Cooking fume analysis system and method based on CNN model

Abstract

The invention provides an oil fume analysis system and method based on a CNN model, and relates to the technical field of oil fume analysis, and the system comprises a data acquisition module which obtains oil fume emission data at a preset catering site and surrounding environment data of the preset catering site, and outputs initial data; the list construction module is used for carrying out feature extraction and feature analysis on the initial data and constructing a cooking fume analysis list based on a feature analysis result; the component analysis module is used for performing component quantitative analysis on the lampblack analysis list and outputting a component quantitative analysis result in combination with statistical characteristic data corresponding to each component; and the correlation analysis module is used for performing environment correlation analysis on the component quantitative analysis result based on a preset analysis model in combination with the surrounding environment data, and outputting an oil smoke-environment correlation analysis result. According to the method, the analysis precision and efficiency of the cooking fume components can be improved, the influence of the cooking fume components on the surrounding environment is evaluated, and the evaluation efficiency and quality of the cooking fume are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of oil fume analysis, and in particular to an oil fume analysis system and method based on a CNN model. Background Art

[0002] In recent years, restaurant fume pollution has been one of the most popular air pollution issues, and has remained high for a long time. Restaurant fume has become another major air pollution source after industrial waste gas and automobile exhaust. On the other hand, there are a huge number of catering companies, which are scattered, disorderly and miscellaneous. It is impossible to achieve full coverage by human supervision, and relying solely on on-site inspections by supervisors can no longer meet today's environmental protection requirements.

[0003] At present, traditional analysis methods and technologies for restaurant fumes have limitations and are unable to accurately analyze and evaluate the complex fume components and their impact on the surrounding environment.

[0004] Therefore, the present invention provides an oil fume analysis system and method based on a CNN model. Summary of the invention

[0005] The present invention provides a fume analysis system and method based on a CNN model, which are used to improve the analysis accuracy and efficiency of restaurant fume components, evaluate the impact of fume components on the surrounding environment, and improve the evaluation efficiency and quality of restaurant fume.

[0006] On the one hand, the present invention provides a fume analysis system based on a CNN model, comprising: A data acquisition module, used to acquire the oil fume emission data at a preset dining point and the surrounding environment data of the preset dining point, and output initial data; A list construction module, used for performing feature extraction and feature analysis on the initial data based on a preset algorithm, and constructing a fume analysis list based on the feature analysis results; A component analysis module, used to perform component quantitative analysis on the oil fume analysis list using a preset component analysis method, and output component quantitative analysis results in combination with statistical characteristic data corresponding to each component; The correlation analysis module is used to combine the surrounding environment data and perform environmental correlation analysis on the component quantitative analysis results based on a preset analysis model, and output the oil smoke-environment correlation analysis results.

[0007] Preferably, the data acquisition module includes: The emission data acquisition submodule is used to capture the emission data at each preset catering point and output the oil fume emission data; An environmental data acquisition submodule is used to acquire environmental data within a preset range at each of the preset dining spots and output surrounding environmental data; The data summary submodule is used to summarize the oil fume emission data and surrounding environment data at each preset catering point and output the initial data.

[0008] Preferably, the emission data acquisition submodule includes: The emission data at the preset catering points in each main catering system are captured to obtain the sub-emission data corresponding to each system, and the oil fume emission data is output, wherein the main catering system includes a commercial catering system, a home cooking system and a canteen catering system.

[0009] Preferably, it also includes: A method selection module is used to perform noise recognition on the oil fume emission data and the surrounding environment data in combination with the noise database before outputting the initial data, and select a noise reduction method corresponding to each noise type from the method database based on the recognition result, and output a noise-method comparison table; The noise reduction module is used to perform noise reduction processing on the oil fume emission data and the surrounding environment data based on the noise-method comparison table, and summarize and output them as initial data.

[0010] Preferably, the list building module includes: A feature extraction submodule, used to extract features from the initial data using a preset feature extraction algorithm, and construct a fume feature set and an environmental data feature set based on the extracted data features; The feature analysis submodule is used to perform feature analysis on the oil fume feature set and the environmental data feature set using a preset analysis algorithm, and obtain a first list corresponding to the VOCs source data and a second list corresponding to the VOCs component emission data, and summarize them to obtain an oil fume analysis list.

[0011] Preferably, the component analysis module comprises: The demand acquisition submodule is used to obtain the user's analysis requirements for oil fume component analysis and output the analysis requirement data; A method selection submodule, for selecting a component analysis method matching the second list from a method database based on the analysis requirement data, and outputting the method as a preset component analysis method; A component analysis submodule, for analyzing the emission data of each VOCs component in the second list based on the preset component analysis method, and outputting a first analysis result in combination with the chemical characteristics and physical characteristics corresponding to each VOCs component; An active feature matching submodule, for performing feature matching on the emission data of each VOCs component in the second list based on the first analysis result and in combination with a preset active feature database, and outputting the VOCs components that meet the preset matching conditions and the emission data under the corresponding VOCs components as an active feature-data comparison table, and at the same time, outputting the VOCs components that do not meet the preset matching conditions in the second list and the emission data under the corresponding VOCs components as an inactive feature-data comparison table; The spatiotemporal data submodule is used to obtain the timestamp and positioning beacon corresponding to each data in the active feature-data comparison table and the inactive feature-data comparison table, and output real-time spatiotemporal data; A historical data acquisition submodule, used to acquire, from a historical database, first historical data that matches the active feature-data comparison table, and second historical data that matches the inactive feature-data comparison table; A spatiotemporal analysis submodule is used to combine the first historical data and the second historical data to respectively perform spatiotemporal analysis on the data in the active feature-data comparison table and the inactive feature-data comparison table, determine the spatiotemporal distribution law corresponding to each VOCs active component and each VOCs inactive component in the table, and summarize and output the spatiotemporal analysis results, wherein the spatiotemporal distribution law includes the indoor spatiotemporal variation law and the outdoor spatiotemporal variation law of each component; The result output submodule is used to output the component quantitative analysis results based on the statistical feature data corresponding to each component in the active feature-data comparison table and the inactive feature-data comparison table and in combination with the spatiotemporal analysis results.

[0012] Preferably, the correlation analysis module includes: The emission characteristic analysis submodule is used to analyze the emission characteristics of each VOCs component in the component quantitative analysis result and output the emission characteristic analysis result; The emission evolution characteristic analysis submodule is used to analyze the emission evolution characteristics of each VOCs component in the component quantitative analysis result and output the evolution characteristic analysis result; The source-driving factor analysis submodule is used to perform source tracing analysis on the component quantitative analysis results using a preset source analysis method, determine the pollution sources and driving factors of each VOCs component emission, and output the source-driving factor analysis results; The influencing factor analysis submodule is used to obtain other source data related to VOCs component emissions, and perform correlation analysis with the component quantitative analysis results to determine the influencing factors of various types of factors in the other source data on the emissions of each VOCs component, and output the analysis results of other source influencing factors; An atmospheric activity feature matching submodule is used to identify the atmospheric activity features of the surrounding environment data and extract data, and output an atmospheric activity feature analysis table; The model analysis submodule is used to use a preset analysis model to comprehensively analyze the emission characteristic analysis results, evolution characteristic analysis results, source-driving factor analysis results, other source influencing factor analysis results and gas activity characteristic analysis table, determine the correlation between each VOCs component in the component quantitative analysis results and the evolution of the surrounding environment, and output the oil smoke-environment correlation analysis results; The algorithm used to analyze the impact between the VOCs in the component quantitative analysis results and the surrounding atmospheric environment is as follows: ; in, Represents the correlation influence coefficient of the i-th VOCs component in the component quantitative analysis result on the surrounding atmospheric environment; represents the exponential function; Represents the adjustment coefficient corresponding to the i-th VOCs component co-existing in the exhaust gas and the atmosphere; It represents the weight coefficient corresponding to the potential of the i-th VOCs component to generate ozone in the atmosphere when calculating the correlation influence coefficient; It represents the weight coefficient corresponding to the contribution of the i-th VOCs component to the formation of secondary organic aerosol in the atmosphere when calculating the correlation influence coefficient; represents the reaction rate constant corresponding to the i-th VOCs component, which is used to characterize the reaction rate between the i-th VOCs component and NOx; Represents the average concentration of the i-th VOCs component co-existing in the exhaust gas and the atmosphere; represents the average concentration of nitrogen oxides corresponding to the i-th VOCs component in the atmosphere; It represents the impact value of the concentration of the i-th VOCs component on the ozone generation potential in the atmosphere; It represents the contribution factor of other factors corresponding to the i-th VOCs component to the formation of secondary organic aerosol; It represents the influence index of the i-th VOCs component in flue gas and atmosphere on the formation of secondary organic aerosol; An index representing the impact of nitrogen oxides on the formation of secondary organic aerosols; It represents the contribution of the concentration of the i-th VOCs component to the formation of secondary organic aerosol in the atmosphere; The emission reduction analysis submodule is used to optimize the oil smoke-environment related analysis results, determine the optimizable items for emission reduction, and output optimization suggestions corresponding to each of the optimizable items to obtain emission reduction analysis results.

[0013] On the other hand, the present invention also provides a fume analysis method based on a CNN model, comprising: Step 1: Obtaining the fume emission data at a preset catering point and the surrounding environment data of the preset catering point, and outputting the initial data; Step 2: extracting and analyzing the features of the initial data based on a preset algorithm, and constructing a fume analysis list based on the feature analysis results; Step 3: Perform component quantitative analysis on the oil fume analysis list using a preset component analysis method, and output component quantitative analysis results in combination with statistical characteristic data corresponding to each component; Step 4: Combine the surrounding environment data and perform environmental correlation analysis on the component quantitative analysis results based on a preset analysis model, and output the oil smoke-environment correlation analysis results.

[0014] The present invention provides a fume analysis system and method based on the CNN model, which can accurately identify the content and characteristics of different components through the synergy of data acquisition, list construction, component analysis and correlation analysis modules, provide reliable data support for subsequent environmental correlation analysis, comprehensively analyze the fume emission data, and reveal the correlation between the fume and the surrounding environment, providing important support for environmental protection decision-making and promoting the reduction of the impact of fume emissions on the environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 Schematic diagram of a CNN-based oil smoke analysis system provided by an embodiment of the present invention; Figure 2 It is a flow chart of a fume analysis method based on a CNN model provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] like Figure 1As shown, an oil smoke analysis system based on a CNN model provided by an embodiment of the present invention includes: A data acquisition module, used to acquire the oil fume emission data at the preset catering point and the surrounding environment data of the preset catering point, and output the initial data; A list building module is used to extract and analyze features of the initial data based on a preset algorithm, and to build a fume analysis list based on the feature analysis results; The component analysis module is used to perform component quantitative analysis on the oil fume analysis list using a preset component analysis method, and output the component quantitative analysis results in combination with the statistical characteristic data corresponding to each component; The correlation analysis module is used to combine the surrounding environment data and perform environmental correlation analysis on the component quantitative analysis results based on the preset analysis model, and output the oil smoke-environment correlation analysis results.

[0019] In this embodiment, preset dining point: a predetermined dining place or location for monitoring and analyzing oil fume emissions; In this embodiment, the oil fume emission data: oil fume emission related data collected at the preset catering point, including information such as oil fume composition, concentration, and emission amount; In this embodiment, surrounding environment data: data related to the surrounding environment of the preset dining point, such as meteorological data, air quality data, etc.; In this embodiment, initial data: raw data output from the data acquisition module, including oil fume emission data and surrounding environment data; In this embodiment, the preset algorithm is a predetermined algorithm or method used to extract and analyze features of the initial data; In this embodiment, the oil fume analysis list: the oil fume analysis list constructed according to the characteristic analysis results, which may include oil fume components, characteristics, etc.; In this embodiment, the preset component analysis method: a pre-set method for quantitatively analyzing the components in the oil fume analysis list; In this embodiment, statistical characteristic data: statistical data related to oil smoke components, such as mean value, standard deviation, etc.; In this embodiment, component quantitative analysis results: quantitative analysis results of each component in the oil fume analysis list; In this embodiment, the preset analysis model: a preset analysis model is used to perform environmental relevance analysis on the component quantitative analysis results; In this embodiment, environmental correlation analysis: analyzing the correlation between the oil smoke components and the surrounding environment data according to a preset analysis model; In this embodiment, the oil fume-environment correlation analysis result refers to the correlation analysis result between the oil fume components and the environmental data.

[0020] The implementation principle and beneficial effects of this embodiment: Through the synergistic effect of data acquisition, inventory construction, component analysis and correlation analysis modules, the present invention can accurately identify the content and characteristics of different components, provide reliable data support for subsequent environmental correlation analysis, comprehensively analyze oil fume emission data, and reveal the correlation between oil fume and the surrounding environment, providing important support for environmental protection decision-making and promoting the reduction of the impact of oil fume emissions on the environment.

[0021] The embodiment of the present invention provides a fume analysis system based on a CNN model, and a data acquisition module, comprising: The emission data acquisition submodule is used to capture the emission data at each preset catering point and output the oil fume emission data; An environmental data acquisition submodule is used to acquire environmental data within a preset range at each preset dining point and output surrounding environmental data; The data summary submodule is used to summarize the oil fume emission data and surrounding environment data at each preset catering point and output the initial data.

[0022] In this embodiment, emission data: the fume emission data generated by each preset catering point, such as the concentration of volatile organic compounds (VOCs), the concentration of particulate matter, etc. Specifically, the fume emission data of restaurant A shows that the VOCs concentration is 50 ppm and the particulate matter concentration is 100 μg / m³; In this embodiment, preset range: the range for obtaining environmental data, for example, environmental data within a preset radius centered at a preset dining spot; In this embodiment, environmental data: surrounding environmental data, including atmospheric composition, atmospheric concentration, wind direction, wind speed, temperature, humidity, etc. Specifically, the surrounding environmental data of restaurant B shows that the wind direction is north wind, the wind speed is 2 m / s, the temperature is 25°C, and the humidity is 60%.

[0023] The implementation principle and beneficial effects of this embodiment: The present invention obtains the oil fume emission data and surrounding environment data of the catering point through the data acquisition module, and then summarizes and inputs these data into the subsequent analysis module, and uses the CNN model to perform oil fume analysis and environmental correlation analysis. The present invention improves the efficiency of obtaining and analyzing oil fume emission data, helps to more comprehensively understand the impact of oil fume emissions on the surrounding environment, and provides a scientific basis for subsequent environmental management and protection.

[0024] An oil fume analysis system based on a CNN model provided by an embodiment of the present invention includes an emission data acquisition submodule, including: The emission data at the preset catering points in each main catering system are captured to obtain the sub-emission data corresponding to each system, and the oil fume emission data is output. Among them, the main catering system includes commercial catering system, home cooking system and canteen catering system.

[0025] In this embodiment, sub-emission data: that is, emission data corresponding to each system. For example, the sub-emission data of a fast food restaurant in the commercial catering system shows that the VOCs concentration is 70 ppm and the particulate matter concentration is 150 μg / m³; In this embodiment, commercial catering system refers to the fume emissions of commercial catering service institutions, such as restaurants, fast food restaurants, etc. In this embodiment, the household cooking system refers to the fume emissions generated by cooking activities in the household; In this embodiment, the canteen catering system refers to the fume emissions generated by places that provide catering services within institutions such as schools and enterprises.

[0026] The implementation principle and beneficial effects of this embodiment: The present invention aims at catering points in different main catering systems, and by capturing emission data and outputting oil fume emission data, it can achieve the acquisition and analysis of emission data of different types of catering, which is convenient for the subsequent use of CNN model to conduct in-depth oil fume analysis and environmental correlation analysis. The present invention can more accurately understand the situation of oil fume emissions in different main catering systems, and provide data support for the formulation of targeted environmental management and protection measures.

[0027] An oil fume analysis system based on a CNN model provided by an embodiment of the present invention further includes: The method selection module is used to identify the noise of the oil fume emission data and the surrounding environment data in combination with the noise database before outputting the initial data, and select the noise reduction method corresponding to each noise type from the method database based on the identification result, and output the noise-method comparison table; The noise reduction module is used to perform noise reduction on the oil fume emission data and the surrounding environment data based on the noise-method comparison table, and summarize and output them as initial data.

[0028] In this embodiment, the noise database: stores data of various noise types, and is used to identify noise in oil fume emission data and surrounding environment data; In this embodiment, noise identification: analyzing the oil fume emission data and the surrounding environment data to identify various types of noise present therein; In this embodiment, the method database: a database storing noise reduction methods corresponding to various noise types, used to reduce the impact of noise on data; In this embodiment, the noise reduction method: a noise reduction processing method adopted for different noise types to improve the accuracy and reliability of data; In this embodiment, the noise-method comparison table: according to the noise identification result, the noise reduction method corresponding to each noise type is selected to form a comparison table for subsequent processing; In this embodiment, noise reduction processing: applying the selected noise reduction method to process the oil fume emission data and the surrounding environment data to remove noise interference and improve the quality and reliability of the data. Assume that in the oil fume analysis system, through noise identification it is found that some oil fume emission data is interfered by traffic noise. Subsequently, a noise reduction method suitable for reducing the impact of traffic noise is selected according to the method database, and then this method is applied to process the affected data, thereby removing the impact of traffic noise and improving the accuracy of the data.

[0029] The implementation principle and beneficial effects of this embodiment: The present invention combines the noise database to perform noise identification, selects a noise reduction method and generates a noise-method comparison table, and then performs noise reduction processing on the oil fume emission data and the surrounding environment data to improve the accuracy and reliability of the data. The present invention can effectively reduce the impact of noise on data analysis and model training, and improve the performance of the system and the credibility of the results.

[0030] The embodiment of the present invention provides a fume analysis system based on a CNN model, and a list building module, including: The feature extraction submodule is used to extract features from the initial data using a preset feature extraction algorithm, and to construct a fume feature set and an environmental data feature set based on the extracted data features; The feature analysis submodule is used to perform feature analysis on the oil fume feature set and the environmental data feature set using a preset analysis algorithm, and obtain a first list corresponding to the VOCs source data and a second list corresponding to the VOCs component emission data, and summarize them to obtain an oil fume analysis list.

[0031] In this embodiment, the preset feature extraction algorithm: a preset algorithm for extracting features from raw data; In this embodiment, the oil smoke feature set: a feature set extracted from the oil smoke data; In this embodiment, the environmental data feature set: a feature set extracted from the environmental data; In this embodiment, the preset analysis algorithm: an algorithm for analyzing the feature set is preset; In this embodiment, feature analysis: analyzing the feature set to obtain a list of VOCs source data and VOCs component emission data; In this embodiment, VOCs source data: that is, data on the sources of VOCs, for example, including cooking fumes, cooking waste gas, detergent volatiles, and food volatiles; In this embodiment, the first list: a list or a list containing VOCs source data; In this embodiment, VOCs component emission data: that is, emission data of each VOCs component, for example, formaldehyde, benzene, aldehyde compounds, etc.; In this embodiment, the second inventory: an inventory or list containing VOCs component emission data.

[0032] Implementation principle and beneficial effects of this embodiment: The present invention extracts features from initial data through the feature extraction submodule, and then analyzes the feature set through the feature analysis submodule to obtain a fume analysis list. The present invention can better understand the features of fume data and environmental data, thereby more accurately analyzing fume emissions.

[0033] The embodiment of the present invention provides a fume analysis system based on a CNN model, and a component analysis module, comprising: The demand acquisition submodule is used to obtain the user's analysis requirements for oil fume component analysis and output the analysis requirement data; A method selection submodule is used to select a component analysis method matching the second list from a method database based on the analysis requirement data, and output it as a preset component analysis method; The component analysis submodule is used to analyze the emission data of each VOCs component in the second list based on a preset component analysis method, and output a first analysis result in combination with the chemical characteristics and physical characteristics corresponding to each VOCs component; An active feature matching submodule is used to perform feature matching on the emission data of each VOCs component in the second list based on the first analysis result and in combination with a preset active feature database, and output the VOCs components that meet the preset matching conditions and the emission data under the corresponding VOCs components as an active feature-data comparison table, and at the same time, output the VOCs components that do not meet the preset matching conditions in the second list and the emission data under the corresponding VOCs components as an inactive feature-data comparison table; The spatiotemporal data submodule is used to obtain the timestamp and positioning beacon corresponding to each data in the active feature-data comparison table and the inactive feature-data comparison table, and output real-time spatiotemporal data; A historical data acquisition submodule, used to acquire, from a historical database, first historical data that matches the active feature-data comparison table, and second historical data that matches the inactive feature-data comparison table; The spatiotemporal analysis submodule is used to combine the first historical data and the second historical data to perform spatiotemporal analysis on the data in the active feature-data comparison table and the inactive feature-data comparison table, respectively, to determine the spatiotemporal distribution law corresponding to each VOCs active component and each VOCs inactive component in the table, and to summarize and output the spatiotemporal analysis results, wherein the spatiotemporal distribution law includes the indoor spatiotemporal variation law and the outdoor spatiotemporal variation law of each component; The result output submodule is used to output the component quantitative analysis results based on the statistical feature data corresponding to each component in the active feature-data comparison table and the inactive feature-data comparison table, and in combination with the spatiotemporal analysis results.

[0034] In this embodiment, analysis requirements: that is, the user's requirements for oil fume component analysis, such as resolution accuracy, precision rate, etc.; In this embodiment, the analysis demand data: the output user demand data describes the user's specific requirements for the fume component analysis; In this embodiment, component analysis method: a specific method or algorithm for analyzing oil fume components selected according to analysis requirements and data characteristics, for example, a mass spectrometry analysis method is used to identify and quantify the concentration of VOCs components such as styrene and acetaldehyde; In this embodiment, chemical characteristics: chemical structural characteristics of oil smoke components, such as molecular structure, chemical bonds, etc.; In this embodiment, physical characteristics: physical properties of the oil smoke components, such as density, boiling point, solubility, etc. For example, the physical characteristics of acetaldehyde include its low boiling point and pungent odor, which helps the system to analyze and identify the component; In this embodiment, the first analysis result is the result obtained by analyzing the VOCs component emission data based on a preset component analysis method. For example, after analysis by mass spectrometry, the emission concentration of styrene is 100 μg / m³; In this embodiment, an active characteristic database is preset: a database containing chemical characteristics and physical characteristics of active components in various VOCs, for example, VOCs components and their characteristics that are irritating to the human respiratory system. Specifically, formaldehyde is a common VOCs, which has a high content in oil smoke, has a pungent odor, and is irritating to the respiratory tract. Long-term exposure may cause discomfort to the eyes, nose and throat, and even cause allergic reactions; In this embodiment, feature matching: a method for determining active components and inactive components in smoke; In this embodiment, the preset matching conditions: the characteristic matching conditions for determining whether each component in the smoke is active or inactive are preset; In this embodiment, the activity characteristic-data comparison table: a comparison table including active components and data corresponding to the active components; In this embodiment, the inactive feature-data comparison table: a comparison table including inactive components and data corresponding to the inactive components; In this embodiment, timestamp: the exact time when the oil fume emission data is recorded by the system, usually expressed in a certain standard format, such as Unix timestamp (in seconds) or date and time format; In this embodiment, the positioning beacon is used to determine the source location of the oil smoke emission data, for example, GPS coordinates or a geocode of a specific location; In this embodiment, real-time spatiotemporal data: real-time data containing time and space information, used to describe the emission of oil smoke at different times and locations, including the emission concentration of each VOCs component at different times and locations, and the timestamps and positioning beacons associated therewith; In this embodiment, the historical database is a database for storing oil fume emission data in the past time period, which is used for searching and analyzing the historical data. For example, the historical database may contain records of emission data of various VOCs components in the past year, which can be used for comparative analysis with current data. In this embodiment, the first historical data: historical data matched with the activity characteristic-data comparison table, used for historical analysis related to the active component; In this embodiment, the second historical data: historical data matched with the inactive feature-data comparison table, is used to perform historical analysis related to the inactive component; In this embodiment, spatiotemporal analysis: analyzes the changes in data in time and space dimensions to reveal the law of data changes over time and place. For example, the system performs spatiotemporal analysis on oil fume emission data to understand the emission of specific VOCs components at different times and places, thereby inferring its change law; In this embodiment, the temporal and spatial distribution law refers to the distribution of different components in time and space, including the indoor and outdoor change laws; In this embodiment, spatiotemporal analysis results: conclusions and summaries of the spatiotemporal analysis process, including the temporal and spatial variation patterns of each component. For example, the spatiotemporal analysis results of the system may include the emission concentration variation trend of component A in different time periods and locations, as well as the differences between indoor and outdoor emissions; In this embodiment, the indoor spatiotemporal variation law: the variation law of each component in the indoor environment in time and space, for example, the variation law of the emission concentration of component A in the kitchen over time, reveals the variation trend of the component in the indoor environment; In this embodiment, outdoor spatiotemporal variation patterns: the temporal and spatial variation patterns of various components in the outdoor environment, for example, the variation patterns of the emission concentration of component A in different areas of the city with seasons and time, reveal the temporal and spatial distribution of the component in the outdoor environment.

[0035] Implementation principle and beneficial effects of this embodiment: The present invention realizes comprehensive analysis of oil fume components and detection of spatiotemporal variation laws through the synergistic effect of submodules such as demand acquisition, method selection, component analysis, active feature matching, spatiotemporal data acquisition, historical data acquisition, spatiotemporal analysis and result output. The present invention can more comprehensively understand the component characteristics and spatiotemporal variation laws of oil fume emissions, and provide a more accurate basis for environmental protection decision-making.

[0036] The embodiment of the present invention provides a fume analysis system based on a CNN model, and a related analysis module, including: The emission characteristic analysis submodule is used to analyze the emission characteristics of each VOCs component in the component quantitative analysis results and output the emission characteristic analysis results; The emission evolution characteristic analysis submodule is used to analyze the emission evolution characteristics of each VOCs component in the component quantitative analysis results and output the evolution characteristic analysis results; The source-driving factor analysis submodule is used to trace the component quantitative analysis results using the preset source analysis method, determine the pollution sources and driving factors of each VOCs component emission, and output the source-driving factor analysis results; The influencing factor analysis submodule is used to obtain other source data related to VOCs component emissions, and perform correlation analysis with the component quantitative analysis results to determine the influencing factors of various types of factors in the other source data on the emissions of each VOCs component, and output the analysis results of other source influencing factors; The atmospheric activity feature matching submodule is used to identify the atmospheric activity features of the surrounding environment data and extract data, and output the atmospheric activity feature analysis table; The model analysis submodule is used to use the preset analysis model to comprehensively analyze the emission characteristic analysis results, evolution characteristic analysis results, source-driving factor analysis results, other source influencing factor analysis results and gas activity characteristic analysis table, determine the correlation between each VOCs component in the component quantitative analysis results and the evolution of the surrounding environment, and output the oil smoke-environment correlation analysis results; Among them, the algorithm used to analyze the impact between VOCs in the component quantitative analysis results and the surrounding atmospheric environment is as follows: ; in, It represents the correlation influence coefficient of the i-th VOCs component in the component quantitative analysis results on the surrounding atmospheric environment; represents the exponential function; Represents the adjustment coefficient corresponding to the i-th VOCs component co-existing in the exhaust gas and the atmosphere; It represents the weight coefficient corresponding to the potential of the i-th VOCs component to generate ozone in the atmosphere when calculating the correlation influence coefficient; It represents the weight coefficient corresponding to the contribution of the i-th VOCs component to the formation of secondary organic aerosol in the atmosphere when calculating the correlation influence coefficient; represents the reaction rate constant corresponding to the i-th VOCs component, which is used to characterize the reaction rate between the i-th VOCs component and NOx; Represents the average concentration of the i-th VOCs component co-existing in the exhaust gas and the atmosphere; represents the average concentration of nitrogen oxides corresponding to the i-th VOCs component in the atmosphere; It represents the impact value of the concentration of the i-th VOCs component on the ozone generation potential in the atmosphere; It represents the contribution factor of other factors corresponding to the i-th VOCs component to the formation of secondary organic aerosol; It represents the influence index of the i-th VOCs component in flue gas and atmosphere on the formation of secondary organic aerosol; An index representing the impact of nitrogen oxides on the formation of secondary organic aerosols; It represents the contribution of the concentration of the i-th VOCs component to the formation of secondary organic aerosol in the atmosphere; The emission reduction analysis submodule is used to optimize the results of the oil smoke-environment related analysis, determine the optimizable items for emission reduction, and output optimization suggestions corresponding to each optimizable item to obtain the emission reduction analysis results.

[0037] In this embodiment, emission characteristics: the characteristics of various VOCs components in the smoke during the emission process, such as emission concentration, emission rate, emission mode, etc.; In this embodiment, the emission characteristic analysis result: the result obtained after analyzing the emission characteristics of each VOCs component, including emission peak value, emission amount and other information; In this embodiment, evolution characteristics: the change characteristics of each VOCs component in the smoke during the emission process, such as the change law and trend of the emission; In this embodiment, the evolution characteristic analysis result: the result obtained after analyzing the emission evolution characteristics of each VOCs component may include information such as emission change trend and periodic change. For example, the different emission trends of component A on weekends and weekdays are one of the evolution characteristic analysis results; In this embodiment, the preset source analysis method is: a preset method or algorithm for tracing the source of emission sources of oil fume components, for example, using a chemical mass balance model to analyze the source and emission amount of styrene and acetaldehyde in oil fume; In this embodiment, source tracing analysis: a method or process of tracing and analyzing the emission sources of oil fume components to determine the pollution sources; In this embodiment, pollution source: the specific source or substance that causes the emission of VOCs in the fume; In this embodiment, driving factors: the main factors affecting the emission and evolution of oil smoke, including temperature, humidity, cooking method, etc.; In this embodiment, the source-driving factor analysis result: the analysis result of the pollution source and driving factor obtained by tracing the component quantitative analysis result according to the preset source analysis method; In this embodiment, the external source data refers to external data related to the emission of VOCs components, which is used to analyze influencing factors and correlation analysis. For example, meteorological data, cooking methods, types of ingredients, etc. can be used as external source data, which are external source factors affecting VOCs emissions; In this embodiment, correlation analysis: correlation analysis is performed on the component quantitative analysis results and other source data to determine the influence of each factor on the emission of VOCs components; In this embodiment, influencing factors refer to the degree of influence of various factors on the emission of VOCs components, which can be a single factor or a comprehensive influence of multiple factors; In this embodiment, the analysis result of the influencing factors of other sources is the result obtained after analyzing the influencing factors of various types of factors in the other source data on the emission of VOCs components; In this embodiment, atmospheric activity characteristics refer to chemical activity characteristics in the atmospheric environment, such as oxidant concentration, organic aerosol generation potential, etc.; In this embodiment, the atmospheric activity characteristic analysis table: a data table extracted after the atmospheric activity characteristic recognition of the surrounding environment data is performed, and is used to analyze the chemical activity characteristics of the atmospheric environment, for example, including data such as ozone concentration and particulate matter concentration; In this embodiment, the optimizable items are: aspects or factors that can be improved or reduced according to the analysis results; In this embodiment, optimization suggestions: specific improvement or reduction suggestions made by the system for the optimizable items, such as suggestions to reduce cooking time, change cooking methods, etc.; In this embodiment, the emission reduction analysis result refers to the emission reduction strategy and recommended results obtained after optimizing the oil smoke-environment related analysis results.

[0038] Implementation principle and beneficial effects of this embodiment: The present invention comprehensively analyzes the oil fume components through various analysis modules, including emission characteristics, evolution characteristics, source-driving factor analysis, influencing factor analysis, etc., and finally determines the correlation between oil fume and the surrounding environment, and puts forward emission reduction suggestions. The present invention can improve the accuracy and efficiency of environmental monitoring and management, provide a scientific basis for environmental protection decision-making, and promote the reduction of the impact of oil fume emissions on the environment.

[0039] like Figure 2 As shown, an oil smoke analysis method based on a CNN model provided by an embodiment of the present invention includes: Step 1: Obtain the fume emission data at the preset catering point and the surrounding environment data of the preset catering point, and output the initial data; Step 2: Extract and analyze the features of the initial data based on the preset algorithm, and construct a fume analysis list based on the feature analysis results; Step 3: Use the preset component analysis method to perform component quantitative analysis on the fume analysis list, and output the component quantitative analysis results in combination with the statistical characteristic data corresponding to each component; Step 4: Combine the surrounding environment data and perform environmental correlation analysis on the component quantitative analysis results based on the preset analysis model, and output the oil smoke-environment correlation analysis results.

[0040] The implementation principle and beneficial effects of this embodiment: Through the synergistic effect of data acquisition, inventory construction, component analysis and correlation analysis modules, the present invention can accurately identify the content and characteristics of different components, provide reliable data support for subsequent environmental correlation analysis, comprehensively analyze oil fume emission data, and reveal the correlation between oil fume and the surrounding environment, providing important support for environmental protection decision-making and promoting the reduction of the impact of oil fume emissions on the environment.

[0041] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A fume analysis system based on CNN model, characterized in that: include: A data acquisition module, used to acquire the oil fume emission data at a preset dining point and the surrounding environment data of the preset dining point, and output initial data; A list construction module, used for performing feature extraction and feature analysis on the initial data based on a preset algorithm, and constructing a fume analysis list based on the feature analysis results; A component analysis module, used to perform component quantitative analysis on the oil fume analysis list using a preset component analysis method, and output component quantitative analysis results in combination with statistical characteristic data corresponding to each component; The correlation analysis module is used to combine the surrounding environment data and perform environmental correlation analysis on the component quantitative analysis results based on a preset analysis model, and output the oil smoke-environment correlation analysis results.

2. The oil fume analysis system based on the CNN model according to claim 1 is characterized in that: The data acquisition module comprises: The emission data acquisition submodule is used to capture the emission data at each preset catering point and output the oil fume emission data; An environmental data acquisition submodule, for acquiring environmental data within a preset range at each of the preset dining spots, and outputting surrounding environmental data; The data summary submodule is used to summarize the oil fume emission data and surrounding environment data at each preset catering point and output the initial data.

3. The oil fume analysis system based on the CNN model according to claim 2 is characterized in that: The emission data acquisition submodule includes: The emission data at the preset catering points in each main catering system are captured to obtain the sub-emission data corresponding to each system, and the oil fume emission data is output, wherein the main catering system includes a commercial catering system, a home cooking system and a canteen catering system.

4. The oil fume analysis system based on the CNN model according to claim 1 is characterized in that: Also includes: A method selection module is used to perform noise recognition on the oil fume emission data and the surrounding environment data in combination with the noise database before outputting the initial data, and select a noise reduction method corresponding to each noise type from the method database based on the recognition result, and output a noise-method comparison table; The noise reduction module is used to perform noise reduction processing on the oil fume emission data and the surrounding environment data based on the noise-method comparison table, and summarize and output them as initial data.

5. The oil fume analysis system based on the CNN model according to claim 1 is characterized in that: The list building module includes: A feature extraction submodule, used to extract features from the initial data using a preset feature extraction algorithm, and construct a fume feature set and an environmental data feature set based on the extracted data features; The feature analysis submodule is used to perform feature analysis on the oil fume feature set and the environmental data feature set using a preset analysis algorithm, and obtain a first list corresponding to the VOCs source data and a second list corresponding to the VOCs component emission data, and summarize them to obtain an oil fume analysis list.

6. The oil fume analysis system based on the CNN model according to claim 5 is characterized in that: The component analysis module comprises: The demand acquisition submodule is used to obtain the user's analysis requirements for oil fume component analysis and output the analysis requirement data; A method selection submodule, for selecting a component analysis method matching the second list from a method database based on the analysis requirement data, and outputting the method as a preset component analysis method; A component analysis submodule, for analyzing the emission data of each VOCs component in the second list based on the preset component analysis method, and outputting a first analysis result in combination with the chemical characteristics and physical characteristics corresponding to each VOCs component; An active feature matching submodule, for performing feature matching on the emission data of each VOCs component in the second list based on the first analysis result and in combination with a preset active feature database, and outputting the VOCs components that meet the preset matching conditions and the emission data under the corresponding VOCs components as an active feature-data comparison table, and at the same time, outputting the VOCs components that do not meet the preset matching conditions in the second list and the emission data under the corresponding VOCs components as an inactive feature-data comparison table; The spatiotemporal data submodule is used to obtain the timestamp and positioning beacon corresponding to each data in the active feature-data comparison table and the inactive feature-data comparison table, and output real-time spatiotemporal data; A historical data acquisition submodule, used to acquire, from a historical database, first historical data that matches the active feature-data comparison table, and second historical data that matches the inactive feature-data comparison table; A spatiotemporal analysis submodule is used to combine the first historical data and the second historical data to respectively perform spatiotemporal analysis on the data in the active feature-data comparison table and the inactive feature-data comparison table, determine the spatiotemporal distribution law corresponding to each VOCs active component and each VOCs inactive component in the table, and summarize and output the spatiotemporal analysis results, wherein the spatiotemporal distribution law includes the indoor spatiotemporal variation law and the outdoor spatiotemporal variation law of each component; The result output submodule is used to output the component quantitative analysis results based on the statistical feature data corresponding to each component in the active feature-data comparison table and the inactive feature-data comparison table and in combination with the spatiotemporal analysis results.

7. The oil fume analysis system based on the CNN model according to claim 1 is characterized in that: The related analysis module, include: The emission characteristic analysis submodule is used to analyze the emission characteristics of each VOCs component in the component quantitative analysis result and output the emission characteristic analysis result; The emission evolution characteristic analysis submodule is used to analyze the emission evolution characteristics of each VOCs component in the component quantitative analysis result and output the evolution characteristic analysis result; The source-driving factor analysis submodule is used to perform source tracing analysis on the component quantitative analysis results using a preset source analysis method, determine the pollution sources and driving factors of each VOCs component emission, and output the source-driving factor analysis results; The influencing factor analysis submodule is used to obtain other source data related to VOCs component emissions, and perform correlation analysis with the component quantitative analysis results to determine the influencing factors of various types of factors in the other source data on the emissions of each VOCs component, and output the analysis results of other source influencing factors; An atmospheric activity feature matching submodule is used to identify the atmospheric activity features of the surrounding environment data and extract data, and output an atmospheric activity feature analysis table; The model analysis submodule is used to use a preset analysis model to comprehensively analyze the emission characteristic analysis results, evolution characteristic analysis results, source-driving factor analysis results, other source influencing factor analysis results and gas activity characteristic analysis table, determine the correlation between each VOCs component in the component quantitative analysis results and the evolution of the surrounding environment, and output the oil smoke-environment correlation analysis results; The algorithm used to analyze the impact between the VOCs in the component quantitative analysis results and the surrounding atmospheric environment is as follows: ； in, Represents the correlation influence coefficient of the i-th VOCs component in the component quantitative analysis result on the surrounding atmospheric environment; represents the exponential function; Represents the adjustment coefficient corresponding to the i-th VOCs component co-existing in the exhaust gas and the atmosphere; It represents the weight coefficient corresponding to the potential of the i-th VOCs component to generate ozone in the atmosphere when calculating the correlation influence coefficient; It represents the weight coefficient corresponding to the contribution of the i-th VOCs component to the formation of secondary organic aerosol in the atmosphere when calculating the correlation influence coefficient; represents the reaction rate constant corresponding to the i-th VOCs component, which is used to characterize the reaction rate between the i-th VOCs component and NOx; Represents the average concentration of the i-th VOCs component co-existing in the exhaust gas and the atmosphere; represents the average concentration of nitrogen oxides corresponding to the i-th VOCs component in the atmosphere; It represents the impact value of the concentration of the i-th VOCs component on the ozone generation potential in the atmosphere; It represents the contribution factor of other factors corresponding to the i-th VOCs component to the formation of secondary organic aerosol; It represents the influence index of the i-th VOCs component in flue gas and atmosphere on the formation of secondary organic aerosol; An index representing the impact of nitrogen oxides on the formation of secondary organic aerosols; It represents the contribution of the concentration of the i-th VOCs component to the formation of secondary organic aerosol in the atmosphere; The emission reduction analysis submodule is used to optimize the oil smoke-environment related analysis results, determine the optimizable items for emission reduction, and output optimization suggestions corresponding to each of the optimizable items to obtain emission reduction analysis results.

8. A fume analysis method based on a CNN model, characterized in that: include: Step 1: Obtaining the fume emission data at a preset catering point and the surrounding environment data of the preset catering point, and outputting the initial data; Step 2: extracting and analyzing the features of the initial data based on a preset algorithm, and constructing a fume analysis list based on the feature analysis results; Step 3: Perform component quantitative analysis on the oil fume analysis list using a preset component analysis method, and output component quantitative analysis results in combination with statistical characteristic data corresponding to each component; Step 4: Combine the surrounding environment data and perform environmental correlation analysis on the component quantitative analysis results based on a preset analysis model, and output the oil smoke-environment correlation analysis results.

Images