CN115099684A - Enterprise safety production management system and management method thereof - Google Patents
Enterprise safety production management system and management method thereof Download PDFInfo
- Publication number
- CN115099684A CN115099684A CN202210841063.4A CN202210841063A CN115099684A CN 115099684 A CN115099684 A CN 115099684A CN 202210841063 A CN202210841063 A CN 202210841063A CN 115099684 A CN115099684 A CN 115099684A
- Authority
- CN
- China
- Prior art keywords
- feature
- enterprise
- vector
- matrix
- production system
- 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
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 296
- 238000007726 management method Methods 0.000 title claims abstract description 103
- 238000009826 distribution Methods 0.000 claims abstract description 27
- 238000012937 correction Methods 0.000 claims abstract description 23
- 238000003062 neural network model Methods 0.000 claims abstract description 20
- 239000013598 vector Substances 0.000 claims description 227
- 239000011159 matrix material Substances 0.000 claims description 165
- 238000013527 convolutional neural network Methods 0.000 claims description 15
- 238000012545 processing Methods 0.000 claims description 15
- 238000010586 diagram Methods 0.000 claims description 13
- 238000010276 construction Methods 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 6
- 230000004913 activation Effects 0.000 claims description 4
- 238000011176 pooling Methods 0.000 claims description 4
- 238000000034 method Methods 0.000 abstract description 12
- 238000005516 engineering process Methods 0.000 abstract description 7
- 238000013467 fragmentation Methods 0.000 abstract description 7
- 238000006062 fragmentation reaction Methods 0.000 abstract description 7
- 230000008569 process Effects 0.000 abstract description 6
- 238000013135 deep learning Methods 0.000 abstract description 5
- 238000012544 monitoring process Methods 0.000 description 10
- 238000003860 storage Methods 0.000 description 9
- 230000008859 change Effects 0.000 description 6
- 238000004590 computer program Methods 0.000 description 6
- 230000001788 irregular Effects 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 238000013507 mapping Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000007792 addition Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012806 monitoring device Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000006798 recombination Effects 0.000 description 1
- 238000005215 recombination Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- 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
-
- 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)
- Economics (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Primary Health Care (AREA)
- Manufacturing & Machinery (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (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 application relates to the field of intelligent production management, and particularly discloses an enterprise safety production management system and a management method thereof, wherein hidden associated features of running state information of each mechanical device at a plurality of preset time points in the safety production process of an enterprise production system are deeply mined by utilizing a deep neural network model of a deep learning technology so as to judge the production risk of the enterprise production system at the current time point, and in the process, a characteristic distribution manifold correction system is added so as to carry out boundary constraint of a characteristic value set, so that fragmentation of the characteristic value set in a decision area in a classification target area due to characteristic values outside the distribution of the set is avoided, and the classification accuracy is improved. In this way, the production risk of the enterprise production system at the current time point can be more accurately judged.
Description
Technical Field
The present invention relates to the field of intelligent production management, and more particularly, to an enterprise safety production management system and a management method thereof.
Background
At present, the safety production management of enterprises mainly depends on a few monitoring personnel, and is responsible for executing corresponding emergency treatment after a safety accident occurs. In this supervision mode, the safety production status of the enterprise cannot be safely grasped in time. In recent years, the development of advanced technologies such as the internet of things, 5G and big data provides a new solution and scheme for enterprise safety production management.
Therefore, an enterprise safety production management system is expected to perform intelligent management on safety production.
Disclosure of Invention
The present application is proposed to solve the above-mentioned technical problems. The embodiment of the application provides an enterprise safety production management system and a management method thereof, which deeply excavate implicit associated characteristics of running state information of each mechanical device at a plurality of preset time points in the safety production process of an enterprise production system by utilizing a deep neural network model of a deep learning technology so as to judge the production risk of the enterprise production system at the current time point, and in the process, a characteristic distribution manifold correction system is added so as to carry out boundary constraint of a characteristic value set, so that fragmentation of the characteristic value set in a decision area in a classification target area caused by the characteristic value outside the distribution of the set is avoided, and the classification accuracy is improved. In this way, the production risk of the enterprise production system at the current time point can be more accurately judged.
According to an aspect of the present application, there is provided an enterprise safe production management system, comprising:
the system comprises a historical data acquisition module, a data processing module and a data processing module, wherein the historical data acquisition module is used for acquiring historical data generated by an enterprise production system in a production process, and the historical data is the running state information of each mechanical device of the enterprise production system at a plurality of preset time points in a safety production process;
the historical data structuring module is used for arranging the historical data generated by the enterprise production system in the production process into a two-dimensional input state matrix according to the time dimension and the mechanical equipment sample dimension;
the historical data feature extraction module is used for carrying out explicit spatial coding on the two-dimensional input state matrix by using the first convolution neural network model to obtain an operation state association feature matrix;
the current state data acquisition module is used for acquiring the running state information of each mechanical device of the enterprise production system at the current time point;
the current state data encoding module is used for enabling the running state information of each mechanical device of the enterprise production system at the current time point to pass through a sequence encoder containing a one-dimensional convolutional layer so as to obtain a current state feature vector;
the characteristic distribution correction module is used for correcting the current state characteristic vector and the equipment operation correlation characteristic matrix respectively to obtain a corrected current state characteristic vector and a corrected equipment operation correlation characteristic matrix;
the vector query module is used for multiplying the corrected current state characteristic vector serving as a query vector by the corrected equipment operation association characteristic matrix to obtain a classification characteristic vector; and
and the safety management result generation module is used for enabling the classified characteristic vectors to pass through a classifier to obtain a classification result, and the classification result is used for indicating whether the production of the enterprise production system at the current time point has risks or not.
In the enterprise safety production management system, in the historical data, if the operating state of each mechanical device of the enterprise production system at each predetermined time point is an open state, the operating power of the mechanical device is used as the operating state information of the mechanical device at the predetermined time point; and if the running state of each mechanical device of the enterprise production system at each preset time point is in a closed state, taking a zero value as the running state information of the mechanical device at the preset time point.
In the above system for managing enterprise safety production, the historical data structuring module includes: the row vector construction unit is used for respectively arranging historical data generated by the enterprise production system in the production process into row vectors according to the time dimension to obtain a plurality of row vectors; and the two-dimensional arrangement unit is used for arranging the plurality of row vectors into the two-dimensional input state matrix according to the sample dimension of the mechanical equipment.
In the above enterprise safety production management system, the historical data feature extraction module is further configured to perform, in layer forward transmission, using each layer of the first convolutional neural network model: performing convolution processing on input data to obtain a convolution characteristic diagram; performing mean pooling based on local channel dimensions on the convolution feature map to obtain a pooled feature map; and performing nonlinear activation on the pooled feature map to obtain an activated feature map; and the output of the last layer of the first convolution neural network model is the running state associated characteristic matrix, and the input of the first layer of the first convolution neural network model is the two-dimensional input state matrix.
In the above enterprise safety production management system, the current state data encoding module includes: the input vector construction unit is used for arranging the running state information of each mechanical device of the enterprise production system at the current time point into one-dimensional input vectors corresponding to each mechanical device of the enterprise production system according to the time dimension; a full-connection coding unit, configured to perform full-connection coding on the input vector by using a full-connection layer of the time sequence encoder according to the following formula to extract a high-dimensional implicit feature of a feature value at each position in the input vector, where the formula is:wherein X is the input vector, Y is the output vector, W is the weight matrix, B is the offset vector,represents a matrix multiplication; a one-dimensional convolution coding unit, configured to perform one-dimensional convolution coding on the input vector by using a one-dimensional convolution layer of the time-series encoder according to the following formula to extract a high-dimensional implicit correlation feature between feature values of each position in the input vector, where the formula is:
wherein, a is the width of the convolution kernel in the x direction, F is the parameter vector of the convolution kernel, G is the local vector matrix operated with the convolution kernel function, and w is the size of the convolution kernel. In the above enterprise safety production management system, the characteristic distribution correction module includes: the vector correction unit is used for correcting the current state feature vector according to the following formula to obtain the corrected current state feature vector;
wherein the formula is:
wherein v is i A feature value, v ', representing the ith position of the current state feature vector' i A feature value representing an ith position of the corrected current state feature vector; the matrix correction unit is used for correcting the equipment operation correlation characteristic matrix according to the following formula to obtain the corrected equipment operation correlation characteristic matrix;
wherein the formula is:
wherein m is i,j Running a feature value, m ', for the device associated with a (i, j) th location of a feature matrix' i,j And (c) representing the characteristic value of the (i, j) th position of the corrected device operation correlation characteristic matrix.
In the above enterprise secure production management system, the secure management result generating module is further configured to: processing the classification feature vector using the classifier to obtain the classification result with the following formula: softmax { (W) n ,B n ):...:(W 1 ,B 1 ) I X }, wherein W 1 To W n As a weight matrix, B 1 To B n Is biased toAnd X is the classification feature vector.
According to another aspect of the present application, a management method of an enterprise safety production management system includes:
acquiring historical data generated by an enterprise production system in a production process, wherein the historical data is the running state information of each mechanical device at a plurality of preset time points in the safety production process of the enterprise production system;
arranging historical data generated by the enterprise production system in a production process into a two-dimensional input state matrix according to a time dimension and a mechanical equipment sample dimension;
performing explicit spatial coding on the two-dimensional input state matrix by using the first convolutional neural network model to obtain an operation state correlation characteristic matrix;
acquiring the running state information of each mechanical device of the enterprise production system at the current time point;
running state information of each mechanical device of the enterprise production system at the current time point is processed through a sequence encoder comprising a one-dimensional convolution layer to obtain a current state feature vector;
respectively correcting the current state feature vector and the equipment operation association feature matrix to obtain a corrected current state feature vector and a corrected equipment operation association feature matrix;
multiplying the corrected current state feature vector serving as a query vector by the corrected equipment operation association feature matrix to obtain a classification feature vector; and
and passing the classified feature vectors through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the production of the enterprise production system at the current time point has risks.
In the management method of the enterprise safety production management system, in the historical data, if the operation state of each mechanical device of the enterprise production system at each predetermined time point is an on state, the operation power of the mechanical device is used as the operation state information of the mechanical device at the predetermined time point; and if the running state of each mechanical device of the enterprise production system at each preset time point is in a closed state, taking a zero value as the running state information of the mechanical device at the preset time point.
In the management method of the enterprise safety production management system, the arrangement of the historical data generated by the enterprise production system in the production process into the two-dimensional input state matrix according to the time dimension and the mechanical equipment sample dimension includes: respectively arranging historical data generated by the enterprise production system in the production process into row vectors according to the time dimension to obtain a plurality of row vectors; and arranging the plurality of row vectors into the two-dimensional input state matrix according to the sample dimension of the mechanical equipment.
In the management method of the enterprise safety production management system, performing explicit spatial coding on the two-dimensional input state matrix by using the first convolutional neural network model to obtain an operation state associated feature matrix, including: using the layers of the first convolutional neural network model in layer forward pass: carrying out convolution processing on input data to obtain a convolution characteristic diagram; performing mean pooling based on local channel dimensions on the convolution feature map to obtain a pooled feature map; and performing nonlinear activation on the pooled feature map to obtain an activated feature map; and the output of the last layer of the first convolution neural network model is the running state associated characteristic matrix, and the input of the first layer of the first convolution neural network model is the two-dimensional input state matrix.
In the management method of the enterprise safety production management system, the step of obtaining the current state feature vector by passing the operation state information of each mechanical device of the enterprise production system at the current time point through a sequence encoder including a one-dimensional convolution layer includes: arranging the running state information of each mechanical device of the enterprise production system at the current time point into one-dimensional input vectors corresponding to each mechanical device of the enterprise production system according to the time dimension; using a full-connected layer of the sequential encoder to sum the input vector with the following formulaConnecting codes to extract high-dimensional implicit features of feature values of all positions in the input vector, wherein the formula is as follows:wherein X is the input vector, Y is the output vector, W is the weight matrix, B is the offset vector,represents a matrix multiplication; performing one-dimensional convolutional coding on the input vector by using a one-dimensional convolutional layer of the time sequence encoder according to the following formula to extract high-dimensional implicit correlation characteristics among characteristic values of all positions in the input vector, wherein the formula is as follows:
wherein, a is the width of the convolution kernel in the x direction, F is the parameter vector of the convolution kernel, G is the local vector matrix operated with the convolution kernel function, and w is the size of the convolution kernel.
In the management method of the enterprise safety production management system, the correcting the current state feature vector and the equipment operation associated feature matrix respectively to obtain a corrected current state feature vector and a corrected equipment operation associated feature matrix includes: correcting the current state feature vector according to the following formula to obtain the corrected current state feature vector;
wherein the formula is:
wherein v is i A feature value, v ', representing the ith position of the current state feature vector' i A feature value representing the ith position of the corrected current state feature vector; and correcting the device operation correlation characteristic matrix according to the following formula to obtain the corrected correlation characteristic matrixThe equipment operation correlation characteristic matrix;
wherein the formula is:
wherein m is i,j Running a feature value, m ', for the device associated with a (i, j) th location of a feature matrix' i,j And (3) representing the characteristic value of the (i, j) th position of the corrected device operation correlation characteristic matrix.
In the management method of the enterprise safety production management system, the classifying feature vector is passed through a classifier to obtain a classification result, and the method includes: processing the classification feature vector using the classifier to obtain the classification result with the following formula: softmax { (W) n ,B n ):...:(W 1 ,B 1 ) I X }, wherein W 1 To W n As a weight matrix, B 1 To B n Is a bias vector and X is the classification feature vector.
According to yet another aspect of the present application, there is provided a computer readable medium having stored thereon computer program instructions, which, when executed by a processor, cause the processor to execute the management method of the enterprise safety production management system as described above.
Compared with the prior art, the enterprise safe production management system and the management method thereof deeply mine the implicit associated characteristics of the running state information of each mechanical device at a plurality of preset time points in the safe production process of the enterprise production system by utilizing the deep neural network model of the deep learning technology so as to judge the production risk of the enterprise production system at the current time point, and in the process, a characteristic distribution manifold correction system is added to carry out boundary constraint of the characteristic value set, so that fragmentation of the characteristic value set in a decision area in a classification target area caused by the characteristic values outside the distribution of the set is avoided, and the classification accuracy is improved. In this way, the production risk of the enterprise production system at the current time point can be more accurately judged.
Drawings
The above and other objects, features and advantages of the present application will become more apparent by describing in more detail embodiments of the present application with reference to the attached drawings. The accompanying drawings are included to provide a further understanding of the embodiments of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the principles of the application. In the drawings, like reference numbers generally indicate like parts or steps.
Fig. 1 is an application scenario diagram of an enterprise safety production management system according to an embodiment of the present application.
Fig. 2 is a block diagram of an enterprise safety production management system according to an embodiment of the present application.
Fig. 3 is a block diagram of a feature distribution correction module in an enterprise safety production management system according to an embodiment of the present application.
Fig. 4 is a flowchart of a management method of an enterprise safety production management system according to an embodiment of the application.
Fig. 5 is a schematic architecture diagram of a management method of an enterprise safety production management system according to an embodiment of the present application.
Detailed Description
Hereinafter, example embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be understood that the described embodiments are only some embodiments of the present application and not all embodiments of the present application, and that the present application is not limited by the example embodiments described herein.
Overview of a scene
As mentioned above, currently, the safety production management of enterprises mainly depends on a few monitoring personnel, which are responsible for executing corresponding emergency treatment after a safety accident occurs. In this supervision mode, the safety production status of the enterprise cannot be safely grasped in time. In recent years, the development of advanced technologies such as internet of things, 5G and big data provides a new solution and scheme for enterprise safety production management.
Therefore, an enterprise safety production management system is expected to perform intelligent management on safety production.
Accordingly, the inventor of the present application considers that since the enterprises need to cooperate to complete specific tasks during the production activities of the enterprises, the operation status information of the mechanical devices can be monitored to intelligently manage the safety production. However, when the mechanical devices cooperate to complete the task, there may be an association relationship between the mechanical devices, and the task completion of the same device at each time point may also have an association relationship with each other, and a link error may cause a safety accident. Therefore, when monitoring information of the operation state of each mechanical device is used to intelligently manage the safety production, the factors in the aspects need to be considered.
Based on this, in the technical scheme of the application, firstly, historical data generated by the enterprise production system in the production process is obtained, and the historical data is operation state information of each mechanical device at a plurality of preset time points when the enterprise production system is in safe production. In particular, in a specific example, if the operation state of each mechanical device of the enterprise production system at each predetermined time point is an on state, the operation power of the mechanical device is used as the operation state information of the mechanical device at the predetermined time point; and if the running state of each mechanical device of the enterprise production system at each preset time point is in a closed state, taking a zero value as the running state information of the mechanical device at the preset time point.
Then, in order to fully extract the associated characteristic information of the operation state information of each mechanical device at a plurality of preset time points in the safety production process of the enterprise production system by using the deep neural network model, furthermore, historical data generated by the enterprise production system in the production process is used for constructing a two-dimensional input state matrix according to the time dimension and the mechanical device dimension. And then, carrying out spatial display coding on the two-dimensional input state matrix through a first convolutional neural network so as to extract implicit associated characteristic information of the running state information of each mechanical device at a plurality of preset time points in time and sample dimensions, namely the dynamic change characteristics of each mechanical device, thereby obtaining the running state associated matrix.
When actually monitoring the operating state of each mechanical device of the enterprise production system, if it is desired to determine whether the enterprise production system at the current time point has a risk, further, first, the operating state information of each mechanical device of the enterprise production system at the current time point is obtained. Then, considering that the running state information of each mechanical device of the enterprise production system has a dynamic rule in time sequence, in order to more fully extract a dynamic change implicit rule of the running state information of each mechanical device, the obtained data information is input into a sequence encoder with a one-dimensional convolutional layer for encoding processing, so as to obtain a current state feature vector.
Therefore, the current state feature vector and the equipment operation association feature matrix can be further fused to perform feature mapping, and then classification is performed to obtain a classification result for representing whether the production of the enterprise production system at the current time point has risks. However, when the current state feature vector is used as a query vector to be multiplied by the device operation associated feature matrix to obtain a classification feature vector, since the current state feature vector represents semantic associated information of the operation state information of each mechanical device at the current time point, and the device operation associated feature matrix represents two-dimensional associated information of a plurality of mechanical devices in a sample-time dimension at a plurality of time points, if the feature distribution of the vector and the matrix in the classification target domain is irregular, the manifold represented by the feature distribution in the classification target space after the multiplication of the vector and the matrix is performed is more irregular, thereby affecting the classification accuracy.
Based on this, before multiplying the current state feature vector by the device operation correlation feature matrix, it is preferably corrected, as:
wherein v is i A feature value, v 'representing the ith position of the current state feature vector' i A feature value, m, representing the ith position of the corrected current state feature vector i,j Running a feature value, m ', for the device associated with a (i, j) th location of a feature matrix' i,j And (3) representing the characteristic value of the (i, j) th position of the corrected device operation correlation characteristic matrix.
Therefore, through the correction, the feature values of the feature vector and the feature matrix and the class conditions of the feature vector and the feature matrix can be structurally understood based on rules respectively to carry out boundary constraint of the feature value set, so that fragmentation of the feature value set in a decision area in a classification target area caused by the feature values outside the distribution of the feature value set is avoided, a more stable conditional class boundary of the current state feature vector and the equipment operation associated feature matrix is given, the classification decision precision of the classification feature vector in the classification target area is improved, and the classification accuracy can be improved, so that the production risk of an enterprise production system at the current time point can be judged more accurately.
Based on this, this application has proposed an enterprise safety production management system, it includes: the system comprises a historical data acquisition module, a data processing module and a data processing module, wherein the historical data acquisition module is used for acquiring historical data generated by an enterprise production system in a production process, and the historical data is the running state information of each mechanical device of the enterprise production system at a plurality of preset time points in a safety production process; the historical data structuring module is used for arranging the historical data generated by the enterprise production system in the production process into a two-dimensional input state matrix according to the time dimension and the mechanical equipment sample dimension; the historical data feature extraction module is used for carrying out explicit spatial coding on the two-dimensional input state matrix by using the first convolution neural network model to obtain an operation state association feature matrix; the current state data acquisition module is used for acquiring the running state information of each mechanical device of the enterprise production system at the current time point; the current state data coding module is used for enabling the running state information of each mechanical device of the enterprise production system at the current time point to pass through a sequence coder containing a one-dimensional convolution layer so as to obtain a current state feature vector; the characteristic distribution correction module is used for correcting the current state characteristic vector and the equipment operation association characteristic matrix respectively to obtain a corrected current state characteristic vector and a corrected equipment operation association characteristic matrix; the vector query module is used for multiplying the corrected current state feature vector serving as a query vector by the corrected equipment operation association feature matrix to obtain a classification feature vector; and the safety management result generation module is used for enabling the classification characteristic vectors to pass through a classifier to obtain a classification result, and the classification result is used for indicating whether the production of the enterprise production system at the current time point has risks or not.
Fig. 1 illustrates an application scenario of an enterprise safety production management system according to an embodiment of the present application. As shown in fig. 1, in this application scenario, first, the operating state information of each mechanical device (e.g., E1-En as illustrated in fig. 1) of the enterprise production system at a plurality of historical predetermined time points in the safety production process is obtained through a cloud storage (e.g., T as illustrated in fig. 1) query, and the operating state information of each mechanical device (e.g., E1-En as illustrated in fig. 1) of the enterprise production system at a current time point is obtained through a monitoring device (e.g., M as illustrated in fig. 1). Then, the obtained operation state information of each mechanical device of the enterprise production system at the plurality of historical predetermined time points in the safety production process and the operation state information of each mechanical device of the enterprise production system at the current time point are input into a server (for example, a cloud server S as illustrated in fig. 1) deployed with an enterprise safety production management algorithm, wherein the server can process the operation state information of each mechanical device of the enterprise production system at the plurality of historical predetermined time points in the safety production process and the operation state information of each mechanical device of the enterprise production system at the current time point by using the enterprise safety production management algorithm to generate a classification result for representing whether the production of the enterprise production system at the current time point is at risk.
Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described with reference to the accompanying drawings.
Exemplary System
Fig. 2 illustrates a block diagram of an enterprise secure production management system according to an embodiment of the present application. As shown in fig. 2, an enterprise secure production management system 200 according to an embodiment of the present application includes: a historical data acquisition module 210, configured to acquire historical data generated by an enterprise production system in a production process, where the historical data is information of operating states of each mechanical device at multiple predetermined time points in a safety production process of the enterprise production system; a historical data structuring module 220, configured to arrange historical data generated by the enterprise production system in a production process into a two-dimensional input state matrix according to a time dimension and a mechanical device sample dimension; a historical data feature extraction module 230, configured to perform explicit spatial coding on the two-dimensional input state matrix using the first convolutional neural network model to obtain an operation state associated feature matrix; a current state data acquisition module 240, configured to acquire operation state information of each mechanical device of the enterprise production system at a current time point; a current state data encoding module 250, configured to pass the running state information of each mechanical device of the enterprise production system at the current time point through a sequence encoder that includes a one-dimensional convolutional layer to obtain a current state feature vector; a feature distribution correction module 260, configured to correct the current state feature vector and the device operation association feature matrix respectively to obtain a corrected current state feature vector and a corrected device operation association feature matrix; a vector query module 270, configured to multiply the corrected current state feature vector, which is used as a query vector, with the corrected device operation association feature matrix to obtain a classification feature vector; and a safety management result generating module 280, configured to pass the classification feature vector through a classifier to obtain a classification result, where the classification result is used to indicate whether there is a risk in the production of the enterprise production system at the current time point.
Specifically, in the embodiment of the present application, the historical data collecting module 210 is configured to obtain historical data generated by an enterprise production system in a production process, where the historical data is operation state information of each mechanical device of the enterprise production system at a plurality of predetermined time points in a safety production process. As described above, it should be understood that, since an enterprise needs to perform a cooperative operation on each mechanical device during a production activity to complete a specific operation task, in the technical solution of the present application, intelligent management may be performed on safety production by monitoring operation state information of each mechanical device. However, when the mechanical devices cooperate to complete the task, there may be an association relationship between the mechanical devices, and there may also be an association relationship between task completion degrees of the same device at various time points, and a link error may cause a safety accident. Therefore, when monitoring information of the operation state of each mechanical device is used to intelligently manage the safety production, the factors in the aspects need to be considered.
Therefore, in the technical solution of the present application, specifically, first, historical data generated by the enterprise production system in the production process is obtained, where the historical data is operation state information of each mechanical device at a plurality of predetermined time points when the enterprise production system is in safety production. In particular, in a specific example, if the operation state of each mechanical device of the enterprise production system at each predetermined time point is an on state, the operation power of the mechanical device is used as the operation state information of the mechanical device at the predetermined time point; and if the running state of each mechanical device of the enterprise production system at each preset time point is in a closed state, taking a zero value as the running state information of the mechanical device at the preset time point.
Specifically, in this embodiment of the present application, the historical data structuring module 220 and the historical data feature extracting module 230 are configured to arrange historical data generated by the enterprise production system in a production process into a two-dimensional input state matrix according to a time dimension and a mechanical device sample dimension, and perform explicit spatial coding on the two-dimensional input state matrix by using the first convolutional neural network model to obtain an operation state associated feature matrix. It should be understood that, in order to fully extract the relevant characteristic information of the operation state information of each mechanical device at a plurality of preset time points in the safety production process of the enterprise production system by using the deep neural network model, furthermore, historical data generated in the production process of the enterprise production system is used for constructing a two-dimensional input state matrix according to the time dimension and the mechanical device dimension. And then, carrying out spatial display coding on the two-dimensional input state matrix through a first convolutional neural network so as to extract implicit associated characteristic information of the running state information of each mechanical device at a plurality of preset time points on time and sample dimensions, namely the dynamic change characteristics of each mechanical device, thereby obtaining the running state associated matrix. Accordingly, in one particular example, using the layers of the first convolutional neural network model in a layer forward pass separately: carrying out convolution processing on input data to obtain a convolution characteristic diagram; performing mean pooling based on local channel dimensions on the convolution feature map to obtain a pooled feature map; and performing nonlinear activation on the pooled feature map to obtain an activated feature map; and the output of the last layer of the first convolution neural network model is the running state associated characteristic matrix, and the input of the first layer of the first convolution neural network model is the two-dimensional input state matrix.
More specifically, in this embodiment of the present application, the historical data structuring module includes: the row vector construction unit is used for respectively arranging historical data generated by the enterprise production system in the production process into row vectors according to the time dimension to obtain a plurality of row vectors; and the two-dimensional arrangement unit is used for arranging the plurality of row vectors into the two-dimensional input state matrix according to the sample dimension of the mechanical equipment.
Specifically, in this embodiment of the application, the current state data collecting module 240 and the current state data encoding module 250 are configured to obtain operation state information of each mechanical device of the enterprise production system at a current time point, and obtain a current state feature vector by passing the operation state information of each mechanical device of the enterprise production system at the current time point through a sequence encoder including a one-dimensional convolutional layer. It should be understood that, when actually monitoring the operation state of each mechanical device of the enterprise production system, if it is desired to determine whether the enterprise production system at the current time point has a risk, further, first, the operation state information of each mechanical device of the enterprise production system at the current time point is obtained. Then, considering that the running state information of each mechanical device of the enterprise production system has a dynamic rule in time sequence, in order to more fully extract a dynamic change implicit rule of the running state information of each mechanical device, the obtained data information is input into a sequence encoder with a one-dimensional convolutional layer for encoding processing, so as to obtain a current state feature vector.
More specifically, in this embodiment of the present application, the current state data encoding module includes: the input vector construction unit is used for arranging the running state information of each mechanical device of the enterprise production system at the current time point into one-dimensional input vectors corresponding to each mechanical device of the enterprise production system according to the time dimension; a full-concatenation coding unit, configured to perform full-concatenation coding on the input vector using a full-concatenation layer of the time sequence encoder according to the following formula to extract a high-dimensional implicit feature of a feature value at each position in the input vector, where the formula is:wherein X is the input vector, Y is the output vector, W is the weight matrix, B is the offset vector,represents a matrix multiplication; a one-dimensional convolution coding unit, configured to perform one-dimensional convolution coding on the input vector by using a one-dimensional convolution layer of the time-series encoder according to the following formula to extract a high-dimensional implicit correlation feature between feature values of each position in the input vector, where the formula is:
wherein, a is the width of the convolution kernel in the x direction, F is the parameter vector of the convolution kernel, G is the local vector matrix operated with the convolution kernel function, and w is the size of the convolution kernel.
Specifically, in this embodiment of the present application, the feature distribution correction module 260 is configured to correct the current state feature vector and the device operation associated feature matrix respectively to obtain a corrected current state feature vector and a corrected device operation associated feature matrix. It should be understood that after the current state feature vector and the device operation associated feature matrix are obtained, the current state feature vector and the device operation associated feature matrix may be further fused to perform feature mapping, and then classification is performed to obtain a classification result indicating whether there is a risk in the production of the enterprise production system at the current time point. However, when the current state feature vector is used as a query vector to be multiplied by the device operation correlation feature matrix to obtain a classification feature vector, since the current state feature vector represents semantic correlation information of the operation state information of each mechanical device at a current time point, and the device operation correlation feature matrix represents two-dimensional correlation information of the multiple mechanical devices in sample-time dimensions at multiple time points, if the feature distribution of the vector and the matrix in the classification target domain is irregular, after the multiplication of the vector and the matrix is performed, the manifold represented by the feature distribution in the classification target space is more irregular, thereby affecting the classification accuracy. Therefore, in the technical solution of the present application, before the current state feature vector is further multiplied and mapped by the device operation associated feature matrix, it is preferably corrected.
More specifically, in this embodiment of the present application, the feature distribution correction module includes: correcting the current state feature vector according to the following formula to obtain the corrected current state feature vector; wherein the formula is:
wherein v is i A feature value, v 'representing the ith position of the current state feature vector' i And representing the characteristic value of the ith position of the corrected current state characteristic vector. Correcting the equipment operation correlation characteristic matrix according to the following formula to obtain the corrected equipment operation correlation characteristic matrix;
wherein the formula is:
wherein m is i,j Running a feature value, m ', for the device associated with a (i, j) th location of a feature matrix' i,j And (3) representing the characteristic value of the (i, j) th position of the corrected device operation correlation characteristic matrix. It should be understood that, by means of the modification, the feature values of the current state feature vector and the device operation associated feature matrix and the class conditions to which the current state feature vector and the device operation associated feature matrix belong can be subjected to rule-based structural understanding respectively to perform boundary constraint of the feature value set, so that fragmentation of the feature value set in a decision area in a classification target area due to the distributed feature values of the set is avoided, a more robust conditional class boundary of the current state feature vector and the device operation associated feature matrix is given, classification decision precision of the classification feature vector in the classification target area is improved, and classification accuracy can be improved, and production risk of an enterprise production system at the current time point can be judged more accurately.
Fig. 3 illustrates a block diagram of a feature distribution correction module in an enterprise safety production management system according to an embodiment of the application. As shown in fig. 3, the feature distribution correction module 260 includes: a vector correction unit 261, configured to correct the current-state feature vector according to the following formula to obtain the corrected current-state feature vector;
wherein the formula is:
wherein v is i A feature value, v ', representing the ith position of the current state feature vector' i A feature value representing an ith position of the corrected current state feature vector; the matrix correction unit 262 is configured to correct the device operation correlation feature matrix according to the following formula to obtain the corrected device operation correlation feature matrix;
wherein the formula is:
wherein m is i,j Running a feature value, m ', for the device associated with a (i, j) th location of a feature matrix' i,j And (3) representing the characteristic value of the (i, j) th position of the corrected device operation correlation characteristic matrix.
Specifically, in this embodiment of the present application, the vector query module 270 is configured to take the corrected current state feature vector as a query vector to be multiplied by the corrected device operation association feature matrix to obtain a classification feature vector. That is, further, the corrected current state feature vector is used as a query vector to be multiplied by the corrected device operation associated feature matrix, so as to map the corrected current state feature vector to a high-dimensional feature space of the corrected device operation associated feature matrix, thereby obtaining a classification feature vector.
Specifically, in this embodiment, the security management result generating module 280 is configured to pass the classification feature vector through a classifier to obtain a classification result, where the classification result is used to indicate whether there is a risk in the production of the enterprise production system at the current time point. Accordingly, in one specific example, the classification feature vector is processed using the classifier to obtain the classification result with the following formula: softmax { (W) n ,B n ):...:(W 1 ,B 1 ) I X }, wherein W 1 To W n As a weight matrix, B 1 To B n Is a bias vector and X is the classification feature vector.
In summary, the enterprise safety production management system 200 based on the embodiment of the present application is clarified, which deeply mines implicit associated features of operating state information of each mechanical device at a plurality of predetermined time points in a safety production process of an enterprise production system by using a deep neural network model of a deep learning technology to determine a production risk of the enterprise production system at a current time point, and in the process, a correction system of a feature distribution manifold is further added to perform boundary constraint on a feature value set, so that fragmentation of the feature value set in a decision area in a classification target area due to an external feature value of a set is avoided, and further classification accuracy is improved. In this way, the production risk of the enterprise production system at the current time point can be more accurately judged.
As described above, the enterprise secure production management system 200 according to the embodiment of the present application may be implemented in various terminal devices, such as a server of an enterprise secure production management algorithm. In one example, the enterprise secure production management system 200 according to the embodiment of the present application may be integrated into the terminal device as a software module and/or a hardware module. For example, the enterprise security production management system 200 may be a software module in the operating system of the terminal device, or may be an application developed for the terminal device; of course, the enterprise security production management system 200 may also be one of many hardware modules of the terminal device.
Alternatively, in another example, the enterprise secure production management system 200 and the terminal device may be separate devices, and the enterprise secure production management system 200 may be connected to the terminal device through a wired and/or wireless network and transmit the interaction information according to an agreed data format.
Exemplary method
Fig. 4 illustrates a flow chart of a management method of the enterprise secure production management system. As shown in fig. 4, a management method of an enterprise safety production management system according to an embodiment of the present application includes the steps of: s110, acquiring historical data generated by an enterprise production system in a production process, wherein the historical data is the running state information of each mechanical device of the enterprise production system at a plurality of preset time points in a safety production process; s120, arranging historical data generated by the enterprise production system in the production process into a two-dimensional input state matrix according to the time dimension and the mechanical equipment sample dimension; s130, performing explicit spatial coding on the two-dimensional input state matrix by using the first convolution neural network model to obtain an operation state association characteristic matrix; s140, acquiring the running state information of each mechanical device of the enterprise production system at the current time point; s150, enabling the running state information of each mechanical device of the enterprise production system at the current time point to pass through a sequence encoder containing a one-dimensional convolutional layer to obtain a current state feature vector; s160, correcting the current state characteristic vector and the equipment operation association characteristic matrix respectively to obtain a corrected current state characteristic vector and a corrected equipment operation association characteristic matrix; s170, multiplying the corrected current state feature vector serving as a query vector by the corrected equipment operation association feature matrix to obtain a classification feature vector; and S180, enabling the classification feature vector to pass through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the production of the enterprise production system at the current time point has risks.
Fig. 5 is a schematic diagram illustrating an architecture of a management method of an enterprise secure production management system according to an embodiment of the present application. As shown IN fig. 5, IN the network architecture of the management method of the enterprise secure production management system, firstly, the obtained historical data (e.g., IN1 as illustrated IN fig. 5) generated by the enterprise production system during the production process is arranged into a two-dimensional input state matrix (e.g., M as illustrated IN fig. 5) according to the time dimension and the mechanical equipment sample dimension; then, explicitly spatially encoding the two-dimensional input state matrix using the first convolutional neural network model (e.g., CNN as illustrated in fig. 5) to obtain a running state associated feature matrix (e.g., MF as illustrated in fig. 5); then, passing the obtained operation state information (e.g., IN2 as illustrated IN fig. 5) of each mechanical device of the enterprise production system at the current time point through a sequence encoder (e.g., E as illustrated IN fig. 5) containing one-dimensional convolutional layers to obtain a current state feature vector (e.g., VF as illustrated IN fig. 5); then, respectively correcting the current state feature vector and the device operation correlation feature matrix to obtain a corrected current state feature vector (for example, VC as illustrated in fig. 5) and a corrected device operation correlation feature matrix (for example, MC as illustrated in fig. 5); then, multiplying the corrected current state feature vector as a query vector by the corrected device operation association feature matrix to obtain a classification feature vector (for example, V as illustrated in fig. 5); and finally, passing the classification feature vector through a classifier (e.g., circle S as illustrated in fig. 5) to obtain a classification result, wherein the classification result is used for indicating whether the production of the enterprise production system at the current time point is at risk.
More specifically, in step S110, historical data generated by the enterprise production system during the production process is obtained, where the historical data is information of operating states of various mechanical devices of the enterprise production system at a plurality of predetermined time points during the safety production process. It should be understood that, since an enterprise needs to cooperate with each mechanical device in a production activity to complete a specific task, in the technical solution of the present application, intelligent management can be performed on safety production by monitoring the operating state information of each mechanical device. However, when the mechanical devices cooperate to complete the task, there may be an association relationship between the mechanical devices, and there may also be an association relationship between task completion degrees of the same device at various time points, and a link error may cause a safety accident. Therefore, when monitoring information of the operating states of the mechanical devices is used for intelligent management of safety production, the factors in the aspects need to be considered.
Therefore, in the technical solution of the present application, specifically, first, historical data generated by the enterprise production system in the production process is obtained, where the historical data is information of operating states of each mechanical device at a plurality of predetermined time points when the enterprise production system is in safety production. Specifically, in a specific example, if the operation state of each mechanical device of the enterprise production system at each predetermined time point is an on state, the operation power of the mechanical device is used as the operation state information of the mechanical device at the predetermined time point; and if the running state of each mechanical device of the enterprise production system at each preset time point is in a closed state, taking a zero value as the running state information of the mechanical device at the preset time point.
More specifically, in step S120 and step S130, the historical data generated by the enterprise production system in the production process is arranged into a two-dimensional input state matrix according to the time dimension and the mechanical equipment sample dimension, and the two-dimensional input state matrix is explicitly and spatially encoded by using the first convolutional neural network model to obtain an operation state association feature matrix. It should be understood that, in order to fully extract the relevant characteristic information of the operating state information of each mechanical device at a plurality of predetermined time points in the safety production process of the enterprise production system by using the deep neural network model, further, the historical data generated by the enterprise production system in the production process is used for constructing a two-dimensional input state matrix according to the time dimension and the mechanical device dimension. And then, carrying out spatial display coding on the two-dimensional input state matrix through a first convolutional neural network so as to extract implicit associated characteristic information of the running state information of each mechanical device at a plurality of preset time points in time and sample dimensions, namely the dynamic change characteristics of each mechanical device, thereby obtaining the running state associated matrix.
More specifically, in step S140 and step S150, the operation state information of each mechanical device of the enterprise production system at the current time point is obtained, and the operation state information of each mechanical device of the enterprise production system at the current time point is passed through a sequence encoder including a one-dimensional convolutional layer to obtain a current state feature vector. It should be understood that, when actually monitoring the operation states of the mechanical devices of the enterprise production system, if it is desired to determine whether the enterprise production system at the current time point has a risk, further, first, the operation state information of the mechanical devices of the enterprise production system at the current time point is obtained. Then, considering that the running state information of each mechanical device of the enterprise production system has a dynamic rule in time sequence, in order to more fully extract a dynamic change implicit rule of the running state information of each mechanical device, the obtained data information is input into a sequence encoder of which the value contains a one-dimensional convolutional layer to be encoded, so as to obtain a current state feature vector.
More specifically, in step S160, the current-state feature vector and the device-operation-related feature matrix are respectively corrected to obtain a corrected current-state feature vector and a corrected device-operation-related feature matrix. It should be understood that after the current state feature vector and the device operation associated feature matrix are obtained, the current state feature vector and the device operation associated feature matrix may be further fused to perform feature mapping, and then classification is performed to obtain a classification result indicating whether there is a risk in the production of the enterprise production system at the current time point. However, when the current state feature vector is used as a query vector to be multiplied by the device operation associated feature matrix to obtain a classification feature vector, since the current state feature vector represents semantic associated information of the operation state information of each mechanical device at a current time point, and the device operation associated feature matrix represents two-dimensional associated information of the multiple mechanical devices in sample-time dimensions at multiple time points, if the feature distribution of the vector and the matrix themselves in a classification target domain is irregular, the manifold represented by the feature distribution in the classification target space is more irregular after the multiplication of the vector and the matrix is performed, thereby affecting the classification accuracy. Therefore, in the technical solution of the present application, before the current state feature vector is further multiplied and mapped by the device operation associated feature matrix, it is preferably corrected.
More specifically, in step S170, the corrected current state feature vector is used as a query vector to be multiplied by the corrected device operation correlation feature matrix to obtain a classification feature vector. That is, further, the corrected current state feature vector is used as a query vector to be multiplied by the corrected device operation associated feature matrix, so as to map the corrected current state feature vector to a high-dimensional feature space of the corrected device operation associated feature matrix, thereby obtaining a classification feature vector.
More specifically, in step S180, the classified feature vector is passed through a classifier to obtain a classification result, and the classification result is used to indicate whether there is a risk in the production of the enterprise production system at the current time point.
In summary, the management method of the enterprise safety production management system based on the embodiment of the present application is clarified, which deeply mines implicit associated features of operating state information of each mechanical device at a plurality of predetermined time points in a safety production process of an enterprise production system by using a deep neural network model of a deep learning technology to judge a production risk of the enterprise production system at a current time point, and in the process, a correction system of a feature distribution manifold is further added to perform boundary constraint of a feature value set, so that fragmentation of the feature value set in a decision area in a classification target area due to an out-of-distribution feature value of the set is avoided, and further, the classification accuracy is improved. In this way, the production risk of the enterprise production system at the current time point can be more accurately judged.
Exemplary computer program product and computer-readable storage Medium
In addition to the above-described methods and apparatus, embodiments of the present application may also be a computer program product comprising computer program instructions that, when executed by a processor, cause the processor to perform the steps in the functions of the management method of an enterprise secure production management system according to the various embodiments of the present application described in the above-mentioned "exemplary methods" section of this specification.
The computer program product may be written with program code for performing the operations of embodiments of the present application in any combination of one or more programming languages, including an object oriented programming language such as Java, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device, or entirely on the remote computing device or server.
Furthermore, an embodiment of the present application may also be a computer-readable storage medium, on which computer program instructions are stored, which, when executed by a processor, cause the processor to perform the steps in the management method of the enterprise secure production management system described in the above section "exemplary method" of this specification.
The computer-readable storage medium may take any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may include, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
The basic principles of the present application have been described above with reference to specific embodiments, but it should be noted that advantages, effects, etc. mentioned in the present application are only examples and are not limiting, and the advantages, effects, etc. must not be considered to be possessed by various embodiments of the present application. Furthermore, the foregoing disclosure of specific details is for the purpose of illustration and description and is not intended to be limiting, since the foregoing disclosure is not intended to be exhaustive or to limit the disclosure to the precise details disclosed.
The block diagrams of devices, apparatuses, systems referred to in this application are only given as illustrative examples and are not intended to require or imply that the connections, arrangements, configurations, etc. must be made in the manner shown in the block diagrams. These devices, apparatuses, devices, systems may be connected, arranged, configured in any manner, as will be appreciated by those skilled in the art. Words such as "including," "comprising," "having," and the like are open-ended words that mean "including, but not limited to," and are used interchangeably therewith. The words "or" and "as used herein mean, and are used interchangeably with, the word" and/or, "unless the context clearly dictates otherwise. The word "such as" is used herein to mean, and is used interchangeably with, the phrase "such as but not limited to".
It should also be noted that in the devices, apparatuses, and methods of the present application, the components or steps may be decomposed and/or recombined. These decompositions and/or recombinations are to be considered as equivalents of the present application.
The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
The foregoing description has been presented for purposes of illustration and description. Furthermore, the description is not intended to limit embodiments of the application to the form disclosed herein. While a number of example aspects and embodiments have been discussed above, those of skill in the art will recognize certain variations, modifications, alterations, additions and sub-combinations thereof.
Claims (10)
1. An enterprise safety production management system, comprising:
the system comprises a historical data acquisition module, a data processing module and a data processing module, wherein the historical data acquisition module is used for acquiring historical data generated by an enterprise production system in a production process, and the historical data is the running state information of each mechanical device of the enterprise production system at a plurality of preset time points in a safety production process;
the historical data structuring module is used for arranging the historical data generated by the enterprise production system in the production process into a two-dimensional input state matrix according to the time dimension and the mechanical equipment sample dimension;
the historical data feature extraction module is used for carrying out explicit spatial coding on the two-dimensional input state matrix by using the first convolution neural network model to obtain an operation state association feature matrix;
the current state data acquisition module is used for acquiring the running state information of each mechanical device of the enterprise production system at the current time point;
the current state data coding module is used for enabling the running state information of each mechanical device of the enterprise production system at the current time point to pass through a sequence coder containing a one-dimensional convolution layer so as to obtain a current state feature vector;
the characteristic distribution correction module is used for correcting the current state characteristic vector and the equipment operation association characteristic matrix respectively to obtain a corrected current state characteristic vector and a corrected equipment operation association characteristic matrix;
the vector query module is used for multiplying the corrected current state characteristic vector serving as a query vector by the corrected equipment operation association characteristic matrix to obtain a classification characteristic vector; and
and the safety management result generation module is used for enabling the classified characteristic vectors to pass through a classifier to obtain a classification result, and the classification result is used for indicating whether the production of the enterprise production system at the current time point has risks or not.
2. The enterprise safety production management system according to claim 1, wherein in the historical data, if the operation state of each mechanical device of the enterprise production system at each predetermined time point is an on state, the operation power of the mechanical device is used as the operation state information of the mechanical device at the predetermined time point; and if the running state of each mechanical device of the enterprise production system at each preset time point is in a closed state, taking a zero value as the running state information of the mechanical device at the preset time point.
3. The enterprise safety production management system of claim 2, wherein the historical data structuring module comprises:
the row vector construction unit is used for respectively arranging historical data generated by the enterprise production system in the production process into row vectors according to the time dimension so as to obtain a plurality of row vectors;
and the two-dimensional arrangement unit is used for arranging the plurality of row vectors into the two-dimensional input state matrix according to the sample dimension of the mechanical equipment.
4. The enterprise safety production management system of claim 3, wherein the historical data feature extraction module is further configured to perform in-layer forward pass using the layers of the first convolutional neural network model, respectively:
performing convolution processing on input data to obtain a convolution characteristic diagram;
performing mean pooling based on local channel dimensions on the convolution feature map to obtain a pooled feature map; and
performing nonlinear activation on the pooled feature map to obtain an activated feature map;
and the output of the last layer of the first convolution neural network model is the running state associated characteristic matrix, and the input of the first layer of the first convolution neural network model is the two-dimensional input state matrix.
5. The enterprise safety production management system of claim 4, wherein the current state data encoding module comprises:
the input vector construction unit is used for arranging the running state information of each mechanical device of the enterprise production system at the current time point into one-dimensional input vectors corresponding to each mechanical device of the enterprise production system according to the time dimension;
a full-concatenation coding unit, configured to perform full-concatenation coding on the input vector using a full-concatenation layer of the time sequence encoder according to the following formula to extract a high-dimensional implicit feature of a feature value at each position in the input vector, where the formula is:wherein X is the input vector, Y is the output vector, W is the weight matrix, B is the offset vector,represents a matrix multiplication;
a one-dimensional convolution coding unit, configured to perform one-dimensional convolution coding on the input vector by using a one-dimensional convolution layer of the time sequence encoder according to the following formula to extract a high-dimensional implicit correlation feature between feature values of each position in the input vector, where the formula is:
wherein, a is the width of the convolution kernel in the x direction, F is the parameter vector of the convolution kernel, G is the local vector matrix operated with the convolution kernel function, and w is the size of the convolution kernel.
6. The enterprise safety production management system of claim 5, wherein the feature distribution correction module comprises:
the vector correction unit is used for correcting the current state feature vector according to the following formula to obtain the corrected current state feature vector;
wherein the formula is:
wherein v is i A feature value, v ', representing the ith position of the current state feature vector' i A feature value representing an ith position of the corrected current state feature vector; and
the matrix correction unit is used for correcting the equipment operation correlation characteristic matrix according to the following formula so as to obtain the corrected equipment operation correlation characteristic matrix;
wherein the formula is:
7. According to claimThe enterprise safety production management system of claim 6, wherein the safety management result generating module is further configured to: processing the classification feature vector using the classifier to obtain the classification result with the following formula: softmax { (W) n ,B n ):...:(W 1 ,B 1 ) I X }, wherein W 1 To W n As a weight matrix, B 1 To B n Is a bias vector and X is the classification feature vector.
8. A management method of an enterprise safety production management system is characterized by comprising the following steps:
acquiring historical data generated by an enterprise production system in a production process, wherein the historical data is the running state information of each mechanical device at a plurality of preset time points in the safety production process of the enterprise production system;
arranging historical data generated by the enterprise production system in a production process into a two-dimensional input state matrix according to a time dimension and a mechanical equipment sample dimension;
performing explicit spatial coding on the two-dimensional input state matrix by using the first convolutional neural network model to obtain an operation state correlation characteristic matrix;
acquiring the running state information of each mechanical device of the enterprise production system at the current time point;
running state information of each mechanical device of the enterprise production system at the current time point is processed through a sequence encoder comprising a one-dimensional convolution layer to obtain a current state feature vector;
respectively correcting the current state feature vector and the equipment operation association feature matrix to obtain a corrected current state feature vector and a corrected equipment operation association feature matrix;
multiplying the corrected current state feature vector serving as a query vector by the corrected equipment operation association feature matrix to obtain a classification feature vector; and
and passing the classified feature vectors through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the production of the enterprise production system at the current time point has risks.
9. The management method of the enterprise safety production management system according to claim 8, wherein the arranging the historical data generated by the enterprise production system in the production process into a two-dimensional input state matrix according to the time dimension and the mechanical equipment sample dimension comprises:
respectively arranging historical data generated by the enterprise production system in the production process into row vectors according to the time dimension to obtain a plurality of row vectors;
and arranging the plurality of row vectors into the two-dimensional input state matrix according to the mechanical equipment sample dimension.
10. The management method of the enterprise safety production management system according to claim 9, wherein the correcting the current state feature vector and the device operation associated feature matrix to obtain a corrected current state feature vector and a corrected device operation associated feature matrix respectively comprises:
correcting the current state feature vector according to the following formula to obtain the corrected current state feature vector;
wherein the formula is:
wherein v is i A feature value representing an ith position of the current-state feature vector,a feature value representing an ith position of the corrected current state feature vector; and
correcting the equipment operation correlation characteristic matrix according to the following formula to obtain the corrected equipment operation correlation characteristic matrix;
wherein the formula is:
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210841063.4A CN115099684B (en) | 2022-07-18 | 2022-07-18 | Enterprise safety production management system and management method thereof |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210841063.4A CN115099684B (en) | 2022-07-18 | 2022-07-18 | Enterprise safety production management system and management method thereof |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115099684A true CN115099684A (en) | 2022-09-23 |
CN115099684B CN115099684B (en) | 2023-04-07 |
Family
ID=83298934
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210841063.4A Active CN115099684B (en) | 2022-07-18 | 2022-07-18 | Enterprise safety production management system and management method thereof |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115099684B (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115879777A (en) * | 2023-03-03 | 2023-03-31 | 吉林大学 | Intelligent petroleum safety deduction system and method based on space-time big data |
CN116026528A (en) * | 2023-01-14 | 2023-04-28 | 慈溪市远辉照明电器有限公司 | High waterproof safe type tri-proof light |
CN115759658B (en) * | 2022-11-24 | 2023-07-18 | 浙江智慧信息产业有限公司 | Enterprise energy consumption data management system suitable for smart city |
CN116975728A (en) * | 2023-08-07 | 2023-10-31 | 山东省地质矿产勘查开发局第八地质大队(山东省第八地质矿产勘查院) | Safety management method and system for coal bed methane drilling engineering |
CN117251718A (en) * | 2023-11-20 | 2023-12-19 | 吉林省拓达环保设备工程有限公司 | Intelligent aeration management system based on artificial intelligence |
Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104348829A (en) * | 2014-09-26 | 2015-02-11 | 智慧城市信息技术有限公司 | Network security situation sensing system and method |
CN107920055A (en) * | 2017-09-27 | 2018-04-17 | 中国银联股份有限公司 | A kind of IP risk evaluating methods and IP Risk Evaluating Systems |
CN108256401A (en) * | 2016-12-29 | 2018-07-06 | 杭州海康威视数字技术股份有限公司 | A kind of method and device for obtaining objective attribute target attribute Feature Semantics |
CN110995475A (en) * | 2019-11-20 | 2020-04-10 | 国网湖北省电力有限公司信息通信公司 | Power communication network fault detection method based on transfer learning |
CN111461392A (en) * | 2020-01-23 | 2020-07-28 | 华中科技大学 | Power failure prediction method and system based on graph neural network |
CN114211975A (en) * | 2021-11-28 | 2022-03-22 | 厦门任鑫实业有限公司 | Charging alarm system for electric automobile and working method thereof |
CN114519394A (en) * | 2022-02-21 | 2022-05-20 | 杭州谣佐科技有限公司 | Router-based configuration method and system of Internet of things equipment |
CN114564993A (en) * | 2022-02-28 | 2022-05-31 | 上海济垚软件科技有限公司 | Fatigue detection system based on multi-source information fusion and operation method thereof |
CN114626448A (en) * | 2022-03-03 | 2022-06-14 | 上海土蜂科技有限公司 | Drawing robot and control method thereof |
CN114641051A (en) * | 2022-03-14 | 2022-06-17 | 杭州岐航科技有限公司 | System using unmanned aerial vehicle as networking relay in smart city and working method thereof |
CN114757659A (en) * | 2022-05-19 | 2022-07-15 | 浙江大学 | Intelligent management system and management method for research and development projects |
-
2022
- 2022-07-18 CN CN202210841063.4A patent/CN115099684B/en active Active
Patent Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104348829A (en) * | 2014-09-26 | 2015-02-11 | 智慧城市信息技术有限公司 | Network security situation sensing system and method |
CN108256401A (en) * | 2016-12-29 | 2018-07-06 | 杭州海康威视数字技术股份有限公司 | A kind of method and device for obtaining objective attribute target attribute Feature Semantics |
CN107920055A (en) * | 2017-09-27 | 2018-04-17 | 中国银联股份有限公司 | A kind of IP risk evaluating methods and IP Risk Evaluating Systems |
CN110995475A (en) * | 2019-11-20 | 2020-04-10 | 国网湖北省电力有限公司信息通信公司 | Power communication network fault detection method based on transfer learning |
CN111461392A (en) * | 2020-01-23 | 2020-07-28 | 华中科技大学 | Power failure prediction method and system based on graph neural network |
CN114211975A (en) * | 2021-11-28 | 2022-03-22 | 厦门任鑫实业有限公司 | Charging alarm system for electric automobile and working method thereof |
CN114519394A (en) * | 2022-02-21 | 2022-05-20 | 杭州谣佐科技有限公司 | Router-based configuration method and system of Internet of things equipment |
CN114564993A (en) * | 2022-02-28 | 2022-05-31 | 上海济垚软件科技有限公司 | Fatigue detection system based on multi-source information fusion and operation method thereof |
CN114626448A (en) * | 2022-03-03 | 2022-06-14 | 上海土蜂科技有限公司 | Drawing robot and control method thereof |
CN114641051A (en) * | 2022-03-14 | 2022-06-17 | 杭州岐航科技有限公司 | System using unmanned aerial vehicle as networking relay in smart city and working method thereof |
CN114757659A (en) * | 2022-05-19 | 2022-07-15 | 浙江大学 | Intelligent management system and management method for research and development projects |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115759658B (en) * | 2022-11-24 | 2023-07-18 | 浙江智慧信息产业有限公司 | Enterprise energy consumption data management system suitable for smart city |
CN116026528A (en) * | 2023-01-14 | 2023-04-28 | 慈溪市远辉照明电器有限公司 | High waterproof safe type tri-proof light |
CN115879777A (en) * | 2023-03-03 | 2023-03-31 | 吉林大学 | Intelligent petroleum safety deduction system and method based on space-time big data |
CN115879777B (en) * | 2023-03-03 | 2023-05-12 | 吉林大学 | Petroleum safety intelligent deduction system and method based on space-time big data |
CN116975728A (en) * | 2023-08-07 | 2023-10-31 | 山东省地质矿产勘查开发局第八地质大队(山东省第八地质矿产勘查院) | Safety management method and system for coal bed methane drilling engineering |
CN116975728B (en) * | 2023-08-07 | 2024-01-26 | 山东省地质矿产勘查开发局第八地质大队(山东省第八地质矿产勘查院) | Safety management method and system for coal bed methane drilling engineering |
CN117251718A (en) * | 2023-11-20 | 2023-12-19 | 吉林省拓达环保设备工程有限公司 | Intelligent aeration management system based on artificial intelligence |
CN117251718B (en) * | 2023-11-20 | 2024-02-13 | 吉林省拓达环保设备工程有限公司 | Intelligent aeration management system based on artificial intelligence |
Also Published As
Publication number | Publication date |
---|---|
CN115099684B (en) | 2023-04-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN115099684B (en) | Enterprise safety production management system and management method thereof | |
US20190219994A1 (en) | Feature extractions to model large-scale complex control systems | |
Chen et al. | An efficient network behavior anomaly detection using a hybrid DBN-LSTM network | |
CN115796173A (en) | Data processing method and system for supervision submission requirements | |
CN115834433B (en) | Data processing method and system based on Internet of things technology | |
CN115759658B (en) | Enterprise energy consumption data management system suitable for smart city | |
CN115324843A (en) | Wind generating set fault diagnosis system and method based on monitoring data | |
CN116579618B (en) | Data processing method, device, equipment and storage medium based on risk management | |
CN116015837A (en) | Intrusion detection method and system for computer network information security | |
CN115147655A (en) | Oil gas gathering and transportation monitoring system and method thereof | |
CN116403253A (en) | Face recognition monitoring management system and method based on convolutional neural network | |
CN116759053A (en) | Medical system prevention and control method and system based on Internet of things system | |
CN115600140A (en) | Fan variable pitch system fault identification method and system based on multi-source data fusion | |
CN116247824A (en) | Control method and system for power equipment | |
Ma et al. | Detecting anomalies in small unmanned aerial systems via graphical normalizing flows | |
CN116785967B (en) | Automatic batching system for electronic grade lithium hexafluorophosphate preparation | |
CN116049905B (en) | Tamper-proof system based on detecting system file change | |
Abdulova | Estimating Risk Potential of Processes and Detailing Coefficients of Multiple-Scale Wavelet Transform | |
CN114912109B (en) | Abnormal behavior sequence identification method and system based on graph embedding | |
CN113781845B (en) | Electronic fence establishing method and system for unmanned aerial vehicle and electronic equipment | |
CN115661758A (en) | Public place crowd density monitoring method and system based on artificial intelligence | |
CN115453976A (en) | Equipment state monitoring method and system | |
MAndelli et al. | Mining nuclear transient data through symbolic conversion | |
US20240103920A1 (en) | Method and system for accelerating the convergence of an iterative computation code of physical parameters of a multi-parameter system | |
CN115410064A (en) | Equipment state monitoring system under intelligent manufacturing and monitoring method thereof |
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 |