CN117391451A - Digital safety control system for steel - Google Patents
Digital safety control system for steel Download PDFInfo
- Publication number
- CN117391451A CN117391451A CN202311465458.XA CN202311465458A CN117391451A CN 117391451 A CN117391451 A CN 117391451A CN 202311465458 A CN202311465458 A CN 202311465458A CN 117391451 A CN117391451 A CN 117391451A
- Authority
- CN
- China
- Prior art keywords
- data
- model
- safety
- historical
- platform
- 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.)
- Pending
Links
- 229910000831 Steel Inorganic materials 0.000 title claims abstract description 21
- 239000010959 steel Substances 0.000 title claims abstract description 20
- 238000012549 training Methods 0.000 claims abstract description 42
- 238000011156 evaluation Methods 0.000 claims abstract description 21
- 238000012502 risk assessment Methods 0.000 claims abstract description 12
- 230000009471 action Effects 0.000 claims description 24
- 238000004519 manufacturing process Methods 0.000 claims description 24
- 238000012545 processing Methods 0.000 claims description 16
- 238000004422 calculation algorithm Methods 0.000 claims description 10
- 238000001914 filtration Methods 0.000 claims description 10
- 238000012544 monitoring process Methods 0.000 claims description 10
- 238000009499 grossing Methods 0.000 claims description 9
- 238000011217 control strategy Methods 0.000 claims description 6
- 238000012216 screening Methods 0.000 claims description 6
- 230000002159 abnormal effect Effects 0.000 claims description 4
- 230000007613 environmental effect Effects 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 4
- 238000012217 deletion Methods 0.000 claims description 3
- 230000037430 deletion Effects 0.000 claims description 3
- 239000002184 metal Substances 0.000 claims description 3
- 229910052751 metal Inorganic materials 0.000 claims description 3
- 231100000572 poisoning Toxicity 0.000 claims description 3
- 230000000607 poisoning effect Effects 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 230000009469 supplementation Effects 0.000 claims description 3
- 238000001931 thermography Methods 0.000 claims description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 3
- 238000007726 management method Methods 0.000 description 35
- 238000000034 method Methods 0.000 description 20
- 238000002485 combustion reaction Methods 0.000 description 11
- 238000004364 calculation method Methods 0.000 description 6
- 230000004048 modification Effects 0.000 description 6
- 238000012986 modification Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 3
- 230000001276 controlling effect Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 239000007789 gas Substances 0.000 description 3
- 239000001301 oxygen Substances 0.000 description 3
- 229910052760 oxygen Inorganic materials 0.000 description 3
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000007123 defense Effects 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000002265 prevention Effects 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000000540 analysis of variance Methods 0.000 description 1
- 238000012098 association analyses Methods 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013523 data management Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 229910052742 iron Inorganic materials 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000003032 molecular docking Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000013450 outlier detection Methods 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 238000011282 treatment Methods 0.000 description 1
- 238000012800 visualization Methods 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
- G06Q50/04—Manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Data Mining & Analysis (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Game Theory and Decision Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Operations Research (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Quality & Reliability (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Manufacturing & Machinery (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of steel safety, in particular to a steel digital safety management and control system, which comprises a management and control platform, a database module, an Internet of things platform, a model training platform, a data center platform, a data acquisition module, an edge control module and an execution module, wherein the management and control platform, the database module, the Internet of things platform and the model training platform jointly form a cloud side frame, the data center platform and the edge control module jointly form an edge side frame, the data acquisition module and the execution module jointly form an end side frame, and the cloud side frame and the edge side frame interact through a virtual private network VPN; the cloud edge end architecture is integrally adopted, and can effectively utilize resources trained by the model on the cloud; the system introduces a security case database, provides support for security risk assessment and evaluation, and can form a dynamic security risk evaluation method.
Description
Technical Field
The invention relates to the technical field of steel safety, in particular to a steel digital safety control system.
Background
In 2020, an action plan of 'industrial Internet+safe production' is issued, and the action plan indicates that 'industrial Internet+safe production' is used in the safe production through the fusion of the industrial Internet, so that the sensing, monitoring, early warning, disposal and evaluation capabilities of industrial safe production are enhanced, the transition from static analysis to dynamic sensing, post emergency to pre-prevention and single-point prevention and control to global joint defense of the safe production is accelerated, and the intrinsic safety level of the industrial production is improved.
At present, a plurality of domestic iron and steel enterprises start to build intelligent security protection management and control platforms, but the intelligent security protection management and control platforms are different in architecture, are mostly deployed locally, lack cloud edge end architecture, lack cloud training means based on visual analysis of security production scenes, are single in security state data sensing means, lack security association analysis, lack dynamic risk assessment of global joint defense and are combined with fusion analysis of an industry security case library.
Disclosure of Invention
Object of the invention
The invention aims to provide a steel digital safety control system which adopts a cloud edge architecture, can train and update a visual analysis model of a safety production scene and performs data comparison by combining a safety case.
(II) technical scheme
In order to solve the above problems, the present invention provides a steel digital safety control system, comprising: the system comprises a management and control platform, a database module, an internet of things platform, a model training platform, a data center platform, a data acquisition module, an edge control module and an execution module, wherein the management and control platform, the database module, the internet of things platform and the model training platform jointly form a cloud side framework, the data center platform and the edge control module jointly form an edge side framework, the data acquisition module and the execution module jointly form an end side framework, and the cloud side framework and the edge side framework interact through a virtual private network VPN;
the data acquisition module can acquire the safety state data of the man-machine object in real time;
the data center station can receive the safety state data and perform standardized processing on the safety state data;
the internet of things platform can receive and manage the standardized safety state data;
the database module can store all historical safety cases;
the management and control platform can receive the safety state data managed by the Internet of things platform and compare, analyze and output an management and control strategy with the corresponding historical safety cases which are called from the database module;
the model training platform can receive the safety state data managed by the Internet of things platform, perform model training on the safety state data and generate a personalized custom model;
the edge control module can receive the personalized customization model to control the execution module;
the execution module executes an action based on the instruction sent by the edge control module.
In another aspect of the present invention, preferably, the method for performing normalization processing on the security state data by the data center includes: and carrying out data deduplication, data deletion supplementation, noise data filtering, abnormal data processing, data smoothing processing and characteristic data selection on the safety state data, and carrying out formatting unified processing on all available data to form standard data.
In another aspect of the present invention, the safety status data preferably includes equipment status data, personnel and vehicle positioning data, ring monitoring data, vision monitoring based data, molten metal thermal imaging data, poisoning media, fire protection data, road traffic data, environmental data, and data for a digital twinning system.
In another aspect of the present invention, the execution module preferably includes an alarm, an electrically operated valve, a fan, a water pump, a fire protection facility, and a safety interlock.
In another aspect of the present invention, preferably, the model training platform includes a model training unit, a model management unit, a model deployment unit, a model display unit, and a model application feedback unit;
the model training unit can receive data managed by the Internet of things platform, perform model training on the data and generate a personalized custom model;
the model management unit can manage a model library, receive the personalized customization model and store the personalized customization model;
the model deployment unit can call the personalized custom model of the model management unit and send the personalized custom model to the edge control module;
the model display unit can call the personalized customization model of the model management unit so as to share in the cloud;
the model application feedback unit can enable the model management module to update a module stored by the model management module and enable the data center platform to optimize a standardized mode of the model application feedback unit based on feedback information of the execution module.
In another aspect of the present invention, preferably, the manner in which the management platform retrieves the corresponding historical security cases from the database module includes:
first screening is carried out on all the historical safety cases stored in the database module by using a first-level identifier, all the first-screened historical safety cases are recorded as first-level historical safety cases, and the first-level identifier at least comprises a target name identifier, a target category identifier and a target address identifier;
performing secondary screening on all the primary historical safety cases by using a secondary identifier, and marking all the secondary screened historical safety cases as secondary historical safety cases, wherein the secondary identifier at least comprises a target scene identifier, a target object identifier, a time identifier and a season identifier;
if the number of the secondary historical safety cases is greater than 1, calling out the secondary historical safety case with the highest similarity by using a preset similarity algorithm;
if the number of the secondary historical security cases is equal to 1, the secondary historical security cases are called;
and if the number of the second-level historical safety cases is equal to 0, sequencing the priority of the first-level historical safety cases, and calling the first-level historical safety case with the highest priority.
In another aspect of the present invention, preferably, the method for retrieving the second-level historical security case with the highest similarity by using a preset similarity algorithm includes:
respectively carrying out keyword division on the secondary historical safety cases, carrying out keyword division on the standardized data,
and calculating the similarity between the standardized data and the secondary historical safety case by using the following formula, and calling the secondary historical safety case with the highest similarity:
wherein S is k Representing the similarity of the normalized data to the kth secondary historical security case, cos (c) i ,d kj ) And (5) representing cosine similarity between the ith keyword of the standardized data and the jth keyword of the kth secondary historical safety case.
In another aspect of the present invention, preferably, the prioritizing the first-level historical security cases, and the manner of retrieving the first-level historical security case with the highest priority includes:
wherein w is r Expressed as the r-th level of historyThe priority score for the whole case, m represents the total number of secondary identifications,the z-th secondary identifier expressed as the r-th primary historical security case is present, and the z-th secondary identifier is present +.>1, absent->Is 0.
In another aspect of the present invention, preferably, the output management and control strategy is to output a production risk evaluation value, and a manner of outputting the production risk evaluation value includes:
acquiring a security model of the retrieved historical security case, the security model consisting of a plurality of features;
carrying out feature extraction on the standardized data according to the security model;
the production risk evaluation value is calculated and obtained by using the following formula:
wherein Q represents a production risk evaluation value;mean value of each characteristic of the normalized data, < > is represented>Representing the average value of each characteristic of the security model; x is X k Representing the value of the characteristic k in the normalized data, Y k Representing the value of the feature k in the security model, n representing the total number of features.
In another aspect of the present invention, preferably, the output control strategy is to output an action of a target execution subject, and a manner of outputting the action of the target execution subject includes:
extracting action standard data of the target execution subject from the retrieved historical security cases;
comparing action data corresponding to the target execution subject in the standardized data with the action standard data;
and outputting the specific action mode of the target execution body according to the compared difference value.
(III) beneficial effects
The technical scheme of the invention has the following beneficial technical effects:
the invention integrally adopts a cloud side end architecture, wherein the data acquisition module and the execution module form an end side architecture together, so as to realize acquisition of field data and execution according to control instructions; the data center table and the edge control module form an edge frame structure together, so that the data can be subjected to standardized processing and the execution module can be controlled; the management and control platform, the database module, the Internet of things platform and the model training platform jointly form a cloud side frame, model training can be carried out, a personalized custom model can be generated, and the adaptability of the model is higher; the cloud side end architecture can effectively utilize resources trained by the model on the cloud; and a security case database is introduced to provide support for security risk assessment and evaluation, and a dynamic security risk evaluation method can be formed.
Drawings
FIG. 1 is a schematic diagram of the overall structure of a management and control system according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of the architecture of a model training platform according to one embodiment of the present invention.
Detailed Description
The objects, technical solutions and advantages of the present invention will become more apparent by the following detailed description of the present invention with reference to the accompanying drawings. It should be understood that the description is only illustrative and is not intended to limit the scope of the invention. In addition, in the following description, descriptions of well-known structures and techniques are omitted so as not to unnecessarily obscure the present invention.
It will be apparent that the described embodiments are some, but not all, embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In addition, the technical features of the different embodiments of the present invention described below may be combined with each other as long as they do not collide with each other.
Examples
Fig. 1 shows an overall structure diagram of an embodiment of the present invention, as shown in fig. 1, including: the system comprises a management and control platform, a database module, an Internet of things platform, a model training platform, a data center platform, a data acquisition module, an edge control module and an execution module;
wherein the management and control platform, the database module, the Internet of things platform and the model training platform jointly form a cloud side frame,
the specific content of the management and control platform is not limited herein, and optionally, in this embodiment, the management and control platform may receive the security status data managed by the internet of things platform and compare, analyze, and output a management and control policy with the corresponding historical security cases invoked from the database module; the intelligent management of the safety of the whole factory is realized;
the specific content of the database module is not limited, in this embodiment, optionally, the database module can store all historical safety cases, and the specific source of all the historical safety cases is not limited, in this embodiment, the safety case big database of the Wuhan safety Environment protection institute of Steel group in the introduction of all the historical safety cases provides support for the safety risk assessment and evaluation of the platform;
the specific content of the internet of things platform is not limited, and optionally, in this embodiment, the internet of things platform can receive and manage security state data subjected to standardized processing; the safety state data comprise safety state data of various man-machine objects in a factory; the safety state data source is data acquired by an enterprise side, and the Internet of things platform is responsible for docking with an edge side data center, has the functions of equipment access, equipment management, data management, safety, application enabling and the like, and realizes the acquisition and management of various Internet of things data;
without limiting the specific content of the model training platform, fig. 2 shows a schematic structural diagram of the model training platform according to an embodiment of the present invention, and as shown in fig. 2, optionally, in this embodiment, the model training platform may receive the security state data managed by the internet of things platform, perform model training on the security state data, and generate a personalized custom model; the model training platform can preprocess mass data, label in a semi-automatic mode, train in a large scale in a distributed mode and generate an automatic model, and the model training platform can realize the deployment capability of end-side-cloud according to needs so as to realize quick creation and deployment of the model; further, in this embodiment, the model training platform includes a model training unit, a model management unit, a model deployment unit, a model display unit, and a model application feedback unit;
the model training unit can receive data managed by the Internet of things platform, perform model training on the data and generate a personalized custom model; the data can be subjected to model training and a personalized custom model can be generated through online coding, a preset algorithm, a distributed cluster, model visualization and an automatic learning function;
the model management unit can manage a model library, receive the personalized customization model and store the personalized customization model; model tracing and precision tracking can be realized;
the model deployment unit can call the personalized custom model of the model management unit and send the personalized custom model to the edge control module; online service, edge service and batch service can be realized;
the model display unit can call the personalized customization model of the model management unit so as to share in the cloud; algorithm sharing, model sharing and data sharing can be realized;
the model application feedback unit can enable the model management module to update a module stored by the model management module and enable the data center platform to optimize a standardized mode of the model application feedback unit based on feedback information of the execution module;
taking a valve for controlling the gas combustion of the hot blast stove as an example, a model training platform acquires data such as the pressure, the temperature, the flow, the oxygen content, the residual oxygen content and the like of a gas pipe network on the hot blast stove site from an internet platform, and the model training platform further trains according to the combustion data in the selected historical safety case and the corresponding safety model and the combustion model, realizes the processes of optimizing, iterating and simulating simulation, trains a better combustion model, updates the model, and issues the updated model to a site edge controller to control the opening of the valve, thereby realizing a better combustion effect. The method comprises the specific steps that under the condition of the same combustion temperature, the lower the consumed energy in unit time is, the better the effect is, and the worse the effect is, the existing model is executed in an execution module, the model training platform obtains a better combustion model through distributed training on the cloud, the better combustion model is issued to an edge controller to be executed, the continuously calculated consumed energy in unit time is used as a feedback value, the model is issued again during training on the cloud, and the effect of closed loop optimization is achieved repeatedly in the mode. When the industrial control of the hot blast furnace changes, the combustion model is also regulated to update, one path of feedback is fed back to a data center platform to propose data optimization suggestions, for example, the conventional exhaust gas temperature is used as data with larger weight value for controlling the combustion model, the residual oxygen value for lifting is found to be used as the weight value for controlling the combustion model to be input after continuous model training, and the input factors of the model are corrected to form a dynamic model framework.
The data center table and the edge control module form an edge frame structure together;
the specific content of the data center is not limited herein, and in this embodiment, optionally, the data center can receive the security state data and perform standardization processing on the security state data; further, the data center processes the collected multi-source and multi-protocol data to form standardized data and stores the standardized data into a database module; further, the data center station performs data deduplication, data deletion supplementation, noise data filtering, abnormal data processing, data smoothing processing and characteristic data selection on the safety state data, and performs formatting unified processing on all available data to form standard data;
specifically, performing the deduplication operation on the security state data includes removing duplicate data records; processing missing data, wherein the missing data is data acquired according to a time sequence, and the data lacks data in a certain time period, namely the missing data, and the missing data can be filled by interpolation, average value or other related data;
the filtering of the safety state data noise comprises filtering the collected data by utilizing a noise filtering method comprising smoothing filtering, median filtering, gaussian filtering and the like, so as to remove noise.
Performing outlier detection on the safety state data comprises a statistical-based method, a Z-Score-based method, a distance-based method, a KNN (K-nearest neighbor) method and a model-based method, and a box graph, wherein certain outliers possibly existing in the data, such as extreme values which are obviously different from other data, are removed; the outliers may be caused by equipment failure, operational errors, and the like. In order to ensure the accuracy of data analysis and modeling, abnormal values need to be detected and processed;
the data smoothing process is performed on the safety state data, including smoothing process by using a sliding average and exponential smoothing, and some device data may have noise or fluctuation characteristics, so that in order to improve the stability and the readability of the data, a data smoothing technology may be used to perform a smoothing operation on the data.
And selecting characteristic data of the safety state data: the method comprises the steps of selecting features with higher correlation or importance from original data, wherein the feature selection mainly adopts statistical methods such as correlation coefficients, analysis of variance and machine learning methods such as decision trees, random forests and the like.
Further, the normalizing the security state data includes: and (3) formulating a standard of data acquisition data format, converting different types of data into standard data through a format converter, performing data format unification operation on the data, and unifying the different types of data into the same data format.
The specific content of the edge control module is not limited herein, and optionally, in this embodiment, the edge control module can receive the personalized customization model to control the execution module; the edge controller has an edge calculation function, has the attribute of a thick client relative to the fact that the edge controller sinks to the field side under the action of calculation force, can control an actuator, and completes various production safety protection actions;
the data acquisition module and the execution module form an end side frame together, and the cloud side frame and the side frame interact through a virtual private network VPN;
the specific content of the data acquisition module is not limited herein, and optionally, in this embodiment, the data acquisition module may acquire the security state data in real time; the specific content of the safety state data is not limited herein, and includes equipment state data, personnel and vehicle positioning data, moving ring monitoring data, vision monitoring-based data, molten metal thermal imaging data, poisoning media, fire protection data, road traffic data, environmental data, and data of a digital twin system;
the specific content of the execution module is not limited herein, and optionally, in this embodiment, the execution module performs an action based on an instruction sent by the edge control module, and further, the execution module includes an alarm, an electric valve, a fan, a water pump, a fire protection facility, and a safety interlock device.
Further, in an embodiment of the present invention, the manner in which the management and control platform retrieves the corresponding historical security case from the database module includes:
first screening is carried out on all the historical safety cases stored in the database module by using a first-level identifier, all the first-screened historical safety cases are recorded as first-level historical safety cases, and the first-level identifier at least comprises a target name identifier, a target category identifier and a target address identifier;
performing secondary screening on all the primary historical safety cases by using a secondary identifier, and marking all the secondary screened historical safety cases as secondary historical safety cases, wherein the secondary identifier at least comprises a target scene identifier, a target object identifier, a time identifier and a season identifier;
if the number of the secondary historical safety cases is greater than 1, calling out the secondary historical safety case with the highest similarity by using a preset similarity algorithm;
if the number of the secondary historical security cases is equal to 1, the secondary historical security cases are called;
and if the number of the second-level historical safety cases is equal to 0, sequencing the priority of the first-level historical safety cases, and calling the first-level historical safety case with the highest priority.
Optionally, in this embodiment, the method for calling out the second-level historical security case with the highest similarity by using a preset similarity algorithm includes:
respectively carrying out keyword division on the secondary historical safety cases, carrying out keyword division on the standardized data,
and calculating the similarity between the standardized data and the secondary historical safety case by using the following formula, and calling the secondary historical safety case with the highest similarity:
wherein S is k Representing the similarity of the normalized data to the kth secondary historical security case, cos (c) i ,d kj ) And (5) representing cosine similarity between the ith keyword of the standardized data and the jth keyword of the kth secondary historical safety case. The two-level historical safety cases with highest similarity are selected through semantic similarity calculation, matching accuracy is better, focuses of the cases and standardized data can be well mastered, calculation is simple, and online calculation time can be shortened.
Optionally, in this embodiment, the priority sorting of the first-level historical security cases, and the manner of calling the first-level historical security case with the highest priority includes:
wherein w is r A priority score expressed as an r-th primary historical security case, m represents the total number of secondary identifications,the z-th secondary identifier expressed as the r-th primary historical security case is present, and the z-th secondary identifier is present +.>1, absent->Is 0. By selecting the case with the most secondary identifications among the primary historical security cases without all the secondary identifications, the algorithm is simple and the storage is carried out according to the identifications. The accuracy of sorting is high in whether to carry out the sorting or not, and the algorithm is simple.
Optionally, in this embodiment, the outputting the output management and control policy is outputting a production risk evaluation value, and a manner of outputting the production risk evaluation value includes:
acquiring a security model of the retrieved historical security case, the security model consisting of a plurality of features;
carrying out feature extraction on the standardized data according to the security model;
the production risk evaluation value is calculated and obtained by using the following formula:
wherein Q represents a production risk evaluation value;mean value of each characteristic of the normalized data, < > is represented>Representing the average value of each characteristic of the security model; x is X k Representing the value of the characteristic k in the normalized data, Y k Representing the value of the feature k in the security model, n representing the total number of features. The integral judgment is carried out by comparing and calculating with the characteristics of the safety model, and the integral judgment is not carried out by a single characteristicJudging that the integrity is higher, and the acquired production risk evaluation value has better reference value.
Optionally, in this embodiment, the outputting the output control policy is outputting an action of the target execution body, and a manner of outputting the action of the target execution body includes:
extracting action standard data of the target execution subject from the retrieved historical security cases;
comparing action data corresponding to the target execution subject in the standardized data with the action standard data;
and outputting the specific action mode of the target execution body according to the compared difference value.
Specifically, the flow is as follows: the method comprises the steps that vision acquisition is carried out on a certain operation platform through a camera of a data acquisition module, acquired data are sent to a data center platform, the data center platform carries out cleaning filtering and other treatments on the data, the treated data are sent to an Internet of things platform, a management and control platform is used for adjusting an adapted safety case according to the steps from historical safety cases in a database module, after a corresponding safety model of the adjusted safety case is obtained, feature extraction and calculation are carried out to obtain a production risk evaluation value, the production risk evaluation value can indicate that the position, the temperature and the like exceed the range, if the position of the operation platform is found to deviate and exceeds a set threshold, a system gives pre-judgment, an accident risk such as falling exists, and the system gives pre-warning, wherein the pre-warning can be in the form of pre-warning through sound and light, or APP and the like, and personnel are reminded of the attention.
For example, the data of the motor bearing Wen Zhen is collected to form a history curve and trend, the vibration value is found to reach a certain set value, the risk of bearing ball fracture can be considered, and the system gives pre-judgment and early warning. And notifying operation staff that the bearing reaches the replacement period.
Further, key equipment and key technological parameters are monitored and early-warned intelligently, and various safety risk monitoring data acquired on the cleaned site are acquired from an internet platform, so that the real and accurate requirements are met.
The data in the safety case is that accident cases happen in the past year, the safety case has stronger reference property, actual various safety risk scenes in the field are compared with case big data, the most probable accident type and accident outbreak point in the scenes can be analyzed, the system further emphasizes that the pre-judging weight value is increased for the risks, more accurate risk assessment and judgment are formed, and safety production is guided. The safety risk level has a certain standard in the industrial safety field, and is corrected by referring to the standard and combining with the characteristics of high cold and high dust and the like in northeast areas of metallurgical enterprises, so that the safety risk level suitable for the enterprises and the northeast metallurgical industry is formed.
Further, the safety risk monitoring content and the emphasis point can be found to be changed under different working conditions, for example, the blast furnace tuyere inspection system, and the infrared thermometer is used for finding that the temperature of a certain blast furnace air supply device has a tendency to be increased, namely, the blast furnace air supply device is predicted to be burnt. In this case, the influence factor is increased for the safety risk assessment of the blast furnace tuyere, the risk level is promoted in the security risk management and control item of the whole plant and is displayed on the main picture, so that a dynamic safety risk assessment method is formed, and when a certain procedure is overhauled, the emphasis of the safety risk assessment is found to change. The safety risk monitoring to equipment is weakened when equipment is shut down and overhauled, the safety risk factors of personnel overhaul are emphasized more, different working condition scenes are formed, and different safety risk coping strategies, namely dynamic evaluation, are adopted.
According to the embodiment, resources trained by the model on the cloud can be effectively utilized through the cloud side framework; the adaptation degree of the model is improved, a safety case big database of the medium steel group Wuhan safety environmental protection institute is introduced, the accuracy of safety risk assessment and evaluation is effectively improved, and quick response and decision guidance are provided for safety management and control; an Internet of things platform and a model training platform are adopted to solve the problem of short plates locally applied to the traditional safety control platform;
and a dynamic safety risk assessment method can be formed by combining the working condition change and the maintenance plan.
It is to be understood that the above-described embodiments of the present invention are merely illustrative of or explanation of the principles of the present invention and are in no way limiting of the invention. Accordingly, any modification, equivalent replacement, improvement, etc. made without departing from the spirit and scope of the present invention should be included in the scope of the present invention. Furthermore, the appended claims are intended to cover all such changes and modifications that fall within the scope and boundary of the appended claims, or equivalents of such scope and boundary.
The invention has been described above with reference to the embodiments thereof. However, these examples are for illustrative purposes only and are not intended to limit the scope of the present invention. The scope of the invention is defined by the appended claims and equivalents thereof. Various alternatives and modifications can be made by those skilled in the art without departing from the scope of the invention, and such alternatives and modifications are intended to fall within the scope of the invention.
Although embodiments of the present invention have been described in detail, it should be understood that various changes, substitutions, and alterations can be made hereto without departing from the spirit and scope of the invention.
It is apparent that the above examples are given by way of illustration only and are not limiting of the embodiments. Other variations or modifications of the above teachings will be apparent to those of ordinary skill in the art. It is not necessary here nor is it exhaustive of all embodiments. While still being apparent from variations or modifications that may be made by those skilled in the art are within the scope of the invention.
Claims (10)
1. A steel digital safety control system, comprising: the system comprises a management and control platform, a database module, an internet of things platform, a model training platform, a data center platform, a data acquisition module, an edge control module and an execution module, wherein the management and control platform, the database module, the internet of things platform and the model training platform jointly form a cloud side framework, the data center platform and the edge control module jointly form an edge side framework, the data acquisition module and the execution module jointly form an end side framework, and the cloud side framework and the edge side framework interact through a virtual private network VPN;
the data acquisition module can acquire the safety state data of the man-machine object in real time;
the data center station can receive the safety state data and perform standardized processing on the safety state data;
the internet of things platform can receive and manage the standardized safety state data;
the database module can store all historical safety cases;
the management and control platform can receive the safety state data managed by the Internet of things platform and compare, analyze and output an management and control strategy with the corresponding historical safety cases which are called from the database module;
the model training platform can receive the safety state data managed by the Internet of things platform, perform model training on the safety state data and generate a personalized custom model;
the edge control module can receive the personalized customization model to control the execution module;
the execution module executes an action based on the instruction sent by the edge control module.
2. The steel digitized safety control system of claim 1 wherein the means for normalizing the safety state data by the data center station comprises: and carrying out data deduplication, data deletion supplementation, noise data filtering, abnormal data processing, data smoothing processing and characteristic data selection on the safety state data, and carrying out formatting unified processing on all available data to form standard data.
3. The steel digitized safety management and control system of claim 1 wherein the safety status data comprises equipment status data, personnel and vehicle positioning data, ring monitoring data, vision monitoring based data, molten metal thermal imaging data, poisoning media, fire protection data, road traffic data, environmental data, and data for a digital twinning system.
4. The steel digital safety control system of claim 1, wherein the execution module comprises an alarm, an electrically operated valve, a blower, a water pump, a fire protection facility, and a safety interlock.
5. The steel digitized safety control system of claim 1 wherein the model training platform comprises a model training unit, a model management unit, a model deployment unit, a model display unit, and a model application feedback unit;
the model training unit can receive data managed by the Internet of things platform, perform model training on the data and generate a personalized custom model;
the model management unit can manage a model library, receive the personalized customization model and store the personalized customization model;
the model deployment unit can call the personalized custom model of the model management unit and send the personalized custom model to the edge control module;
the model display unit can call the personalized customization model of the model management unit so as to share in the cloud;
the model application feedback unit can enable the model management module to update a module stored by the model management module and enable the data center platform to optimize a standardized mode of the model application feedback unit based on feedback information of the execution module.
6. The steel digitized security management and control system of claim 1 wherein the manner in which the management and control platform retrieves the corresponding historical security case from the database module comprises:
first screening is carried out on all the historical safety cases stored in the database module by using a first-level identifier, all the first-screened historical safety cases are recorded as first-level historical safety cases, and the first-level identifier at least comprises a target name identifier, a target category identifier and a target address identifier;
performing secondary screening on all the primary historical safety cases by using a secondary identifier, and marking all the secondary screened historical safety cases as secondary historical safety cases, wherein the secondary identifier at least comprises a target scene identifier, a target object identifier, a time identifier and a season identifier;
if the number of the secondary historical safety cases is greater than 1, calling out the secondary historical safety case with the highest similarity by using a preset similarity algorithm;
if the number of the secondary historical security cases is equal to 1, the secondary historical security cases are called;
and if the number of the second-level historical safety cases is equal to 0, sequencing the priority of the first-level historical safety cases, and calling the first-level historical safety case with the highest priority.
7. The digitized steel safety control system of claim 6 wherein the means for retrieving the highest similarity secondary historical safety case using a predetermined similarity algorithm comprises:
respectively carrying out keyword division on the secondary historical safety cases, carrying out keyword division on the standardized data,
and calculating the similarity between the standardized data and the secondary historical safety case by using the following formula, and calling the secondary historical safety case with the highest similarity:
wherein S is k Representing the similarity of the normalized data to the kth secondary historical security case, cos (c) i ,d kj ) And (5) representing cosine similarity between the ith keyword of the standardized data and the jth keyword of the kth secondary historical safety case.
8. The steel digitized security management system of claim 6 wherein prioritizing the first level historical security cases and retrieving the highest priority first level historical security cases comprises:
wherein w is r A priority score expressed as an r-th primary historical security case, m represents the total number of secondary identifications,the z-th secondary identifier expressed as the r-th primary historical security case is present, and the z-th secondary identifier is present +.>1, in the absence ofIs 0.
9. The steel digitized safety control system of claim 1 wherein the output management and control strategy is to output a production risk assessment value, the way to output the production risk assessment value comprising:
acquiring a security model of the retrieved historical security case, the security model consisting of a plurality of features;
carrying out feature extraction on the standardized data according to the security model;
the production risk evaluation value is calculated and obtained by using the following formula:
wherein Q represents a production risk evaluation value;mean value of each characteristic of the normalized data, < > is represented>Representing the average value of each characteristic of the security model; x is X k Representing the value of the characteristic k in the normalized data, Y k Representing the value of the feature k in the security model, n representing the total number of features.
10. The steel digitized safety control system of claim 1 wherein the output control strategy is to output the actions of the target execution subject in a manner that includes:
extracting action standard data of the target execution subject from the retrieved historical security cases;
comparing action data corresponding to the target execution subject in the standardized data with the action standard data;
and outputting the specific action mode of the target execution body according to the compared difference value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311465458.XA CN117391451A (en) | 2023-11-07 | 2023-11-07 | Digital safety control system for steel |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311465458.XA CN117391451A (en) | 2023-11-07 | 2023-11-07 | Digital safety control system for steel |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117391451A true CN117391451A (en) | 2024-01-12 |
Family
ID=89466446
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311465458.XA Pending CN117391451A (en) | 2023-11-07 | 2023-11-07 | Digital safety control system for steel |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117391451A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117692610A (en) * | 2024-02-02 | 2024-03-12 | 建龙西林钢铁有限公司 | AR workshop inspection system |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109066819A (en) * | 2018-09-25 | 2018-12-21 | 中国人民解放军军事科学院国防工程研究院 | A kind of idle work optimization method of the power distribution network based on case reasoning |
| CN110719210A (en) * | 2019-12-05 | 2020-01-21 | 赣江新区智慧物联研究院有限公司 | Industrial equipment predictive maintenance method based on cloud edge cooperation |
| CN111784191A (en) * | 2020-07-17 | 2020-10-16 | 塔盾信息技术(上海)有限公司 | Safety management aid decision-making system based on industrial intelligence |
| CN113627735A (en) * | 2021-07-16 | 2021-11-09 | 宁夏建设投资集团有限公司 | Early warning method and system for safety risk of engineering construction project |
| CN115412947A (en) * | 2022-08-26 | 2022-11-29 | 武汉烽火技术服务有限公司 | Fault simulation method and system based on digital twin and AI algorithm |
| CN115456842A (en) * | 2022-09-14 | 2022-12-09 | 国网辽宁省电力有限公司信息通信分公司 | Emergency scheme generation method, system and storage medium under federated learning framework |
| CN116896495A (en) * | 2023-07-17 | 2023-10-17 | 上海电器科学研究所（集团）有限公司 | Bian Yun collaborative model dynamic training method |
| CN116957341A (en) * | 2023-08-11 | 2023-10-27 | 华北理工大学 | Intelligent safety risk management and control system based on steel mill |
| CN116993151A (en) * | 2023-09-25 | 2023-11-03 | 中钢集团武汉安全环保研究院有限公司 | Blast furnace tapping area personnel risk management and control system based on Internet of things |
-
2023
- 2023-11-07 CN CN202311465458.XA patent/CN117391451A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109066819A (en) * | 2018-09-25 | 2018-12-21 | 中国人民解放军军事科学院国防工程研究院 | A kind of idle work optimization method of the power distribution network based on case reasoning |
| CN110719210A (en) * | 2019-12-05 | 2020-01-21 | 赣江新区智慧物联研究院有限公司 | Industrial equipment predictive maintenance method based on cloud edge cooperation |
| CN111784191A (en) * | 2020-07-17 | 2020-10-16 | 塔盾信息技术(上海)有限公司 | Safety management aid decision-making system based on industrial intelligence |
| CN113627735A (en) * | 2021-07-16 | 2021-11-09 | 宁夏建设投资集团有限公司 | Early warning method and system for safety risk of engineering construction project |
| CN115412947A (en) * | 2022-08-26 | 2022-11-29 | 武汉烽火技术服务有限公司 | Fault simulation method and system based on digital twin and AI algorithm |
| CN115456842A (en) * | 2022-09-14 | 2022-12-09 | 国网辽宁省电力有限公司信息通信分公司 | Emergency scheme generation method, system and storage medium under federated learning framework |
| CN116896495A (en) * | 2023-07-17 | 2023-10-17 | 上海电器科学研究所（集团）有限公司 | Bian Yun collaborative model dynamic training method |
| CN116957341A (en) * | 2023-08-11 | 2023-10-27 | 华北理工大学 | Intelligent safety risk management and control system based on steel mill |
| CN116993151A (en) * | 2023-09-25 | 2023-11-03 | 中钢集团武汉安全环保研究院有限公司 | Blast furnace tapping area personnel risk management and control system based on Internet of things |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117692610A (en) * | 2024-02-02 | 2024-03-12 | 建龙西林钢铁有限公司 | AR workshop inspection system |
| CN117692610B (en) * | 2024-02-02 | 2024-04-26 | 建龙西林钢铁有限公司 | AR workshop inspection system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108154291B (en) | Power supply plant performance index evaluation method and system based on strategy library | |
| CN117391451A (en) | Digital safety control system for steel | |
| CN107885170A (en) | The method of safety in production control device, system and management and control of keeping the safety in production | |
| CN104267711A (en) | Running state monitoring and failure diagnosis method for tobacco logistics system | |
| CN116543531B (en) | Data center system for centralized management and displaying underground monitoring and control data | |
| CN108776452B (en) | Special equipment field maintenance monitoring method and system | |
| CN115330129A (en) | Enterprise safety risk early warning analysis method and system | |
| CN117523451A (en) | Video data analysis system and method based on intelligent security technology | |
| CN114581046A (en) | Data processing method based on big data and cloud computing and big data service platform | |
| CN109460403A (en) | The real-time dynamic quantitative scaling method of the stifled ash of air preheater | |
| CN112232235A (en) | Intelligent factory remote monitoring method and system based on 5G | |
| CN107191666A (en) | Towards the safety cut-off valve performance monitoring method and system of Natural Gas Station | |
| CN115574945B (en) | Control system and method for inspection robot of submerged arc furnace | |
| CN116055521A (en) | Inspection system and image recognition method for electric inspection robot | |
| CN112966867B (en) | Transformer early warning method based on PSO-BP neural network and quartile method | |
| CN114413055A (en) | Operation and maintenance device and method for subway tunnel fan interlocking air valve | |
| CN117391455B (en) | Major hazard source safety control system for iron and steel enterprises | |
| CN117078022B (en) | Steel plant personnel risk early warning system and method based on multi-source data fusion | |
| CN112949986A (en) | Steam power system operation stability evaluation method and system, electronic equipment and storage medium | |
| CN106909976A (en) | A kind of intelligent building facilities management system | |
| CN117853078B (en) | Unmanned aerial vehicle-based dynamic safety alarm system and method for construction site of building engineering | |
| CN116821403B (en) | Intelligent operation and maintenance method and system for factory equipment | |
| Kabir et al. | Combining drone-based monitoring and machine learning for online reliability evaluation of wind turbines | |
| CN117711065A (en) | Power consumption operation prediction protection method based on behavior habit analysis | |
| CN116844300A (en) | Hidden danger tracing system and method for current overload of metal material |
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 |