Prediction analysis method and system for mixed micro-plastic, electronic equipment and medium
Through the proposed prediction and analysis method of hybrid microplastics, the problem of time-consuming and costly prediction of microplastics in the prior art is solved, the prediction accuracy is improved, and the impact of microplastics on the environment is comprehensively evaluated.
Patent Information
- Application Number
- CN202510059999.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-15
- Publication Date
- 2025-05-30
AI Technical Summary
The prior art has problems of long and high cost in the classification prediction of microplastics, and the assessment of the potential impact of microplastics on the environment and humans is not comprehensive enough.
A predictive analysis method for hybrid microplastics is proposed. By obtaining the spectral data and element composition data of the target hybrid microplastics, extracting spectral descriptors and element composition descriptors, identifying the types and proportions of microplastics, and generating an environmental impact assessment report.
Improve the accuracy of predictive values for identifying the proportion of microplastics, and more comprehensively evaluate the environmental impact of microplastics, reducing costs and time expenditures.
Smart Images
Figure CN120069135A_ABST
Abstract
Description
Technical Field
[0001] The embodiments of the present application relate to the technical field of microplastic treatment, and in particular, to a prediction analysis method, system, electronic device, and medium for mixed microplastics. Background Art
[0002] With the increasing severity of microplastic pollution, the recycling of microplastics has become an important issue that urgently needs to be solved. As an important and basic step in microplastic recycling, the classification prediction of microplastics has become the focus of current research. In the current research on garbage classification prediction technology, a variety of analysis techniques are widely used, including spectral analysis, thermal analysis, mass spectrometry analysis, chemical analysis, and image analysis, etc. However, there are generally problems of long time consumption and high cost, and it is restricted by conditions such as equipment. Moreover, with the increasing attention to the potential environmental impacts of new pollutants, the potential impacts of microplastics on the environment and humans are often ignored, and the corresponding assessment methods and basic environmental potential impact data should be known to more people. Summary of the Invention
[0003] The following is an overview of the subject matter described in detail in this article. This overview is not intended to limit the scope of protection of the claims.
[0004] The main purpose of the embodiments of the present disclosure is to propose a prediction analysis method, system, electronic device, and medium for mixed microplastics, which can improve the accuracy of the predicted value of identifying the proportion of microplastics and evaluate the impact on the environment.
[0005] The first aspect of the embodiments of the present application proposes a prediction analysis method for mixed microplastics, and the method includes:
[0006] Obtain the target spectral data and target element composition data of the target mixed microplastics; the target mixed microplastics are composed of multiple microplastics, the target spectral data includes the spectral values of the target mixed microplastics in any spectral mode, and the target element composition data includes the content values of the mixed monomer elements in the target mixed microplastics and the corresponding chemical molecular formulas;
[0007] Extract a plurality of target spectral descriptors from the target spectral data, and extract a target element composition descriptor from the target element composition data; the target spectral descriptors include target spectral features and their values, and the target element composition descriptors include target element composition features and their values;
[0008] According to the plurality of target spectral descriptors, identify the microplastic type information in the target mixed microplastics, and select several target spectral descriptors from the plurality of target spectral descriptors whose importance for identifying the microplastic type information is higher than the first threshold;
[0009] Based on the plurality of target spectral descriptors and the target elemental composition descriptor, identify the predicted values of the proportion of each microplastic in the target mixed microplastics;
[0010] Generate an environmental impact assessment report for the target mixed microplastics according to the microplastic type information and the predicted values.
[0011] The present application proposes a predictive analysis method for mixed microplastics, which has the following beneficial effects:
[0012] This method first obtains the target spectral data and target elemental composition data of the target mixed microplastics, then extracts the target spectral descriptors that can characterize the target spectral data and the target elemental composition descriptors that can characterize the target elemental composition data from them. Then, through a plurality of target spectral descriptors, identify the microplastic type information in the target mixed microplastics, and determine a plurality of target spectral descriptors whose importance for identifying the microplastic type information is higher than the first threshold, which are used for subsequent identification of the predicted values of the proportion of each microplastic in the target mixed microplastics according to the plurality of target spectral descriptors and the target elemental composition descriptor. Compared with the scheme of solely identifying the predicted values of the microplastic proportion based on the target elemental composition descriptor, several target spectral descriptors with higher importance in identifying the microplastic type information are selected here. Such target spectral descriptors play an important decisive role in the identification of the predicted values. The accuracy of the predicted values for identifying the proportion of microplastics can be improved through the selected target spectral descriptors and the target elemental composition descriptor; finally, generate an environmental impact assessment report for the target mixed microplastics according to the microplastic type information and the predicted values to better understand the impact of the target mixed microplastics on the environment.
[0013] In some embodiments, before obtaining the target spectral data and target elemental composition data of the target mixed microplastics, it further includes:
[0014] Obtain the initial spectral data and initial elemental composition data of the target mixed microplastics;
[0015] Perform a first preprocessing on the initial spectral data to obtain the target spectral data; the first preprocessing includes at least one of peak feature extraction, area feature extraction, and principal component analysis;
[0016] Perform a second preprocessing on the initial elemental composition data to obtain the target elemental composition data; the second preprocessing includes at least one of missing value processing, outlier processing, and data standardization.
[0017] In some embodiments, according to the plurality of target spectral descriptors, identifying the microplastic type information in the target mixed microplastics includes:
[0018] Input the multiple target spectral descriptors into a preset classification model to identify the types of microplastics in the target mixed microplastics through the classification model;
[0019] Among them, the training process of the classification model includes the following:
[0020] Construct a first machine learning model and construct a spectral database; the spectral database contains a preset multiple theoretical spectral descriptors and their corresponding theoretical microplastic types;
[0021] Input the multiple theoretical spectral descriptors in the spectral database into the first machine learning model, and make the first machine learning model output the corresponding microplastic types;
[0022] Optimize the first machine learning model according to the output corresponding microplastic types and the theoretical microplastic types in the spectral database until the classification model is obtained.
[0023] In some embodiments, constructing the spectral database includes:
[0024] Based on quantum theory, construct spectral images of multiple single microplastics;
[0025] Overlay the spectral images of the multiple single microplastics to construct a spectral image of mixed microplastics;
[0026] Generate theoretical spectral descriptors according to the spectral image of the mixed microplastics, and determine the microplastic types corresponding to the theoretical spectral descriptors;
[0027] Construct the spectral database according to the theoretical spectral descriptors and the corresponding microplastic types.
[0028] In some embodiments, selecting several target spectral descriptors from the multiple target spectral descriptors whose importance for identifying the microplastic type information is higher than a first threshold includes:
[0029] Analyze the importance of the multiple target spectral descriptors for the classification model to output microplastic type information according to the feature importance analysis method;
[0030] Sort the multiple target spectral descriptors according to the importance to obtain a sorting result;
[0031] Select several target spectral descriptors whose importance is higher than the first threshold according to the sorting result.
[0032] In some embodiments, according to the several target spectral descriptors and the target element composition descriptors, identifying the predicted values of the proportions of each microplastic in the target mixed microplastics includes:
[0033] Input the several target spectral descriptors and the target element composition descriptor into a preset prediction model, and obtain the predicted values of the proportion of each microplastic in the target mixed microplastics recognized by the prediction model;
[0034] Among them, the training process of the prediction model includes the following:
[0035] Construct a second machine learning model and construct an element composition database; the element composition database contains multiple sets of input data and output data. Each set of input data includes the theoretical element composition descriptors and their spectral descriptors corresponding to all microplastic combination modes of a mixed microplastic, and the output data is the theoretical prediction value;
[0036] Input the multiple sets of input data in the element composition database into the second machine learning model, and enable the second machine learning model to output the corresponding prediction data;
[0037] Optimize the second machine learning model according to the output prediction data and the output data in the element composition database until the prediction model is obtained.
[0038] In some embodiments, the constructing the element composition database includes:
[0039] Determine the elements and molecular formulas corresponding to all microplastic combination modes of the mixed microplastics;
[0040] Calculate the theoretical element composition descriptors and their corresponding theoretical prediction values according to the elements and molecular formulas corresponding to all microplastic combination modes of the mixed microplastics;
[0041] Determine the theoretical spectral descriptors corresponding to the theoretical element composition descriptors;
[0042] Select the theoretical element composition descriptors and theoretical spectral descriptors with corresponding relationships as a set of input data, and use the theoretical prediction values corresponding to the theoretical element descriptors as output data;
[0043] Construct the element composition database according to the input data and the output data.
[0044] To achieve the above object, a second aspect of the embodiments of the present invention provides a prediction analysis system for mixed microplastics, and the system includes:
[0045] A data acquisition module for acquiring target spectral data and target elemental composition data of target mixed microplastics; the target mixed microplastics are composed of multiple types of microplastics, the target spectral data includes spectral values of the target mixed microplastics in any spectral mode, and the target elemental composition data includes the content values of mixed monomer elements in the target mixed microplastics and corresponding chemical formulas;
[0046] A descriptor acquisition module for extracting a plurality of target spectral descriptors from the target spectral data and extracting a target elemental composition descriptor from the target elemental composition data; the target spectral descriptors include target spectral features and their values, and the target elemental composition descriptors include target elemental composition features and their values;
[0047] A classification module for identifying the types of microplastics in the target mixed microplastics according to the plurality of target spectral descriptors, and selecting from the plurality of target spectral descriptors several target spectral descriptors whose importance for identifying the types of microplastics is higher than a first threshold;
[0048] A prediction module for identifying predicted values of the proportion of each microplastic in the target mixed microplastics according to the several target spectral descriptors and the target elemental composition descriptor;
[0049] An environmental assessment module for generating an environmental impact assessment report of the target mixed microplastics according to the types of microplastics information and the predicted values.
[0050] To achieve the above object, a third aspect of the embodiments of the present invention provides an electronic device, including: at least one control processor and a memory for communicatively connecting with the at least one control processor; the memory stores instructions executable by the at least one control processor, and the instructions are executed by the at least one control processor so that the at least one control processor can execute the above-mentioned prediction analysis method of mixed microplastics.
[0051] To achieve the above object, a fourth aspect of the embodiments of the present invention provides a computer-readable storage medium, the computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to make a computer execute the above-mentioned prediction analysis method of mixed microplastics.
[0052] It can be understood that the beneficial effects of the above second aspect to the fourth aspect compared with the related art are the same as the beneficial effects of the above first aspect compared with the related art. For the relevant descriptions, reference can be made to the relevant descriptions in the above first aspect, and details will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS
[0053] To more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the accompanying drawings required for use in the embodiments or the description of related technologies. Obviously, the accompanying drawings in the following description are only some embodiments of the embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.
[0054] Figure 1 is a schematic flowchart of the prediction analysis method for mixed microplastics provided by the embodiments of the present application;
[0055] Figure 2 is a schematic diagram of the feature importance of the spectral output provided by the embodiments of the present application;
[0056] Figure 3 is a schematic diagram of the prediction effect of the prediction model provided by the embodiments of the present application;
[0057] Figure 4 is a schematic diagram of an embodiment of the prediction analysis system for mixed microplastics provided by the present application;
[0058] Figure 5 is a schematic diagram of an embodiment of the electronic device provided by the present application. Detailed implementation manners
[0059] In order to make the purpose, technical solutions and advantages of the present application more clear, the following further details the present application in combination with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.
[0060] It should be noted that although the functional modules are divided in the device schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order from the module division in the device or the flowchart. Terms such as "first" and "second" in the specification, claims and the above accompanying drawings are used to distinguish similar objects and do not necessarily need to describe a specific order or sequence.
[0061] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.
[0062] With the increasing severity of microplastic pollution, the recycling of microplastics has become an important issue that urgently needs to be solved. As an important and fundamental step in microplastic recycling, the classification prediction of microplastics has become the focus of current research. In the current research on waste classification prediction technology, a variety of analysis techniques are widely used, including spectral analysis, thermal analysis, mass spectrometry analysis, chemical analysis, and image analysis, etc. However, there are generally problems of long time consumption and high cost, and it is restricted by conditions such as equipment. Moreover, with the increasing attention to the potential environmental impacts of new pollutants, the potential impacts of microplastics on the environment and humans are often ignored, and the corresponding assessment methods and basic environmental potential impact data should be known to more people.
[0063] To solve the above technical deficiencies, such as Figure 1 , an embodiment of the present application provides a prediction analysis method for mixed microplastics, and the method includes the following steps:
[0064] Step S110, obtaining the target spectral data and target element composition data of the target mixed microplastics.
[0065] Step S120, extracting a plurality of target spectral descriptors from the target spectral data, and extracting a target element composition descriptor from the target element composition data.
[0066] Step S130, based on the plurality of target spectral descriptors, identifying the types of microplastics in the target mixed microplastics, and selecting from the plurality of target spectral descriptors several target spectral descriptors whose importance for identifying the types of microplastics is higher than the first threshold.
[0067] Step S140, based on the several target spectral descriptors and the target element composition descriptor, identifying the predicted values of the proportions of the respective microplastics in the target mixed microplastics.
[0068] Step S150, generating an environmental impact assessment report for the target mixed microplastics according to the types of microplastics information and the predicted values.
[0069] In step S110 of this embodiment, the target mixed microplastics refer to the mixed microplastics to be tested, which are composed of various microplastics.
[0070] Among them, the target spectral data of the target mixed microplastics refers to the spectral values of the target mixed microplastics in any spectral mode, and the spectral modes include but are not limited to infrared spectroscopy, ultraviolet-visible spectroscopy, and Raman spectroscopy, such as the spectral values of the target mixed microplastics under infrared spectroscopy. In addition, the spectral values include but are not limited to important information such as the characteristic peak position, peak value, and bandwidth.
[0071] Among them, the target element composition data includes the content values of the mixed elements in the target mixed microplastics and the corresponding chemical molecular formulas. The elements include, but are not limited to, carbon, hydrogen, oxygen, nitrogen, chlorine, etc. The molecular formulas of each element are well-known in the field and will not be elaborated here.
[0072] In some embodiments, before step S110, the method further includes the following steps:
[0073] Step S210, obtaining the initial spectral data and the initial element composition data of the target mixed microplastics;
[0074] Step S220, performing a first preprocessing on the initial spectral data to obtain the target spectral data; the first preprocessing includes at least one of peak feature extraction, area feature extraction, and principal component analysis;
[0075] Step S230, performing a second preprocessing on the initial element composition data to obtain the target element composition data; the second preprocessing includes at least one of missing value processing, outlier processing, and data standardization.
[0076] In this embodiment, a preprocessing process is added, which can filter out impurities in the initial spectral data and the initial element composition data, ensuring the operability of the subsequent identification and prediction process. In addition, according to the data distribution of the spectral data, some spectral data with uneven distribution ranges can be normalized.
[0077] In step S120, following the above introduction, a plurality of target spectral descriptors are extracted from the target spectral data. Here, the target spectral descriptors include the target spectral features and their spectral values, and the target element composition descriptors are extracted from the target element composition data. The target element composition descriptors include the target element composition features and their values.
[0078] In step S130 of this embodiment, according to a plurality of target spectral descriptors, the microplastic type information in the target mixed microplastics is identified, including the following steps:
[0079] Step S1310, according to a plurality of target spectral descriptors, identifying the microplastic type information in the target mixed microplastics, including:
[0080] Step S1320, inputting a plurality of target spectral descriptors into a preset classification model to identify the microplastic type information in the target mixed microplastics through the classification model;
[0081] Among them, the training process of the classification model includes the following steps:
[0082] Step S1321: Construct the first machine learning model and build a spectral database; the spectral database contains multiple preset theoretical spectral descriptors and their corresponding theoretical microplastic types; the theoretical spectral descriptors include spectral features and their numerical values. It should be noted that this descriptor can be generated based on images, as detailed in the following description. Step S1322: Input the multiple theoretical spectral descriptors in the spectral database into the first machine learning model, and enable the first machine learning model to output the corresponding microplastic types;
[0083] Step S1323: Optimize the first machine learning model according to the output corresponding microplastic types and the theoretical microplastic types in the spectral database until a classification model is obtained.
[0084] Among them, the construction of the spectral database in step S1321 includes the following steps:
[0085] (1) Based on quantum theory, construct spectral images of multiple single microplastics.
[0086] (2) Superimpose the spectral images of multiple single microplastics to construct a spectral image of mixed microplastics.
[0087] (3) Generate theoretical spectral descriptors according to the spectral image of mixed microplastics, and determine the microplastic types corresponding to the theoretical spectral descriptors.
[0088] (4) Construct a spectral database according to the theoretical spectral descriptors and their corresponding microplastic types.
[0089] In the process of the above steps (1) to (4), by using the method of quantum computing, it is possible to obtain the spectral images of microplastics without experimental methods, saving the cumbersome steps and conditional limitations required by experiments. By constructing the atomic structure relationship of single microplastics and their stacked mixed microplastics, and using quantum computing software, such as Gaussian and Multiwfn, calculate various properties of molecules, such as molecular energy, structure, vibration frequency, molecular orbit, etc., to obtain spectral data such as infrared spectra and Raman spectra, and draw them into spectral images through GaussView.
[0090] After obtaining the spectral images of single microplastics based on quantum computing, spectral images under different types of microplastics can be obtained through superposition. The reason for this superposition is that the quantum computing modeling of mixed microplastics has high requirements for computing power and there may be mutual molecular influences. Moreover, in actual mixed microplastics, there are few intermolecular interactions. Considering the above two factors, superposition calculation is selected.
[0091] Before step (3), data enhancement methods can also be applied to the spectral image of mixed microplastics to achieve data expansion.
[0092] Finally, generate theoretical spectral descriptors based on the spectral images of the mixed microplastics, determine the types of microplastics corresponding to the theoretical spectral descriptors, and construct a spectral database.
[0093] In step S1321 of this embodiment, the first machine learning model includes, but is not limited to, decision trees, random forests, support vector machines, and convolutional networks.
[0094] In step S1323 of this embodiment, cross-validation and random search can be used to calculate the optimal parameters of the classification model to obtain the optimal model parameters, and the parameters of the cross-validation are selected according to the actual situation.
[0095] In step S130 of this embodiment, it further includes:
[0096] Select several target spectral descriptors from multiple target spectral descriptors whose importance for identifying the microplastic type information is higher than the first threshold.
[0097] Specifically, step S130 includes:
[0098] (1) According to the feature importance analysis method, analyze the importance of multiple target spectral descriptors for the classification model to output microplastic type information.
[0099] (2) Sort the multiple target spectral descriptors according to the importance to obtain a sorting result.
[0100] (3) Select several target spectral descriptors whose importance is higher than the first threshold according to the sorting result.
[0101] In this embodiment, by analyzing the influence of different target spectral descriptors on the classification results of the classification model, the respective decisive roles of each target spectral descriptor on the output of the model can be determined. Therefore, it can be selected that the higher the feature importance analysis is (the higher the feature importance value), the higher the decisiveness for the classification results of the target mixed microplastics. Preferably, generally two to three target spectral descriptors are selected to be used to assist in the generation of subsequent prediction values.
[0102] In some embodiments, step S140 includes the following steps:
[0103] Step S1410, input the descriptors composed of several target spectral descriptors and target elements into a preset prediction model to obtain the predicted values of the proportions of each microplastic in the target mixed microplastics identified by the prediction model.
[0104] Among them, the training process of the prediction model includes the following:
[0105] Step S1411: Construct a second machine learning model and a database of element compositions; the database of element compositions contains multiple sets of input data and output data. Each set of input data includes the theoretical element composition descriptors and their spectral descriptors corresponding to all the microplastic combination ways of a mixed microplastic, and the output data is the theoretical prediction value.
[0106] Step S1412: Input the multiple sets of input data in the database of element compositions into the second machine learning model, and make the second machine learning model output the corresponding prediction data.
[0107] Step S1413: Optimize the second machine learning model according to the output prediction data and the output data in the database of element compositions until a prediction model is obtained.
[0108] In some embodiments of the present application, constructing the database of element compositions includes:
[0109] (1) Determine the elements and molecular formulas corresponding to all the microplastic combination ways of the mixed microplastic.
[0110] (2) Calculate the theoretical element composition descriptors and their corresponding theoretical prediction values according to the elements and molecular formulas corresponding to all the microplastic combination ways of the mixed microplastic.
[0111] (3) Determine the theoretical spectral descriptors corresponding to the theoretical element composition descriptors. Here, it means to match according to the types of mixed microplastics to ensure finding the theoretical element composition descriptors and spectral descriptors corresponding to the same type of mixed microplastics.
[0112] (4) Select the theoretical element composition descriptors and theoretical spectral descriptors with corresponding relationships as a set of input data, and use the theoretical prediction value with a corresponding relationship with the theoretical element descriptor as the output data.
[0113] (5) Construct a database of element compositions according to the input data and output data.
[0114] In step S1413, cross-validation and random search can be used to calculate the optimal parameters of the classification model to obtain the optimal model parameters, and the parameters of cross-validation are selected according to the actual situation.
[0115] In step S1411, the second machine learning model includes, but is not limited to, random forest, XGBoost, and multi-layer perceptron neural network algorithms. Here, the data from these two sources, namely the spectral descriptors and element composition descriptors, are integrated and used as the input of the second machine learning model.
[0116] In step S150, the environmental impact assessment report of the target mixed microplastic is output in text form, aiming to help understand the potential environmental impact of microplastics and related impact parameters.
[0117] Among them, the environmental impact assessment report includes, but is not limited to, the types of microplastics, size distribution, degradability, degradation time required, ecological toxicity, and mobility, etc. The report generation can be processed using an assessment model, and its training process is as follows:
[0118] Based on the potential environmental impact data of microplastics, construct an assessment database;
[0119] Construct a third machine learning model;
[0120] Call the theoretical type information and theoretical prediction numerical data of the mixed microplastics as the input of the model, and use the collected environmental impact data of the microplastics as the output of the model; finally, obtain the assessment model.
[0121] The method provided in the embodiment of the present application has at least the following beneficial effects:
[0122] This method first obtains the target spectral data and target element composition data of the target mixed microplastics, then extracts the target spectral descriptors that can characterize the target spectral data and the target element composition descriptors that can characterize the target element composition data, and then through multiple target spectral descriptors, identifies the types of microplastics in the target mixed microplastics, and determines several target spectral descriptors whose importance for identifying the types of microplastics information is higher than the first threshold, for subsequent prediction of the proportion of each microplastic in the target mixed microplastics based on several target spectral descriptors and target element composition descriptors. Compared with the scheme of simply predicting the proportion of microplastics based on the target element composition descriptors, several target spectral descriptors with higher importance in identifying the types of microplastics information are selected here. Such target spectral descriptors can play an important decisive role in the identification of the prediction value. The accuracy of the prediction value for identifying the proportion of microplastics can be improved through the selected target spectral descriptors and target element composition descriptors; finally, an environmental impact assessment report of the target mixed microplastics is generated based on the types of microplastics information and the prediction value to better understand the impact of the target mixed microplastics on the environment.
[0123] Such as Figure 2 and Figure 3 , in some embodiments of the present invention, a prediction analysis method for mixed microplastics is provided, including steps S910 to S940:
[0124] Step S910, data acquisition and data preprocessing.
[0125] Obtain the original input data of the target mixed microplastics, and the original input data includes initial spectral data and initial element composition data;
[0126] Perform data preprocessing on the input data to obtain target spectral data and target elemental composition data, and then extract target spectral descriptors and target elemental composition descriptors.
[0127] Step S920, the classification process of the classification model.
[0128] Input the target spectral descriptors into a preset classification module to obtain the type information of the target mixed microplastics output by it, and at the same time output several of the most important target spectral descriptors based on feature importance analysis.
[0129] The training process of the classification model includes:
[0130] Construct a machine learning model;
[0131] Construct a spectral database: Calculate the spectral images of each microplastic based on quantum theory, including constructing the spectral images of single microplastic types using the standard curve method; constructing the spectral images of mixed microplastics using image superposition technology, performing data augmentation on the data of the spectral images through data enhancement methods, and processing the spectral image data through data preprocessing to obtain theoretical spectral descriptors, and forming a spectral database with all the theoretical spectral descriptors.
[0132] Use the theoretical spectral descriptors in the spectral database as the input of the machine learning model, and use the microplastic type data as the output of the machine learning model;
[0133] Train the optimal parameters of the machine learning model to obtain the classification model.
[0134] The feature importance of the spectral output is as Figure 2 shown.
[0135] Step S930, the prediction process of the prediction model;
[0136] Input the filtered target spectral descriptors and target elemental composition descriptors into a pre-constructed prediction model to obtain the predicted values of the target mixed microplastics output by it.
[0137] The training process of the prediction model includes:
[0138] Construct a machine learning model;
[0139] Construct a prediction database: Correlate the spectral descriptors with the elemental composition descriptors one by one; Based on the elemental information and molecular formula information of the mixed microplastics, calculate all possible mixing methods of all mixed microplastics to obtain the theoretical prediction values and their corresponding theoretical elemental compositions of all possible combinations of the mixed microplastics; Use the theoretical elemental compositions and their corresponding predicted spectral descriptors as the input that can be called by the prediction database, and use the theoretical prediction values as the output that can be called by the prediction database.
[0140] Call the theoretical element composition descriptor and spectral descriptor of the prediction database as the input of the machine learning model, and use the theoretical prediction value as the output of the machine learning model;
[0141] Train the optimal parameters of the machine learning model to obtain a prediction model.
[0142] The effect of the prediction model is as Figure 3 shown.
[0143] Step S940, the evaluation model of the target mixed microplastics.
[0144] Input the type information and prediction value of the target mixed microplastics into the pre-constructed evaluation model to obtain an environmental impact assessment report of the target mixed microplastics.
[0145] As Figure 4 shown, an embodiment of the present application provides a prediction analysis system for mixed microplastics, and the system includes:
[0146] The data acquisition module 1100 is used to acquire the target spectral data and target element composition data of the target mixed microplastics; the target mixed microplastics are composed of multiple types of microplastics, the target spectral data includes the spectral values of the target mixed microplastics in any spectral mode, and the target element composition data includes the content values of the mixed elements in the target mixed microplastics and the corresponding chemical molecular formulas;
[0147] The descriptor acquisition module 1200 is used to extract multiple target spectral descriptors from the target spectral data, and extract the target element composition descriptor from the target element composition data; the target spectral descriptors include the target spectral features and their values, and the target element composition descriptors include the target element composition features and their values;
[0148] The classification module 1300 is used to identify the type information of the microplastics in the target mixed microplastics according to multiple target spectral descriptors, and select several target spectral descriptors whose importance for identifying the type information of the microplastics is higher than the first threshold from the multiple target spectral descriptors;
[0149] The prediction module 1400 is used to identify the predicted values of the proportions of the respective microplastics in the target mixed microplastics according to several target spectral descriptors and the target element composition descriptor;
[0150] The environmental assessment module 1500 is used to generate an environmental impact assessment report of the target mixed microplastics according to the microplastic type information and the prediction value.
[0151] It should be noted that the prediction and analysis system of mixed microplastics provided in this embodiment and the above-mentioned prediction and analysis method of mixed microplastics are based on the same inventive concept. Therefore, the relevant content of the above-mentioned prediction and analysis method of mixed microplastics also applies to the content of the prediction and analysis system of mixed microplastics. Therefore, it will not be elaborated here.
[0152] As Figure 5 , this embodiment of the present application also provides an electronic device, which includes:
[0153] At least one memory;
[0154] At least one processor;
[0155] At least one program;
[0156] The program is stored in the memory, and the processor executes at least one program to implement the above-mentioned prediction and analysis method of mixed microplastics in the present disclosure.
[0157] The electronic device can be any intelligent terminal including a mobile phone, a tablet computer, a personal digital assistant (PDA), an in-vehicle computer, etc.
[0158] The following will introduce the electronic device in this embodiment of the present application in detail.
[0159] The processor 1600 can be implemented by using a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, etc., and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present invention;
[0160] The memory 1700 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM), etc. The memory 1700 can store an operating system and other application programs. When implementing the technical solutions provided in the embodiments of this specification through software or firmware, the relevant program codes are stored in the memory 1700 and are called by the processor 1600 to execute the prediction and analysis method of mixed microplastics in the embodiments of the present invention.
[0161] The input / output interface 1800 is used to implement information input and output;
[0162] A communication interface 1900 is used to implement communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.);
[0163] A bus 2000 transmits information between various components of the device (such as a processor 1600, a memory 1700, an input / output interface 1800, and a communication interface 1900);
[0164] Among them, the processor 1600, the memory 1700, the input / output interface 1800, and the communication interface 1900 achieve communication connections with each other inside the device through the bus 2000.
[0165] An embodiment of the present invention also provides a storage medium. This storage medium is a computer-readable storage medium, and this computer-readable storage medium stores computer-executable instructions. These computer-executable instructions are used to cause a computer to execute the above-mentioned prediction analysis method for hybrid microplastics.
[0166] As a non-transitory computer-readable storage medium, the memory can be used to store non-transitory software programs and non-transitory computer-executable programs. In addition, the memory can include high-speed random access memory, and can also include non-transitory memory, such as at least one disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some embodiments, the memory can optionally include a memory that is remotely set relative to the processor, and these remote memories can be connected to the processor through a network. Examples of the above-mentioned network include but are not limited to the Internet, an enterprise intranet, a local area network, a mobile communication network, and combinations thereof.
[0167] The embodiments described in the present invention are for more clearly illustrating the technical solutions of the embodiments of the present invention, and do not constitute a limitation on the technical solutions provided by the embodiments of the present invention. Those skilled in the art know that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of the present invention are equally applicable to similar technical problems.
[0168] Those skilled in the art can understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present invention, and may include more or fewer steps than those shown, or combine certain steps, or different steps.
[0169] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0170] Those of ordinary skill in the art will understand that all or some of the steps in the methods disclosed above, and the functional modules / units in systems and devices, can be implemented as software, firmware, hardware, or a suitable combination thereof.
[0171] As used in the specification of this application and the above-mentioned drawings, the terms "first", "second", "third", "fourth", etc. (if any) are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances so that the embodiments of this application described herein can be implemented in an order different from those illustrated or described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.
[0172] It should be understood that in this application, "at least one (item)" means one or more, and "a plurality" means two or more. "And / or" is used to describe the association relationship of associated objects and indicates that there can be three relationships. For example, "A and / or B" can mean: only A exists, only B exists, and both A and B exist at the same time. Here, A and B can be singular or plural. The character " / " generally means that the associated objects before and after are in an "or" relationship. "At least one (one) of the following" or a similar expression means any combination of these items, including any combination of single items (ones) or plural items (ones). For example, at least one (one) of a, b, or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.
[0173] In several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of units is only a logical function division, and there can be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection to each other can be through some interfaces, and the indirect coupling or communication connection of devices or units can be in electrical, mechanical, or other forms.
[0174] The unit described as a separation component may or may not be physically separated. The component displayed as a unit may or may not be a physical unit, that is, it may be located in one place or distributed across multiple network units. Some or all of these units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0175] In addition, each functional unit in various embodiments of the present application can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.
[0176] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions for causing an electronic device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in various embodiments of the present application. The aforementioned storage medium includes: various media that can store programs such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs.
[0177] The above has specifically described the preferred implementation of the embodiments of the present application, but the embodiments of the present application are not limited to the above implementation manners. Those skilled in the art can also make various equivalent deformations or substitutions without departing from the spirit of the embodiments of the present application, and these equivalent deformations or substitutions are all included within the scope defined by the claims of the embodiments of the present application.
Claims
1. A method for predicting and analyzing mixed microplastics, characterized in that: The method comprises: Obtaining target spectral data and target elemental composition data of a target mixed microplastic; the target mixed microplastic is composed of a plurality of microplastics, the target spectral data includes spectral values of the target mixed microplastic under any spectral mode, and the target elemental composition data includes mixed monomer element content values and corresponding chemical molecular formulas in the target mixed microplastic; Extracting a plurality of target spectrum descriptors from the target spectrum data, and extracting a target element composition descriptor from the target element composition data; the target spectrum descriptors include target spectrum features and their values, and the target element composition descriptors include target element composition features and their values; According to the multiple target spectral descriptors, the microplastic type information in the target mixed microplastics is identified, and from the multiple target spectral descriptors, a number of target spectral descriptors whose importance for identifying the microplastic type information is higher than a first threshold are selected; Identify predicted values of the proportions of each microplastic in the target mixed microplastic according to the plurality of target spectral descriptors and the target elemental composition descriptors; An environmental impact assessment report of the target mixed microplastics is generated based on the microplastic type information and the predicted value.
2. The prediction and analysis method for mixed microplastics according to claim 1, characterized in that: Before obtaining the target spectrum data and target element composition data of the target mixed microplastics, the method further includes: Obtaining initial spectral data and initial elemental composition data of the target mixed microplastics; Performing a first preprocessing on the initial spectral data to obtain the target spectral data; the first preprocessing includes at least one of peak feature extraction, area feature extraction and principal component analysis; The initial element composition data is subjected to a second preprocessing to obtain the target element composition data; the second preprocessing includes at least one of missing value processing, outlier processing and data standardization.
3. The prediction and analysis method of mixed microplastics according to claim 1, characterized in that: Identifying information about the types of microplastics in the target mixed microplastics according to the multiple target spectral descriptors includes: Inputting the multiple target spectral descriptors into a preset classification model to identify the type information of microplastics in the target mixed microplastics through the classification model; The training process of the classification model includes the following: Constructing a first machine learning model and constructing a spectral database; the spectral database contains a plurality of preset theoretical spectral descriptors and their corresponding theoretical microplastic types; Inputting a plurality of theoretical spectral descriptors in the spectral database into a first machine learning model, and causing the first machine learning model to output corresponding microplastic types; The first machine learning model is optimized according to the output corresponding microplastic types and the theoretical microplastic types in the spectral database until the classification model is obtained.
4. The prediction and analysis method for mixed microplastics according to claim 3, characterized in that: The constructing of the spectrum database comprises: Based on quantum theory, multiple spectral images of single microplastics were constructed; Overlaying the spectral images of the plurality of single microplastics to construct a spectral image of mixed microplastics; Generating a theoretical spectral descriptor according to the mixed microplastic spectral image, and determining the type of microplastic corresponding to the theoretical spectral descriptor; The spectral database is constructed based on the theoretical spectral descriptors and the corresponding microplastic types.
5. The prediction and analysis method for mixed microplastics according to claim 3, characterized in that: From the plurality of target spectral descriptors, a plurality of target spectral descriptors whose importance for identifying the type information of the microplastics is higher than a first threshold are selected, including: Analyzing the importance of the multiple target spectral descriptors to the microplastic type information output by the classification model according to a feature importance analysis method; sorting the plurality of target spectral descriptors according to the importance to obtain a sorting result; A plurality of target spectral descriptors whose importance is higher than the first threshold are selected according to the ranking result.
6. The prediction and analysis method for mixed microplastics according to claim 1, characterized in that: According to the plurality of target spectral descriptors and the target elemental composition descriptors, a predicted value of the proportion of each microplastic in the target mixed microplastic is identified, including: Inputting the target spectral descriptors and the target elemental composition descriptors into a preset prediction model to obtain a predicted value of the proportion of each microplastic in the target mixed microplastic identified by the prediction model; The training process of the prediction model includes the following: Constructing a second machine learning model and constructing an element composition database; the element composition database contains multiple sets of input data and output data, each set of input data includes theoretical element composition descriptors and spectral descriptors corresponding to all microplastic combinations of a mixed microplastic, and the output data is a theoretical predicted value; Inputting multiple groups of input data in the element composition database into a second machine learning model, and causing the second machine learning model to output corresponding prediction data; The second machine learning model is optimized according to the output prediction data and the output data in the element composition database until the prediction model is obtained.
7. The prediction and analysis method for mixed microplastics according to claim 6, characterized in that: The building blocks constitute a database, including: Determine the elements and molecular formulas corresponding to all microplastic combinations of mixed microplastics; Calculate theoretical element composition descriptors and their corresponding theoretical predicted values according to the elements and molecular formulas corresponding to all microplastic combinations of the mixed microplastics; determining theoretical spectral descriptors corresponding to theoretical elemental composition descriptors; Selecting theoretical element composition descriptors and theoretical spectrum descriptors with corresponding relationships as a set of input data, and taking theoretical predicted values with corresponding relationships with the theoretical element descriptors as output data; The element composition database is constructed according to the input data and the output data.
8. A prediction and analysis system for mixed microplastics, characterized in that: The system comprises: A data acquisition module, used to acquire target spectral data and target elemental composition data of a target mixed microplastic; the target mixed microplastic is composed of a plurality of microplastics, the target spectral data includes the spectral value of the target mixed microplastic under any spectral mode, and the target elemental composition data includes the content value of the mixed monomer elements in the target mixed microplastic and the corresponding chemical molecular formula; A descriptor acquisition module, used to extract a plurality of target spectrum descriptors from the target spectrum data, and to extract a target element composition descriptor from the target element composition data; the target spectrum descriptor includes a target spectrum feature and its value, and the target element composition descriptor includes a target element composition feature and its value; A classification module, for identifying the type information of microplastics in the target mixed microplastics according to the multiple target spectral descriptors, and selecting a number of target spectral descriptors whose importance for identifying the type information of the microplastics is higher than a first threshold from the multiple target spectral descriptors; A prediction module, for identifying a predicted value of the proportion of each microplastic in the target mixed microplastic according to the plurality of target spectral descriptors and the target elemental composition descriptors; The environmental assessment module is used to generate an environmental impact assessment report of the target mixed microplastics based on the microplastic type information and the predicted value.
9. An electronic device, characterized in that: It includes at least one control processor and a memory for communicating with the at least one control processor; the memory stores instructions that can be executed by the at least one control processor, and the instructions are executed by the at least one control processor so that the at least one control processor can perform the predictive analysis method for mixed microplastics as described in any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the predictive analysis method for mixed microplastics according to any one of claims 1 to 7.