Method, system, equipment and medium for detecting soluble volatile pollutants in water
Through the combination of GC-IMS technology and multimodal pre-training neural network, the spectrum characteristics of solubility volatile pollutants in water bodies are obtained, and the problems of low accuracy, high cost and low efficiency of existing detection methods are solved, and the problem of low accuracy, high cost and low efficiency of existing detection methods is achieved. Fast and efficient detection of volatile pollutants for water solubility.
Patent Information
- Application Number
- CN202411118533.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-15
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2044-08-15
AI Technical Summary
The detection results of solubility volatile pollutants in existing water bodies have low accuracy, high detection cost and low detection efficiency, which cannot meet the needs of large-scale water bodies for testing.
GC-IMS technology is used to obtain saturated steam concentration and spectrum characteristics of dissolved volatile pollutants, and combined with multimodal pre-trained neural network to build a detection model, and detect it through text and image features fusion to achieve fast and efficient pollutant identification. .
It improves the detection accuracy, reduces the detection cost and operation complexity, and meets the needs of large-scale water body inspection.
Smart Images

Figure CN118641675B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of water body detection, and in particular to a method, system, equipment and medium for detecting soluble volatile pollutants in water. Background Art
[0002] In recent years, domestic water odor incidents have occurred frequently. Water odor has become an important factor affecting the quality of human life, just like air pollution. With the development of urbanization and industrial production, the source and quality of domestic drinking water have further deteriorated. The water purification process of the tap water treatment plant is still mainly based on conventional treatment, and the main removal targets are clay particles and pathogenic microorganisms. The removal capacity of soluble volatile pollutants is limited. There are many types of soluble volatile pollutants, and the removal methods are different. If you want to quickly and efficiently remove soluble volatile pollutants, you need to clarify the types of soluble volatile pollutants in order to select targeted removal methods.
[0003] The existing detection of soluble volatile pollutants in water bodies generally adopts sensory analysis, quantitative analysis, semi-quantitative analysis and non-target screening. Sensory analysis relies on the personal ability of the evaluator, and the accuracy of the test results is low. Quantitative analysis, semi-quantitative analysis and non-target screening need to rely on gas chromatography-mass spectrometry (GC-MS), which has cumbersome operation process, high detection cost and low detection efficiency, and cannot meet the needs of large-scale water body detection.
[0004] It can be seen that how to solve the problems of low detection accuracy, high detection cost and low detection efficiency in the detection of soluble volatile pollutants in water bodies, as well as the inability to meet the needs of large-scale water body detection has become a technical problem that technical personnel in this field need to solve urgently. Summary of the invention
[0005] The present invention provides a method, system, equipment and medium for detecting soluble volatile pollutants in water, so as to solve the technical problems of low detection accuracy, high detection cost and low detection efficiency in the detection of soluble volatile pollutants in water, and the inability to meet the needs of large-scale water body detection, so as to realize simple and efficient detection of soluble volatile pollutants in water and improve the detection accuracy.
[0006] In a first aspect, the present invention provides a method for detecting soluble volatile pollutants in water, the method comprising:
[0007] Obtain the saturated vapor concentration of typical dissolved volatile pollutants in water at a specific temperature;
[0008] Using GC-IMS technology to obtain a first GC-IMS spectrum of the soluble volatile pollutant in a saturated vapor pressure state at a specific temperature, and using text to describe the first GC-IMS spectrum to obtain text features of the first GC-IMS spectrum;
[0009] Constructing a water body detection model, and training the water body detection model based on the saturated vapor concentration corresponding to the soluble volatile pollutant, the first GC-IMS spectrum, and the text feature;
[0010] In the actual detection process, different specific times are set so that the water body to be detected is at the specific temperature and specific atmospheric pressure, and the GC-IMS technology is used to obtain the second GC-IMS spectra corresponding to different specific times;
[0011] The second GC-IMS spectrum is input into the trained water body detection model to obtain the first type result and the first type concentration of the soluble volatile pollutants in the water body to be detected.
[0012] Preferably, the water body detection model includes a text feature extraction module, an image feature extraction module, a feature fusion module and a multimodal task processing module;
[0013] The text feature extraction module is used to extract the first feature vector of the text feature;
[0014] The image feature extraction module is used to extract a second feature vector of the first GC-IMS spectrum;
[0015] The feature fusion module is used to fuse the first feature vector and the second feature vector to obtain a fused feature vector;
[0016] The multimodal task processing module is used to jointly learn the fused feature vector and output a first type result and a first type concentration of soluble volatile pollutants present in the water body to be detected.
[0017] Preferably, the water body detection model is constructed using a multimodal pre-trained neural network.
[0018] Preferably, the method further comprises:
[0019] Uploading the second GC-IMS spectrum to the cloud for big data analysis, and extracting pollution attribute features corresponding to each of the second GC-IMS spectra;
[0020] Based on the pollution attribute characteristics, screening and comparison are performed in the historical database to obtain a second type result and a second type concentration;
[0021] The first category result is corrected with the second category result, and the first category concentration is corrected with the second category concentration.
[0022] Preferably, the method for obtaining the specific atmospheric pressure includes:
[0023] The saturated vapor pressure of the soluble volatile pollutant at a specific temperature is obtained, and the peak critical value of the specific atmospheric pressure is determined according to the minimum value of all the saturated vapor pressures.
[0024] Preferably, the method further comprises:
[0025] Establishing a mapping relationship between the specific atmospheric pressure, the corresponding specific time, the first type of result and the first type of concentration to construct a first detection result data set;
[0026] Performing an interpretability analysis on the first detection result data set, and correcting the first detection result data set according to the interpretability analysis result to obtain a corrected second detection result data set;
[0027] The interpretability analysis method includes:
[0028] Based on the relationship between the saturated vapor pressure of the soluble volatile pollutant and the concentration of the first type, performing an interpretability analysis on the first detection result data set; or,
[0029] Based on the variation trend of the concentration of the first type of the soluble volatile pollutants at different specific times, performing an interpretability analysis on the first detection result data set; or,
[0030] An interpretability analysis is performed on the first detection result data set based on the source of the water body to be detected.
[0031] Preferably, the method further comprises:
[0032] A detection log is generated based on the first type result and the first type concentration, and the detection log is sent to an environmental protection terminal, a water treatment personnel mobile terminal, and a user query terminal respectively.
[0033] In a second aspect, the present invention further provides a system for detecting soluble volatile pollutants in water, the system comprising: a saturated vapor concentration acquisition unit, a first GC-IMS spectrum acquisition unit, a model building training unit, a detection data acquisition unit and a detection unit;
[0034] The saturated steam concentration acquisition unit is used to obtain the saturated steam concentration of typical soluble volatile pollutants in the water body at a specific temperature;
[0035] The first GC-IMS spectrum acquisition unit is used to use GC-IMS technology to obtain a first GC-IMS spectrum of the soluble volatile pollutant in a saturated vapor pressure state at a specific temperature, and use text to describe the first GC-IMS spectrum to obtain text features of the first GC-IMS spectrum;
[0036] The model building training unit is used to build a water body detection model, and train the water body detection model based on the saturated vapor concentration corresponding to the soluble volatile pollutant, the first GC-IMS spectrum and the text feature;
[0037] The detection data acquisition unit is used to set different specific times during the actual detection process, so that the water body to be detected is at the specific temperature and specific atmospheric pressure, and use the GC-IMS technology to obtain the second GC-IMS spectra corresponding to different specific times;
[0038] The detection unit is used to input the second GC-IMS spectrum into the trained water body detection model to obtain the first type result and the first type concentration of the soluble volatile pollutants in the water body to be detected.
[0039] In a third aspect, the present invention also provides a computer device, comprising a memory, a processor and a transceiver, which are connected via a bus; the memory is used to store a set of computer program instructions and data, and transmit the stored data to the processor, and the processor executes the program instructions stored in the memory to execute the above-mentioned method for detecting soluble volatile pollutants in water.
[0040] In a fourth aspect, the present invention further provides a computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium. When the computer program is executed, the above-mentioned method for detecting soluble volatile pollutants in water is implemented.
[0041] The present invention provides a method, system, device and medium for detecting soluble volatile pollutants in water. Compared with the prior art, the beneficial effects of the embodiments of the present invention are at least one of the following:
[0042] (1) Simple and efficient operation. All spectra are obtained through GC-IMS technology, combined with multimodal pre-trained neural networks. No large-scale instruments or equipment are required, and the operation process is simple and efficient.
[0043] (2) Detection accuracy: In the model training stage, the first GC-IMS spectrum under the saturated vapor pressure state is used, which has more obvious image features. In the actual detection process, the peak critical value of a specific atmospheric pressure is determined according to the minimum value of the saturated vapor pressure of all soluble volatile pollutants at a specific temperature, which can ensure the smooth overflow of all soluble volatile pollutants. At this time, the concentration of soluble volatile pollutants is also relatively large, and the second GC-IMS spectrum corresponding to more soluble volatile pollutants can be captured, as well as more significant image features, further improving the accuracy of water body detection. BRIEF DESCRIPTION OF THE DRAWINGS
[0044] Figure 1 It is a schematic diagram of the steps of a method for detecting soluble volatile pollutants in water provided by a preferred embodiment of the present invention;
[0045] Figure 2 is an example of a GC-IMS spectrum provided by a preferred embodiment of the present invention;
[0046] Figure 3 It is a schematic diagram of the steps of a method for correcting a detection result provided by a preferred embodiment of the present invention;
[0047] Figure 4 It is a schematic diagram of the method steps for performing interpretable analysis of the detection results provided by a preferred embodiment of the present invention;
[0048] Figure 5 It is a schematic diagram of the structure of a water-soluble volatile pollutant detection system provided by a preferred embodiment of the present invention;
[0049] Figure 6 It is a schematic diagram of a computer device provided by a preferred embodiment of the present invention. DETAILED DESCRIPTION
[0050] The following is a detailed explanation of the implementation mode of the present invention in conjunction with the accompanying drawings. The embodiments are provided only for illustrative purposes and cannot be understood as limiting the present invention. The accompanying drawings are only for reference and illustration purposes and do not constitute a limitation on the scope of patent protection of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. In the description of the present invention, the terms "first", "second", "third", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first", "second", "third", etc. may explicitly or implicitly include one or more of the features. In the description of the present invention, unless otherwise specified, the meaning of "multiple" is two or more.
[0051] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected" and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be a connection between the two elements. The terms "vertical", "horizontal", "left", "right", "upper", "lower" and similar expressions used herein are only for illustrative purposes, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.
[0052] In the description of the present invention, it should be noted that, unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as those commonly understood by those skilled in the art. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood by specific circumstances.
[0053] See also Figure 1 In an embodiment of the present invention, a method for detecting soluble volatile pollutants in water is provided, the method comprising:
[0054] S1. Obtain the saturated vapor concentration of typical soluble volatile pollutants in water at a specific temperature.
[0055] S2. Using GC-IMS technology to obtain a first GC-IMS spectrum of the soluble volatile pollutant in a saturated vapor pressure state at a specific temperature, and using text to describe the first GC-IMS spectrum to obtain text features of the first GC-IMS spectrum.
[0056] S3. Construct a water body detection model, and train the water body detection model based on the saturated vapor concentration corresponding to the soluble volatile pollutants, the first GC-IMS spectrum, and the text features.
[0057] S4. In the actual detection process, different specific times are set so that the water body to be detected is at the specific temperature and specific atmospheric pressure, and the GC-IMS technology is used to obtain the second GC-IMS spectra corresponding to different specific times.
[0058] S5. Input the second GC-IMS spectrum into the trained water body detection model to obtain the first type result and the first type concentration of the soluble volatile pollutants in the water body to be detected.
[0059] According to the database of odorous chemicals recorded in "Water Odor Chemicals: Categories, Sources, Analysis Methods and Control" (Liu Zehua, Tan Qifeng, Dang Zhi and Yi Hua, Science Press, March 2019), the types of typical soluble volatile pollutants in water bodies are obtained, and the saturated vapor pressure of each typical soluble volatile pollutant at a specific temperature is obtained. In a preferred embodiment of the present invention, the specific temperature is 25°C. The normal temperature of 25°C has relatively low environmental requirements during the detection process. At this time, the overflow of soluble volatile pollutants in water bodies is also faster, and more significant features can be captured.
[0060] Based on the saturated vapor pressure of the soluble volatile pollutant at a specific temperature, the saturated vapor concentration of the corresponding soluble volatile pollutant at a specific temperature can be calculated. The saturated vapor concentration at a specific temperature is a purely physical calculation and will not be described in detail in the embodiments of the present invention.
[0061] In a preferred embodiment of the present invention, the saturated vapor concentration of typical soluble volatile pollutants in water at a specific temperature is used as sample data, which can be obtained only through physical calculations. There is no need to rely on conventional measurement methods to obtain the concentration data of soluble volatile pollutants, and the operation process is simple.
[0062] In a preferred embodiment of the present invention, the GC-IMS technology is used to obtain the first GC-IMS spectrum of each soluble volatile pollutant in a saturated vapor pressure state at a specific temperature. Different from the conventional water body detection using gas chromatography / mass spectrometry, the detection process is complicated and the detection efficiency is low. The present invention uses GC-IMS technology to obtain the first GC-IMS spectrum of each soluble volatile pollutant in a saturated vapor pressure state at a specific temperature. GC-IMS is a gas chromatography ion mobility spectrometer, which is an instrument that combines a rapid separation gas chromatograph with a high-sensitivity ion mobility spectrometer, such as Figure 2 The following is an example of a GC-IMS spectrum. The red vertical line at 0 on the horizontal axis is the reaction ion peak (RIP peak), which has been normalized. The vertical axis represents the retention time of the gas chromatography, and the horizontal axis represents the ion migration time, which has been normalized. Each point on both sides of the reaction ion peak represents a volatile organic compound. The color represents the concentration of the substance, white indicates a lower concentration, red indicates a higher concentration, and the darker the color, the greater the concentration.
[0063] Furthermore, the first GC-IMS spectrum is described by text to obtain the text features of the first GC-IMS spectrum. The text features can be as follows: the red vertical line at the horizontal axis 0 is the reaction ion peak, the vertical axis represents the retention time of the gas chromatography, the horizontal axis represents the ion migration time, and each point on both sides of the reaction ion peak represents a volatile organic compound. The color represents the concentration of the substance, white indicates a lower concentration, red indicates a higher concentration, and the darker the color, the greater the concentration. The specific presentation form of the text features can be changed according to the needs to make it more in line with the detection needs.
[0064] In the preferred real-time example of the present invention, a water body detection model is constructed based on a multimodal pre-trained neural network. Different modal information is expressed in different ways, and the angles of reflecting the characteristics of things are also different. In the preferred embodiment of the present invention, a multimodal pre-trained neural network that can process both text features and image features is selected to construct a water body detection model. While the features constrain each other, more abundant feature information can be obtained to improve the accuracy of the detection results of soluble volatile pollutants in water bodies.
[0065] In a preferred embodiment of the present invention, the water body detection model includes a text feature extraction module, an image feature extraction module, a feature fusion module and a multimodal task processing module. The text feature extraction module is used to extract the first feature vector of the text feature; the image feature extraction module is used to extract the second feature vector of the first GC-IMS spectrum; the feature fusion module is used to fuse the first feature vector and the second feature vector to obtain a fused feature vector; the multimodal task processing module is used to jointly learn the fused feature vector and output the first type result and the first type concentration of the soluble volatile pollutants in the water body to be detected.
[0066] In a preferred embodiment of the present invention, the water body detection model is constructed using the CLIP pre-training model. The CLIP pre-training model is a typical multimodal pre-training neural network, which uses a self-supervised learning method that does not require a large amount of artificially labeled data. Based on a contrastive learning module, images and texts can be jointly encoded to achieve cross-retrieval between images and texts. CLIP will encode the text features and the first GC-IMS spectrum into a first feature vector and a second feature vector, respectively, and learn the matching relationship between the spectrum and the text by comparing the similarity of the first feature vector and the second feature vector. In the CLIP model, the text feature extraction module converts text features into a first feature vector and uses a deep learning model to capture the semantic information of the language; the image feature extraction module converts image features into a second feature vector and uses a convolutional neural network to capture visual content; the feature fusion module adjusts the text feature extraction module and the image feature extraction module so that they fuse the first feature vector and the second feature vector in the same high-dimensional feature space and output a fused feature vector, thereby aligning the first feature vector and the second feature vector in the same space and ensuring that the text features and the first GC-IMS spectrum can be effectively compared in the same multidimensional space; the multimodal task processing module identifies the first type of results and the first type of concentration of soluble volatile pollutants in the corresponding water body to be tested through joint learning of the fused feature vector.
[0067] Compared with the traditional classification and recognition model, the water body detection model of the present invention can simultaneously recognize and fuse the text features and the first GC-IMS spectrum, realize the complementarity and constraint of features, and improve the accuracy of recognition. In addition, the water body detection model of the present invention does not require a large amount of manual labeling, is simple to operate, and saves human resources.
[0068] Furthermore, the water body detection model is trained to construct a training data set using the saturated vapor concentration corresponding to the soluble volatile pollutants, the first GC-IMS spectrum, and the text features. The training data set is divided into a training set, a test set, and a validation set in a ratio of 7:2:1. The text features are input into this feature extraction module, and the first GC-IMS spectrum is input into the image feature extraction module, so that the text features and the first GC-IMS spectrum are mutually constrained and feature complementarity is achieved. The water body detection model is trained using the types and saturated vapor concentrations corresponding to the soluble volatile pollutants as outputs to obtain a trained water body detection model.
[0069] After the water body detection model is built and trained, in the actual detection process, it is necessary to set the environment of the water body to be detected, which includes temperature requirements and atmospheric pressure requirements. Among them, the temperature requirement is that the temperature of the water body to be detected remains unchanged at a specific temperature. The atmospheric pressure requirement is less than the minimum value of the saturated vapor pressure of all soluble volatile pollutants at a specific temperature. Therefore, the method for obtaining a specific atmospheric pressure includes: obtaining the saturated vapor pressure of soluble volatile pollutants at a specific temperature, and determining the peak critical value of the specific atmospheric pressure based on the minimum value of all saturated vapor pressures.
[0070] In the actual detection process, it is also necessary to set different specific times and obtain the corresponding second GC-IMS spectra after different specific times to obtain multiple different detection results for further interpretable analysis. For example, if the specific time is 5 minutes, 10 minutes and 15 minutes, interpretable analysis is performed on the corresponding detection results after 5 minutes, 10 minutes and 15 minutes to screen the detection results and obtain more accurate detection results.
[0071] In a preferred embodiment of the present invention, the temperature of the water body to be detected during the actual detection process is kept constant at a specific temperature consistent with the training set obtained during the model training phase, thereby reducing changes in the environment during the detection process and making the operation simpler.
[0072] In the actual detection process, GC-IMS technology is used to obtain the second GC-IMS spectrum corresponding to different specific times, and the second GC-IMS spectrum is input into the trained water body detection model to obtain the first type result and the first type concentration of the soluble volatile pollutants in the water body to be detected.
[0073] Vapor pressure refers to the presence of vapor of the substance on the surface of a liquid or solid, and the pressure exerted by these vapors on the surface of the liquid. It reflects the ability of molecules of an object to escape from the surface of an object. Saturated vapor pressure refers to the size of the vapor pressure when the equilibrium state is reached. After the temperature is specified, the greater the difference between the atmospheric pressure and the saturated vapor pressure, the greater the evaporation rate. The GC-IMS spectra of soluble volatile pollutants under saturated vapor pressure have more obvious image features, so in a preferred embodiment of the present invention, the first GC-IMS spectra under saturated vapor pressure are used in the model training stage. In the actual detection process, the atmospheric pressure is set below the minimum value of the saturated vapor pressure of all soluble volatile pollutants at a specific temperature, which can ensure the smooth overflow of all soluble volatile pollutants. At this time, the concentration of soluble volatile pollutants is also relatively large, and more soluble volatile pollutants can be captured. The second GC-IMS spectra corresponding to the pollutants and more significant spectra features can be captured to improve the accuracy of water body detection.
[0074] In a preferred embodiment of the present invention, Figure 3 As shown, the method for detecting soluble volatile pollutants in water provided by the present invention also includes:
[0075] S01. Upload the second GC-IMS spectrum to the cloud for big data analysis, and extract the pollution attribute characteristics corresponding to each of the second GC-IMS spectra.
[0076] S02. Screening and comparing are performed in a historical database based on the pollution attribute characteristics to obtain a second type result and a second type concentration.
[0077] S03. Correct the first type result with the second type result, and correct the first type concentration with the second type concentration.
[0078] In a preferred embodiment of the present invention, the second GC-IMS spectrum can be uploaded to the cloud, and features of each second GC-IMS spectrum can be extracted in the cloud to obtain corresponding pollution attribute features, which can be screened and compared with the historical database pre-built based on the first GC-IMS spectrum to obtain the second category result and the second category concentration, and the first category result can be corrected with the second category result, and the first category concentration can be corrected with the second category concentration. The final water body detection result can be accurately determined by the features.
[0079] In a preferred embodiment of the present invention, Figure 4 As shown, the method for detecting soluble volatile pollutants in water provided by the present invention also includes:
[0080] S001. Establish a mapping relationship between the specific atmospheric pressure, the corresponding specific time, the first type of result and the first type of concentration to construct a first detection result data set.
[0081] S002, performing an explainability analysis on the first detection result data set, and correcting the first detection result data set according to the explainability analysis result to obtain a corrected second detection result data set;
[0082] The interpretability analysis method includes:
[0083] Based on the relationship between the saturated vapor pressure of the soluble volatile pollutant and the concentration of the first type, performing an interpretability analysis on the first detection result data set; or,
[0084] Based on the variation trend of the concentration of the first type of the soluble volatile pollutants at different specific times, performing an interpretability analysis on the first detection result data set; or,
[0085] An interpretability analysis is performed on the first detection result data set based on the source of the water body to be detected.
[0086] At a specific temperature and a specific atmospheric pressure, the difference between the saturated vapor pressure and the atmospheric pressure of the soluble volatile pollutants with low saturated vapor pressure is small, the overflow speed is suppressed, and the overflow of the soluble volatile pollutants with high saturated vapor pressure is faster than that of the soluble volatile pollutants with low saturated vapor pressure. Then, the content of the soluble volatile pollutants with high saturated vapor pressure in the mixed vapor of the water body to be detected is relatively high, and the corresponding characteristics in the GC-IMS spectrum are more obvious, and the concentration is relatively high. According to this feature, the interpretability analysis can be carried out in the following way: based on the relationship between the saturated vapor pressure of the soluble volatile pollutants and the concentration of the first type, the interpretability analysis of the first test result data set is carried out. Specifically, it is determined whether the concentration of the soluble volatile pollutants with high saturated vapor pressure in the first test result data set is greater than the concentration of the soluble volatile pollutants with low saturated vapor pressure. If not, the interpretability analysis result is unreasonable, and the corresponding data may have problems, and the problematic data will be deleted. Because the first test result data set of the present invention includes multiple mapping information, deleting one of them will not affect the final test result.
[0087] In addition, the interpretability analysis method also includes: based on the change trend of the first type concentration of the soluble volatile pollutants at different specific times, the interpretability analysis of the first detection result data set is performed. Specifically: as the specific time changes, the longer the specific time is, the concentration of the first type of soluble volatile pollutants in the mixed steam of the water body to be detected increases. If the first detection result data set does not conform to this rule, it indicates that the interpretability analysis result of the detection result is unreasonable, and the corresponding detection result needs to be deleted.
[0088] Furthermore, the interpretability analysis method also includes: performing interpretability analysis on the first detection result based on the source of the water body to be detected. Specifically, after a water pollution incident occurs, a preliminary investigation should be conducted on the scene to record the surrounding conditions, as shown in Table 1, which is a field sampling information record table.
[0089] Table 1
[0090]
[0091] The types of soluble volatile pollutants produced by different pollution sources also have certain regular characteristics. Table 2 shows the types of soluble volatile pollutants produced by different pollution sources.
[0092] Table 2
[0093]
[0094] An interpretability analysis is performed on the first test result data set in combination with Table 1 and Table 2. If the types of soluble volatile pollutants present in the water body to be tested are consistent with Table 1 and Table 2, the interpretability analysis result is reasonable, indicating that the first test result data set is accurate; if the types of soluble volatile pollutants present in the water body to be tested are inconsistent with Table 1 and Table 2, the interpretability analysis result is unreasonable, indicating that the first test result data set is inaccurate, and the problematic data can be deleted or retested.
[0095] Through interpretable analysis, inaccurate first test results can be screened out, which can further improve the accuracy of the test results.
[0096] In a preferred embodiment of the present invention, after obtaining the first type result and the first type concentration, a detection log is generated based on the first type result and the first type concentration, and the detection log is sent to the environmental protection terminal, the water treatment personnel mobile terminal and the user query terminal respectively. The environmental protection terminal is a department with supervisory responsibilities for water quality, such as the Environmental Protection Bureau. The water treatment personnel are the staff who perform subsequent treatment on the water body. The user query terminal is the mobile terminal of each drinking water user.
[0097] In a preferred embodiment of the present invention, by obtaining the saturated vapor concentration of typical soluble volatile pollutants in water at a specific temperature, using GC-IMS technology to obtain the first GC-IMS spectrum of soluble volatile pollutants in a saturated vapor pressure state at a specific temperature, and using text to describe the first GC-IMS spectrum, a text feature of the first GC-IMS spectrum is obtained, a water body detection model is constructed based on a multimodal pre-trained neural network, and the water body detection model is trained based on the saturated vapor concentration, the first GC-IMS spectrum and the text feature. In the actual detection process, different specific times are set so that the water body to be detected is at a specific temperature and a specific atmospheric pressure, and the GC-IMS technology is used to obtain the corresponding second GC-IMS spectrum after different specific times, and the second spectrum is input into the trained water body detection model to obtain the first type result and the first type concentration of the pollutants in the water body to be detected. The method for detecting soluble volatile pollutants in water provided by the present invention combines GC-IMS technology with multimodal pre-trained neural networks, does not require large-scale instruments and equipment and cumbersome operating procedures, and fuses text features and image features to obtain richer feature vectors and improve the accuracy of detection.
[0098] Accordingly, if Figure 5As shown, based on a method for detecting volatile pollutants soluble in water, an embodiment of the present invention further provides a system for detecting volatile pollutants soluble in water, comprising: a saturated steam concentration acquisition unit 1, a first GC-IMS spectrum acquisition unit 2, a model building and training unit 3, a detection data acquisition unit 4 and a detection unit 5;
[0099] The saturated steam concentration acquisition unit 1 is used to acquire the saturated steam concentration of typical soluble volatile pollutants in the water body at a specific temperature.
[0100] The first GC-IMS spectrum acquisition unit 2 is used to use GC-IMS technology to obtain a first GC-IMS spectrum of the soluble volatile pollutants in a saturated vapor pressure state at a specific temperature, and use text to describe the first GC-IMS spectrum to obtain text features of the first GC-IMS spectrum.
[0101] The model building training unit 3 is used to build a water body detection model, and train the water body detection model based on the saturated vapor concentration corresponding to the soluble volatile pollutants, the first GC-IMS spectrum and the text features.
[0102] The detection data acquisition unit 4 is used to set different specific times during the actual detection process, so that the water body to be detected is at a specific temperature and a specific atmospheric pressure, and use the GC-IMS technology to obtain the second GC-IMS spectra corresponding to different specific times.
[0103] The detection unit 5 is used to input the second GC-IMS spectrum into the trained water body detection model to obtain the first type result and the first type concentration of the soluble volatile pollutants in the water body to be detected.
[0104] For the specific definition of a water-soluble volatile pollutant detection system, please refer to the above-mentioned definition of a water-soluble volatile pollutant detection method, which will not be repeated here. A person of ordinary skill in the art can appreciate that the various modules and steps described in conjunction with the embodiments disclosed in the present invention can be implemented in hardware, software, or a combination of both. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.
[0105] like Figure 6As shown, a computer device provided in an embodiment of the present invention includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the steps in the above-mentioned method for detecting soluble volatile pollutants in water are implemented, for example Figure 1 Steps S1 to S5 described in .
[0106] Those skilled in the art will understand that the Figure 6 It is only an example of a computer device and does not constitute a limitation of the computer device. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the computer device may also include input and output devices, network access devices, buses, etc.
[0107] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the computer device, and uses various interfaces and lines to connect various parts of the entire computer device.
[0108] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the computer device by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.
[0109] Wherein, if the module integrated in the computer device 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 such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium can include: any entity or device that can carry the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal and software distribution medium, etc.
[0110] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.
[0111] Accordingly, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to perform the steps in the method for detecting soluble volatile pollutants in water bodies as in the above embodiment, for example Figure 1 Steps S1 to S5 described in .
[0112] A method, system, computer device and storage medium for detecting soluble volatile pollutants in water provided in this embodiment are aimed at the technical problems of low accuracy of detection results, high detection cost and low detection efficiency in detecting soluble volatile pollutants in water, and the inability to meet the needs of large-scale water body detection. The present invention obtains the saturated vapor concentration of typical soluble volatile pollutants in water at a specific temperature, uses GC-IMS technology to obtain a first GC-IMS spectrum of soluble volatile pollutants in a saturated vapor pressure state at a specific temperature, and uses text to describe the first GC-IMS spectrum to obtain text features of the first GC-IMS spectrum, constructs a water body detection model based on a multimodal pre-trained neural network, trains the water body detection model based on saturated vapor concentration, the first GC-IMS spectrum and text features, sets different specific times in the actual detection process, makes the water body to be detected at a specific temperature and a specific atmospheric pressure, uses GC-IMS technology to obtain the corresponding second GC-IMS spectrum after different specific times, inputs the second spectrum into the trained water body detection model, and obtains the first type result and the first type concentration of the pollutants in the water body to be detected. The method for detecting soluble volatile pollutants in water provided by the present invention combines GC-IMS technology with a multimodal pre-trained neural network. It does not require large-scale instruments and equipment and cumbersome operating procedures. It fuses text features and image features to obtain richer feature vectors and improve the accuracy of detection.
[0113] The above-mentioned embodiments only express several preferred implementation modes of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this technical field, several improvements and substitutions can be made without departing from the technical principle of the present invention, and these improvements and substitutions should also be regarded as the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be based on the protection scope of the claims.
Claims
1. A method for detecting soluble volatile pollutants in water, characterized in that: The method comprises: Obtain the saturated vapor concentration of typical dissolved volatile pollutants in water at a specific temperature; Using GC-IMS technology to obtain a first GC-IMS spectrum of the soluble volatile pollutant in a saturated vapor pressure state at a specific temperature, and using text to describe the first GC-IMS spectrum to obtain text features of the first GC-IMS spectrum; Constructing a water body detection model, and training the water body detection model based on the saturated vapor concentration corresponding to the soluble volatile pollutant, the first GC-IMS spectrum, and the text feature; In the actual detection process, different specific times are set so that the water body to be detected is at the specific temperature and specific atmospheric pressure, and the GC-IMS technology is used to obtain the second GC-IMS spectra corresponding to different specific times; Inputting the second GC-IMS spectrum into the trained water body detection model to obtain the first type result and the first type concentration of the soluble volatile pollutants in the water body to be detected; The water body detection model includes a text feature extraction module, an image feature extraction module, a feature fusion module and a multimodal task processing module; The text feature extraction module is used to extract the first feature vector of the text feature; The image feature extraction module is used to extract a second feature vector of the first GC-IMS spectrum; The feature fusion module is used to fuse the first feature vector and the second feature vector to obtain a fused feature vector; The multimodal task processing module is used to jointly learn the fused feature vector and output a first type result and a first type concentration of soluble volatile pollutants present in the water body to be detected.
2. The method for detecting soluble volatile pollutants in water as claimed in claim 1, characterized in that: The water body detection model is constructed using a multimodal pre-trained neural network.
3. The method for detecting soluble volatile pollutants in water as claimed in claim 1, characterized in that: The method further comprises: Uploading the second GC-IMS spectra to the cloud for big data analysis, and extracting pollution attribute features corresponding to each of the second GC-IMS spectra; Based on the pollution attribute characteristics, screening and comparison are performed in the historical database to obtain a second type result and a second type concentration; The first category result is corrected with the second category result, and the first category concentration is corrected with the second category concentration.
4. The method for detecting soluble volatile pollutants in water as claimed in claim 1, characterized in that: The method for obtaining the specific atmospheric pressure comprises: The saturated vapor pressure of the soluble volatile pollutant at a specific temperature is obtained, and the peak critical value of the specific atmospheric pressure is determined according to the minimum value of all the saturated vapor pressures.
5. The method for detecting soluble volatile pollutants in water as claimed in claim 1, characterized in that: The method further comprises: Establishing a mapping relationship between the specific atmospheric pressure, the corresponding specific time, the first type of result and the first type of concentration to construct a first detection result data set; Performing an interpretability analysis on the first detection result data set, and correcting the first detection result data set according to the interpretability analysis result to obtain a corrected second detection result data set; The interpretability analysis method includes: Based on the relationship between the saturated vapor pressure of the soluble volatile pollutant and the concentration of the first type, performing an interpretability analysis on the first detection result data set; or, Based on the variation trend of the concentration of the first type of the soluble volatile pollutants at different specific times, performing an interpretability analysis on the first detection result data set; or, An interpretability analysis is performed on the first detection result data set based on the source of the water body to be detected.
6. The method for detecting soluble volatile pollutants in water according to claim 1, characterized in that: The method further comprises: A detection log is generated based on the first type result and the first type concentration, and the detection log is sent to an environmental protection terminal, a water treatment personnel mobile terminal, and a user query terminal respectively.
7. A water-soluble volatile pollutant detection system, characterized in that: The system comprises: a saturated steam concentration acquisition unit, a first GC-IMS spectrum acquisition unit, a model building training unit, a detection data acquisition unit and a detection unit; The saturated steam concentration acquisition unit is used to obtain the saturated steam concentration of typical soluble volatile pollutants in the water body at a specific temperature; The first GC-IMS spectrum acquisition unit is used to use the GC-IMS technology to acquire a first GC-IMS spectrum of the soluble volatile pollutant in a saturated vapor pressure state at a specific temperature, and use text to describe the first GC-IMS spectrum to obtain text features of the first GC-IMS spectrum; The model building training unit is used to build a water body detection model, and train the water body detection model based on the saturated vapor concentration corresponding to the soluble volatile pollutant, the first GC-IMS spectrum and the text feature; The detection data acquisition unit is used to set different specific times during the actual detection process, so that the water body to be detected is at the specific temperature and specific atmospheric pressure, and use the GC-IMS technology to obtain the second GC-IMS spectra corresponding to different specific times; The detection unit is used to input the second GC-IMS spectrum into the trained water body detection model to obtain the first type result and the first type concentration of the soluble volatile pollutants in the water body to be detected; The water body detection model includes a text feature extraction module, an image feature extraction module, a feature fusion module and a multimodal task processing module; The text feature extraction module is used to extract the first feature vector of the text feature; The image feature extraction module is used to extract a second feature vector of the first GC-IMS spectrum; The feature fusion module is used to fuse the first feature vector and the second feature vector to obtain a fused feature vector; The multimodal task processing module is used to jointly learn the fused feature vector and output a first type result and a first type concentration of soluble volatile pollutants present in the water body to be detected.
8. A computer device, characterized in that: The computer device includes a memory, a processor and a transceiver, which are connected via a bus; the memory is used to store a set of computer program instructions and data, and transmit the stored data to the processor, and the processor executes the program instructions stored in the memory to execute the method for detecting soluble volatile pollutants in water as described in any one of claims 1 to 6.
9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed, the method for detecting soluble volatile pollutants in water as claimed in any one of claims 1 to 6 is implemented.
