CN116384754A - Deep learning-based environmental pollution risk assessment method in chemical industry park - Google Patents
Deep learning-based environmental pollution risk assessment method in chemical industry park Download PDFInfo
- Publication number
- CN116384754A CN116384754A CN202310645532.XA CN202310645532A CN116384754A CN 116384754 A CN116384754 A CN 116384754A CN 202310645532 A CN202310645532 A CN 202310645532A CN 116384754 A CN116384754 A CN 116384754A
- Authority
- CN
- China
- Prior art keywords
- pollutant
- chemical industry
- park
- industry park
- pollution
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 239000000126 substance Substances 0.000 title claims abstract description 122
- 238000003912 environmental pollution Methods 0.000 title claims abstract description 56
- 238000000034 method Methods 0.000 title claims abstract description 46
- 238000013135 deep learning Methods 0.000 title claims abstract description 41
- 238000012502 risk assessment Methods 0.000 title claims abstract description 35
- 239000003344 environmental pollutant Substances 0.000 claims abstract description 227
- 231100000719 pollutant Toxicity 0.000 claims abstract description 227
- 238000009826 distribution Methods 0.000 claims abstract description 83
- 230000004927 fusion Effects 0.000 claims abstract description 39
- 238000013528 artificial neural network Methods 0.000 claims abstract description 17
- 238000013210 evaluation model Methods 0.000 claims abstract description 16
- 238000012549 training Methods 0.000 claims abstract description 15
- 238000011156 evaluation Methods 0.000 claims description 47
- 239000011159 matrix material Substances 0.000 claims description 32
- 238000004519 manufacturing process Methods 0.000 claims description 18
- 239000002689 soil Substances 0.000 claims description 17
- 230000007613 environmental effect Effects 0.000 claims description 16
- 230000008859 change Effects 0.000 claims description 14
- 238000000605 extraction Methods 0.000 claims description 14
- 230000007246 mechanism Effects 0.000 claims description 12
- 230000000007 visual effect Effects 0.000 claims description 12
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 12
- 230000002776 aggregation Effects 0.000 claims description 10
- 238000004220 aggregation Methods 0.000 claims description 10
- 238000012824 chemical production Methods 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 9
- 238000004458 analytical method Methods 0.000 claims description 8
- 238000007781 pre-processing Methods 0.000 claims description 8
- 239000013598 vector Substances 0.000 claims description 8
- 238000007405 data analysis Methods 0.000 claims description 6
- 230000006870 function Effects 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000013523 data management Methods 0.000 claims description 4
- 238000013507 mapping Methods 0.000 claims description 4
- 238000012544 monitoring process Methods 0.000 claims description 4
- 230000008569 process Effects 0.000 claims description 4
- 230000011218 segmentation Effects 0.000 claims description 4
- 238000007619 statistical method Methods 0.000 claims description 4
- 238000012546 transfer Methods 0.000 claims description 4
- 230000002265 prevention Effects 0.000 abstract description 3
- 238000003860 storage Methods 0.000 description 8
- 230000008439 repair process Effects 0.000 description 4
- 238000007726 management method Methods 0.000 description 3
- 238000004140 cleaning Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000036541 health Effects 0.000 description 2
- 229910001385 heavy metal Inorganic materials 0.000 description 2
- 238000013508 migration Methods 0.000 description 2
- 230000005012 migration Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 230000001502 supplementing effect Effects 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000006227 byproduct Substances 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 239000000047 product Substances 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000002912 waste gas Substances 0.000 description 1
- 239000002351 wastewater Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
- G06N3/0442—Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Human Resources & Organizations (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Biomedical Technology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Software Systems (AREA)
- Tourism & Hospitality (AREA)
- Strategic Management (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Economics (AREA)
- Evolutionary Biology (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Game Theory and Decision Science (AREA)
- Primary Health Care (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
Abstract
The invention discloses a deep learning-based environmental pollution risk assessment method in a chemical industry park, which comprises the following steps: acquiring multi-source data of a target in a chemical industry park, acquiring pollutant class information according to the multi-source data, and reading the spatial distribution of pollution sources corresponding to various pollutants; acquiring main pollutants according to a multi-source data time sequence in preset time, and extracting space-time fusion characteristics of pollution in a park range based on a graph convolution neural network; constructing an environmental pollution evaluation model in a chemical industry park based on deep learning, and training according to space-time fusion characteristics; and acquiring real-time multi-source data of the target in the chemical industry park, and generating the risk assessment grade of the pollution of the current target in the chemical industry park. According to the invention, by acquiring the space-time fusion characteristics and analyzing the potential association of each pollution characteristic, the accurate environmental pollution risk assessment in the chemical industry park is realized, a data basis is provided for early pollution prevention and pollution treatment in the chemical industry park, and deeper pollution to the environment is prevented.
Description
Technical Field
The invention relates to the technical field of environmental pollution risk assessment, in particular to a method for assessing environmental pollution risk in a chemical industry park based on deep learning.
Background
In recent years, with the high-speed development of the economy of China, the number of chemical enterprises is continuously increased, a large number of chemical parks are established, and as chemical enterprises use a large number of chemical objects, the production process is complex, and besides a large number of chemical raw materials, various byproducts are generated in the production process. The three wastes generated in the production process have complex components and large quantity, and the generated waste water, waste residue, waste gas and the like can cause long-term and potential harm to the environment or directly influence the health of people. The environmental pollution risk in the chemical industry park is evaluated in time, technical support can be provided for early pollution prevention and pollution control in the chemical industry park, and deeper pollution to the environment is prevented. Because of the complexity, comprehensiveness and uncertainty of the environmental risk of the chemical industry park, no report on the environmental pollution risk assessment method of the chemical industry park is currently available.
For the environmental pollution risk assessment in the chemical industry park, the environmental condition in the chemical industry park area can be known, and important data reference is provided for the environmental management in the area. In the environmental pollution risk assessment, only the characteristics of the indexes are often considered in the pollution risk assessment according to the existing characteristic indexes, and a plurality of potential correlations exist among a large number of pollution characteristics, so that the potential correlations are beneficial to realizing more accurate pollution risk assessment. Therefore, how to obtain the relevant characteristics of each evaluation data through deep learning to realize accurate evaluation in the environmental pollution risk evaluation of the chemical industry park is an urgent problem which needs to be solved.
Disclosure of Invention
In order to solve the technical problems, the invention provides a method for evaluating the environmental pollution risk in a chemical industry park based on deep learning.
The first aspect of the invention provides a deep learning-based environmental pollution risk assessment method in a chemical industry park, which comprises the following steps:
acquiring multi-source data of a target in a chemical industry park, acquiring pollutant class information according to the multi-source data, and reading the spatial distribution of pollution sources corresponding to various pollutants;
preprocessing the multi-source data, generating a multi-source data time sequence in preset time to obtain main pollutants, extracting time sequence features and space features of pollution in a park range based on a graph convolution neural network, and fusing the time sequence features and the space features to obtain space-time fusion features;
constructing an environmental pollution evaluation model in a chemical industry park based on deep learning, and training according to the space-time fusion characteristics;
and acquiring real-time multi-source data of the target in the chemical industry park, performing data analysis processing to finish feature extraction and classification, and generating the risk assessment grade of the pollution of the current target in the chemical industry park.
In this scheme, according to multisource data acquisition pollutant class information to read each class pollutant and correspond the spatial distribution of pollution source, specifically do:
Collecting multi-source data of a target in a chemical industry park, wherein the multi-source data comprise park enterprise types, enterprise production information, enterprise numbers and environment data of park water resources, soil and atmosphere, and the environment data comprise pollutant and pollutant concentration information in the water resources, the soil and the atmosphere;
reading enterprise related information in multi-source data, obtaining a plurality of word vectors through word segmentation, obtaining pollutant keywords according to the word vectors, obtaining pollutant category information according to the pollutant keywords to be associated with enterprises, and setting pollutant category labels for the enterprises;
the method comprises the steps of obtaining geographic position information of each enterprise, dividing a park area into a plurality of grid areas, judging the grid areas according to the geographic position information of the enterprise, marking the grid areas, generating pollution source position marks, and obtaining the space distribution of pollution corresponding to each pollutant;
and performing de-duplication operation on the pollutant class labels of enterprises in the grid area, using the pollutant class labels as the pollutant class labels of the grid area, and verifying and correcting the pollutant class labels according to the environmental data of water resources, soil and atmosphere of the park.
In this scheme, acquire the main pollutant in each grid region, specifically:
Acquiring multi-source data time sequence within preset time of each grid area, and extracting concentration time sequence of different types of pollutants in each grid area;
calculating pearson correlation coefficients of different grating areas according to concentration time sequence sequences of different types of pollutants in the different grating areas, and when a certain pollutant concentration in the two grating areas corresponds to the pearson correlation coefficient and is larger than a preset threshold value, and a pollutant type label of one grating area does not comprise the pollutant, adding the pollutant type label in the grating area;
traversing all the grating areas, updating the pollutant class labels of the grating areas, and determining main pollutants of each grating area according to the updated pollutant class labels.
In this scheme, extract time sequence characteristic and space characteristic that pollutes in garden scope based on drawing convolution neural network, specifically do:
constructing an undirected graph in each grid area in the chemical industry park according to the target, and taking main pollutants, soil lithology and meteorological features corresponding to the grid area nodes as additional features of each grid area node in the undirected graph;
calculating the feature similarity of each grid area node, selecting adjacent grid area nodes, constructing a similarity adjacent matrix, calculating Euclidean distance of each grid area node, acquiring, and selecting grid area nodes meeting the distance requirement to construct a distance adjacent matrix;
According to the Pearson correlation coefficient between different grating areas as the aggregation weight of the grating area nodes, respectively generating the space characteristics of main pollutants by using a similarity adjacent matrix and a distance adjacent matrix through a characteristic transfer and characteristic aggregation mechanism;
setting a time sequence feature extraction layer by using a time convolution neural network, introducing hole convolution in the time sequence feature extraction layer, connecting by using a residual error network, setting convolution kernels identical to nodes of a grid area, and extracting time sequence features of a main pollutant concentration time sequence by using different adjacent matrixes to perform graph convolution;
and introducing an attention mechanism, setting an attention mechanism fusion module, fusing the time sequence features and the space features, and combining fusion features corresponding to the similarity adjacent matrix and the distance adjacent matrix to generate space-time fusion features.
In this scheme, construct the environmental pollution evaluation model in producing chemical industry garden based on degree of depth study, specifically do:
acquiring the pollutant distribution of the target in the chemical industry park according to the main pollutants in each grid area and the corresponding pollutant concentration information;
acquiring historical pollutant distribution and corresponding space-time fusion characteristics according to a multi-source data time sequence in preset time through data retrieval, constructing a training set according to the historical pollutant distribution and the corresponding space-time fusion characteristics of a target in a chemical production park, and training an LSTM network;
When the output precision of the LSTM network meets the preset standard, outputting the trained LSTM network, and acquiring the pollutant distribution after the preset time based on the current pollutant distribution of the target in the chemical industry park.
In this scheme, construct the environmental pollution evaluation model in producing chemical industry garden based on degree of depth study, still include:
constructing an environmental pollution evaluation model in a chemical industry park based on LSTM (least squares) network and fuzzy comprehensive analysis, and acquiring main pollutants, geographical position information and geological hydrologic information of a target in the chemical industry park as park characteristics;
acquiring a sensitive target according to the park characteristics, constructing a data retrieval task based on the sensitive target, acquiring an environment risk factor data set through similarity calculation in a preset data retrieval space, and carrying out statistical analysis on the environment risk factor data set;
selecting the similarity and the environmental risk factors with the use frequency larger than a preset threshold according to the preset quantity, determining an evaluation index, and generating weight information of the evaluation index according to a analytic hierarchy process;
forming a risk evaluation set for fuzzy comprehensive evaluation according to the evaluation indexes, presetting risk evaluation grades of pollution of a target in a chemical industry park, and judging membership of evaluation factors to each evaluation grade according to a trigonometric function to obtain a membership matrix;
And calculating a fuzzy comprehensive evaluation result in the target layer according to the membership matrix and the index weight, and generating a risk evaluation grade of pollution of the target in the chemical production park.
In this scheme, the pollutant distribution after according to the preset time generates early warning information, specifically:
acquiring pollutant distribution change characteristics of a preset time step through historical pollutant distribution of a target in a chemical industry park, and acquiring standard values of pollutant distribution after preset time according to the pollutant distribution change characteristics of the preset time step;
acquiring predicted pollutant distribution of preset time, and comparing the predicted pollutant distribution with a standard value of the pollutant distribution to acquire pollutant distribution deviation;
when the distribution deviation of the pollutants is larger than a preset deviation threshold, generating early warning information of the corresponding pollutants;
and determining a corresponding grid area according to the early warning information, and generating pollutant treatment measures according to the geographical hydrologic information and the position information of the grid area.
In this scheme, still include: mapping a physical model of a target in a chemical industry park to a three-dimensional twin body model based on digital twin, constructing a three-dimensional visual model, and carrying out feature analysis, environmental risk evaluation and data management according to the three-dimensional visual model;
And monitoring the target in the chemical industry park through the three-dimensional visual model to acquire early warning information of each category and newly-added pollutant and pollution source information.
The second aspect of the present invention also provides a deep learning-based environmental pollution risk assessment system in a chemical industry park, the system comprising: the system comprises a memory and a processor, wherein the memory comprises a deep learning-based on-production chemical industry park environmental pollution risk assessment method program, and the deep learning-based on-production chemical industry park environmental pollution risk assessment method program realizes the following steps when being executed by the processor:
acquiring multi-source data of a target in a chemical industry park, acquiring pollutant class information according to the multi-source data, and reading the spatial distribution of pollution sources corresponding to various pollutants;
preprocessing the multi-source data, generating a multi-source data time sequence in preset time to obtain main pollutants, extracting time sequence features and space features of pollution in a park range based on a graph convolution neural network, and fusing the time sequence features and the space features to obtain space-time fusion features;
constructing an environmental pollution evaluation model in a chemical industry park based on deep learning, and training according to the space-time fusion characteristics;
And acquiring real-time multi-source data of the target in the chemical industry park, performing data analysis processing to finish feature extraction and classification, and generating the risk assessment grade of the pollution of the current target in the chemical industry park.
The invention discloses a deep learning-based environmental pollution risk assessment method in a chemical industry park, which comprises the following steps: acquiring multi-source data of a target in a chemical industry park, acquiring pollutant class information according to the multi-source data, and reading the spatial distribution of pollution sources corresponding to various pollutants; preprocessing multi-source data, generating a multi-source data time sequence within preset time to obtain main pollutants, extracting time sequence characteristics and space characteristics of pollution in a park range based on a graph convolution neural network, and obtaining space-time fusion characteristics after fusion; constructing an environmental pollution evaluation model in a chemical industry park based on deep learning, and training according to the space-time fusion characteristics; and acquiring real-time multi-source data of the target in the chemical industry park, performing data analysis processing to finish feature extraction and classification, and generating the risk assessment grade of the pollution of the current target in the chemical industry park. According to the invention, by acquiring the space-time fusion characteristics and analyzing the potential association of each pollution characteristic, the accurate environmental pollution risk assessment in the chemical industry park is realized, a data basis is provided for early pollution prevention and pollution treatment in the chemical industry park, and deeper pollution to the environment is prevented.
Drawings
FIG. 1 shows a flow chart of a method for evaluating environmental pollution risk in a chemical industry park based on deep learning of the invention;
FIG. 2 illustrates a flow chart of a method of extracting pollution spatiotemporal fusion features in the context of a campus in accordance with the present invention;
FIG. 3 is a flow chart of a method for generating risk assessment grades by constructing an environmental pollution assessment model according to the present invention;
FIG. 4 shows a block diagram of an environmental pollution risk assessment system in a chemical industry park based on deep learning.
Detailed Description
In order that the above-recited objects, features and advantages of the present invention will be more clearly understood, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description. It should be noted that, in the case of no conflict, the embodiments of the present application and the features in the embodiments may be combined with each other.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, however, the present invention may be practiced in other ways than those described herein, and therefore the scope of the present invention is not limited to the specific embodiments disclosed below.
FIG. 1 shows a flow chart of a method for evaluating environmental pollution risk in a chemical industry park based on deep learning.
As shown in fig. 1, the first aspect of the present invention provides a method for evaluating environmental pollution risk in a chemical industry park based on deep learning, comprising:
s102, multi-source data of a target in a chemical industry park are acquired, pollutant class information is acquired according to the multi-source data, and the spatial distribution of pollution sources corresponding to various pollutants is read;
s104, preprocessing the multi-source data, generating a multi-source data time sequence in preset time to obtain main pollutants, extracting time sequence features and space features of pollution in a park range based on a graph convolution neural network, and fusing the time sequence features and the space features to obtain space-time fusion features;
s106, constructing an environmental pollution evaluation model in a chemical industry park based on deep learning, and training according to the space-time fusion characteristics;
s108, acquiring real-time multi-source data of the target in the chemical industry park, performing data analysis processing to finish feature extraction and classification, and generating a risk assessment grade of pollution of the current target in the chemical industry park.
The method comprises the steps of collecting multi-source data of a target in a chemical industry park, wherein the multi-source data comprise park enterprise types, enterprise production information, enterprise numbers and environment data of park water resources, soil and atmosphere, and the environment data comprise the water resources, the soil and pollutants in the atmosphere and pollutant concentration information; reading enterprise related information in multi-source data, namely data such as enterprise types, enterprise basic information and the like, obtaining a plurality of word vectors through word segmentation processing, obtaining pollutant keywords according to the word vectors, obtaining pollutant category information according to the pollutant keywords to be associated with enterprises, setting pollutant category labels for the enterprises, and enabling one enterprise to comprise a plurality of pollutant category labels; the method comprises the steps of obtaining geographic position information of each enterprise, dividing a park area into a plurality of grid areas, judging the grid areas according to the geographic position information of the enterprise, marking the grid areas, generating pollution source position marks, and obtaining the space distribution of pollution corresponding to each pollutant; and performing repeated label removal operation on the pollutant class labels of enterprises in the grid area, using the repeated label removal operation as the pollutant class labels of the grid area, verifying and correcting the pollutant class labels according to the environmental data of water resources, soil and atmosphere in the park, and if the environmental data contains pollutants corresponding to a certain pollutant class label, verifying successfully, and if not, supplementing or deleting.
Figure 2 shows a flow chart of a method of extracting pollution spatiotemporal fusion features on a campus scale in accordance with the present invention.
According to the embodiment of the invention, the time sequence characteristics and the space characteristics of pollution in the park are extracted based on the graph convolution neural network, and the method specifically comprises the following steps:
s202, constructing an undirected graph in each grid area in the chemical industry park according to the target, and taking main pollutants, soil lithology and meteorological features corresponding to the grid area nodes as additional features of each grid area node in the undirected graph;
s204, calculating the feature similarity of each grid region node, selecting adjacent grid region nodes, constructing a similarity adjacent matrix, calculating the Euclidean distance of each grid region node, and selecting the grid region nodes meeting the distance requirement to construct a distance adjacent matrix;
s206, according to the Pearson correlation coefficient among different grid areas as the aggregation weight of the grid area nodes, respectively generating the space characteristics of the main pollutants by using a similarity adjacent matrix and a distance adjacent matrix through a characteristic transfer and characteristic aggregation mechanism;
s208, setting a time sequence feature extraction layer by using a time convolution neural network, introducing hole convolution into the time sequence feature extraction layer, connecting by using a residual error network, setting convolution kernels identical to nodes of a grid area, and extracting time sequence features of a main pollutant concentration time sequence by using different adjacent matrixes to perform graph convolution;
S210, introducing an attention mechanism, setting an attention mechanism fusion module, fusing time sequence features and space features, and combining fusion features corresponding to a similarity adjacent matrix and a distance adjacent matrix to generate space-time fusion features.
It should be noted that, each grid after rasterization is regarded as a node on the graph, each grid region corresponds to a specific region of the target in the chemical production park, different similarity adjacent matrixes and distance adjacent matrixes are set to fully extract the space relation of pollutants, and in addition, the pearson correlation coefficient is used as an aggregation weight in the space relation simulation to characterize the characteristic contribution of the nodes of the adjacent grid regions, so that the model is more focused on the grid region with large space association. In addition, cavity convolution and residual connection are introduced into the time convolution neural network, so that the phenomenon of overfitting is prevented, the receptive field is enlarged, and the time correlation of the pollutant concentration time sequence is greatly reserved.
Preprocessing multi-source data such as data cleaning and screening to obtain multi-source data time sequence within preset time of each grid area, and extracting concentration time sequence of different pollutants in each grid area; calculating pearson correlation coefficients of different grating areas according to concentration time sequence sequences of different types of pollutants in the different grating areas, and when a certain pollutant concentration in the two grating areas corresponds to the pearson correlation coefficient and is larger than a preset threshold value, and a pollutant type label of one grating area does not comprise the pollutant, adding the pollutant type label in the grating area; traversing all the grating areas, updating the pollutant class labels of the grating areas, and determining main pollutants of each grating area according to the updated pollutant class labels.
FIG. 3 is a flow chart of a method for generating risk assessment grades by constructing an environmental pollution assessment model according to the present invention.
According to the embodiment of the invention, the environmental pollution evaluation model constructed in the chemical industry park based on the deep learning further comprises:
s302, constructing an environmental pollution evaluation model in a chemical industry park based on LSTM network combined with fuzzy comprehensive analysis, and acquiring main pollutants, geographical position information and geological hydrologic information of a target in the chemical industry park as park characteristics;
s304, acquiring a sensitive target according to the park characteristics, constructing a data retrieval task based on the sensitive target, acquiring an environment risk factor data set through similarity calculation in a preset data retrieval space, and carrying out statistical analysis on the environment risk factor data set;
s306, selecting environment risk factors with similarity and use frequency larger than a preset threshold according to the preset quantity, determining an evaluation index, and generating weight information of the evaluation index according to an analytic hierarchy process;
s308, forming a risk evaluation set for fuzzy comprehensive evaluation according to the evaluation indexes, presetting risk evaluation grades of pollution of a target in a chemical industry park, and judging membership of evaluation factors to each evaluation grade according to a trigonometric function to obtain a membership matrix;
And S310, calculating a comprehensive fuzzy evaluation result in the target layer according to the membership matrix and the index weight, and generating a risk evaluation grade of pollution of the target in the chemical production park.
It should be noted that, according to the park characteristics, the sensitive target is obtained, for example, the target is in the residential area of the chemical industry park accessory, the health risk of the human body needs to be considered, the heavy metal pollution in the soil, the migration attribute of part of the heavy metal has, the migration risk needs to be considered, and the like.
Acquiring the pollutant distribution of the target in the chemical industry park according to the main pollutants in each grid area and the corresponding pollutant concentration information; acquiring historical pollutant distribution and corresponding space-time fusion characteristics according to a multi-source data time sequence in preset time through data retrieval, constructing a training set according to the historical pollutant distribution and the corresponding space-time fusion characteristics of a target in a chemical production park, and training an LSTM network; when the output precision of the LSTM network meets the preset standard, outputting the trained LSTM network, and acquiring the pollutant distribution after the preset time based on the current pollutant distribution of the target in the chemical industry park.
It should be noted that, the early warning information is generated according to the pollutant distribution after the preset time, specifically: acquiring pollutant distribution change characteristics of a preset time step through historical pollutant distribution of a target in a chemical industry park, and acquiring standard values of pollutant distribution after preset time according to the pollutant distribution change characteristics of the preset time step; acquiring predicted pollutant distribution of preset time, and comparing the predicted pollutant distribution with a standard value of the pollutant distribution to acquire pollutant distribution deviation; when the distribution deviation of the pollutants is larger than a preset deviation threshold, generating early warning information of the corresponding pollutants; and determining a corresponding grid area according to the early warning information, and generating pollutant treatment measures by utilizing a big data means according to the geographical hydrological information and the position information of the grid area.
Acquiring a historical pollutant distribution time sequence, splitting a target time period according to a preset time step, acquiring pollutant distribution change characteristics of each time step, acquiring a starting time stamp of each time step, and judging the production working condition of an enterprise corresponding to the pollutant of the starting time stamp;
matching the production working condition with the pollutant distribution change characteristics of the time step, storing the pollutant distribution change characteristics into a database, acquiring the current production working condition of an enterprise corresponding to the pollutant, and calculating the similarity in the database by utilizing the current production working condition;
obtaining data with similarity meeting a preset standard, extracting corresponding pollutant distribution change characteristics, reading a time stamp of the pollutant distribution change characteristics, and selecting the pollutant distribution change characteristics of the latest time stamp to calculate the standard value of the pollutant distribution after the preset time;
if the data with the similarity meeting the preset standard does not exist, the category information and the characteristic information of the pollutants are obtained, and the similar pollutants are selected for re-searching.
The method comprises the steps of mapping a physical model of a target in a chemical industry park to a three-dimensional twin body model based on digital twin, constructing a three-dimensional visual model, and carrying out feature analysis, environmental risk evaluation and data management according to the three-dimensional visual model; and monitoring the target in the chemical industry park through the three-dimensional visual model to acquire early warning information of each category and newly-added pollutant and pollution source information. The management personnel can intuitively analyze the basic information, the pollutant content, the pollutant space distribution condition and the site risk condition of the target in the chemical industry park, digitally manage the site information, and based on the system, propose related suggestions and measures for follow-up site repair and treatment for the pollution condition, and provide assistance for site repair and treatment work.
FIG. 4 shows a block diagram of an environmental pollution risk assessment system in a chemical industry park based on deep learning.
The second aspect of the present invention also provides a deep learning-based environmental pollution risk assessment system 4 for a chemical industry park, which comprises: the memory 41 and the processor 42, wherein the memory comprises a deep learning-based on-production chemical industry park environmental pollution risk assessment method program, and the deep learning-based on-production chemical industry park environmental pollution risk assessment method program realizes the following steps when being executed by the processor:
acquiring multi-source data of a target in a chemical industry park, acquiring pollutant class information according to the multi-source data, and reading the spatial distribution of pollution sources corresponding to various pollutants;
preprocessing the multi-source data, generating a multi-source data time sequence in preset time to obtain main pollutants, extracting time sequence features and space features of pollution in a park range based on a graph convolution neural network, and fusing the time sequence features and the space features to obtain space-time fusion features;
constructing an environmental pollution evaluation model in a chemical industry park based on deep learning, and training according to the space-time fusion characteristics;
And acquiring real-time multi-source data of the target in the chemical industry park, performing data analysis processing to finish feature extraction and classification, and generating the risk assessment grade of the pollution of the current target in the chemical industry park.
The method comprises the steps of collecting multi-source data of a target in a chemical industry park, wherein the multi-source data comprise park enterprise types, enterprise production information, enterprise numbers and environment data of park water resources, soil and atmosphere, and the environment data comprise the water resources, the soil and pollutants in the atmosphere and pollutant concentration information; reading enterprise related information in multi-source data, namely data such as enterprise types, enterprise basic information and the like, obtaining a plurality of word vectors through word segmentation processing, obtaining pollutant keywords according to the word vectors, obtaining pollutant category information according to the pollutant keywords to be associated with enterprises, setting pollutant category labels for the enterprises, and enabling one enterprise to comprise a plurality of pollutant category labels; the method comprises the steps of obtaining geographic position information of each enterprise, dividing a park area into a plurality of grid areas, judging the grid areas according to the geographic position information of the enterprise, marking the grid areas, generating pollution source position marks, and obtaining the space distribution of pollution corresponding to each pollutant; and performing repeated label removal operation on the pollutant class labels of enterprises in the grid area, using the repeated label removal operation as the pollutant class labels of the grid area, verifying and correcting the pollutant class labels according to the environmental data of water resources, soil and atmosphere in the park, and if the environmental data contains pollutants corresponding to a certain pollutant class label, verifying successfully, and if not, supplementing or deleting.
According to the embodiment of the invention, the time sequence characteristics and the space characteristics of pollution in the park are extracted based on the graph convolution neural network, and the method specifically comprises the following steps:
constructing an undirected graph in each grid area in the chemical industry park according to the target, and taking main pollutants, soil lithology and meteorological features corresponding to the grid area nodes as additional features of each grid area node in the undirected graph;
calculating the feature similarity of each grid area node, selecting adjacent grid area nodes, constructing a similarity adjacent matrix, calculating Euclidean distance of each grid area node, acquiring, and selecting grid area nodes meeting the distance requirement to construct a distance adjacent matrix;
according to the Pearson correlation coefficient between different grating areas as the aggregation weight of the grating area nodes, respectively generating the space characteristics of main pollutants by using a similarity adjacent matrix and a distance adjacent matrix through a characteristic transfer and characteristic aggregation mechanism;
setting a time sequence feature extraction layer by using a time convolution neural network, introducing hole convolution in the time sequence feature extraction layer, connecting by using a residual error network, setting convolution kernels identical to nodes of a grid area, and extracting time sequence features of a main pollutant concentration time sequence by using different adjacent matrixes to perform graph convolution;
And introducing an attention mechanism, setting an attention mechanism fusion module, fusing the time sequence features and the space features, and combining fusion features corresponding to the similarity adjacent matrix and the distance adjacent matrix to generate space-time fusion features.
It should be noted that, each grid after rasterization is regarded as a node on the graph, each grid region corresponds to a specific region of the target in the chemical production park, different similarity adjacent matrixes and distance adjacent matrixes are set to fully extract the space relation of pollutants, and in addition, the pearson correlation coefficient is used as an aggregation weight in the space relation simulation to characterize the characteristic contribution of the nodes of the adjacent grid regions, so that the model is more focused on the grid region with large space association. In addition, cavity convolution and residual connection are introduced into the time convolution neural network, so that the phenomenon of overfitting is prevented, the receptive field is enlarged, and the time correlation of the pollutant concentration time sequence is greatly reserved.
Preprocessing multi-source data such as data cleaning and screening to obtain multi-source data time sequence within preset time of each grid area, and extracting concentration time sequence of different pollutants in each grid area; calculating pearson correlation coefficients of different grating areas according to concentration time sequence sequences of different types of pollutants in the different grating areas, and when a certain pollutant concentration in the two grating areas corresponds to the pearson correlation coefficient and is larger than a preset threshold value, and a pollutant type label of one grating area does not comprise the pollutant, adding the pollutant type label in the grating area; traversing all the grating areas, updating the pollutant class labels of the grating areas, and determining main pollutants of each grating area according to the updated pollutant class labels.
According to the embodiment of the invention, the environmental pollution evaluation model constructed in the chemical industry park based on the deep learning further comprises:
constructing an environmental pollution evaluation model in a chemical industry park based on LSTM (least squares) network and fuzzy comprehensive analysis, and acquiring main pollutants, geographical position information and geological hydrologic information of a target in the chemical industry park as park characteristics;
acquiring a sensitive target according to the park characteristics, constructing a data retrieval task based on the sensitive target, acquiring an environment risk factor data set through similarity calculation in a preset data retrieval space, and carrying out statistical analysis on the environment risk factor data set;
selecting the similarity and the environmental risk factors with the use frequency larger than a preset threshold according to the preset quantity, determining an evaluation index, and generating weight information of the evaluation index according to a analytic hierarchy process;
forming a risk evaluation set for fuzzy comprehensive evaluation according to the evaluation indexes, presetting risk evaluation grades of pollution of a target in a chemical industry park, and judging membership of evaluation factors to each evaluation grade according to a trigonometric function to obtain a membership matrix;
and calculating a fuzzy comprehensive evaluation result in the target layer according to the membership matrix and the index weight, and generating a risk evaluation grade of pollution of the target in the chemical production park.
The method comprises the steps of acquiring the pollutant distribution of a target in a chemical industry park according to main pollutants in each grid area and corresponding pollutant concentration information; acquiring historical pollutant distribution and corresponding space-time fusion characteristics according to a multi-source data time sequence in preset time through data retrieval, constructing a training set according to the historical pollutant distribution and the corresponding space-time fusion characteristics of a target in a chemical production park, and training an LSTM network; when the output precision of the LSTM network meets the preset standard, outputting the trained LSTM network, and acquiring the pollutant distribution after the preset time based on the current pollutant distribution of the target in the chemical industry park.
It should be noted that, the early warning information is generated according to the pollutant distribution after the preset time, specifically: acquiring pollutant distribution change characteristics of a preset time step through historical pollutant distribution of a target in a chemical industry park, and acquiring standard values of pollutant distribution after preset time according to the pollutant distribution change characteristics of the preset time step; acquiring predicted pollutant distribution of preset time, and comparing the predicted pollutant distribution with a standard value of the pollutant distribution to acquire pollutant distribution deviation; when the distribution deviation of the pollutants is larger than a preset deviation threshold, generating early warning information of the corresponding pollutants; and determining a corresponding grid area according to the early warning information, and generating pollutant treatment measures by utilizing a big data means according to the geographical hydrological information and the position information of the grid area.
The method comprises the steps of mapping a physical model of a target in a chemical industry park to a three-dimensional twin body model based on digital twin, constructing a three-dimensional visual model, and carrying out feature analysis, environmental risk evaluation and data management according to the three-dimensional visual model; and monitoring the target in the chemical industry park through the three-dimensional visual model to acquire early warning information of each category and newly-added pollutant and pollution source information. The management personnel can intuitively analyze the basic information, the pollutant content, the pollutant space distribution condition and the site risk condition of the target in the chemical industry park, digitally manage the site information, and based on the system, propose related suggestions and measures for follow-up site repair and treatment for the pollution condition, and provide assistance for site repair and treatment work.
The third aspect of the present invention also provides a computer readable storage medium, including a deep learning-based on-production chemical industrial park environmental pollution risk assessment method program, where the deep learning-based on-production chemical industrial park environmental pollution risk assessment method program is executed by a processor, to implement the steps of the deep learning-based on-production chemical industrial park environmental pollution risk assessment method according to any one of the above.
In the several embodiments provided in this application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The above described device embodiments are only illustrative, e.g. the division of the units is only one logical function division, and there may be other divisions in practice, such as: multiple units or components may be combined or may be integrated into another system, or some features may be omitted, or not performed. In addition, the various components shown or discussed may be coupled or directly coupled or communicatively coupled to each other via some interface, whether indirectly coupled or communicatively coupled to devices or units, whether electrically, mechanically, or otherwise.
The units described above as separate components may or may not be physically separate, and components shown as units may or may not be physical units; can be located in one place or distributed to a plurality of network units; some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in each embodiment of the present invention may be integrated in one processing unit, or each unit may be separately used as one unit, or two or more units may be integrated in one unit; the integrated units may be implemented in hardware or in hardware plus software functional units.
Those of ordinary skill in the art will appreciate that: all or part of the steps for implementing the above method embodiments may be implemented by hardware related to program instructions, and the foregoing program may be stored in a computer readable storage medium, where the program, when executed, performs steps including the above method embodiments; and the aforementioned storage medium includes: a mobile storage device, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk or an optical disk, or the like, which can store program codes.
Alternatively, the above-described integrated units of the present invention may be stored in a computer-readable storage medium if implemented in the form of software functional modules and sold or used as separate products. Based on such understanding, the technical solutions of the embodiments of the present invention may be embodied in essence or a part contributing to the prior art in the form of a software product stored in a storage medium, including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in the embodiments of the present invention. And the aforementioned storage medium includes: a removable storage device, ROM, RAM, magnetic or optical disk, or other medium capable of storing program code.
The foregoing is merely illustrative of the present invention, and the present invention is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (8)
1. The method for evaluating the environmental pollution risk of the chemical industry park on the basis of deep learning is characterized by comprising the following steps of:
acquiring multi-source data of a target in a chemical industry park, acquiring pollutant class information according to the multi-source data, and reading the spatial distribution of pollution sources corresponding to various pollutants;
preprocessing the multi-source data, generating a multi-source data time sequence in preset time to obtain main pollutants, extracting time sequence features and space features of pollution in a park range based on a graph convolution neural network, and fusing the time sequence features and the space features to obtain space-time fusion features;
constructing an environmental pollution evaluation model in a chemical industry park based on deep learning, and training according to the space-time fusion characteristics;
and acquiring real-time multi-source data of the target in the chemical industry park, performing data analysis processing to finish feature extraction and classification, and generating the risk assessment grade of the pollution of the current target in the chemical industry park.
2. The method for evaluating environmental pollution risk in a chemical industry park based on deep learning according to claim 1, wherein the method is characterized by obtaining pollutant class information according to the multi-source data and reading the spatial distribution of pollution sources corresponding to each class of pollutants, and specifically comprises the following steps:
collecting multi-source data of a target in a chemical industry park, wherein the multi-source data comprise park enterprise types, enterprise production information, enterprise numbers and environment data of park water resources, soil and atmosphere, and the environment data comprise pollutant and pollutant concentration information in the water resources, the soil and the atmosphere;
reading enterprise related information in multi-source data, obtaining a plurality of word vectors through word segmentation, obtaining pollutant keywords according to the word vectors, obtaining pollutant category information according to the pollutant keywords to be associated with enterprises, and setting pollutant category labels for the enterprises;
the method comprises the steps of obtaining geographic position information of each enterprise, dividing a park area into a plurality of grid areas, judging the grid areas according to the geographic position information of the enterprise, marking the grid areas, generating pollution source position marks, and obtaining the space distribution of pollution corresponding to each pollutant;
And performing de-duplication operation on the pollutant class labels of enterprises in the grid area, using the pollutant class labels as the pollutant class labels of the grid area, and verifying and correcting the pollutant class labels according to the environmental data of water resources, soil and atmosphere of the park.
3. The method for evaluating environmental pollution risk in a chemical industry park based on deep learning according to claim 2, wherein the main pollutants in each grid area are obtained specifically as follows:
acquiring multi-source data time sequence within preset time of each grid area, and extracting concentration time sequence of different types of pollutants in each grid area;
calculating pearson correlation coefficients of different grating areas according to concentration time sequence sequences of different types of pollutants in the different grating areas, and when a certain pollutant concentration in the two grating areas corresponds to the pearson correlation coefficient and is larger than a preset threshold value, and a pollutant type label of one grating area does not comprise the pollutant, adding the pollutant type label in the grating area;
traversing all the grating areas, updating the pollutant class labels of the grating areas, and determining main pollutants of each grating area according to the updated pollutant class labels.
4. The method for evaluating environmental pollution risk in a chemical industry park based on deep learning according to claim 1, wherein the method is characterized by extracting time sequence features and space features of pollution in the park based on a graph convolution neural network, and specifically comprises the following steps:
constructing an undirected graph in each grid area in the chemical industry park according to the target, and taking main pollutants, soil lithology and meteorological features corresponding to the grid area nodes as additional features of each grid area node in the undirected graph;
calculating the feature similarity of each grid area node, selecting adjacent grid area nodes, constructing a similarity adjacent matrix, calculating Euclidean distance of each grid area node, acquiring, and selecting grid area nodes meeting the distance requirement to construct a distance adjacent matrix;
according to the Pearson correlation coefficient between different grating areas as the aggregation weight of the grating area nodes, respectively generating the space characteristics of main pollutants by using a similarity adjacent matrix and a distance adjacent matrix through a characteristic transfer and characteristic aggregation mechanism;
setting a time sequence feature extraction layer by using a time convolution neural network, introducing hole convolution in the time sequence feature extraction layer, connecting by using a residual error network, setting convolution kernels identical to nodes of a grid area, and extracting time sequence features of a main pollutant concentration time sequence by using different adjacent matrixes to perform graph convolution;
And introducing an attention mechanism, setting an attention mechanism fusion module, fusing the time sequence features and the space features, and combining fusion features corresponding to the similarity adjacent matrix and the distance adjacent matrix to generate space-time fusion features.
5. The method for evaluating the environmental pollution risk in the chemical industry park based on the deep learning according to claim 1, wherein the environmental pollution evaluation model in the chemical industry park based on the deep learning is constructed, specifically:
acquiring the pollutant distribution of the target in the chemical industry park according to the main pollutants in each grid area and the corresponding pollutant concentration information;
acquiring historical pollutant distribution and corresponding space-time fusion characteristics according to a multi-source data time sequence in preset time through data retrieval, constructing a training set according to the historical pollutant distribution and the corresponding space-time fusion characteristics of a target in a chemical production park, and training an LSTM network;
when the output precision of the LSTM network meets the preset standard, outputting the trained LSTM network, and acquiring the pollutant distribution after the preset time based on the current pollutant distribution of the target in the chemical industry park.
6. The deep learning-based environmental pollution risk assessment method for a chemical industry park of claim 5, wherein the deep learning-based environmental pollution assessment model is constructed for the chemical industry park, and further comprising:
Constructing an environmental pollution evaluation model in a chemical industry park based on LSTM (least squares) network and fuzzy comprehensive analysis, and acquiring main pollutants, geographical position information and geological hydrologic information of a target in the chemical industry park as park characteristics;
acquiring a sensitive target according to the park characteristics, constructing a data retrieval task based on the sensitive target, acquiring an environment risk factor data set through similarity calculation in a preset data retrieval space, and carrying out statistical analysis on the environment risk factor data set;
selecting the similarity and the environmental risk factors with the use frequency larger than a preset threshold according to the preset quantity, determining an evaluation index, and generating weight information of the evaluation index according to a analytic hierarchy process;
forming a risk evaluation set for fuzzy comprehensive evaluation according to the evaluation indexes, presetting risk evaluation grades of pollution of a target in a chemical industry park, and judging membership of evaluation factors to each evaluation grade according to a trigonometric function to obtain a membership matrix;
and calculating a fuzzy comprehensive evaluation result in the target layer according to the membership matrix and the index weight, and generating a risk evaluation grade of pollution of the target in the chemical production park.
7. The method for evaluating environmental pollution risk in a chemical industry park based on deep learning according to claim 6, wherein the method is characterized in that the early warning information is generated according to the pollutant distribution after the preset time, and specifically comprises the following steps:
Acquiring pollutant distribution change characteristics of a preset time step through historical pollutant distribution of a target in a chemical industry park, and acquiring standard values of pollutant distribution after preset time according to the pollutant distribution change characteristics of the preset time step;
acquiring predicted pollutant distribution of preset time, and comparing the predicted pollutant distribution with a standard value of the pollutant distribution to acquire pollutant distribution deviation;
when the distribution deviation of the pollutants is larger than a preset deviation threshold, generating early warning information of the corresponding pollutants;
and determining a corresponding grid area according to the early warning information, and generating pollutant treatment measures according to the geographical hydrologic information and the position information of the grid area.
8. The deep learning-based environmental pollution risk assessment method for a chemical industrial park of claim 1, further comprising:
mapping a physical model of a target in a chemical industry park to a three-dimensional twin body model based on digital twin, constructing a three-dimensional visual model, and carrying out feature analysis, environmental risk evaluation and data management according to the three-dimensional visual model;
and monitoring the target in the chemical industry park through the three-dimensional visual model to acquire early warning information of each category and newly-added pollutant and pollution source information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310645532.XA CN116384754B (en) | 2023-06-02 | 2023-06-02 | Deep learning-based environmental pollution risk assessment method in chemical industry park |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310645532.XA CN116384754B (en) | 2023-06-02 | 2023-06-02 | Deep learning-based environmental pollution risk assessment method in chemical industry park |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116384754A true CN116384754A (en) | 2023-07-04 |
| CN116384754B CN116384754B (en) | 2023-08-04 |
Family
ID=86967889
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310645532.XA Active CN116384754B (en) | 2023-06-02 | 2023-06-02 | Deep learning-based environmental pollution risk assessment method in chemical industry park |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116384754B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116754499A (en) * | 2023-08-23 | 2023-09-15 | 天津中科谱光信息技术有限公司 | Multi-topology node hyperspectral water quality parameter joint inversion method and related equipment |
| CN116821386A (en) * | 2023-08-30 | 2023-09-29 | 深圳巨湾科技有限公司 | Single characteristic evaluation control method based on multi-source heterogeneous data structuring |
| CN117171675A (en) * | 2023-11-02 | 2023-12-05 | 北京建工环境修复股份有限公司 | Water environment microorganism detection method, system and medium based on multi-source data |
| CN117493777A (en) * | 2023-12-29 | 2024-02-02 | 成都秦川物联网科技股份有限公司 | Ultrasonic flowmeter data cleaning method, system and device based on Internet of things |
| CN117787570A (en) * | 2024-02-26 | 2024-03-29 | 山东公用环保科技集团有限公司 | Comprehensive evaluation method and system for heavy metal pollution of water environment |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110598953A (en) * | 2019-09-23 | 2019-12-20 | 哈尔滨工程大学 | Space-time correlation air quality prediction method |
| CN112229952A (en) * | 2020-09-30 | 2021-01-15 | 重庆科技学院 | Method for detecting toxic and harmful gas in chemical industrial park |
| WO2021051609A1 (en) * | 2019-09-20 | 2021-03-25 | 平安科技（深圳）有限公司 | Method and apparatus for predicting fine particulate matter pollution level, and computer device |
| WO2021174751A1 (en) * | 2020-03-02 | 2021-09-10 | 平安国际智慧城市科技股份有限公司 | Method, apparatus and device for locating pollution source on basis of big data, and storage medium |
| CN114997499A (en) * | 2022-06-07 | 2022-09-02 | 浙江工业大学 | Urban particulate matter concentration space-time prediction method under semi-supervised learning |
| CN116109462A (en) * | 2023-04-11 | 2023-05-12 | 乐百氏(广东)饮用水有限公司 | Pollution monitoring and early warning method and system for drinking water source area after natural disaster |
| CN116186566A (en) * | 2023-04-27 | 2023-05-30 | 北京建工环境修复股份有限公司 | Diffusion prediction method and system based on deep learning |
-
2023
- 2023-06-02 CN CN202310645532.XA patent/CN116384754B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021051609A1 (en) * | 2019-09-20 | 2021-03-25 | 平安科技（深圳）有限公司 | Method and apparatus for predicting fine particulate matter pollution level, and computer device |
| CN110598953A (en) * | 2019-09-23 | 2019-12-20 | 哈尔滨工程大学 | Space-time correlation air quality prediction method |
| WO2021174751A1 (en) * | 2020-03-02 | 2021-09-10 | 平安国际智慧城市科技股份有限公司 | Method, apparatus and device for locating pollution source on basis of big data, and storage medium |
| CN112229952A (en) * | 2020-09-30 | 2021-01-15 | 重庆科技学院 | Method for detecting toxic and harmful gas in chemical industrial park |
| CN114997499A (en) * | 2022-06-07 | 2022-09-02 | 浙江工业大学 | Urban particulate matter concentration space-time prediction method under semi-supervised learning |
| CN116109462A (en) * | 2023-04-11 | 2023-05-12 | 乐百氏(广东)饮用水有限公司 | Pollution monitoring and early warning method and system for drinking water source area after natural disaster |
| CN116186566A (en) * | 2023-04-27 | 2023-05-30 | 北京建工环境修复股份有限公司 | Diffusion prediction method and system based on deep learning |
Non-Patent Citations (1)
| Title |
|---|
| 王娜;张越非;龚有国;黄李洲;池汝安;: "化工园区大气污染事故风险模拟研究", 企业科技与发展, no. 01 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116754499A (en) * | 2023-08-23 | 2023-09-15 | 天津中科谱光信息技术有限公司 | Multi-topology node hyperspectral water quality parameter joint inversion method and related equipment |
| CN116754499B (en) * | 2023-08-23 | 2023-10-31 | 天津中科谱光信息技术有限公司 | Multi-topology node hyperspectral water quality parameter joint inversion method and related equipment |
| CN116821386A (en) * | 2023-08-30 | 2023-09-29 | 深圳巨湾科技有限公司 | Single characteristic evaluation control method based on multi-source heterogeneous data structuring |
| CN116821386B (en) * | 2023-08-30 | 2023-11-21 | 深圳巨湾科技有限公司 | Single characteristic evaluation control method based on multi-source heterogeneous data structuring |
| CN117171675A (en) * | 2023-11-02 | 2023-12-05 | 北京建工环境修复股份有限公司 | Water environment microorganism detection method, system and medium based on multi-source data |
| CN117171675B (en) * | 2023-11-02 | 2024-01-12 | 北京建工环境修复股份有限公司 | Water environment microorganism detection method, system and medium based on multi-source data |
| CN117493777A (en) * | 2023-12-29 | 2024-02-02 | 成都秦川物联网科技股份有限公司 | Ultrasonic flowmeter data cleaning method, system and device based on Internet of things |
| CN117493777B (en) * | 2023-12-29 | 2024-03-15 | 成都秦川物联网科技股份有限公司 | Ultrasonic flowmeter data cleaning method, system and device based on Internet of things |
| CN117787570A (en) * | 2024-02-26 | 2024-03-29 | 山东公用环保科技集团有限公司 | Comprehensive evaluation method and system for heavy metal pollution of water environment |
| CN117787570B (en) * | 2024-02-26 | 2024-05-17 | 山东公用环保科技集团有限公司 | Comprehensive evaluation method and system for heavy metal pollution of water environment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116384754B (en) | 2023-08-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116384754B (en) | Deep learning-based environmental pollution risk assessment method in chemical industry park | |
| CN116186566B (en) | Diffusion prediction method and system based on deep learning | |
| CN116385689B (en) | Visual information management method, system and medium for site pollution data | |
| CN116362570B (en) | Multi-dimensional pollution analysis method and system based on big data platform | |
| CN116362569B (en) | Multi-dimensional evaluation method and system for site pollution | |
| Soroudi et al. | A comprehensive multi-criteria decision making-based land capability assessment for municipal solid waste landfill sitting | |
| CN115034600A (en) | Early warning method and system for geological disaster monitoring | |
| CN116485202B (en) | Industrial pollution real-time monitoring method and system based on Internet of things | |
| CN116611546B (en) | Knowledge-graph-based landslide prediction method and system for target research area | |
| CN117171223A (en) | Microorganism culture scheme recommendation method and system in microorganism repair process | |
| CN115982646B (en) | Management method and system for multisource test data based on cloud platform | |
| Ramadan et al. | New environmental prediction model using fuzzy logic and neural networks | |
| CN116384775A (en) | Management method, system and storage medium of pollution site monitoring equipment | |
| CN113971213A (en) | Smart city management public information sharing system | |
| Wang et al. | Research on grey relational clustering model of multiobjective human resources based on time constraint | |
| CN116189010B (en) | Mine ecological identification early warning method and system based on satellite map | |
| CN116955648B (en) | Knowledge graph analysis method based on non-privacy data association | |
| CN117114105B (en) | Target object recommendation method and system based on scientific research big data information | |
| CN116341990B (en) | Knowledge management evaluation method and system for infrastructure engineering | |
| CN117633249B (en) | Method and device for constructing basic variable for SDGs space type monitoring index | |
| Wang et al. | A study of applications in archive management empowered by artificial intelligence | |
| CN118113736A (en) | Method and device for improving data retrieval performance stability of airborne database | |
| Pakhmode et al. | Regional Crime Data Analysis and Insights Using Fb Prophet | |
| Bertholdo et al. | Data mining techniques for water ecotoxicity classification for application on water resources management | |
| Qiao et al. | Energy Reports |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |