the total X-ray count, it is set to zero.

[0062] T3. Data transformation: The data were logarithmically transformed to compress the distribution, and then the X-ray spectrum of each particle data was normalized so that all chemical element variables had equal weight in the classification.

[0063] T4. Cluster analysis: All sample data are integrated and cluster analysis is performed using the k-means method. The optimal k value is determined by the elbow method, that is, by selecting different k values, observing the changes in the sum of squared errors (SSE). k It represents the error of clustering samples into k categories, specifically the sum of the squares of the distances from each point in the cluster to the cluster center.

[0064]

[0065] In the formula, X i is the i-th cluster; j is X i Sample points in m i Yes X i The center of mass, X i The mean of all samples in .

[0066] As the number of clusters k increases, the sample division becomes more and more refined, and SSE k Gradually decreases. When k is greater than a certain number, SSE k As the k value continues to increase, it tends to be flat, and the k=K value corresponding to the elbow is taken as the number of clusters in this embodiment. In order to make the particle classification more detailed, the K value is taken as 100 here.

[0067] T5. Manual classification: After clustering, the 100 clusters were divided into 33 categories according to the similarity of their EDS spectra with those of different types of particles in related studies and the morphological characteristics of single particles, namely Si-Al, Si-rich, Si-Al-Ca, Si-Al-Fe, Si-Al-K-Mg, Si-Al-Mg, Si-Al-Na, Si-Al-Ca-Mg-Fe, Si-Al-Ca-Mg, Si-Ca, Al-rich, Al-Ca, Ca-Mg, Ca-Mg-Cl, P-containing, K-containing, Fly ash, C-rich, Fe-rich, Ca-rich, S-Ca, Ti-rich, Fe-Ca-S, Fe-K-Cl, Mn-rich, K-Cl, Pb-Cl, Pb-Fe-Cl, Pb-Cl-K, S-Al-Ca-Mg, S-Ca-Pb, S-Ca-K-Cl and Other.

[0068] The prediction accuracy and model confusion matrix of different types of particulate matter sources are shown as follows: Figure 4 and Figure 5 The results show that the source analysis model in the present invention can accurately identify different types of particulate matter to the corresponding pollution sources, further proving that the model has excellent performance.

[0069] In this embodiment, in step S5, while inputting the image, component data and particle size parameters of the target single particle into the trained source analysis model for processing, and obtaining the probability value that the target single particle belongs to the set pollution source, the image features of the target single particle are extracted through the residual network model, and the source of the target single particle is preliminarily predicted based on the image features of the target single particle, so as to obtain the preliminary probability value that the target single particle belongs to the set pollution source.

[0070] Specifically, data enhancement techniques are applied to enhance the diversity of data sets. Random rotation, flipping and scaling of single-particle microscopic images are used to reduce the impact of variability in particle position and size on prediction results. At the same time, image brightness, contrast, saturation and hue are randomly adjusted to reduce the differences in microscopic images of different samples caused by differences in CCSEM analysis conditions. In terms of image feature extraction, a residual network (ResNet) is used to extract image features and identify the source of particles based on image features. Compared with traditional convolutional neural networks (CNNs), the residual blocks inside ResNet use jump connections, which alleviates the problem of gradient disappearance caused by the deepening of the neural network. Traditional CNN architectures require a large amount of training data to achieve high prediction accuracy. The present invention adopts a transfer learning strategy, and the initial parameters of the ResNet model use parameters trained on a large ImageNet dataset. In order to make the pre-trained features and adapt to the image dataset of a single particle, the last residual block is fine-tuned, and the pre-trained weights of other blocks remain unchanged. After using the ResNet model for feature extraction, a fully connected layer is also added for particle source prediction. During the training process, the Adam optimizer was used, with a learning rate of 1e-2, a step size of 7, a decay rate of 0.1, and a negative log-likelihood loss function to calculate the loss and optimize the model performance. Finally, the preliminary probability values ​​of each particle coming from different sources were recorded.

[0071] In this embodiment, in step S5, the image, component data and particle size parameters of the target single particle are input into the trained source analysis model for processing to obtain the probability value of the target single particle belonging to the set pollution source, and the preliminary probability value of the target single particle belonging to the set pollution source is merged with the particle size parameters and component data of the target single particle, and input into the extreme gradient boosting model, and the probability value of the target single particle belonging to the set pollution source is obtained through prediction by the extreme gradient boosting model.

[0072] Specifically, the preliminary probability value of the single particle belonging to the set pollution source predicted by the residual network model is combined with the particle size parameters and composition data of the single particle as the input of the extreme gradient boosting (XGBoost) model, and the particle source label is the dependent variable. The extreme gradient boosting model training is completed, and the probability of each particle belonging to a different pollution source calculated by the extreme gradient boosting model is output. In the optimization process of the extreme gradient boosting model, grid search is used to select the most appropriate hyperparameters, and five-fold cross validation is used to evaluate the performance of the XGBoost model.

[0073] In summary, the present invention uses a computer-controlled scanning electron microscope to analyze the collected samples of the set pollution source, obtains the image, component data and particle size parameters of the single particles in the sample of the set pollution source, and constructs a single particle data set of the pollution source sample; randomly divides the single particle data set into a first data set and a second data set according to a set ratio; the first data set is used for training the source analysis model; the second data set is used for verifying the source analysis model; the source analysis model is constructed according to the residual network model and the extreme gradient lifting model; the source analysis model is iteratively trained through the first data set; the prediction result of the source analysis model is verified through the second data set to obtain the trained source analysis model; the image, component data and particle size parameters of the target single particle are input into the trained source analysis model for processing, and the probability value of the target single particle belonging to the set pollution source is obtained, and the source of the target single particle is determined. The present invention can fully explore the morphology, particle size and component characteristics of particulate matter, identify the source of particulate matter and quantify the contribution of different pollution sources. At the same time, it is helpful to further study the physical and chemical properties of ambient atmospheric particulate matter and the source analysis of atmospheric particulate matter.

[0074] It should be noted that the method of the embodiment of the present disclosure can be performed by a single device, such as a computer or a server. The method of the present embodiment can also be applied in a distributed scenario and completed by multiple devices cooperating with each other. In the case of such a distributed scenario, one of the multiple devices can only perform one or more steps in the method of the embodiment of the present disclosure, and the multiple devices will interact with each other to complete the described method.

[0075] It should be noted that the above describes some embodiments of the present disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the above embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0076] Example 2

[0077] See also Figure 6 Embodiment 2 of the present invention provides an atmospheric single particle source identification device based on morphology, particle size, and composition, including:

[0078] The single particle data set acquisition module 001 is used to analyze the collected samples of the set pollution source by using a scanning electron microscope controlled by a computer, obtain the image, component data and particle size parameters of the single particles in the samples of the set pollution source, and construct a single particle data set of the pollution source samples;

[0079] The single particle data set division module 002 is used to randomly divide the single particle data set into a first data set and a second data set according to a set ratio; the first data set is used for training the source analysis model; the second data set is used for verifying the source analysis model;

[0080] The source analysis model construction module 003 is used to construct a source analysis model according to the residual network model and the extreme gradient boosting model;

[0081] The source analysis model training module 004 is used to iteratively train the source analysis model using the first data set; verify the prediction result of the source analysis model using the second data set to obtain the trained source analysis model;

[0082] The single particle source determination module 005 is used to input the image, component data and particle size parameters of the target single particle into the trained source analysis model for processing, obtain the probability value of the target single particle belonging to the set pollution source, and determine the source of the target single particle.

[0083] In this embodiment, in the single particle data set acquisition module 001, in the process of analyzing the collected samples of the set pollution source by applying a computer-controlled scanning electron microscope, the set pollution source includes: soil dust, road dust, construction dust, coal smoke dust, biomass combustion dust and steel smelting dust.

[0084] In this embodiment, in the single particle data acquisition module 001, the image of the single particle includes: a high-definition microscopic photo of the single particle;

[0085] The composition data of the single particle includes: carbon, oxygen, sodium, magnesium, aluminum, silicon, phosphorus, sulfur, chlorine, potassium, calcium, zinc, barium, vanadium, chromium, cobalt, selenium, tin, titanium, manganese, nickel, copper and lead;

[0086] The particle size parameters of the single particle include: maximum diameter, minimum diameter, average diameter, vertical diameter and equivalent circle diameter.

[0087] In this embodiment, in the single particle source determination module 005, in the process of inputting the image, component data and particle size parameters of the target single particle into the trained source analysis model for processing, and obtaining the probability value that the target single particle belongs to the set pollution source, the image features of the target single particle are extracted through the residual network model, and the source of the target single particle is preliminarily predicted based on the image features of the target single particle, so as to obtain the preliminary probability value that the target single particle belongs to the set pollution source.

[0088] In this embodiment, in the single particle source determination module 005, in the process of inputting the image, component data and particle size parameters of the target single particle into the trained source analysis model for processing, and obtaining the probability value of the target single particle belonging to the set pollution source, the preliminary probability value of the target single particle belonging to the set pollution source is merged with the particle size parameters and component data of the target single particle, and input into the extreme gradient boosting model, and the probability value of the target single particle belonging to the set pollution source is obtained through prediction by the extreme gradient boosting model.

[0089] It should be noted that the information interaction, execution process, etc. between the modules / units of the above-mentioned system are based on the same concept as the method embodiment in Example 1 of the present application, and the technical effects they bring are the same as those of the method embodiment of the present application. For specific contents, please refer to the description in the method embodiment shown in the previous part of the present application, and will not be repeated here.

[0090] Example 3

[0091] Embodiment 3 of the present invention provides a non-transitory computer-readable storage medium, in which a program code for a method for identifying the source of a single atmospheric particle based on morphology, particle size, and composition is stored. The program code includes instructions for executing the method for identifying the source of a single atmospheric particle based on morphology, particle size, and composition of embodiment 1 or any possible implementation thereof.

[0092] The computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that includes one or more available media. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0093] Example 4

[0094] Embodiment 4 of the present invention provides an electronic device, including: a memory and a processor;

[0095] The processor and the memory communicate with each other via a bus; the memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to execute the atmospheric single particle source identification method based on morphology, particle size, and composition of Example 1 or any possible implementation thereof.

[0096] Specifically, the processor can be implemented by hardware or by software. When implemented by hardware, the processor can be a logic circuit, an integrated circuit, etc.; when implemented by software, the processor can be a general-purpose processor implemented by reading software codes stored in a memory. The memory can be integrated in the processor or can be located outside the processor and exist independently.

[0097] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions can be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center.

[0098] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, and optionally, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0099] Although the present invention has been described in detail above by general description and specific embodiments, it is obvious to those skilled in the art that some modifications or improvements can be made on the basis of the present invention. Therefore, these modifications or improvements made on the basis of not departing from the spirit of the present invention all belong to the scope of protection claimed by the present invention.

Claims

1. A method for identifying the source of atmospheric single particles based on morphology, particle size and composition, characterized in that: include: Analyzing the collected samples of the set pollution source by using a computer-controlled scanning electron microscope to obtain images, composition data and particle size parameters of single particles in the samples of the set pollution source, and constructing a single particle data set of the pollution source samples; The single particle data set is randomly divided into a first data set and a second data set according to a set ratio; the first data set is used for training a source analysis model; and the second data set is used for verifying the source analysis model; Construct a source analysis model based on the residual network model and the extreme gradient boosting model; Iteratively training the source analysis model using the first data set; Verifying the prediction result of the source analysis model by using the second data set to obtain the trained source analysis model; The image, component data and particle size parameters of the target single particle are input into the trained source analysis model for processing, the probability value of the target single particle belonging to the set pollution source is obtained, and the source of the target single particle is determined.

2. The method for identifying the source of single atmospheric particles based on morphology, particle size and composition according to claim 1 is characterized in that: In the process of analyzing the collected samples of the set pollution sources by applying a computer-controlled scanning electron microscope, the set pollution sources include: soil dust, road dust, construction dust, coal smoke dust, biomass combustion dust and steel smelting dust.

3. The method for identifying the source of single atmospheric particles based on morphology, particle size and composition according to claim 2 is characterized in that: The image of the single particle includes: a high-definition microscopic photo of the single particle; The composition data of the single particle includes: carbon, oxygen, sodium, magnesium, aluminum, silicon, phosphorus, sulfur, chlorine, potassium, calcium, zinc, barium, vanadium, chromium, cobalt, selenium, tin, titanium, manganese, nickel, copper and lead; The particle size parameters of the single particle include: maximum diameter, minimum diameter, average diameter, vertical diameter and equivalent circle diameter.

4. The method for identifying the source of single atmospheric particles based on morphology, particle size and composition according to claim 3 is characterized in that: In the process of inputting the image, component data and particle size parameters of the target single particle into the trained source analysis model for processing to obtain the probability value of the target single particle belonging to the set pollution source, the image features of the target single particle are extracted through the residual network model, and the source of the target single particle is preliminarily predicted based on the image features of the target single particle to obtain the preliminary probability value of the target single particle belonging to the set pollution source.

5. The method for identifying the source of single atmospheric particles based on morphology, particle size and composition according to claim 4 is characterized in that: In the process of inputting the image, component data and particle size parameters of the target single particle into the trained source analysis model for processing to obtain the probability value of the target single particle belonging to the set pollution source, the preliminary probability value of the target single particle belonging to the set pollution source is merged with the particle size parameters and component data of the target single particle, and inputted into the extreme gradient boosting model, and the probability value of the target single particle belonging to the set pollution source is obtained through prediction by the extreme gradient boosting model.

6. The atmospheric single particle source identification device based on morphology, particle size, and composition adopts the atmospheric single particle source identification method based on morphology, particle size, and composition as described in claims 1-5 above, characterized in that: include: A single particle data set acquisition module is used to analyze the collected samples of the set pollution source by using a scanning electron microscope controlled by a computer, obtain the image, composition data and particle size parameters of the single particles in the samples of the set pollution source, and construct a single particle data set of the pollution source samples; A single particle data set division module, used to randomly divide the single particle data set into a first data set and a second data set according to a set ratio; the first data set is used for training the source analysis model; the second data set is used for verifying the source analysis model; The source analysis model building module is used to build a source analysis model based on the residual network model and the extreme gradient boosting model; A source analysis model training module, configured to iteratively train the source analysis model using the first data set; Verifying the prediction result of the source analysis model by using the second data set to obtain the trained source analysis model; The single particle source determination module is used to input the image, component data and particle size parameters of the target single particle into the trained source analysis model for processing, obtain the probability value that the target single particle belongs to the set pollution source, and determine the source of the target single particle.

7. The atmospheric single particle source identification device based on morphology, particle size and composition according to claim 6 is characterized in that: In the single particle data set acquisition module, in the process of analyzing the collected samples of the set pollution sources by applying a computer-controlled scanning electron microscope, the set pollution sources include: soil dust, road dust, construction dust, coal smoke dust, biomass combustion dust and steel smelting dust.

8. The atmospheric single particle source identification device based on morphology, particle size and composition according to claim 7 is characterized in that: The image of the single particle includes: a high-definition microscopic photo of the single particle; In the single particle data acquisition module, the component data of the single particle includes: carbon, oxygen, sodium, magnesium, aluminum, silicon, phosphorus, sulfur, chlorine, potassium, calcium, zinc, barium, vanadium, chromium, cobalt, selenium, tin, titanium, manganese, nickel, copper and lead; The particle size parameters of the single particle include: maximum diameter, minimum diameter, average diameter, vertical diameter and equivalent circle diameter.

9. The atmospheric single particle source identification device based on morphology, particle size and composition according to claim 8 is characterized in that: In the single particle source determination module, in the process of inputting the image, component data and particle size parameters of the target single particle into the trained source analysis model for processing, and obtaining the probability value that the target single particle belongs to the set pollution source, the image features of the target single particle are extracted through the residual network model, and the source of the target single particle is preliminarily predicted based on the image features of the target single particle, so as to obtain the preliminary probability value that the target single particle belongs to the set pollution source.

10. The atmospheric single particle source identification device based on morphology, particle size and composition according to claim 9 is characterized in that: In the single particle source determination module, in the process of inputting the image, component data and particle size parameters of the target single particle into the trained source analysis model for processing, and obtaining the probability value of the target single particle belonging to the set pollution source, the preliminary probability value of the target single particle belonging to the set pollution source is merged with the particle size parameters and component data of the target single particle, and inputted into the extreme gradient boosting model, and the probability value of the target single particle belonging to the set pollution source is obtained through prediction by the extreme gradient boosting model.

Images