CN115879777A - Intelligent petroleum safety deduction system and method based on space-time big data - Google Patents
Intelligent petroleum safety deduction system and method based on space-time big data Download PDFInfo
- Publication number
- CN115879777A CN115879777A CN202310194456.5A CN202310194456A CN115879777A CN 115879777 A CN115879777 A CN 115879777A CN 202310194456 A CN202310194456 A CN 202310194456A CN 115879777 A CN115879777 A CN 115879777A
- Authority
- CN
- China
- Prior art keywords
- full
- attribute
- node
- matrix
- data
- 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
- 239000003208 petroleum Substances 0.000 title claims abstract description 235
- 238000000034 method Methods 0.000 title claims abstract description 54
- 239000011159 matrix material Substances 0.000 claims description 223
- 239000013598 vector Substances 0.000 claims description 41
- 230000008676 import Effects 0.000 claims description 34
- 238000012545 processing Methods 0.000 claims description 29
- 230000009467 reduction Effects 0.000 claims description 28
- 238000013527 convolutional neural network Methods 0.000 claims description 27
- 238000010586 diagram Methods 0.000 claims description 26
- 239000003638 chemical reducing agent Substances 0.000 claims description 25
- 230000006870 function Effects 0.000 claims description 20
- 230000004913 activation Effects 0.000 claims description 12
- 238000010276 construction Methods 0.000 claims description 8
- 238000005457 optimization Methods 0.000 claims description 7
- 238000009826 distribution Methods 0.000 claims description 6
- 238000011176 pooling Methods 0.000 claims description 6
- 230000005540 biological transmission Effects 0.000 claims description 3
- 238000013507 mapping Methods 0.000 abstract description 14
- 238000013135 deep learning Methods 0.000 abstract description 12
- 238000005516 engineering process Methods 0.000 abstract description 5
- 238000013473 artificial intelligence Methods 0.000 abstract description 4
- 238000012549 training Methods 0.000 description 14
- 238000004590 computer program Methods 0.000 description 10
- 230000008569 process Effects 0.000 description 10
- 238000013528 artificial neural network Methods 0.000 description 8
- 238000004519 manufacturing process Methods 0.000 description 8
- 238000003062 neural network model Methods 0.000 description 8
- 238000004458 analytical method Methods 0.000 description 7
- 230000007246 mechanism Effects 0.000 description 6
- 230000003993 interaction Effects 0.000 description 5
- 238000007792 addition Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 4
- 238000011161 development Methods 0.000 description 4
- 230000018109 developmental process Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 238000000605 extraction Methods 0.000 description 4
- 238000005065 mining Methods 0.000 description 4
- 238000003860 storage Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 230000004069 differentiation Effects 0.000 description 2
- 230000005284 excitation Effects 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 239000000446 fuel Substances 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000003058 natural language processing Methods 0.000 description 2
- 230000002787 reinforcement Effects 0.000 description 2
- 230000003014 reinforcing effect Effects 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 238000013519 translation Methods 0.000 description 2
- 238000011144 upstream manufacturing Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000006798 recombination Effects 0.000 description 1
- 238000005215 recombination Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000005728 strengthening Methods 0.000 description 1
Images
Abstract
The invention discloses an intelligent petroleum safety deduction system and method based on space-time big data, relating to the technical field of intelligent deduction and obtaining a plurality of nodes related to petroleum safety and attribute data of the nodes; the method comprises the steps of establishing a mapping relation among a plurality of nodes and a plurality of attribute data by adopting an artificial intelligence technology based on deep learning, and accurately evaluating a grade label of the oil supply interruption risk based on the mapping relation, so that measures such as correspondingly releasing reserves, increasing the yield and managing and controlling traffic are taken based on the grade label of the risk, and the safety of oil is ensured.
Description
Technical Field
The present application relates to the field of intelligent deduction technology, and more particularly, to an intelligent deduction system for petroleum safety based on space-time big data and a method thereof.
Background
Petroleum is an important energy source for the blood and national development of industry, and the stable supply of petroleum is the key of economic development. Petroleum production, supply, etc. may be affected by international conditions, geopolitics, energy patterns, economic policies, disasters, weather, etc. In response to the impact of one or more major emergencies on the energy (especially petroleum) supply in our country, region, and the world, many scientists have focused on the impact and risk of various instability factors.
At present, although the yield, consumption, import, export, price and the like of oil are in a constantly fluctuating state for a long time internationally, the supply and demand of the oil are generally kept balanced, namely dynamic oil safety is achieved. When a certain event or events occur, the balance is broken to cause the petroleum safety problem, which relates to the upstream, middle and downstream of the petroleum whole industry chain, that is, each link of production, transportation, processing, consumption and the like of petroleum can be influenced by various factors due to the occurrence of single or multiple events to the whole industry chain, thereby breaking the regional, national and even global petroleum safety balance. However, the traditional petroleum safety strategy analysis method has the defects of untimely data updating, difficult processing of complex data, low efficiency and the like, and is difficult to adapt to the requirement of rapid analysis in the big data era.
Therefore, an optimized intelligent deduction system for petroleum safety based on space-time big data is expected.
Disclosure of Invention
The present application is proposed to solve the above-mentioned technical problems. The embodiment of the application provides an intelligent petroleum safety deduction system and method based on space-time big data, wherein the system and method acquire a plurality of nodes related to petroleum safety and attribute data of the nodes; the method comprises the steps of establishing a mapping relation among a plurality of nodes and a plurality of attribute data by adopting an artificial intelligence technology based on deep learning, and accurately evaluating a grade label of the petroleum supply interruption risk based on the mapping relation, so that corresponding measures such as reserve release, yield increase, traffic control and the like are taken based on the grade label of the risk, and the safety of petroleum is ensured.
In a first aspect, an intelligent petroleum safety deduction system based on space-time big data is provided, which comprises:
the system comprises a data acquisition module, a data processing module and a data processing module, wherein the data acquisition module is used for acquiring a plurality of nodes related to petroleum safety and attribute data of the nodes, the nodes comprise oil producing countries, import countries, pipelines, export ports, import ports, straits, military bases, oil refineries and gas stations, and the attribute data comprise reserves, yields, consumption, export volumes, import volumes, prices, GDP (gross distribution provider), population, urbanization rates, passenger volumes, freight volumes and automobile holding volumes;
the association construction module is used for arranging the attribute data of the nodes into a full-node full-attribute input matrix according to the node dimension and the attribute dimension respectively;
the dimension reduction module is used for enabling the full-node full-attribute input matrix to pass through a data dimension reducer based on an automatic encoder to obtain a full-node full-attribute low-dimensional feature matrix after dimension reduction;
the spatial feature enhancement module is used for enabling the dimensionality-reduced full-node full-attribute low-dimensional feature matrix to pass through a convolutional neural network model using spatial attention so as to obtain a full-node full-attribute petroleum safety feature matrix;
the optimization module is used for highlighting the feature discrimination of the full-node full-attribute petroleum safety feature matrix to obtain an optimized full-node full-attribute petroleum safety feature matrix;
and the risk early warning module is used for enabling the optimized full-node full-attribute petroleum safety feature matrix to pass through a classifier to obtain a classification result, and the classification result is used for representing a petroleum supply interruption risk level label.
In the above intelligent petroleum safety deduction system based on spatio-temporal big data, the correlation construction module includes: the row vector arrangement unit is used for respectively arranging the attribute data of the nodes into attribute input row vectors according to the dimension of the nodes; and the two-dimensional matrixing unit is used for performing two-dimensional arrangement on the attribute input row vectors according to the attribute dimensions to obtain the full-node full-attribute input matrix.
In the above intelligent petroleum safety deduction system based on space-time big data, the dimension reduction module includes: the coding unit is used for inputting the full-node full-attribute input matrix into a coder of the data dimensionality reducer, wherein the coder uses a convolution layer to perform explicit space coding on the full-node full-attribute input matrix to obtain full-node full-attribute characteristics; and the decoding unit is used for inputting the full-node full-attribute feature into a decoder of the data dimensionality reducer, wherein the decoder uses a deconvolution layer to perform deconvolution processing on the full-node full-attribute feature so as to obtain a full-node full-attribute low-dimensional feature matrix after dimensionality reduction.
In the above intelligent petroleum safety deduction system based on spatio-temporal big data, the spatial feature enhancement module is configured to: the layers of the convolutional neural network model using spatial attention perform the following operations on input data in the forward direction of the layers: performing convolution processing on input data to generate a convolution characteristic diagram; pooling the convolved feature map to generate a pooled feature map; performing nonlinear activation on the pooled feature map to generate an activated feature map; calculating the mean value of each position of the activation feature map along the channel dimension to generate a spatial feature matrix; calculating Softmax-like function values of all positions in the spatial feature matrix to obtain a spatial score matrix; calculating the position-based multiplication of the spatial feature matrix and the spatial score matrix to obtain a feature matrix; wherein the feature matrix output by the last layer of the convolutional neural network model using spatial attention is the full-node full-attribute petroleum safety feature matrix.
In the above intelligent deduction system for petroleum safety based on space-time big data, the optimization module is configured to: performing characteristic discrimination highlighting on the full-node full-attribute petroleum safety characteristic matrix by using the following formula to obtain an optimized full-node full-attribute petroleum safety characteristic matrix; wherein the formula is:
wherein ,
is the full-node full-attribute petroleum safety feature matrix, based on the total node attribute>
and />
Is a predetermined hyper-parameter>
and />
Represents a position-wise addition and subtraction of a feature matrix, and>
representing convolution operation by a single convolution layer>
The optimized full-node full-attribute petroleum safety feature matrix is obtained.
In the above-mentioned petroleum safety intelligence deduction system based on big data of space-time, the risk early warning module includes: the matrix expansion unit is used for expanding the optimized full-node full-attribute petroleum safety feature matrix into classification feature vectors according to row vectors or column vectors; a full-concatenation encoding unit, configured to perform full-concatenation encoding on the classification feature vector using a plurality of full-concatenation layers of the classifier to obtain an encoded classification feature vector; and the classification unit is used for enabling the coded classification feature vector to pass through a Softmax classification function of the classifier so as to obtain the classification result.
In a second aspect, an intelligent deduction method for petroleum safety based on space-time big data is provided, which comprises the following steps:
acquiring a plurality of nodes related to petroleum safety and attribute data of the nodes, wherein the nodes comprise oil producing countries, import countries, pipelines, export ports, import ports, straits, military bases, oil refineries and gas stations, and the attribute data comprise reserves, yields, consumption, export volumes, import volumes, prices, GDPs, population, urbanization rates, passenger volumes, freight volumes and automobile reserves;
respectively arranging the attribute data of the plurality of nodes into a full-node full-attribute input matrix according to the node dimension and the attribute dimension;
enabling the full-node full-attribute input matrix to pass through a data dimension reducer based on an automatic encoder to obtain a full-node full-attribute low-dimensional feature matrix after dimension reduction;
obtaining a full-node full-attribute petroleum safety feature matrix by using the dimensionality-reduced full-node full-attribute low-dimensional feature matrix through a convolutional neural network model using space attention;
performing characteristic discrimination highlighting on the full-node full-attribute petroleum safety characteristic matrix to obtain an optimized full-node full-attribute petroleum safety characteristic matrix;
and passing the optimized full-node full-attribute petroleum safety feature matrix through a classifier to obtain a classification result, wherein the classification result is used for representing a petroleum supply failure risk grade label.
In the above intelligent deduction method for petroleum safety based on space-time big data, the arranging the attribute data of the plurality of nodes into a full-node full-attribute input matrix according to the node dimension and the attribute dimension respectively comprises: respectively arranging the attribute data of the plurality of nodes into attribute input row vectors according to the dimension of the nodes; and performing two-dimensional arrangement on the attribute input row vectors according to the attribute dimensions to obtain the full-node full-attribute input matrix.
In the above intelligent petroleum safety deduction method based on space-time big data, the full-node full-attribute input matrix is passed through a data dimension reducer based on an automatic encoder to obtain a dimension-reduced full-node full-attribute low-dimensional feature matrix, which includes: inputting the full-node full-attribute input matrix into an encoder of the data dimensionality reducer, wherein the encoder uses a convolution layer to perform explicit spatial coding on the full-node full-attribute input matrix to obtain full-node full-attribute characteristics; and inputting the full-node full-attribute features into a decoder of the data dimensionality reducer, wherein the decoder performs deconvolution processing on the full-node full-attribute features by using a deconvolution layer to obtain a full-node full-attribute low-dimensional feature matrix after dimensionality reduction.
In the above intelligent petroleum safety deduction method based on space-time big data, the method for obtaining the full-node full-attribute petroleum safety feature matrix by using the convolution neural network model of spatial attention with the full-node full-attribute low-dimensional feature matrix after dimensionality reduction comprises the following steps: the layers of the convolutional neural network model using spatial attention perform the following operations on input data in the forward direction of the layers: performing convolution processing on input data to generate a convolution characteristic diagram; pooling the convolved feature map to generate a pooled feature map; performing nonlinear activation on the pooled feature map to generate an activated feature map; calculating a mean of the positions of the activation feature map along a channel dimension to generate a spatial feature matrix; calculating Softmax-like function values of all positions in the spatial feature matrix to obtain a spatial score matrix; calculating the position-based multiplication of the spatial feature matrix and the spatial score matrix to obtain a feature matrix; wherein the feature matrix output by the last layer of the convolutional neural network model using spatial attention is the full-node full-attribute petroleum safety feature matrix.
Compared with the prior art, the intelligent petroleum safety deduction system and method based on the space-time big data, which are provided by the application, are used for acquiring a plurality of nodes related to petroleum safety and attribute data of the nodes; the method comprises the steps of establishing a mapping relation among a plurality of nodes and a plurality of attribute data by adopting an artificial intelligence technology based on deep learning, and accurately evaluating a grade label of the petroleum supply interruption risk based on the mapping relation, so that corresponding measures such as reserve release, yield increase, traffic control and the like are taken based on the grade label of the risk, and the safety of petroleum is ensured.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is an application scenario diagram of an intelligent petroleum safety deduction system based on space-time big data according to an embodiment of the application.
FIG. 2 is a block diagram of an intelligent petroleum safety deduction system based on space-time big data according to an embodiment of the application.
FIG. 3 is a block diagram of the correlation construction module in the intelligent petroleum safety deduction system based on spatiotemporal big data according to an embodiment of the present application.
FIG. 4 is a block diagram of the dimension reduction module in the intelligent petroleum safety deduction system based on spatiotemporal big data according to the embodiment of the present application.
Fig. 5 is a block diagram of the risk early warning module in an intelligent deduction system for petroleum safety based on spatiotemporal big data according to an embodiment of the present application.
FIG. 6 is a flowchart of an intelligent deduction method for petroleum safety based on space-time big data according to an embodiment of the present application.
FIG. 7 is a schematic diagram of a system architecture of a space-time big data-based intelligent petroleum safety deduction method according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
Unless otherwise defined, all technical and scientific terms used in the examples of this application have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of the present application.
In the description of the embodiments of the present application, it should be noted that, unless otherwise specified and limited, the term "connected" should be interpreted broadly, for example, as an electrical connection, a communication between two elements, a direct connection, or an indirect connection via an intermediate, and the specific meaning of the terms may be understood by those skilled in the art according to specific situations.
It should be noted that the terms "first \ second \ third" referred to in the embodiments of the present application are only used for distinguishing similar objects, and do not represent a specific ordering for the objects, and it should be understood that "first \ second \ third" may interchange a specific order or sequence when allowed. It should be understood that "first \ second \ third" distinct objects may be interchanged under appropriate circumstances such that the embodiments of the application described herein may be implemented in an order other than those illustrated or described herein.
As described above, at present, although international oil production, consumption, import, export, price, etc. are in a state of constant fluctuation for a long time, the oil supply and demand are generally balanced, that is, dynamic oil safety is achieved. When a certain event or events occur, the balance is broken to cause the petroleum safety problem, which relates to the upstream, middle and downstream of the petroleum whole industry chain, that is, each link of production, transportation, processing, consumption and the like of petroleum can be influenced by various factors due to the occurrence of single or multiple events to the whole industry chain, thereby breaking the regional, national and even global petroleum safety balance. However, the traditional petroleum safety strategy analysis method has the defects of untimely data updating, difficult processing of complex data, low efficiency and the like, and is difficult to adapt to the requirement of rapid analysis in the big data era. Therefore, an optimized intelligent deduction system for petroleum safety based on space-time big data is expected.
Specifically, in the technical scheme of the application, under the technical support of big data, cloud computing, the internet of things and the like, the petroleum safety space-time big data is used as a data resource, when one or more events suddenly appear, key abnormal information is rapidly extracted and identified, short-term or medium-term influences generated by one or more links of the whole industrial chain are analyzed, and countermeasures such as reserve release, yield increase, traffic control and the like are taken according to early warning levels, so that a novel 'event-computing-response-countermeasures' working mode facing petroleum safety is realized.
Accordingly, when consideration is given to the fact that risk abnormity early warning is carried out based on petroleum safety space-time big data, and measures such as reserve release, yield increase, traffic control and the like are taken according to early warning levels, most importantly, the petroleum safety space-time big data is analyzed to output a fuel cut-off risk level label. However, it is considered that the petroleum safety space-time big data includes space data and attribute data, wherein the space data is various types of petroleum-related data with space geometric features (points, lines, faces), i.e. node data, which mainly include reserves, output (measured in units of million tons, thousand barrels/day, etc.) consumption, output, input, price (year, month, day), GDP, urbanization rate, passenger transport volume, cargo transport volume, automobile holding volume, etc. of petroleum in various countries around the world. Since the plurality of node data in the petroleum safety space-time big data have spatial node relevance information, each node data includes the plurality of attribute data, and the attribute data of the plurality of nodes also have relevance relations between attributes, the petroleum risk early warning cannot be performed based on single data change. In the process, how to establish the mapping relationship among the nodes and the attribute data to accurately evaluate the grade label of the petroleum outage risk is difficult, so that the corresponding measures such as reserve release, yield increase, traffic control and the like are taken based on the grade label of the risk, and the safety of petroleum is ensured.
In recent years, deep learning and neural networks have been widely used in the fields of computer vision, natural language processing, text signal processing, and the like. In addition, deep learning and neural networks also exhibit a level close to or even exceeding that of humans in the fields of image classification, object detection, semantic segmentation, text translation and the like.
The deep learning and the development of the neural network provide new solution ideas and schemes for mining complex mapping relations between the nodes and the attribute data. Those skilled in the art will appreciate that the deep neural network model based on deep learning can adjust the parameters of the deep neural network model through a proper training strategy, for example, through a gradient descent back propagation algorithm, so as to enable the deep neural network model to simulate complex nonlinear relations between things, and this is obviously suitable for simulating and establishing complex mapping relations between the plurality of nodes and the plurality of attribute data.
Specifically, in the technical solution of the present application, first, a plurality of nodes related to petroleum safety and attribute data of the plurality of nodes are obtained, where the plurality of nodes include a petroleum producing country, an importing country, a pipeline, an export port, an import port, a strait, a military base, a petroleum refinery, and a gas station, and the attribute data include a reserve volume, a production volume, a consumption volume, an export volume, an import volume, a price, a GDP, a population, an urbanization rate, a passenger transportation volume, a cargo transportation volume, and a vehicle holding volume. Namely, petroleum safety space-time big data are used as input data to carry out analysis processing, so as to evaluate a petroleum outage risk level label.
Next, considering that there is a correlation between the nodes related to petroleum safety, for example, there is a linear correlation between the port of export and the port of import, the nodes related to petroleum safety include the attribute data of the nodes, and there is a correlation between the attribute data of the nodes. Therefore, in order to accurately analyze the change characteristic information of the petroleum, it is necessary to arrange the attribute data of the plurality of nodes into a full-node full-attribute input matrix according to the node dimension and the attribute dimension, respectively, so as to integrate the distribution information of the attribute data of each node and each node.
Then, considering that for the petroleum safety space-time big data, the problem that the full-node full-attribute input matrix after fusion and integration often has high sparsity in a high-dimensional space, so that the mode extraction analysis cannot be directly performed, and dimension reduction is usually required. And considering the activity rule of the internal data for accurately capturing the petroleum safety space-time big data due to the potential low-dimensional characteristics of the data which can be acquired by the representation learning based on the self-encoder. It should be appreciated that the autoencoder neural network is an unsupervised learning algorithm that applies back propagation to a set of unlabeled training samples
，/>
Representing each training sample in a set of training samples, the self-encoder continuously learns the function pick>
So that->
Here, is>
Is a constant learning function of the self-encoder>
The weight term W and the bias term b are adjusted such that the target output is equal to the input, i.e. yi = xi, where yi denotes the target output and xi denotes the target input. Therefore, in the technical solution of the present application, the full-node full-attribute input matrix is further passed through a data dimension reducer based on an automatic encoder to obtain a full-node full-attribute low-dimensional feature matrix after dimension reduction.
Further, a convolutional neural network model with excellent performance in the aspect of local implicit associated feature extraction is used for feature mining of the full-node full-attribute low-dimensional feature matrix after dimensionality reduction, and particularly, when the petroleum supply failure risk early warning is actually carried out, the spatial position features of the nodes related to petroleum safety need to be concerned. Since the attention mechanism can select the focus position, a more distinguishing feature representation is generated, and the feature added into the attention module can be adaptively changed along with the deepening of the network. Therefore, in the technical scheme of the application, in order to improve the accuracy of the petroleum supply interruption risk level early warning, a spatial attention mechanism is further used for feature enhancement. Specifically, the dimensionality-reduced full-node full-attribute low-dimensional feature matrix is processed in a convolutional neural network model using space attention to extract relevance feature distribution information focusing on the spatial position of each node among the attribute data of the nodes in the dimensionality-reduced full-node full-attribute low-dimensional feature matrix, so that a full-node full-attribute petroleum safety feature matrix is obtained.
And then, taking the full-node full-attribute petroleum safety feature matrix as a classification feature matrix, and performing classification processing in a classifier to obtain a label classification result for representing the petroleum supply failure risk level. That is, in the technical solution of the present application, the label of the classifier is a petroleum outage risk level label, wherein the classifier determines which classification label the classification feature matrix belongs to through a soft maximum function. Therefore, the grade label of the oil supply interruption risk can be accurately evaluated, so that measures such as correspondingly releasing reserves, increasing the yield and managing and controlling traffic are taken based on the grade label of the risk, and the safety of oil is ensured.
Here, when the dimensionality reduced full-node full-attribute low-dimensional feature matrix is used for obtaining the full-node full-attribute petroleum safety feature matrix through a convolutional neural network model with spatial attention, due to a spatial attention mechanism, the eigenvalue of some positions in the full-node full-attribute petroleum safety feature matrix has more significant importance relative to the eigenvalue of other positions, so if the eigenvalue of the full-node full-attribute petroleum safety feature matrix can be effectively distinguished in a classification task, the training speed of the model and the accuracy of a classification result can be obviously improved.
Therefore, the full-node full-attribute petroleum safety feature matrix is described in the application
Performing interaction strengthening based on distinguishable physical stimulus, specifically:
wherein 
Is the intensified full-node full-attribute petroleum safety characteristic matrix>
and />
Is a predetermined hyper-parameter, is>
and />
Means for adding and subtracting by position, the division means each position of the characteristic matrix divided by the corresponding value and +>
Representing the convolution operation through a single convolutional layer.
Here, the discriminative physical incentive-based interaction reinforcement is used for improving the interaction between the feature space and the solution space of the classification problem in the back propagation process through gradient descent, and extracts and imitates the feasible feature (actionable feature) in a manner similar to the physical incentive, thereby obtaining the physical expression of the feasible feature with gradient discriminative ability by using the general-purpose low-dimensional conductive physical incentive manner, thereby reinforcing the full-node full-attribute petroleum safety feature matrix in the training process
Active part in the system to promote the strengthened full-node full-attribute petroleum safety characteristic matrix
Training speed under a classification task and accuracy of a classification result of the trained classification features. Therefore, the grade label of the oil supply interruption risk can be accurately evaluated, so that measures such as correspondingly releasing reserves, increasing the yield and managing and controlling traffic are taken based on the grade label of the risk, and the safety of oil is ensured.
Fig. 1 is an application scenario diagram of a petroleum safety intelligent deduction system based on spatiotemporal big data according to an embodiment of the present application. As shown in fig. 1, in this application scenario, first, a plurality of nodes related to petroleum safety including a country of oil production, a country of import, a pipeline, an export port, an import port, a strait, a military base, an oil refinery, and a gas station, and attribute data (for example, C as illustrated in fig. 1) of the plurality of nodes including a reserve, a yield, a consumption, an export, an import, a price, a GDP, a population, a urbanization rate, a passenger transportation volume, a cargo transportation volume, and a car holding volume are obtained; then, the acquired attribute data of the plurality of nodes is input into a server (for example, S as illustrated in fig. 1) deployed with a petroleum safety intelligent deduction algorithm based on space-time big data, wherein the server can process the attribute data of the plurality of nodes based on the petroleum safety intelligent deduction algorithm based on space-time big data to generate a label for representing a petroleum outage risk level.
Having described the general principles of the present application, various non-limiting embodiments of the present application will now be described with reference to the accompanying drawings.
In one embodiment of the present application, fig. 2 is a block diagram of an intelligent petroleum safety deduction system based on spatiotemporal big data according to an embodiment of the present application. As shown in fig. 2, the intelligent petroleum safety deduction system 100 based on spatiotemporal big data according to the embodiment of the present application includes: a data acquisition module 110, configured to acquire a plurality of nodes related to petroleum safety and attribute data of the plurality of nodes, where the plurality of nodes include oil country, import country, pipeline, export port, import port, strait, military base, oil refinery, and gas station, and the attribute data includes reserve volume, yield, consumption volume, export volume, import volume, price, GDP, population, urbanization rate, passenger transportation volume, freight volume, and vehicle holding volume; the association structure module 120 is configured to arrange the attribute data of the plurality of nodes into a full-node full-attribute input matrix according to a node dimension and an attribute dimension, respectively; a dimension reduction module 130, configured to pass the full-node full-attribute input matrix through a data dimension reducer based on an automatic encoder to obtain a full-node full-attribute low-dimensional feature matrix after dimension reduction; the spatial feature enhancement module 140 is configured to obtain a full-node full-attribute petroleum safety feature matrix by passing the dimensionality-reduced full-node full-attribute low-dimensional feature matrix through a convolutional neural network model using spatial attention; the optimization module 150 is used for highlighting the feature discrimination of the full-node full-attribute petroleum safety feature matrix to obtain an optimized full-node full-attribute petroleum safety feature matrix; and a risk early warning module 160, configured to pass the optimized full-node full-attribute petroleum safety feature matrix through a classifier to obtain a classification result, where the classification result is used to represent a petroleum outage risk level label.
Specifically, in this embodiment, the data acquisition module 110 is configured to obtain a plurality of nodes related to petroleum safety and attribute data of the plurality of nodes, where the plurality of nodes include oil country, import country, pipeline, export port, import port, strait, military base, oil refinery, and gas station, and the attribute data includes reserve, yield, consumption, export volume, import volume, price, GDP, population, urbanization rate, passenger transportation volume, cargo transportation volume, and automobile holding volume.
Specifically, in the technical scheme of the application, under the technical support of big data, cloud computing, the internet of things and the like, the petroleum safety space-time big data is used as a data resource, when one or more events suddenly appear, key abnormal information is rapidly extracted and identified, short-term or medium-term influences generated by one or more links of the whole industrial chain are analyzed, and countermeasures such as reserve release, yield increase, traffic control and the like are taken according to early warning levels, so that a novel 'event-computing-response-countermeasures' working mode facing petroleum safety is realized.
Accordingly, when consideration is given to the fact that risk abnormity early warning is carried out based on petroleum safety space-time big data, and measures such as reserve release, yield increase, traffic control and the like are taken according to early warning levels, most importantly, the petroleum safety space-time big data is analyzed to output a fuel cut-off risk level label. However, it is considered that the petroleum safety space-time big data includes space data and attribute data, wherein the space data is various types of petroleum-related data with space geometric features (points, lines, faces), i.e. node data, which mainly include reserves, output (measured in units of million tons, thousand barrels/day, etc.) consumption, output, input, price (year, month, day), GDP, urbanization rate, passenger transport volume, cargo transport volume, automobile holding volume, etc. of petroleum in various countries around the world. Because the plurality of node data in the petroleum safety space-time big data have spatial node relevance information, each node data comprises the plurality of attribute data, and the attribute data of the plurality of nodes also have relevance relation among attributes, the risk early warning of petroleum cannot be carried out based on single data change. In the process, how to establish the mapping relationship among the plurality of nodes and the plurality of attribute data so as to accurately evaluate the grade label of the oil outage risk, so that countermeasures such as correspondingly releasing reserves, increasing the yield and managing and controlling traffic are taken based on the grade label of the risk, and the safety of the oil is ensured.
In recent years, deep learning and neural networks have been widely used in the fields of computer vision, natural language processing, text signal processing, and the like. In addition, deep learning and neural networks also exhibit a level close to or even exceeding that of humans in the fields of image classification, object detection, semantic segmentation, text translation, and the like.
Deep learning and development of neural networks provide new solutions and schemes for mining complex mapping relationships between the plurality of nodes and the plurality of attribute data. Those skilled in the art will appreciate that the deep neural network model based on deep learning can adjust the parameters of the deep neural network model through a proper training strategy, for example, through a gradient descent back propagation algorithm, so as to enable the deep neural network model to simulate complex nonlinear relations between things, and this is obviously suitable for simulating and establishing complex mapping relations between the plurality of nodes and the plurality of attribute data.
Specifically, in the technical solution of the present application, first, a plurality of nodes related to petroleum safety and attribute data of the plurality of nodes are obtained, where the plurality of nodes include a petroleum producing country, an importing country, a pipeline, an export port, an import port, a strait, a military base, a petroleum refinery, and a gas station, and the attribute data include a reserve volume, a production volume, a consumption volume, an export volume, an import volume, a price, a GDP, a population, an urbanization rate, a passenger transportation volume, a cargo transportation volume, and a vehicle holding volume. Namely, the petroleum safety space-time big data is used as input data to be analyzed and processed, so as to evaluate the petroleum outage risk level label.
Specifically, in this embodiment of the present application, the association and construction module 120 is configured to arrange the attribute data of the plurality of nodes into a full-node full-attribute input matrix according to a node dimension and an attribute dimension, respectively. Next, considering that there is a correlation between the nodes related to petroleum safety, for example, there is a linear correlation between the port of export and the port of import, the nodes related to petroleum safety include the attribute data of the nodes, and there is a correlation between the attribute data of the nodes. Therefore, in order to accurately analyze the change characteristic information of the petroleum, it is necessary to arrange the attribute data of the plurality of nodes into a full-node full-attribute input matrix according to the node dimension and the attribute dimension, respectively, so as to integrate the distribution information of the attribute data of each node and each node.
Fig. 3 is a block diagram of the correlation construction module in the space-time big data-based intelligent petroleum safety deduction system according to the embodiment of the present application, and as shown in fig. 3, the correlation construction module 120 includes: a row vector arrangement unit 121, configured to arrange the attribute data of the multiple nodes into attribute-used input row vectors according to the node dimensions, respectively; and a two-dimensional matrixing unit 122, configured to perform two-dimensional arrangement on the attribute input row vectors according to the attribute dimensions to obtain the full-node full-attribute input matrix.
Specifically, in this embodiment of the present application, the dimension reduction module 130 is configured to pass the full-node full-attribute input matrix through an auto-encoder-based data dimension reducer to obtain a reduced full-node full attributeAnd (5) a sex low-dimensional feature matrix. Then, considering that for the petroleum safety space-time big data, the problem that the full-node full-attribute input matrix after fusion and integration often has high sparsity in a high-dimensional space, the mode extraction analysis cannot be directly performed, and dimension reduction is usually required. And considering the activity rule of the internal data for accurately capturing the petroleum safety space-time big data due to the potential low-dimensional characteristics of the data which can be obtained by the representation learning based on the self-encoder. It should be appreciated that the autoencoder neural network is an unsupervised learning algorithm that applies back propagation for a set of unlabeled training samples
，/>
Representing individual training samples in a training sample set, the self-encoder continuously learns a function>
So that->
Here, based on>
Is a constant learning function of the self-encoder>
The weight term W and the bias term b are adjusted such that the target output is equal to the input, i.e. yi = xi, where yi denotes the target output and xi denotes the target input. Therefore, in the technical solution of the present application, the full-node full-attribute input matrix is further passed through a data dimension reducer based on an automatic encoder to obtain a full-node full-attribute low-dimensional feature matrix after dimension reduction.
Fig. 4 is a block diagram of the dimension reduction module in the space-time big data-based intelligent petroleum safety deduction system according to the embodiment of the application, and as shown in fig. 4, the dimension reduction module 130 includes: an encoding unit 131, configured to input the full-node full-attribute input matrix into an encoder of the data dimension reducer, where the encoder performs explicit spatial encoding on the full-node full-attribute input matrix using a convolutional layer to obtain a full-node full-attribute feature; and a decoding unit 132, configured to input the full-node full-attribute feature into a decoder of the data dimension reducer, where the decoder performs deconvolution processing on the full-node full-attribute feature by using a deconvolution layer to obtain a full-node full-attribute low-dimensional feature matrix after dimension reduction.
Specifically, in this embodiment of the present application, the spatial feature enhancement module 140 is configured to pass the dimensionality-reduced full-node full-attribute low-dimensional feature matrix through a convolutional neural network model using spatial attention to obtain a full-node full-attribute petroleum safety feature matrix. Further, a convolutional neural network model with excellent performance in the aspect of local implicit associated feature extraction is used for feature mining of the full-node full-attribute low-dimensional feature matrix after dimensionality reduction, and particularly, when the petroleum supply failure risk early warning is actually carried out, the spatial position features of the nodes related to petroleum safety need to be concerned.
Since the attention mechanism can select the focus position, a more distinguishing feature representation is generated, and the feature added into the attention module can be adaptively changed along with the deepening of the network. Therefore, in the technical scheme of the application, in order to improve the accuracy of the petroleum supply interruption risk level early warning, a spatial attention mechanism is further used for feature enhancement. Specifically, the dimensionality-reduced full-node full-attribute low-dimensional feature matrix is processed in a convolutional neural network model using space attention to extract relevance feature distribution information focusing on the spatial position of each node among the attribute data of the nodes in the dimensionality-reduced full-node full-attribute low-dimensional feature matrix, so that a full-node full-attribute petroleum safety feature matrix is obtained.
Wherein the spatial feature enhancement module 140 is configured to: each layer of the convolutional neural network model using spatial attention performs the following operations on input data in the forward transmission process of the layer: performing convolution processing on input data to generate a convolution characteristic diagram; pooling the convolved feature map to generate a pooled feature map; performing nonlinear activation on the pooled feature map to generate an activated feature map; calculating the mean value of each position of the activation feature map along the channel dimension to generate a spatial feature matrix; calculating Softmax-like function values of all positions in the spatial feature matrix to obtain a spatial score matrix; calculating the position-based multiplication of the spatial feature matrix and the spatial score matrix to obtain a feature matrix; wherein the feature matrix output by the last layer of the convolutional neural network model using spatial attention is the full-node full-attribute petroleum safety feature matrix.
Specifically, in this embodiment, the optimization module 150 is configured to perform feature differentiation highlighting on the full-node full-attribute petroleum safety feature matrix to obtain an optimized full-node full-attribute petroleum safety feature matrix. Here, when the dimensionality reduced full-node full-attribute low-dimensional feature matrix is used for obtaining the full-node full-attribute petroleum safety feature matrix through a convolutional neural network model with spatial attention, due to a spatial attention mechanism, the eigenvalue of some positions in the full-node full-attribute petroleum safety feature matrix has more significant importance relative to the eigenvalue of other positions, so if the eigenvalue of the full-node full-attribute petroleum safety feature matrix can be effectively distinguished in a classification task, the training speed of the model and the accuracy of a classification result can be obviously improved.
Accordingly, the present application is directed to the full-node full-attribute petroleum safety feature matrix, i.e., the optimization module 150, configured to: performing characteristic discrimination highlighting on the full-node full-attribute petroleum safety characteristic matrix by using the following formula to obtain the optimized full-node full-attribute petroleum safety characteristic matrix; wherein the formula is:
wherein ,
is the full node full Attribute Petroleum Security feature matrix, based upon>
and />
Is a predetermined hyper-parameter>
and />
Represents a position-wise addition and subtraction of a feature matrix, and>
representing convolution operation by a single convolution layer>
The optimized full-node full-attribute petroleum safety feature matrix is obtained.
Here, the discriminative physical excitation-based interaction reinforcement is used to promote interaction between a feature space and a solution space of a classification problem in a back propagation process through gradient descent, which extracts and imitates actionable features (actionable features) in a physical excitation-like manner, thereby obtaining a physical expression of the actionable features with gradient discriminativity using a general-purpose low-dimensional conductive physical excitation manner, thereby reinforcing the full-node full-attribute petroleum safety feature matrix in a training process
Active part in the interior to improve the strengthened full-node safetyFull feature matrix
Training speed under a classification task and accuracy of a classification result of the trained classification features. Therefore, the grade label of the oil supply interruption risk can be accurately evaluated, so that measures such as correspondingly releasing reserves, increasing the yield and managing and controlling traffic are taken based on the grade label of the risk, and the safety of oil is ensured.
Specifically, in this embodiment of the present application, the risk pre-warning module 160 is configured to pass the optimized full-node full-attribute oil safety feature matrix through a classifier to obtain a classification result, where the classification result is used to represent an oil outage risk level label. And then, taking the full-node full-attribute petroleum safety feature matrix as a classification feature matrix to perform classification processing in a classifier so as to obtain a label classification result for representing the petroleum supply failure risk level. That is, in the technical solution of the present application, the label of the classifier is a petroleum outage risk level label, wherein the classifier determines to which classification label the classification feature matrix belongs through a soft maximum function. Therefore, the grade label of the oil supply interruption risk can be accurately evaluated, so that measures such as correspondingly releasing reserves, increasing the yield and managing and controlling traffic are taken based on the grade label of the risk, and the safety of oil is ensured.
Fig. 5 is a block diagram of the risk early warning module in the intelligent petroleum safety deduction system based on spatiotemporal big data according to an embodiment of the present application, and as shown in fig. 5, the risk early warning module 160 includes: a matrix expansion unit 161, configured to expand the optimized full-node full-attribute petroleum safety feature matrix into classification feature vectors according to row vectors or column vectors; a full-concatenation encoding unit 162, configured to perform full-concatenation encoding on the classification feature vector using a plurality of full-concatenation layers of the classifier to obtain an encoded classification feature vector; and a classification unit 163, configured to pass the encoded classification feature vector through a Softmax classification function of the classifier to obtain the classification result.
In a specific example of the present application, the classifier is used to process the optimized full-node full-attribute petroleum safety feature matrix according to the following formula to obtain the classification result; wherein the formula is:
In conclusion, a space-time big data based intelligent petroleum safety deduction system 100 based on the embodiment of the present application is illustrated, which acquires a plurality of nodes related to petroleum safety and attribute data of the plurality of nodes; the method comprises the steps of establishing a mapping relation among a plurality of nodes and a plurality of attribute data by adopting an artificial intelligence technology based on deep learning, and accurately evaluating a grade label of the oil supply interruption risk based on the mapping relation, so that measures such as correspondingly releasing reserves, increasing the yield and managing and controlling traffic are taken based on the grade label of the risk, and the safety of oil is ensured.
As described above, the petroleum safety intelligence deduction system 100 based on spatiotemporal big data according to the embodiment of the present application may be implemented in various terminal devices, such as a server for petroleum safety intelligence deduction based on spatiotemporal big data, and the like. In one example, the petroleum safety intelligent deduction system 100 based on the spatio-temporal big data according to the embodiment of the present application may be integrated into a terminal device as one software module and/or hardware module. For example, the petroleum safety intelligence deduction system 100 based on spatio-temporal big data may be a software module in the operating system of the terminal device, or may be an application program developed for the terminal device; of course, the petroleum safety intelligent deduction system 100 based on the space-time big data can also be one of numerous hardware modules of the terminal device.
Alternatively, in another example, the spatio-temporal big data-based petroleum safety intelligent deduction system 100 and the terminal device may be separate devices, and the spatio-temporal big data-based petroleum safety intelligent deduction system 100 may be connected to the terminal device through a wired and/or wireless network and transmit the interactive information according to an agreed data format.
In one embodiment of the present application, fig. 6 is a flowchart of an intelligent petroleum safety deduction method based on spatiotemporal big data according to an embodiment of the present application. As shown in fig. 6, an intelligent petroleum safety deduction method based on spatio-temporal big data according to an embodiment of the present application includes: 210, acquiring a plurality of nodes related to petroleum safety and attribute data of the plurality of nodes, wherein the plurality of nodes comprise oil producing countries, import countries, pipelines, export ports, import ports, straits, military bases, oil refineries and gas stations, and the attribute data comprise reserves, yields, consumption, export, import, prices, GDPs, population, urbanization rates, passenger traffic, freight traffic and automobile holding capacity; 220, arranging the attribute data of the plurality of nodes into a full-node full-attribute input matrix according to the node dimension and the attribute dimension respectively; 230, passing the full-node full-attribute input matrix through a data dimension reducer based on an automatic encoder to obtain a full-node full-attribute low-dimensional feature matrix after dimension reduction; 240, obtaining a full-node full-attribute petroleum safety feature matrix by using the dimension-reduced full-node full-attribute low-dimensional feature matrix through a convolutional neural network model using space attention; 250, performing characteristic discrimination highlighting on the full-node full-attribute petroleum safety characteristic matrix to obtain an optimized full-node full-attribute petroleum safety characteristic matrix; and 260, passing the optimized full-node full-attribute petroleum safety feature matrix through a classifier to obtain a classification result, wherein the classification result is used for representing a petroleum supply failure risk grade label.
Fig. 7 is a schematic diagram of a system architecture of a spatiotemporal big data-based petroleum safety intelligent deduction method according to an embodiment of the present application. As shown in fig. 7, in the system architecture of the intelligent petroleum safety deduction method based on spatiotemporal big data, first, a plurality of nodes related to petroleum safety including a petroleum producing country, an importing country, a pipeline, an export port, an import port, a channel, a military base, a petroleum refinery and a gas station and attribute data of the plurality of nodes including a reserve amount, a yield, a consumption amount, an export amount, an import amount, a price, a GDP, a population, a urbanization rate, a passenger transportation amount, a cargo transportation amount and a car holding amount are obtained; then, arranging the attribute data of the plurality of nodes into a full-node full-attribute input matrix according to the node dimension and the attribute dimension respectively; then, the full-node full-attribute input matrix passes through a data dimension reducer based on an automatic encoder to obtain a dimension-reduced full-node full-attribute low-dimensional feature matrix; then, the dimensionality-reduced full-node full-attribute low-dimensional feature matrix is subjected to a convolution neural network model using space attention to obtain a full-node full-attribute petroleum safety feature matrix; then, performing feature discrimination highlighting on the full-node full-attribute petroleum safety feature matrix to obtain an optimized full-node full-attribute petroleum safety feature matrix; and finally, passing the optimized full-node full-attribute petroleum safety feature matrix through a classifier to obtain a classification result, wherein the classification result is used for representing a petroleum supply failure risk grade label.
In one specific example, in the above-mentioned intelligent petroleum safety deduction method based on spatio-temporal big data, arranging the attribute data of the plurality of nodes into a full-node full-attribute input matrix according to the node dimension and the attribute dimension respectively includes: respectively arranging attribute data of the plurality of nodes into attribute input row vectors according to the dimension of the nodes; and performing two-dimensional arrangement on the attribute input row vectors according to the attribute dimensions to obtain the full-node full-attribute input matrix.
In one specific example, in the above intelligent petroleum safety deduction method based on space-time big data, the full-node full-attribute input matrix is passed through an auto-encoder-based data dimensionality reducer to obtain a dimensionality-reduced full-node full-attribute low-dimensional feature matrix, which includes: inputting the full-node full-attribute input matrix into an encoder of the data dimension reducer, wherein the encoder uses a convolutional layer to perform explicit spatial coding on the full-node full-attribute input matrix to obtain a full-node full-attribute feature; and inputting the full-node full-attribute feature into a decoder of the data dimension reducer, wherein the decoder performs deconvolution processing on the full-node full-attribute feature by using a deconvolution layer to obtain a full-node full-attribute low-dimensional feature matrix after dimension reduction.
In one specific example, in the above-mentioned intelligent petroleum safety deduction method based on spatio-temporal big data, passing the dimensionality-reduced full-node full-attribute low-dimensional feature matrix through a convolutional neural network model using spatial attention to obtain a full-node full-attribute petroleum safety feature matrix, including: the layers of the convolutional neural network model using spatial attention perform the following operations on input data in the forward direction of the layers: performing convolution processing on input data to generate a convolution characteristic diagram; pooling the convolved feature map to generate a pooled feature map; performing nonlinear activation on the pooled feature map to generate an activated feature map; calculating the mean value of each position of the activation feature map along the channel dimension to generate a spatial feature matrix; calculating Softmax-like function values of all positions in the spatial feature matrix to obtain a spatial score matrix; calculating the position-based multiplication of the spatial feature matrix and the spatial score matrix to obtain a feature matrix; wherein the feature matrix output by the last layer of the convolutional neural network model using spatial attention is the full-node full-attribute petroleum safety feature matrix.
In a specific example, in the above intelligent deduction method for petroleum safety based on spatio-temporal big data, the method for performing feature differentiation highlighting on the full-node full-attribute petroleum safety feature matrix to obtain an optimized full-node full-attribute petroleum safety feature matrix includes: performing characteristic discrimination highlighting on the full-node full-attribute petroleum safety characteristic matrix by using the following formula to obtain an optimized full-node full-attribute petroleum safety characteristic matrix; wherein the formula is:
wherein ,
is the full-node full-attribute petroleum safety feature matrix, based on the total node attribute>
and />
Is a predetermined hyper-parameter>
and />
Represents a position-wise addition and subtraction of a feature matrix, and>
represents convolution operation by a single convolution layer, <' > or>
And obtaining the optimized full-node full-attribute petroleum safety feature matrix.
In a specific example, in the above intelligent petroleum safety deduction method based on spatio-temporal big data, the method for passing the optimized full-node full-attribute petroleum safety feature matrix through a classifier to obtain a classification result, and the classification result is used for representing a petroleum outage risk level label, includes: the matrix expansion unit is used for expanding the optimized full-node full-attribute petroleum safety feature matrix into classification feature vectors according to row vectors or column vectors; a full-concatenation encoding unit, configured to perform full-concatenation encoding on the classification feature vector using a plurality of full-concatenation layers of the classifier to obtain an encoded classification feature vector; and the classification unit is used for enabling the coded classification feature vector to pass through a Softmax classification function of the classifier so as to obtain the classification result.
It will be understood by those skilled in the art that the detailed operations of the steps of the above-described spatio-temporal big data-based petroleum safety intelligence deduction method have been described in detail in the above description of the spatio-temporal big data-based petroleum safety intelligence deduction system with reference to fig. 1 to 5, and thus, a repeated description thereof will be omitted.
The present application also provides a computer program product comprising instructions that, when executed, cause an apparatus to perform operations corresponding to the above-described methods.
In an embodiment of the present application, there is also provided a computer-readable storage medium storing a computer program for executing the method described above.
It should be understood that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects may be employed. Also, the computer program product may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and so forth) having computer-usable program code embodied therein.
The methods, systems, and computer program products of embodiments of the present application are described in flowchart and/or block diagram form. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
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, devices, systems referred to in this application are only used as illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured 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 one 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 herein. 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, each component or step can be decomposed and/or re-combined. 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.
Finally, it should also be noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "include", "including" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or terminal device including a series of elements includes not only those elements but also other elements not explicitly listed or inherent to such process, method, article, or terminal device. Without further limitation, an element defined by the phrases "comprising one of \ 8230; \8230;" does not exclude the presence of additional like elements in a process, method, article, or terminal device that comprises the element.
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 intelligent petroleum safety deduction system based on space-time big data is characterized by comprising:
the system comprises a data acquisition module, a data processing module and a data processing module, wherein the data acquisition module is used for acquiring a plurality of nodes related to petroleum safety and attribute data of the nodes, the nodes comprise oil producing countries, import countries, pipelines, export ports, import ports, straits, military bases, oil refineries and gas stations, and the attribute data comprise reserves, yields, consumption, export volumes, import volumes, prices, GDP (gross distribution provider), population, urbanization rates, passenger volumes, freight volumes and automobile holding volumes;
the association construction module is used for arranging the attribute data of the nodes into a full-node full-attribute input matrix according to the node dimension and the attribute dimension respectively;
the dimension reduction module is used for enabling the full-node full-attribute input matrix to pass through a data dimension reducer based on an automatic encoder to obtain a full-node full-attribute low-dimensional feature matrix after dimension reduction;
the spatial feature enhancement module is used for enabling the dimensionality-reduced full-node full-attribute low-dimensional feature matrix to pass through a convolutional neural network model using spatial attention so as to obtain a full-node full-attribute petroleum safety feature matrix;
the optimization module is used for highlighting the feature discrimination of the full-node full-attribute petroleum safety feature matrix to obtain an optimized full-node full-attribute petroleum safety feature matrix;
and the risk early warning module is used for enabling the optimized full-node full-attribute petroleum safety feature matrix to pass through a classifier to obtain a classification result, and the classification result is used for representing a petroleum supply interruption risk level label.
2. The spatiotemporal big data-based petroleum safety intelligent deduction system according to claim 1, wherein the correlation construction module comprises:
the row vector arrangement unit is used for respectively arranging the attribute data of the nodes into attribute input row vectors according to the dimension of the nodes;
and the two-dimensional matrixing unit is used for performing two-dimensional arrangement on the attribute input row vectors according to the attribute dimensions to obtain the full-node full-attribute input matrix.
3. The space-time big data-based intelligent petroleum safety deduction system according to claim 2, wherein the dimensionality reduction module comprises:
the coding unit is used for inputting the full-node full-attribute input matrix into a coder of the data dimensionality reducer, wherein the coder uses a convolution layer to perform explicit space coding on the full-node full-attribute input matrix to obtain full-node full-attribute characteristics;
and the decoding unit is used for inputting the full-node full-attribute feature into a decoder of the data dimension reducer, wherein the decoder uses a deconvolution layer to perform deconvolution processing on the full-node full-attribute feature so as to obtain a full-node full-attribute low-dimensional feature matrix after dimension reduction.
4. The spatiotemporal big data-based petroleum safety intelligent deduction system according to claim 3, wherein the spatial feature enhancement module is used for: each layer of the convolutional neural network model using spatial attention performs the following operations on input data in the forward transmission process of the layer:
performing convolution processing on input data to generate a convolution characteristic diagram;
pooling the convolution feature map to generate a pooled feature map;
performing nonlinear activation on the pooled feature map to generate an activated feature map;
calculating a mean of the positions of the activation feature map along a channel dimension to generate a spatial feature matrix;
calculating Softmax-like function values of all positions in the spatial feature matrix to obtain a spatial score matrix;
calculating the position-based point multiplication of the spatial feature matrix and the spatial score matrix to obtain a feature matrix;
wherein the feature matrix output by the last layer of the convolutional neural network model using spatial attention is the full-node full-attribute petroleum safety feature matrix.
5. The spatiotemporal big data-based petroleum safety intelligence deduction system of claim 4, wherein the optimization module is configured to: performing characteristic discrimination highlighting on the full-node full-attribute petroleum safety characteristic matrix by using the following formula to obtain the optimized full-node full-attribute petroleum safety characteristic matrix;
wherein the formula is:
wherein ,
is the full node full Attribute Petroleum Security feature matrix, based upon>
and />
Is a predetermined hyper-parameter>
and />
Represents a position-wise addition and subtraction of a feature matrix, and>
represents convolution operation by a single convolution layer, <' > or>
And obtaining the optimized full-node full-attribute petroleum safety feature matrix.
6. The intelligent petroleum safety deduction system based on space-time big data as claimed in claim 5, wherein the risk early warning module comprises:
the matrix expansion unit is used for expanding the optimized full-node full-attribute petroleum safety feature matrix into classification feature vectors according to row vectors or column vectors;
a full-concatenation encoding unit, configured to perform full-concatenation encoding on the classification feature vector using a plurality of full-concatenation layers of the classifier to obtain an encoded classification feature vector;
and the classification unit is used for enabling the encoding classification characteristic vector to pass through a Softmax classification function of the classifier so as to obtain the classification result.
7. An intelligent petroleum safety deduction method based on space-time big data is characterized by comprising the following steps:
acquiring a plurality of nodes related to petroleum safety and attribute data of the plurality of nodes, wherein the plurality of nodes comprise oil producing countries, import countries, pipelines, export ports, import ports, straits, military bases, oil refineries and gas stations, and the attribute data comprise reserves, yields, consumption, export, import, prices, GDP, population, urbanization rates, passenger traffic, freight traffic and automobile holding capacity;
respectively arranging the attribute data of the plurality of nodes into a full-node full-attribute input matrix according to the node dimension and the attribute dimension;
enabling the full-node full-attribute input matrix to pass through a data dimension reducer based on an automatic encoder to obtain a full-node full-attribute low-dimensional feature matrix after dimension reduction;
obtaining a full-node full-attribute petroleum safety feature matrix by using the dimension-reduced full-node full-attribute low-dimensional feature matrix through a convolutional neural network model using space attention;
performing characteristic discrimination highlighting on the full-node full-attribute petroleum safety characteristic matrix to obtain an optimized full-node full-attribute petroleum safety characteristic matrix;
and passing the optimized full-node full-attribute petroleum safety feature matrix through a classifier to obtain a classification result, wherein the classification result is used for representing a petroleum supply failure risk grade label.
8. The intelligent petroleum safety deduction method based on space-time big data according to claim 7, wherein the arranging the attribute data of the plurality of nodes into a full-node full-attribute input matrix according to the node dimension and the attribute dimension respectively comprises:
respectively arranging the attribute data of the plurality of nodes into attribute input row vectors according to the dimension of the nodes;
and performing two-dimensional arrangement on the attribute input row vectors according to the attribute dimensions to obtain the full-node full-attribute input matrix.
9. The intelligent petroleum safety deduction method based on spatio-temporal big data as claimed in claim 8, wherein passing the full-node full-attribute input matrix through an auto-encoder based data dimension reducer to obtain a dimension-reduced full-node full-attribute low-dimensional feature matrix comprises:
inputting the full-node full-attribute input matrix into an encoder of the data dimensionality reducer, wherein the encoder uses a convolution layer to perform explicit spatial coding on the full-node full-attribute input matrix to obtain full-node full-attribute characteristics;
and inputting the full-node full-attribute features into a decoder of the data dimensionality reducer, wherein the decoder performs deconvolution processing on the full-node full-attribute features by using a deconvolution layer to obtain a full-node full-attribute low-dimensional feature matrix after dimensionality reduction.
10. The intelligent petroleum safety deduction method based on spatio-temporal big data according to claim 9, wherein the dimension-reduced full-node full-attribute low-dimensional feature matrix is passed through a convolutional neural network model using spatial attention to obtain a full-node full-attribute petroleum safety feature matrix, comprising: each layer of the convolutional neural network model using spatial attention performs the following operations on input data in the forward transmission process of the layer:
performing convolution processing on input data to generate a convolution characteristic diagram;
pooling the convolution feature map to generate a pooled feature map;
performing nonlinear activation on the pooled feature map to generate an activated feature map;
calculating the mean value of each position of the activation feature map along the channel dimension to generate a spatial feature matrix;
calculating Softmax-like function values of all positions in the spatial feature matrix to obtain a spatial score matrix;
calculating the position-based point multiplication of the spatial feature matrix and the spatial score matrix to obtain a feature matrix;
wherein the feature matrix output by the last layer of the convolutional neural network model using spatial attention is the full-node full-attribute petroleum safety feature matrix.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310194456.5A CN115879777B (en) | 2023-03-03 | 2023-03-03 | Petroleum safety intelligent deduction system and method based on space-time big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310194456.5A CN115879777B (en) | 2023-03-03 | 2023-03-03 | Petroleum safety intelligent deduction system and method based on space-time big data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115879777A true CN115879777A (en) | 2023-03-31 |
| CN115879777B CN115879777B (en) | 2023-05-12 |
Family
ID=85761858
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310194456.5A Active CN115879777B (en) | 2023-03-03 | 2023-03-03 | Petroleum safety intelligent deduction system and method based on space-time big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115879777B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116309446A (en) * | 2023-03-14 | 2023-06-23 | 浙江固驰电子有限公司 | Method and system for manufacturing power module for industrial control field |
| CN116682068A (en) * | 2023-07-28 | 2023-09-01 | 新疆华屹能源发展有限公司 | Oil well sand prevention operation construction monitoring method and system thereof |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107437137A (en) * | 2016-05-25 | 2017-12-05 | 达索系统美国公司 | Risk identification in supply chain |
| CN111539394A (en) * | 2020-07-08 | 2020-08-14 | 浙江浙能天然气运行有限公司 | Pipeline along-line third-party construction early warning method based on time domain characteristics and space-time information |
| US20200279200A1 (en) * | 2019-02-28 | 2020-09-03 | NB Ventures, Inc., dba GEP | Self-driven system & method for operating enterprise and supply chain applications |
| WO2021204010A1 (en) * | 2020-11-19 | 2021-10-14 | 平安科技（深圳）有限公司 | Time series anomaly detection method and apparatus, and computer device and storage medium |
| CN114064886A (en) * | 2021-11-25 | 2022-02-18 | 天津大学 | Mine project risk response measure recommendation method and system based on deep learning |
| CN114330541A (en) * | 2021-12-28 | 2022-04-12 | 于志青 | Road traffic accident risk prediction deep learning algorithm |
| CN115099684A (en) * | 2022-07-18 | 2022-09-23 | 江西中科冠物联网科技有限公司 | Enterprise safety production management system and management method thereof |
| CN115293910A (en) * | 2022-08-04 | 2022-11-04 | 杭州衡泰技术股份有限公司 | Intelligent enterprise cash flow rating system based on financial big data |
-
2023
- 2023-03-03 CN CN202310194456.5A patent/CN115879777B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107437137A (en) * | 2016-05-25 | 2017-12-05 | 达索系统美国公司 | Risk identification in supply chain |
| US20200279200A1 (en) * | 2019-02-28 | 2020-09-03 | NB Ventures, Inc., dba GEP | Self-driven system & method for operating enterprise and supply chain applications |
| CN111539394A (en) * | 2020-07-08 | 2020-08-14 | 浙江浙能天然气运行有限公司 | Pipeline along-line third-party construction early warning method based on time domain characteristics and space-time information |
| WO2021204010A1 (en) * | 2020-11-19 | 2021-10-14 | 平安科技（深圳）有限公司 | Time series anomaly detection method and apparatus, and computer device and storage medium |
| CN114064886A (en) * | 2021-11-25 | 2022-02-18 | 天津大学 | Mine project risk response measure recommendation method and system based on deep learning |
| CN114330541A (en) * | 2021-12-28 | 2022-04-12 | 于志青 | Road traffic accident risk prediction deep learning algorithm |
| CN115099684A (en) * | 2022-07-18 | 2022-09-23 | 江西中科冠物联网科技有限公司 | Enterprise safety production management system and management method thereof |
| CN115293910A (en) * | 2022-08-04 | 2022-11-04 | 杭州衡泰技术股份有限公司 | Intelligent enterprise cash flow rating system based on financial big data |
Non-Patent Citations (5)
| Title |
|---|
| 张轩诚;王国梁;: "中国海上石油进口的安全风险及对策", 山西师范大学学报(自然科学版) * |
| 杜彦良等: ""基于时空大数据的城市地下空间结构安全服役智慧感知与性态演化预测"", 《中国科学基金》 * |
| 杨扬等: ""基于CNN-PSO的跨境供应链风险预警研究"", 《公路交通科技》 * |
| 赵庆周;李勇;田世明;段义隆;谭益;曹一家;: "基于智能配电网大数据分析的状态监测与故障处理方法", 电网技术 * |
| 赵黎冬等: ""基于时空大数据的石油安全智能分析方法研究"", 《地球物理学进展》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116309446A (en) * | 2023-03-14 | 2023-06-23 | 浙江固驰电子有限公司 | Method and system for manufacturing power module for industrial control field |
| CN116682068A (en) * | 2023-07-28 | 2023-09-01 | 新疆华屹能源发展有限公司 | Oil well sand prevention operation construction monitoring method and system thereof |
| CN116682068B (en) * | 2023-07-28 | 2023-11-10 | 新疆华屹能源发展有限公司 | Oil well sand prevention operation construction monitoring method and system thereof |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115879777B (en) | 2023-05-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Deng et al. | Knowledge-driven stock trend prediction and explanation via temporal convolutional network | |
| CN113822494B (en) | Risk prediction method, device, equipment and storage medium | |
| CN115879777A (en) | Intelligent petroleum safety deduction system and method based on space-time big data | |
| Wang et al. | LightLog: A lightweight temporal convolutional network for log anomaly detection on the edge | |
| Wu | Collaborative filtering recommendation algorithm for MOOC resources based on deep learning | |
| Chen et al. | Multi-modal generative adversarial networks for traffic event detection in smart cities | |
| Mahmud et al. | Sentiment analysis from user-generated reviews of ride-sharing mobile applications | |
| CN116086790B (en) | Performance detection method and system for high-pressure valve of hydrogen fuel cell | |
| CN116383399A (en) | Event public opinion risk prediction method and system | |
| Kamran et al. | Decision support system for the prediction of mine fire levels in underground coal mining using machine learning approaches | |
| Tang et al. | Self-supervised anomaly pattern detection for large scale industrial data | |
| Zhang et al. | Semantic understanding and prompt engineering for large-scale traffic data imputation | |
| CN115935991A (en) | Multitask model generation method and device, computer equipment and storage medium | |
| Ma et al. | GAFM: a knowledge graph completion method based on graph attention faded mechanism | |
| Li | Cross-border E-commerce intelligent information recommendation system based on deep learning | |
| Ma et al. | MLRP-KG: Mine landslide risk prediction based on knowledge graph | |
| Lei et al. | Prediction of PM2. 5 concentration considering temporal and spatial features: A case study of Fushun, Liaoning Province | |
| Feng et al. | A representation learning framework for stock movement prediction | |
| CN113591971A (en) | User individual behavior prediction method based on DPI time series word embedded vector | |
| Shi et al. | Enhancing financial fraud detection with hierarchical graph attention networks: A study on integrating local and extensive structural information | |
| Fang et al. | AutoMSNet: Multi-source spatio-temporal network via automatic neural architecture search for traffic flow prediction | |
| Li et al. | Location and time embedded feature representation for spatiotemporal traffic prediction | |
| Gan et al. | Subsequence-level entity attention lstm for relation extraction | |
| Shi et al. | Improved credit risk prediction based on an integrated graph representation learning approach with graph transformation | |
| Zhang et al. | Credit evaluation of SMEs based on GBDT-CNN-LR hybrid integrated model |
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 |