CN116681292A - Petrochemical harbor security risk analysis and responsibility division method based on deep learning - Google Patents
Petrochemical harbor security risk analysis and responsibility division method based on deep learning Download PDFInfo
- Publication number
- CN116681292A CN116681292A CN202310975656.4A CN202310975656A CN116681292A CN 116681292 A CN116681292 A CN 116681292A CN 202310975656 A CN202310975656 A CN 202310975656A CN 116681292 A CN116681292 A CN 116681292A
- Authority
- CN
- China
- Prior art keywords
- harbor
- risk
- model
- petrochemical
- initial
- 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
- 238000012502 risk assessment Methods 0.000 title claims abstract description 102
- 238000000034 method Methods 0.000 title claims abstract description 87
- 238000013135 deep learning Methods 0.000 title claims abstract description 66
- 238000012544 monitoring process Methods 0.000 claims abstract description 104
- 238000003860 storage Methods 0.000 claims abstract description 54
- 238000001514 detection method Methods 0.000 claims abstract description 41
- 238000009792 diffusion process Methods 0.000 claims abstract description 33
- 230000006870 function Effects 0.000 claims description 60
- 239000011159 matrix material Substances 0.000 claims description 35
- 238000000513 principal component analysis Methods 0.000 claims description 14
- 238000012795 verification Methods 0.000 claims description 13
- 238000012847 principal component analysis method Methods 0.000 claims description 12
- 238000005457 optimization Methods 0.000 claims description 8
- 238000012549 training Methods 0.000 claims description 5
- 230000014509 gene expression Effects 0.000 claims description 3
- 238000005192 partition Methods 0.000 claims description 3
- 230000005855 radiation Effects 0.000 claims description 3
- 238000010276 construction Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 8
- 238000004519 manufacturing process Methods 0.000 description 8
- 238000004364 calculation method Methods 0.000 description 7
- 238000011156 evaluation Methods 0.000 description 4
- 238000004590 computer program Methods 0.000 description 3
- 238000007689 inspection Methods 0.000 description 3
- 238000012797 qualification Methods 0.000 description 3
- 230000001360 synchronised effect Effects 0.000 description 3
- 230000009897 systematic effect Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 239000000428 dust Substances 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 230000001902 propagating effect Effects 0.000 description 1
- 238000009877 rendering Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000005728 strengthening Methods 0.000 description 1
- 238000004800 variational method Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
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
- 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
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/067—Enterprise or organisation modelling
-
- 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
- G06Q50/265—Personal security, identity or safety
-
- 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
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Educational Administration (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- General Health & Medical Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Security & Cryptography (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Primary Health Care (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a petrochemical harbor safety risk analysis and responsibility division method based on deep learning, which solves the problems that the prior art cannot predict the potential diffusion risk of a harbor storage tank region and early warn the risk of related responsibility personnel, thereby influencing the harbor safety operation, and comprises the following steps: acquiring petrochemical harbor operation scene and storage tank area related data, acquiring an operation environment model, and periodically acquiring real-time monitoring data through real-time monitoring points; constructing a harbor area risk analysis model, and identifying and detecting real-time monitoring data through a risk detection model to obtain a risk prediction result; according to the deep learning-based petrochemical harbor safety risk analysis and responsibility division method provided by the invention, potential risks of the petrochemical harbor are pre-warned through the risk detection model, so that the operation safety of petrochemical harbor personnel, equipment and storage tank regions is ensured.
Description
Technical Field
The invention belongs to the technical field of evaluation and analysis, and particularly relates to a petrochemical harbor safety risk analysis and responsibility division method based on deep learning.
Background
At present, casualties caused by production accidents in petrochemical harbor areas often occur, and personnel safety risks are extremely serious in control situation. The safety risk management and control of the port storage tank area are well gripped, the development of the port safety operation is guaranteed to be urgent, and the key of guaranteeing the safety production of the petrochemical port area is that the safety operation responsibility of various related personnel at all levels in the process of gripping the petrochemical port area is realized. In the petrochemical harbor operation process, the safety production responsibility of various personnel and various links of the operation of the safe production of the petrochemical harbor is evaluated and supervised, so that the method is one of effective means for promoting the safety responsibility in place, strengthening the management and control of the operation production risk of the petrochemical harbor and preventing the occurrence of safety accidents.
In the existing production management and safety early warning technical means, a systematic dividing method special for safety responsibilities of petrochemical harbor enterprises is not available, prediction of potential diffusion risks of harbor storage tank areas and risk early warning of related responsibilities cannot be conducted, harbor safety operation is affected, and based on the method, a petrochemical harbor safety risk analysis and responsibilities dividing method based on deep learning is provided.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a petrochemical harbor safety risk analysis and responsibility division method based on deep learning, which solves the problems that the prior art cannot predict the potential diffusion risk of a harbor storage tank region and early warn the risks of related responsibility people, thereby influencing the harbor safety operation.
In the existing production management and safety early warning technical means, a systematic dividing method of safety responsibility of petrochemical harbor enterprises is not specially used, prediction of potential diffusion risks of harbor storage tank areas and risk early warning of related responsibilities cannot be carried out, harbor safety operation is affected, based on the method, a petrochemical harbor safety risk analysis and responsibility dividing method based on deep learning is provided, the petrochemical harbor safety risk analysis and responsibility dividing method based on deep learning comprises the steps of acquiring petrochemical harbor operation scenes and storage tank area related data, obtaining an operation environment model, arranging real-time monitoring points based on the operation environment model, periodically acquiring real-time monitoring data through the real-time monitoring points, optimizing through principal component analysis, carrying out weighting improvement on the harbor risk analysis model, obtaining a risk detection model, loading the real-time monitoring data, and identifying and detecting the real-time monitoring data through the risk detection model, so that a risk prediction result is obtained. According to the deep learning-based petrochemical harbor safety risk analysis and responsibility division method, potential risks of the petrochemical harbor are pre-warned through the risk detection model, meanwhile, the potential risks of operation scenes, operation equipment, operation personnel or storage tank areas with large potential risks can be pre-warned according to the characteristics of multi-work cross operation of the petrochemical harbor, responsibility of relevant responsibility personnel can be pre-warned, and the operation safety of the petrochemical harbor personnel, equipment and storage tank areas is guaranteed.
The invention discloses a petrochemical harbor security risk analysis and responsibility division method based on deep learning, which comprises the following steps:
acquiring petrochemical harbor operation scenes and storage tank area related data, and performing three-dimensional modeling on a petrochemical harbor safety control area based on deep learning to obtain an operation environment model, wherein the operation environment model comprises model topology parameters, operation network nodes and operation network connecting edges;
loading an operation environment model, arranging real-time monitoring points based on the operation environment model, and periodically acquiring real-time monitoring data through the real-time monitoring points;
based on deep learning training, constructing a harbor risk analysis model, optimizing through principal component analysis, and carrying out weighting improvement on the harbor risk analysis model to obtain a risk detection model;
and loading real-time monitoring data, and identifying and detecting the real-time monitoring data through a risk detection model to obtain a risk prediction result, wherein the risk prediction result comprises a current risk value, a responsibility association degree and a risk diffusion trend graph, and pushing the risk prediction result to related responsible persons based on the risk prediction result.
Preferably, the method for three-dimensional modeling of the petrochemical harbor safety control area based on deep learning specifically includes:
acquiring petrochemical harbor operation scenes and storage tank area related data;
extracting three-dimensional coordinate parameters of a petrochemical harbor area operation scene and a storage tank area in the associated data, constructing an initial three-dimensional cloud point frame based on the three-dimensional coordinate parameters, and endowing the initial three-dimensional cloud point frame with initial model topology parameters, wherein the initial model topology parameters comprise initial accident occurrence rate, initial accident propagation rate and an accident initial responsibility dividing function;
loading an initial three-dimensional cloud point frame, traversing a safety accident data set in the associated data, and correcting the initial three-dimensional cloud point frame based on deep learning to obtain a working environment model.
Preferably, the method for correcting the initial three-dimensional cloud point frame based on the deep learning specifically comprises the following steps:
traversing a security incident data set in the associated data, calculating the importance of the nodes of the operation network and the nodes corresponding to the edges of the operation network and the importance of the edges based on the security incident data set, and calculating the importance of the incident based on the security incident data set;
constructing an importance judging matrix according to the node importance, the edge connection importance and the accident importance;
And calculating an importance judging matrix through a least square method, and determining the weight coefficients of the safety accident, the operation network node and the operation network connecting edge in the initial three-dimensional cloud point frame.
Preferably, the method for correcting the initial three-dimensional cloud point frame based on the deep learning specifically further comprises the following steps:
correcting the topological parameters of the initial model based on the safety accidents, the operation network nodes and the weight coefficients of the operation network connecting edges to obtain a corrected initial three-dimensional cloud point frame;
loading the corrected initial three-dimensional cloud point frame, and carrying out space grid division and associated responsibility division on the corrected initial three-dimensional cloud point frame to obtain a space grid set and responsibility division set;
and constructing a working environment model comprising the responsible partition set according to the position information of the space grid set.
Preferably, the method for obtaining the risk detection model by principal component analysis optimization and weighting improvement of the harbor district risk analysis model specifically comprises the following steps:
acquiring a harbor area risk analysis model;
generating a plurality of groups of influence factor initial comparison matrixes based on historical data, wherein the initial influence factor comparison matrixes are obtained based on three-scale comparison, and the initial influence factor comparison matrixes are obtained by combining the three scales Expressed as:
；
wherein the matrix is comparedComprises->Go->Column->For historical data->Number of initial influencing factors in>An impact hierarchy that is an initial impact factor;
calculating initial impact factor comparison matrixImportance index>，/>Calculated by formula (1);
（1）
wherein ,calculating by formula (2) the +_for the initial influence factor comparison value>Representing a comparison matrix->Is a comparison vector upper bound constant of +.>Representing a comparison matrix->Is a constant of the lower bound of the comparison vector;
（2）
constructing a port risk analysis model verification matrix, and verifying the compatibility of the port risk analysis model based on the model verification matrix.
Preferably, the method for obtaining the risk detection model by principal component analysis optimization and weighting improvement on the harbor district risk analysis model specifically further comprises:
verifying whether the compatibility of the harbor risk analysis model exceeds a preset compatibility threshold, and if so, performing weighting improvement on the harbor risk analysis model based on a principal component analysis method;
and if the compatibility of the harbor risk analysis model is smaller than a preset compatibility threshold, improving the model verification matrix, and verifying the compatibility of the harbor risk analysis model based on the model verification matrix again.
Preferably, the method for weighting and improving the harbor risk analysis model based on the principal component analysis method specifically comprises the following steps:
loading historical data and establishing a plurality of model improvement functions based on a principal component analysis method;
weighting and improving the harbor risk analysis model based on the model improvement function.
Preferably, the model improvement function comprises a monitoring point influence function, an accident propagation function and a monitoring point risk function;
the function expressions of the monitoring point influence function, the accident propagation function and the monitoring point risk function are respectively as follows:
（3）
wherein ,indicating the influence value of the monitoring point->To influence the gain factor>For monitoring points->And adjacent monitoring point->Is>For effective radiation distance of monitoring point, and monitoring point +.>The three-dimensional coordinates are obtained based on the operation environment model;
（4）
wherein ,representing accident spread value, ++>Is a propagation gain coefficient;
（5）
wherein ,representing a risk function value;
and weighting the harbor risk analysis model based on the model improvement function, wherein the weighting is improved as follows:
（6）
（7）
wherein ,weight coefficient representing weighted harbor district risk analysis model, < ->Weighting coefficients for monitoring point influence functions, +.>Weighting coefficients representing accident propagation functions, +. >A weighting coefficient representing a monitoring point risk function.
Preferably, the method for identifying and detecting real-time monitoring data through the risk detection model to obtain a risk prediction result specifically includes:
acquiring real-time monitoring data;
calculating a current risk value and a risk diffusion value corresponding to the current data based on the risk detection model;
and loading responsibility division groups in the operation environment model, calling related responsibility people based on the current data type, and calculating the responsibility association degree of the related responsibility people.
Preferably, the method for identifying and detecting real-time monitoring data through the risk detection model to obtain a risk prediction result specifically further includes:
and overlapping the risk diffusion values, and carrying out three-dimensional reconstruction on the risk diffusion values corresponding to the monitoring points to obtain a risk diffusion trend graph.
Compared with the prior art, the embodiment of the application has the following main beneficial effects:
according to the deep learning-based petrochemical harbor safety risk analysis and responsibility division method, potential risks of the petrochemical harbor are pre-warned through the risk detection model, meanwhile, the potential risks of operation scenes, operation equipment, operation personnel or storage tank areas with large potential risks can be pre-warned according to the characteristics of multi-work cross operation of the petrochemical harbor, responsibility of relevant responsibility personnel can be pre-warned, and the operation safety of the petrochemical harbor personnel, equipment and storage tank areas is guaranteed.
According to the method, the space grid division and the associated responsibility division are carried out on the corrected initial three-dimensional cloud point frame, so that construction of an operation environment model is facilitated, and meanwhile, the operation environment model can also provide support for real-time monitoring point arrangement, so that workload during safety risk analysis and responsibility division is remarkably reduced, the accuracy of safety risk analysis and responsibility division is ensured, and the operation environment model comprises a space grid set and responsibility division sets and can also provide support for harbor area risk analysis model construction.
Drawings
Fig. 1 is a schematic implementation flow chart of the petrochemical harbor security risk analysis and responsibility division method based on deep learning.
Fig. 2 shows a schematic implementation flow chart of a three-dimensional modeling method for a petrochemical harbor safety control region based on deep learning according to an embodiment of the present application.
Fig. 3 shows a schematic implementation flow chart of a method for correcting an initial three-dimensional cloud point frame based on deep learning according to an embodiment of the application.
Fig. 4 shows a schematic implementation flow chart of a method for obtaining a risk detection model by principal component analysis optimization and weighting improvement on a harbor risk analysis model according to an embodiment of the present application.
Fig. 5 shows a schematic implementation flow chart of a method for weighting improvement on a harbor risk analysis model based on a principal component analysis method according to an embodiment of the present application.
Fig. 6 is a schematic implementation flow chart of a method for identifying and detecting real-time monitoring data through a risk detection model to obtain a risk prediction result according to an embodiment of the present application.
Fig. 7 shows a schematic structural diagram of a petrochemical harbor safety risk analysis and responsibility division system based on deep learning according to an embodiment of the present application.
Fig. 8 shows a schematic structural diagram of an environmental model building module according to an embodiment of the present application.
Fig. 9 is a schematic structural diagram of a prediction result generation module according to an embodiment of the present application.
Detailed Description
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs; the terminology used in the description of the applications herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having" and any variations thereof in the description of the application and the claims and the description of the drawings above are intended to cover a non-exclusive inclusion. The terms first, second and the like in the description and in the claims or in the above-described figures, are used for distinguishing between different objects and not necessarily for describing a sequential or chronological order.
In the existing production management and safety early warning technical means, a systematic dividing method of safety responsibility of petrochemical harbor enterprises is not specially used, prediction of potential diffusion risks of harbor storage tank areas and risk early warning of related responsibilities cannot be carried out, harbor safety operation is affected, based on the method, a petrochemical harbor safety risk analysis and responsibility dividing method based on deep learning is provided, the petrochemical harbor safety risk analysis and responsibility dividing method based on deep learning comprises the steps of acquiring petrochemical harbor operation scenes and storage tank area related data, obtaining an operation environment model, arranging real-time monitoring points based on the operation environment model, periodically acquiring real-time monitoring data through the real-time monitoring points, optimizing through principal component analysis, carrying out weighting improvement on the harbor risk analysis model, obtaining a risk detection model, loading the real-time monitoring data, and identifying and detecting the real-time monitoring data through the risk detection model, so that a risk prediction result is obtained. According to the deep learning-based petrochemical harbor safety risk analysis and responsibility division method, potential risks of the petrochemical harbor are pre-warned through the risk detection model, meanwhile, the potential risks of operation scenes, operation equipment, operation personnel or storage tank areas with large potential risks can be pre-warned according to the characteristics of multi-work cross operation of the petrochemical harbor, responsibility of relevant responsibility personnel can be pre-warned, and the operation safety of the petrochemical harbor personnel, equipment and storage tank areas is guaranteed.
The embodiment of the invention provides a petrochemical harbor security risk analysis and responsibility division method based on deep learning, and fig. 1 shows a schematic implementation flow diagram of the petrochemical harbor security risk analysis and responsibility division method based on deep learning, wherein the petrochemical harbor security risk analysis and responsibility division method based on deep learning specifically comprises the following steps:
step S10, acquiring petrochemical harbor operation scenes and storage tank zone related data, and carrying out three-dimensional modeling on a petrochemical harbor safety control area based on deep learning to obtain an operation environment model, wherein the operation environment model comprises model topology parameters, operation network nodes and operation network connecting edges;
the petrochemical harbor operation scene association data includes, but is not limited to, operation data of operators using equipment, alarm equipment data, monitoring equipment data, equipment/facility operation data, cargo storage capacity, transportation capacity, petrochemical harbor traffic line configuration data, operation scene lighting data, dust pollution data, vibration, noise collection data and safety accident data sets, and the storage tank region association data includes, but is not limited to, harbor storage tank position data, harbor storage tank deformation data, harbor storage tank storage capacity and harbor storage tank maintenance data.
And S20, loading an operation environment model, arranging real-time monitoring points based on the operation environment model, and periodically acquiring real-time monitoring data through the real-time monitoring points.
In this embodiment, the operation environment model is a three-dimensional virtual model including a petrochemical harbor operation scene and storage tank facilities, operators, harbor storage tank location information and responsibilities, real-time monitoring of the petrochemical harbor operation scene can be achieved by constructing the operation environment model, the responsibilities associated with the current operation scene or storage tank can be presented in real time, meanwhile, the operation environment model can achieve real-time monitoring point arrangement based on the corrected initial three-dimensional cloud point frame, each group of petrochemical harbor operation scene and storage tank is at least correspondingly provided with a group of real-time monitoring points, and the data acquired by the real-time monitoring points areMultiple groups of real-time monitoring points->Composition of the monitoring data set->While monitoring data set->The three-dimensional coordinate of the current monitoring point in the operation environment model is included->。
Step S30, training and constructing a harbor risk analysis model based on deep learning, optimizing through principal component analysis, and carrying out weighting improvement on the harbor risk analysis model to obtain a risk detection model;
And S40, loading real-time monitoring data, and identifying and detecting the real-time monitoring data through a risk detection model to obtain a risk prediction result, wherein the risk prediction result comprises a current risk value, a responsibility association degree and a risk diffusion trend graph, and pushing the risk prediction result to related responsible persons based on the risk prediction result.
In this embodiment, the "responsible person" may be specific operators, management personnel, and inspection personnel of the petrochemical dock facility, or may be a petrochemical dock facility, a dock operator, an execution enterprise unit corresponding to a dock storage tank, a responsible enterprise unit, a management enterprise unit, and a supervision enterprise unit.
According to the deep learning-based petrochemical harbor safety risk analysis and responsibility division method, potential risks of the petrochemical harbor are pre-warned through the risk detection model, meanwhile, the potential risks of operation scenes, operation equipment, operation personnel or storage tank areas with large potential risks can be pre-warned according to the characteristics of multi-work cross operation of the petrochemical harbor, responsibility of relevant responsibility personnel can be pre-warned, and the operation safety of the petrochemical harbor personnel, equipment and storage tank areas is guaranteed.
The embodiment of the invention provides a method for three-dimensionally modeling a petrochemical harbor safety control area based on deep learning, and fig. 2 shows a schematic implementation flow chart of the method for three-dimensionally modeling the petrochemical harbor safety control area based on the deep learning, wherein the method for three-dimensionally modeling the petrochemical harbor safety control area based on the deep learning specifically comprises the following steps:
Step S101, acquiring petrochemical harbor operation scenes and storage tank area related data;
step S102, extracting three-dimensional coordinate parameters of a petrochemical harbor area operation scene and a storage tank area in the associated data, constructing an initial three-dimensional cloud point frame based on the three-dimensional coordinate parameters, and endowing the initial three-dimensional cloud point frame with initial model topology parameters, wherein the initial model topology parameters comprise initial accident occurrence rate, initial accident propagation rate and an accident initial responsibility dividing function;
and step S103, loading an initial three-dimensional cloud point frame, traversing a safety accident data set in the associated data, and correcting the initial three-dimensional cloud point frame based on deep learning to obtain a working environment model.
In this embodiment, the construction of the initial three-dimensional cloud point frame based on the three-dimensional coordinate parameters is implemented based on a communicable three-dimensional reconstruction electronic device, where the electronic device may be any stationary or mobile computing device capable of performing data processing, such as a laptop, a wearable device, or a stationary computing device such as a desktop computer, or other types of computing devices, the initial three-dimensional cloud point frame reconstructed by the three-dimensional reconstruction electronic device is an initial three-dimensional monitoring point rendering frame, which may be rendered in real time through background monitoring, and the initial model topology parameters include, but are not limited to, an initial accident rate, an initial accident propagation rate, and an accident initial responsibility dividing function, and may further include a monitoring point violation operation rate, a violation command rate, a skill level rate, a device maintenance qualification rate, a device failure rate, a device with disease operation rate, a tank damage rate, a tank leakage rate, a dust pollution control qualification rate, and a noise pollution control qualification rate, where the initial accident rate, and an accident initial responsibility dividing function are calculated by using the petrochemical port area operation obtained through deep learning and the associated data of the tank area.
The embodiment of the invention provides a method for correcting an initial three-dimensional cloud point frame based on deep learning, and fig. 3 shows a schematic implementation flow chart of the method for correcting the initial three-dimensional cloud point frame based on the deep learning, wherein the method for correcting the initial three-dimensional cloud point frame based on the deep learning specifically comprises the following steps:
step S1031, traversing a security incident data set in the associated data, calculating the importance of the nodes of the operation network and the nodes and the importance of the edges corresponding to the edges of the operation network based on the security incident data set, and calculating the importance of the incident based on the security incident data set;
in this embodiment, the node importance, the edge connection importance and the accident importance are all obtained by sampling a security accident data set in the self-correlation data, and the accident importance is a ratio of the frequency of occurrence of multiple groups of single accidents in the initial three-dimensional cloud point frame to the total number of nodes and the number of edge connection of the current initial three-dimensional cloud point frame through the sequencing calculation of the frequency of occurrence of the accidents of the nodes and the edge connection of the nodes in the security accident data set and the ratio of the frequency of occurrence of all security accidents in the initial three-dimensional cloud point frame.
Step S1032, constructing an importance judging matrix according to the node importance, the edge connection importance and the accident importance;
It should be noted that, the importance evaluation matrix is a matrix for evaluating the weight of the security accident, the operation network node and the operation network connecting edge in the initial three-dimensional cloud point frame, and in this embodiment, in order to verify the importance evaluation matrix, an variational method is adopted to construct an inspection matrix to inspect the compatibility of the importance evaluation matrix, and the inspection matrix may be a boolean matrix.
And step S1033, calculating an importance judgment matrix through a least square method, and determining the weight coefficients of the safety accident, the operation network node and the operation network connecting edge in the initial three-dimensional cloud point frame.
Step S1034, correcting the topological parameters of the initial model based on the security incidents, the operation network nodes and the weight coefficients of the operation network connecting edges to obtain an initial three-dimensional cloud point frame after correction;
step S1035, loading the corrected initial three-dimensional cloud point frame, and carrying out space grid division and associated responsibility division on the corrected initial three-dimensional cloud point frame to obtain a space grid set and responsibility division set;
step S1036, constructing a working environment model comprising the responsible partition group according to the position information of the space grid group.
In this embodiment, when the corrected initial three-dimensional cloud point frame is spatially grid-divided, the corrected initial three-dimensional cloud point frame is respectively divided along the following directions ，/>，/>The coordinate axes of the three directions are divided into a plurality of three-dimensional space grids, the three-dimensional space grids are numbered from a single direction, all operation network nodes and operation network connecting edge coordinate data are filled in the three-dimensional space grids, a single group of central points of all operation network nodes and operation network connecting edges in the three-dimensional space grids are calculated based on covariance matrixes, the operation network nodes and the operation network connecting edges are replaced based on the central points, real-time monitoring points are arranged based on an operation environment model in the step S20, and the real-time monitoring data are periodically acquired through the real-time monitoring points to provide support, wherein the real-time monitoring points are the central points in the three-dimensional space grids.
The generation of the responsibility division set is based on the calculation and generation of the security accident responsibility division function of the central point in each group of three-dimensional space grids in the space grid set.
Wherein the responsibility division function is presented by formula (8);
（8）
wherein ,is the responsibility value of the central point in the current three-dimensional space grid, and +.>Normalized weights for responsibility division function, +.>Is constant and is->A responsibility division feature vector representing a center point in the three-dimensional space grid;
（9）
wherein ,dividing a matrix for security accident liability people at central points in a three-dimensional space grid, and carrying out +.>The maximum number of safety accidents is the central point in the three-dimensional space grid.
According to the method, the space grid division and the associated responsibility division are carried out on the corrected initial three-dimensional cloud point frame, so that construction of an operation environment model is facilitated, and meanwhile, the operation environment model can also provide support for real-time monitoring point arrangement, so that workload during safety risk analysis and responsibility division is remarkably reduced, the accuracy of safety risk analysis and responsibility division is ensured, and the operation environment model comprises a space grid set and responsibility division sets and can also provide support for harbor area risk analysis model construction.
The embodiment of the application provides a method for obtaining a risk detection model by principal component analysis optimization and weighting improvement on a harbor risk analysis model, and fig. 4 shows a schematic implementation flow diagram of the method for obtaining the risk detection model by principal component analysis optimization and weighting improvement on the harbor risk analysis model, wherein the method for obtaining the risk detection model by principal component analysis optimization and weighting improvement on the harbor risk analysis model specifically comprises the following steps:
Step S201, obtaining a harbor risk analysis model;
in this embodiment, the building and training of the harbor risk analysis model is based on deep learning neural network, and the feature extraction rule of the harbor risk analysis model may be based on sigmoid function to extract features.
Step S202, generating a plurality of groups of influence factor initial comparison matrixes based on historical data, wherein the initial influence factor comparison matrixes are obtained based on three-scale comparison, and the initial influence factor comparison matrixes are obtained by comparing the three scalesExpressed as:
；
wherein the matrix is comparedComprises->Go->Column->For historical data->Number of initial influencing factors in>For the influence level of the initial influence factor, the matrix is compared +.>The calculation of (2) can be obtained by the sum of squares of the residuals of the least square method;
step S203, calculating an initial impact factor comparison matrixImportance index>，/>Calculated by formula (1);
（1）
wherein ,calculating by formula (2) the +_for the initial influence factor comparison value>Representing a comparison matrix->Is a comparison vector upper bound constant of +.>Can be 1, 2, 3 or 4, < >>Representing a comparison matrix->Is a comparison vector lower bound constant of +.>May be 1, 0.5, 0.25, 0.1;
（2）
step S204, constructing a harbor risk analysis model verification matrix, and verifying the compatibility of the harbor risk analysis model based on the model verification matrix.
In this embodiment, the model verification matrix may be a boolean matrix, and the preset compatibility threshold may be 0.8-1.
Step S205, verifying whether the compatibility of the harbor area risk analysis model exceeds a preset compatibility threshold;
step S206, if the number is larger than the preset compatibility threshold, weighting improvement is carried out on the harbor risk analysis model based on the principal component analysis method;
step S207, if the compatibility of the harbor risk analysis model is smaller than the preset compatibility threshold, the model verification matrix is improved, and the compatibility of the harbor risk analysis model is verified again based on the model verification matrix.
The embodiment of the invention provides a method for weighting and improving a harbor risk analysis model based on a principal component analysis method, and fig. 5 shows a schematic implementation flow chart of the method for weighting and improving the harbor risk analysis model based on the principal component analysis method, wherein the method for weighting and improving the harbor risk analysis model based on the principal component analysis method specifically comprises the following steps:
step S2061, loading historical data and establishing a plurality of model improvement functions based on a principal component analysis method;
step S2062, weighting and improving the harbor risk analysis model based on the model improvement function.
In this embodiment, the model improvement function includes a monitoring point impact function, an accident propagation function, a monitoring point risk function;
The function expressions of the monitoring point influence function, the accident propagation function and the monitoring point risk function are respectively as follows:
（3）
wherein ,indicating the influence value of the monitoring point->To influence the gain factor>For monitoring points->And adjacent monitoring point->Is>For effective radiation distance of monitoring point, and monitoring point +.>The three-dimensional coordinates are obtained based on the operation environment model;
（4）
wherein ,Representing accident spread value, ++>For propagating the gain factor, < > in this embodiment>，/>Are constant coefficients;
（5）
wherein ,representing a risk function value;
and weighting the harbor risk analysis model based on the model improvement function, wherein the weighting is improved as follows:
（6）
（7）
wherein ,weight coefficient representing weighted harbor district risk analysis model, < ->Weighting coefficients for monitoring point influence functions, +.>Weighting coefficients representing accident propagation functions, +.>Weighting coefficients representing monitoring point risk functions。
The embodiment of the invention provides a method for identifying and detecting real-time monitoring data through a risk detection model to obtain a risk prediction result, and fig. 6 shows a schematic implementation flow chart of the method for identifying and detecting real-time monitoring data through the risk detection model to obtain the risk prediction result, wherein the method for identifying and detecting real-time monitoring data through the risk detection model to obtain the risk prediction result specifically comprises the following steps:
Step S301, acquiring real-time monitoring data;
the real-time monitoring data is acquired based on the data acquisition terminal arranged at the monitoring point, and the data acquisition terminal is matched with various measuring tools with an automatic data acquisition function, so that automatic data acquisition is realized, and the data acquisition terminal has a wireless communication function.
Step S302, calculating a current risk value and a risk diffusion value corresponding to current data based on a risk detection model;
wherein the current risk value calculation is based on formula (5);
（5）
and risk spread valueCan be calculated according to the formula (6).
（6）
Step S303, loading responsibility division groups in the operation environment model, calling the associated responsibility people based on the current data type, and calculating the responsibility association degree of the associated responsibility people.
And S304, overlapping the risk diffusion values, and performing three-dimensional reconstruction on the risk diffusion values corresponding to the monitoring points to obtain a risk diffusion trend graph.
In the embodiment, the risk diffusion trend graph is obtained by carrying out three-dimensional reconstruction on the corresponding superposition risk diffusion value of the monitoring points, so that the accident risk of the monitoring points in the operation environment model can be presented in real time, and the risk diffusion trend graph is obtained, so that risk early warning and pushing are more visual.
The embodiment of the invention provides a petrochemical harbor safety risk analysis and responsibility division system based on deep learning, and fig. 7 shows a structural schematic diagram of the petrochemical harbor safety risk analysis and responsibility division system based on deep learning, wherein the petrochemical harbor safety risk analysis and responsibility division system based on deep learning specifically comprises:
the environment model construction module 100 is configured to obtain petrochemical harbor operation scene and storage tank area related data, and perform three-dimensional modeling on a petrochemical harbor safety control area based on deep learning to obtain an operation environment model, where the operation environment model includes model topology parameters, operation network nodes and operation network edges;
the monitoring point arrangement module 200 is used for loading an operation environment model, arranging real-time monitoring points based on the operation environment model and periodically acquiring real-time monitoring data through the real-time monitoring points;
the detection model construction module 300 is used for training and constructing a harbor risk analysis model based on deep learning, optimizing through principal component analysis and weighting the harbor risk analysis model to obtain a risk detection model;
The prediction result generation module 400 is configured to load real-time monitoring data, identify and detect the real-time monitoring data through a risk detection model, and obtain a risk prediction result, where the risk prediction result includes a current risk value, a responsibility association degree and a risk diffusion trend graph, and push the risk prediction result to related responsible persons based on the risk prediction result.
The invention provides a petrochemical harbor safety risk analysis and responsibility division method based on deep learning, which is used for pre-warning potential risks of a petrochemical harbor through a risk detection model, and simultaneously carrying out risk analysis and responsibility pre-warning of related responsible persons on a working scene, working equipment, working personnel or a storage tank area with high potential risks according to the characteristics of multi-species cross operation of the petrochemical harbor, so that the operation safety of the petrochemical harbor personnel, equipment and a storage tank area is ensured.
The embodiment of the invention provides an environment model building module 100, fig. 8 shows a schematic structural diagram of the environment model building module 100, and the environment model building module 100 specifically includes:
The relevant data acquisition unit 110 is used for acquiring relevant data of a petrochemical harbor area operation scene and a storage tank area;
the frame construction unit 120 is configured to extract three-dimensional coordinate parameters of a petrochemical harbor operation scene and a storage tank region in the associated data, construct an initial three-dimensional cloud point frame based on the three-dimensional coordinate parameters, and assign an initial model topology parameter to the initial three-dimensional cloud point frame, where the initial model topology parameter includes an initial accident occurrence rate, an initial accident propagation rate, and an accident initial responsibility dividing function;
the frame correction unit 130 is used for loading an initial three-dimensional cloud point frame, traversing a safety accident data set in the associated data, and correcting the initial three-dimensional cloud point frame based on deep learning to obtain a working environment model.
The embodiment of the invention provides a prediction result generation module 400, and fig. 9 shows a schematic structural diagram of the prediction result generation module 400, where the prediction result generation module 400 specifically includes:
a diffusion value calculation unit 410, wherein the diffusion value calculation unit 410 calculates a current risk value and a risk diffusion value corresponding to the current data based on the risk detection model;
The responsibility association degree judging unit 420 is used for loading responsibility division groups in the operation environment model, calling associated responsibility people based on the current data type, and calculating responsibility association degree of the associated responsibility people;
and the diffusion trend graph generating unit 430 is used for superposing the risk diffusion values, and performing three-dimensional reconstruction on the risk diffusion values corresponding to the monitoring points to obtain a risk diffusion trend graph.
In this embodiment, the diffusion value calculating unit 410, the responsibility association degree judging unit 420 and the diffusion trend graph generating unit 430 implement data interaction by adopting a 5G or WiFi communication mode, and the responsibility association degree judging unit 420 calculates the responsibility association degree of the associated responsible person through the responsibility division function calculation, and the responsibility division function is presented through the formula (8);
（8）
wherein ,is the responsibility value of the central point in the current three-dimensional space grid, and +.>Normalized weights for responsibility division function, +.>Is constant and is->The responsibility for representing the center point within the three-dimensional spatial grid divides the feature vector.
In another aspect of the embodiments of the present invention, a computer-readable storage medium is also provided. The computer readable storage medium stores computer program instructions executable by a processor. Which when executed, performs the method of any of the embodiments described above.
Meanwhile, in another aspect, an embodiment of the present application further provides a computer device, where the computer device includes a memory and a processor, and the memory stores a computer program, where the computer program is executed by the processor to implement the method of any one of the foregoing embodiments.
The memory is used as a non-volatile computer readable storage medium and can be used for storing non-volatile software programs, non-volatile computer executable programs and modules, such as program instructions/modules corresponding to the petrochemical harbor safety risk analysis and responsibility division method based on deep learning in the embodiment of the application. The memory may include a memory program area and a memory data area, wherein the memory program area may store an operating system, at least one application program required for a function; the storage data area may store data created by use of a petrochemical harbor security risk analysis and responsibility division method based on deep learning, and the like. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory optionally includes memory remotely located relative to the processor, the remote memory being connectable to the local module through a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
Finally, it should be noted that the computer-readable storage media (e.g., memory) herein can be either volatile memory or nonvolatile memory, or can include both volatile and nonvolatile memory. By way of example, and not limitation, nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM), which acts as external cache memory. By way of example, and not limitation, RAM may be available in a variety of forms such as synchronous RAM (DRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), and Direct Rambus RAM (DRRAM). The storage devices of the disclosed aspects are intended to comprise, without being limited to, these and other suitable types of memory.
In summary, the invention provides a petrochemical harbor safety risk analysis and responsibility division method based on deep learning, which pre-warns potential risks of a petrochemical harbor through a risk detection model, and meanwhile, can pre-warn the risks of operation scenes, operation equipment, operation personnel or storage tank areas with high potential risks and responsibility of related responsible persons according to the characteristics of multi-species cross operation of the petrochemical harbor, thereby ensuring the operation safety of the personnel, equipment and storage tank areas in the petrochemical harbor.
It should be noted that, for simplicity of description, the foregoing embodiments are all illustrated as a series of acts, but it should be understood by those skilled in the art that the present invention is not limited by the order of acts, as some steps may be performed in other order or concurrently in accordance with the present invention. Further, those skilled in the art will also appreciate that the embodiments described in the specification are all preferred embodiments, and that the acts and modules referred to are not necessarily required for the present invention.
The above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the scope of the present invention. It will be apparent that the described embodiments are merely some, but not all, embodiments of the invention. Based on these embodiments, all other embodiments that may be obtained by one of ordinary skill in the art without inventive effort are within the scope of the invention. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art may still combine, add or delete features of the embodiments of the present invention or make other adjustments according to circumstances without any conflict, so as to obtain different technical solutions without substantially departing from the spirit of the present invention, which also falls within the scope of the present invention.
Claims (10)
1. The petrochemical harbor safety risk analysis and responsibility division method based on the deep learning is characterized by comprising the following steps of:
acquiring petrochemical harbor operation scenes and storage tank area related data, and performing three-dimensional modeling on a petrochemical harbor safety control area based on deep learning to obtain an operation environment model, wherein the operation environment model comprises model topology parameters, operation network nodes and operation network connecting edges;
loading an operation environment model, arranging real-time monitoring points based on the operation environment model, and periodically acquiring real-time monitoring data through the real-time monitoring points;
based on deep learning training, constructing a harbor risk analysis model, optimizing through principal component analysis, and carrying out weighting improvement on the harbor risk analysis model to obtain a risk detection model;
and loading real-time monitoring data, and identifying and detecting the real-time monitoring data through a risk detection model to obtain a risk prediction result, wherein the risk prediction result comprises a current risk value, a responsibility association degree and a risk diffusion trend graph, and pushing the risk prediction result to related responsible persons based on the risk prediction result.
2. The petrochemical harbor safety risk analysis and responsibility division method based on deep learning as set forth in claim 1, wherein: the method for carrying out three-dimensional modeling on the petrochemical harbor safety control area based on deep learning specifically comprises the following steps:
acquiring petrochemical harbor operation scenes and storage tank area related data;
extracting three-dimensional coordinate parameters of a petrochemical harbor area operation scene and a storage tank area in the associated data, constructing an initial three-dimensional cloud point frame based on the three-dimensional coordinate parameters, and endowing the initial three-dimensional cloud point frame with initial model topology parameters, wherein the initial model topology parameters comprise initial accident occurrence rate, initial accident propagation rate and an accident initial responsibility dividing function;
loading an initial three-dimensional cloud point frame, traversing a safety accident data set in the associated data, and correcting the initial three-dimensional cloud point frame based on deep learning to obtain a working environment model.
3. The petrochemical harbor safety risk analysis and responsibility division method based on deep learning as set forth in claim 2, wherein: the method for correcting the initial three-dimensional cloud point frame based on the deep learning specifically comprises the following steps:
traversing a security incident data set in the associated data, calculating the importance of the nodes of the operation network and the nodes corresponding to the edges of the operation network and the importance of the edges based on the security incident data set, and calculating the importance of the incident based on the security incident data set;
Constructing an importance judging matrix according to the node importance, the edge connection importance and the accident importance;
and calculating an importance judging matrix through a least square method, and determining the weight coefficients of the safety accident, the operation network node and the operation network connecting edge in the initial three-dimensional cloud point frame.
4. The petrochemical harbor safety risk analysis and responsibility division method based on deep learning as set forth in claim 2, wherein: the method for correcting the initial three-dimensional cloud point frame based on the deep learning specifically further comprises the following steps:
correcting the topological parameters of the initial model based on the safety accidents, the operation network nodes and the weight coefficients of the operation network connecting edges to obtain a corrected initial three-dimensional cloud point frame;
loading the corrected initial three-dimensional cloud point frame, and carrying out space grid division and associated responsibility division on the corrected initial three-dimensional cloud point frame to obtain a space grid set and responsibility division set;
and constructing a working environment model comprising the responsible partition set according to the position information of the space grid set.
5. The petrochemical harbor safety risk analysis and responsibility division method based on deep learning according to claim 4, wherein: the method for obtaining the risk detection model by principal component analysis optimization and weighting improvement of the harbor risk analysis model specifically comprises the following steps:
Acquiring a harbor area risk analysis model;
generating a plurality of groups of influence factor initial comparison matrixes based on historical data, wherein the initial influence factor comparison matrixes are obtained based on three-scale comparison, and the initial influence factor comparison matrixes are obtained by combining the three scalesExpressed as:
；
wherein the matrix is comparedComprises->Go->Column->For historical data->Number of initial influencing factors in>An impact hierarchy that is an initial impact factor;
calculating initial impact factor comparison matrixImportance index>，/>Calculated by formula (1);
（1）
wherein ,calculating by formula (2) the +_for the initial influence factor comparison value>Representing a comparison matrix->Is a comparison vector upper bound constant of +.>Representing a comparison matrix->Is a constant of the lower bound of the comparison vector;
（2）
constructing a port risk analysis model verification matrix, and verifying the compatibility of the port risk analysis model based on the model verification matrix.
6. The petrochemical harbor safety risk analysis and responsibility division method based on deep learning according to claim 5, wherein: the method for obtaining the risk detection model by principal component analysis optimization and weighting improvement on the harbor risk analysis model specifically further comprises the following steps:
verifying whether the compatibility of the harbor risk analysis model exceeds a preset compatibility threshold, and if so, performing weighting improvement on the harbor risk analysis model based on a principal component analysis method;
And if the compatibility of the harbor risk analysis model is smaller than a preset compatibility threshold, improving the model verification matrix, and verifying the compatibility of the harbor risk analysis model based on the model verification matrix again.
7. The petrochemical harbor safety risk analysis and responsibility division method based on deep learning according to claim 6, wherein: the method for carrying out weighting improvement on the harbor district risk analysis model based on the principal component analysis method comprises the following steps:
loading historical data and establishing a plurality of model improvement functions based on a principal component analysis method;
weighting and improving the harbor risk analysis model based on the model improvement function.
8. The petrochemical harbor safety risk analysis and responsibility division method based on deep learning according to claim 7, wherein: the model improvement function comprises a monitoring point influence function, an accident propagation function and a monitoring point risk function;
the function expressions of the monitoring point influence function, the accident propagation function and the monitoring point risk function are respectively as follows:
（3）
wherein ,indicating the influence value of the monitoring point->To influence the gain factor>For monitoring points->And adjacent monitoring point->Is>For effective radiation distance of monitoring point, and monitoring point +. >The three-dimensional coordinates are obtained based on the operation environment model;
（4）
wherein ,representing accident spread value, ++>Is a propagation gain coefficient;
（5）
wherein ,representing a risk function value;
and weighting the harbor risk analysis model based on the model improvement function, wherein the weighting is improved as follows:
（6）
（7）
wherein ,weight coefficient representing weighted harbor district risk analysis model, < ->Weighting coefficients for monitoring point influence functions, +.>Representing accident transmissionWeighting coefficients of the broadcast function->A weighting coefficient representing a monitoring point risk function.
9. The petrochemical harbor safety risk analysis and responsibility division method based on deep learning according to claim 8, wherein: the method for identifying and detecting the real-time monitoring data through the risk detection model to obtain the risk prediction result specifically comprises the following steps:
acquiring real-time monitoring data;
calculating a current risk value and a risk diffusion value corresponding to the current data based on the risk detection model;
and loading responsibility division groups in the operation environment model, calling related responsibility people based on the current data type, and calculating the responsibility association degree of the related responsibility people.
10. The petrochemical harbor security risk analysis and responsibility division method based on deep learning as set forth in claim 9, wherein: the method for identifying and detecting the real-time monitoring data through the risk detection model to obtain the risk prediction result specifically further comprises the following steps:
And overlapping the risk diffusion values, and carrying out three-dimensional reconstruction on the risk diffusion values corresponding to the monitoring points to obtain a risk diffusion trend graph.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310975656.4A CN116681292B (en) | 2023-08-04 | 2023-08-04 | Petrochemical harbor security risk analysis and responsibility division method based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310975656.4A CN116681292B (en) | 2023-08-04 | 2023-08-04 | Petrochemical harbor security risk analysis and responsibility division method based on deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116681292A true CN116681292A (en) | 2023-09-01 |
| CN116681292B CN116681292B (en) | 2024-04-09 |
Family
ID=87784120
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310975656.4A Active CN116681292B (en) | 2023-08-04 | 2023-08-04 | Petrochemical harbor security risk analysis and responsibility division method based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116681292B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117474343A (en) * | 2023-12-27 | 2024-01-30 | 中交第一航务工程勘察设计院有限公司 | Petrochemical harbor danger source safety risk early warning method, petrochemical harbor danger source safety risk early warning device, petrochemical harbor danger source safety risk early warning equipment and storage medium |
| CN117786604A (en) * | 2024-02-23 | 2024-03-29 | 交通运输部水运科学研究所 | Fused safety toughness measurement analysis system based on harbor disaster data |
| CN118095825A (en) * | 2023-12-21 | 2024-05-28 | 华能济南黄台发电有限公司 | Safety risk grading management and control system for power plant production |
| CN118411038A (en) * | 2024-07-03 | 2024-07-30 | 江苏达科数智技术有限公司 | Petrochemical industry data visual interaction management method and system |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105787677A (en) * | 2016-03-31 | 2016-07-20 | 交通运输部水运科学研究所 | Petrochemical port area accident risk area quantitative assessment method |
| KR20220055555A (en) * | 2020-10-26 | 2022-05-04 | 씨드로닉스(주) | Method and device for monitoring harbor and ship |
| CN114862169A (en) * | 2022-04-24 | 2022-08-05 | 东南大学 | BIM-based tunnel engineering safety risk monitoring system and method |
| CN114997757A (en) * | 2022-08-04 | 2022-09-02 | 交通运输部水运科学研究所 | Port petrochemical region transportation risk early warning method and system based on cascading failure |
| CN115600900A (en) * | 2022-10-28 | 2023-01-13 | 交通运输部水运科学研究所（Cn） | Safety risk assessment method, system and storage medium for petrochemical port area |
| WO2023061039A1 (en) * | 2021-10-13 | 2023-04-20 | 中通服和信科技有限公司 | Tailing pond risk monitoring and early-warning system based on internet of things |
| CN116013036A (en) * | 2023-03-08 | 2023-04-25 | 中国水产科学研究院南海水产研究所 | Safe intelligent early warning method, system and medium in fishing port construction process |
| CN116258241A (en) * | 2022-12-13 | 2023-06-13 | 山东大学 | Space-time correlation-based supervision environment risk prediction method and system |
| CN116385690A (en) * | 2023-06-05 | 2023-07-04 | 四川云控交通科技有限责任公司 | BIM model-based three-dimensional operation and maintenance management and control platform and management and control method thereof |
-
2023
- 2023-08-04 CN CN202310975656.4A patent/CN116681292B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105787677A (en) * | 2016-03-31 | 2016-07-20 | 交通运输部水运科学研究所 | Petrochemical port area accident risk area quantitative assessment method |
| KR20220055555A (en) * | 2020-10-26 | 2022-05-04 | 씨드로닉스(주) | Method and device for monitoring harbor and ship |
| WO2023061039A1 (en) * | 2021-10-13 | 2023-04-20 | 中通服和信科技有限公司 | Tailing pond risk monitoring and early-warning system based on internet of things |
| CN114862169A (en) * | 2022-04-24 | 2022-08-05 | 东南大学 | BIM-based tunnel engineering safety risk monitoring system and method |
| CN114997757A (en) * | 2022-08-04 | 2022-09-02 | 交通运输部水运科学研究所 | Port petrochemical region transportation risk early warning method and system based on cascading failure |
| CN115600900A (en) * | 2022-10-28 | 2023-01-13 | 交通运输部水运科学研究所（Cn） | Safety risk assessment method, system and storage medium for petrochemical port area |
| CN116258241A (en) * | 2022-12-13 | 2023-06-13 | 山东大学 | Space-time correlation-based supervision environment risk prediction method and system |
| CN116013036A (en) * | 2023-03-08 | 2023-04-25 | 中国水产科学研究院南海水产研究所 | Safe intelligent early warning method, system and medium in fishing port construction process |
| CN116385690A (en) * | 2023-06-05 | 2023-07-04 | 四川云控交通科技有限责任公司 | BIM model-based three-dimensional operation and maintenance management and control platform and management and control method thereof |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118095825A (en) * | 2023-12-21 | 2024-05-28 | 华能济南黄台发电有限公司 | Safety risk grading management and control system for power plant production |
| CN117474343A (en) * | 2023-12-27 | 2024-01-30 | 中交第一航务工程勘察设计院有限公司 | Petrochemical harbor danger source safety risk early warning method, petrochemical harbor danger source safety risk early warning device, petrochemical harbor danger source safety risk early warning equipment and storage medium |
| CN117786604A (en) * | 2024-02-23 | 2024-03-29 | 交通运输部水运科学研究所 | Fused safety toughness measurement analysis system based on harbor disaster data |
| CN117786604B (en) * | 2024-02-23 | 2024-04-26 | 交通运输部水运科学研究所 | Fused safety toughness measurement analysis system based on harbor disaster data |
| CN118411038A (en) * | 2024-07-03 | 2024-07-30 | 江苏达科数智技术有限公司 | Petrochemical industry data visual interaction management method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116681292B (en) | 2024-04-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116681292B (en) | Petrochemical harbor security risk analysis and responsibility division method based on deep learning | |
| Shadabfar et al. | Resilience-based design of infrastructure: Review of models, methodologies, and computational tools | |
| Chen et al. | A methodology for quantitative vulnerability assessment of coupled multi-hazard in Chemical Industrial Park | |
| US11906112B2 (en) | Methods for safety management of compressors in smart gas pipeline network and internet of things systems thereof | |
| CN116700199B (en) | Factory production control method and system based on digital twin technology | |
| CN111310803B (en) | Environment data processing method and device | |
| KR20220084751A (en) | System and method for comprehensive diagnosis of possible disasters in decayed urban regeneration areas | |
| CN111428917A (en) | Soil pollution prediction method and system for heavy metal stable pollution source | |
| CN116384255B (en) | Park dangerous situation perception method and system based on multi-source data fusion | |
| CN114511243A (en) | Method and system for dynamically evaluating fire risk based on Internet of things monitoring | |
| Sengupta et al. | Challenges for introducing risk assessment into land use planning decisions in an Indian context | |
| CN114445041B (en) | Method and system for emergency treatment of in-transit accidents of hazardous chemical substances | |
| CN110399537B (en) | Artificial intelligence technology-based warning situation space-time prediction method | |
| CN115883981A (en) | Camera layout method and device | |
| CN115689281A (en) | Factory hazard risk management and control method and device | |
| CN117454677B (en) | Wharf construction simulation method and system based on three-dimensional simulation technology | |
| RU2536657C1 (en) | System for evaluating safety and efficiency of design solutions to ensure safety of hazardous production facility | |
| CN111105582B (en) | Forest fire prevention monitoring method and system, computer equipment and readable storage medium | |
| CN117113788A (en) | Method for predicting structural failure of steel normal-pressure storage tank under coupling of fire and explosion | |
| CN114202261B (en) | Village-level industrial park fire risk directed graph depicting method and device | |
| Sun et al. | Coal mine gas safety evaluation based on adaptive weighted least squares support vector machine and improved dempster–shafer evidence theory | |
| CN108804947A (en) | Method and apparatus for determining the white list operated to database | |
| CN118657358B (en) | Automatic inspection plan generation method and device based on artificial intelligence for intelligent construction site | |
| CN115047543B (en) | Power transmission corridor rainfall early warning method and system | |
| Xu et al. | Research on executing equipment control algorithm based on multi-sensor pipe gallery |
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 |