Oil safety supply management method and system based on big data

Abstract

The present application relates to the field of intelligent supply management, and specifically to a method and system for managing safe oil supply based on big data. In the data preparation stage, the time series of node attribute data of multiple nodes are obtained to provide an important data basis for subsequent time series analysis. At the same time, the time series of node attribute data of multiple nodes are understood in the time series context by combining big data and artificial intelligence technology, and a masked reverse clustering mechanism is introduced to aggregate the attribute characteristics of each node to form a full-node clustering representation with spatiotemporal dynamic characteristics, so as to evaluate the level of oil supply interruption risk based on this, and can further release, increase or reduce production capacity accordingly according to the level of oil supply interruption risk to ensure stable and safe oil supply.

Description

Technical Field

[0001] The present application relates to the field of intelligent supply management, and specifically to a method and system for oil safety supply management based on big data. Background Art

[0002] At present, although the international oil production, consumption, import, export, price, etc. have been in a state of constant fluctuation for a long time, the overall supply and demand of oil has maintained a balance, that is, dynamic oil security has been achieved. When one or some events occur, this balance will be broken and oil security issues will be caused, which will involve the upstream, midstream and downstream of the entire oil industry chain. In other words, the production, transportation, processing, consumption and other links of oil may be affected by multiple factors due to a single or multiple events, affecting the entire industry chain, thereby breaking the regional, national and even global oil security balance. However, traditional oil security strategic analysis methods have shortcomings such as untimely data updates, difficulty in processing complex data, and low efficiency, which are difficult to adapt to the requirements of rapid analysis in the big data era.

[0003] Patent CN115879777B provides an intelligent deduction system for oil safety based on spatiotemporal big data and its method, which evaluates the level label of oil supply interruption risk by establishing a mapping relationship between multiple nodes and multiple attribute data related to oil safety. However, the node attribute data obtained does not have time series characteristics, so the time data is not fully utilized in the process of data processing and analysis, resulting in insufficient dynamics in analyzing the oil supply interruption risk level.

[0004] Therefore, an optimized solution is expected to solve the above problems. Summary of the invention

[0005] This application is made in consideration of the above problems. One purpose of this application is to provide a method and system for managing oil safety supply based on big data.

[0006] The embodiment of the present application provides a method for managing safe oil supply based on big data, which includes:

[0007] Acquire a time series of node attribute data of a plurality of nodes related to oil supply security, wherein the node attribute data includes reserves, production, consumption, export, import, price, GDP, population, urbanization rate, passenger volume, freight volume, and vehicle ownership;

[0008] Performing node attribute temporal analysis on the time series of node attribute data of each node to obtain multiple node attribute temporal context semantic encoding vectors;

[0009] Performing dual semantic enhancement and node spatiotemporal clustering representation on a plurality of the node attribute temporal context semantic encoding vectors to obtain a node temporal attribute information spatial clustering representation vector;

[0010] Determine a prediction result based on the spatial clustering representation vector of the node time series attribute information, wherein the prediction result is used to represent a risk level label of oil supply interruption;

[0011] Based on the forecast results, the oil production capacity release strategy is adjusted.

[0012] For example, according to the oil safety supply management method based on big data of an embodiment of the present application, a node attribute time series analysis is performed on the time series of node attribute data of each node to obtain multiple node attribute time series context semantic encoding vectors, including:

[0013] Arranging each node attribute data in the time series of the node attribute data of the plurality of nodes into a node attribute input vector according to the sample dimension and then passing it through a node attribute high-dimensional embedding encoder based on a fully connected layer to obtain a time series of node attribute embedding encoding vectors;

[0014] The time series of the node attributes of each node are embedded in the encoding vector and input into a time series encoder based on a bidirectional long short-term memory neural network model to obtain multiple node attribute temporal context semantic encoding vectors.

[0015] For example, according to the oil safety supply management method based on big data of an embodiment of the present application, dual semantic enhancement and node spatiotemporal clustering representation are performed on multiple node attribute temporal context semantic encoding vectors to obtain a node temporal attribute information spatial clustering representation vector, including:

[0016] Inputting the plurality of node attribute temporal context semantic coding vectors into a saliency enhancement module based on node attribute global context semantics to obtain a plurality of enhanced node attribute temporal context semantic coding vectors;

[0017] A plurality of the enhanced node attribute temporal context semantic encoding vectors are input into a feature distribution mask based reverse clustering analysis network to obtain a spatial clustering representation vector of the node temporal attribute information.

[0018] For example, according to the oil safety supply management method based on big data of an embodiment of the present application, a plurality of the node attribute temporal context semantic coding vectors are input into a saliency enhancement module based on node attribute global context semantics to obtain a plurality of enhanced node attribute temporal context semantic coding vectors, including:

[0019] Calculating the maximum value of each node attribute temporal context semantic coding vector among the plurality of node attribute temporal context semantic coding vectors to obtain a node attribute temporal context semantic salient feature vector composed of each of the maximum values;

[0020] Calculating the average value of each node attribute temporal context semantic coding vector in the plurality of node attribute temporal context semantic coding vectors to obtain a node attribute temporal context semantic global feature vector composed of each of the average values;

[0021] Performing neighborhood encoding and activation on the node attribute temporal context semantic salient feature vector and the node attribute temporal context semantic global feature vector to obtain a node attribute temporal context semantic salient feature context associated activation feature vector and a node attribute temporal context semantic global feature context associated activation feature vector;

[0022] After fusing the node attribute temporal context semantic saliency feature context association activation feature vector and the node attribute temporal context semantic global feature context association activation feature vector to obtain a node attribute temporal context semantic saliency-global feature context association activation feature vector, nonlinearly activate the node attribute temporal context semantic saliency-global feature context association activation feature vector to obtain a node attribute temporal context semantic saliency-global feature context association weight feature vector;

[0023] Taking the node attribute temporal context semantic saliency-global feature context associated weight feature vector as the weight vector, perform positional dot multiplication on each of the multiple node attribute temporal context semantic coding vectors, and add the sequence of vectors obtained by positional dot multiplication to the multiple node attribute temporal context semantic coding vectors element by element to obtain the multiple enhanced node attribute temporal context semantic coding vectors.

[0024] For example, according to the oil safety supply management method based on big data of an embodiment of the present application, the node attribute temporal context semantic significant feature vector and the node attribute temporal context semantic global feature vector are neighborhood encoded and activated to obtain the node attribute temporal context semantic significant feature context associated activation feature vector and the node attribute temporal context semantic global feature context associated activation feature vector, including:

[0025] Performing one-dimensional convolution coding, nonlinear activation and linear transformation on the node attribute temporal context semantically significant feature vector to obtain the node attribute temporal context semantically significant feature context-related activation feature vector;

[0026] One-dimensional convolution coding, nonlinear activation and linear transformation are performed on the node attribute temporal context semantic global feature vector to obtain the node attribute temporal context semantic global feature context association activation feature vector.

[0027] For example, according to the oil safety supply management method based on big data of an embodiment of the present application, the node attribute temporal context semantic significance-global feature context associated activation feature vector is nonlinearly activated to obtain the node attribute temporal context semantic significance-global feature context associated weight feature vector, including:

[0028] After the node attribute temporal context semantic saliency-global feature context associated activation feature vector is respectively passed through the tanh activation function and the sigmoid activation function, the two obtained vectors are point multiplied to obtain the node attribute temporal context semantic saliency-global feature context associated weight feature vector.

[0029] For example, according to the oil safety supply management method based on big data of an embodiment of the present application, a plurality of the enhanced node attribute temporal context semantic encoding vectors are input into a feature distribution mask-based reverse clustering analysis network to obtain the node temporal attribute information space clustering representation vector, including:

[0030] Calculating the positional mean vector of the plurality of enhanced node attribute temporal context semantic coding vectors to cluster the plurality of enhanced node attribute temporal context semantic coding vectors to obtain a node attribute temporal context semantic clustering center vector;

[0031] Calculating the semantic correlation coefficient between each enhanced node attribute temporal context semantic coding vector in the plurality of enhanced node attribute temporal context semantic coding vectors and the node attribute temporal context semantic clustering center vector to obtain a sequence of semantic correlation coefficients;

[0032] Taking the reciprocal of each semantic correlation coefficient in the sequence of semantic correlation coefficients to obtain a sequence of semantic anti-correlation coefficients;

[0033] Using a Sigmoid function to normalize the sequence of semantic anti-correlation coefficients to obtain a sequence of semantic anti-correlation weight coefficients;

[0034] Using each semantic anti-correlation weight coefficient in the sequence of the semantic anti-correlation weight coefficients as a weight, weighting each enhanced node attribute temporal context semantic coding vector in the plurality of enhanced node attribute temporal context semantic coding vectors to obtain a plurality of reverse inhibition node attribute temporal context semantic coding vectors;

[0035] Calculating positional differences between a plurality of the enhanced node attribute temporal context semantic coding vectors and a plurality of the reverse inhibited node attribute temporal context semantic coding vectors to obtain a plurality of clustering optimized node attribute temporal context semantic coding vectors;

[0036] The position-based mean vectors of a plurality of the cluster-optimized node attribute temporal context semantic encoding vectors are calculated to obtain the node temporal attribute information space cluster representation vector.

[0037] For example, according to the oil safety supply management method based on big data of an embodiment of the present application, the semantic correlation coefficient between each enhanced node attribute temporal context semantic coding vector in a plurality of enhanced node attribute temporal context semantic coding vectors and the node attribute temporal context semantic clustering center vector is calculated to obtain a sequence of semantic correlation coefficients, including:

[0038] The matrix product result between the transposed vector of the i-th enhanced node attribute temporal context semantic coding vector and the node attribute temporal context semantic clustering center vector is divided by the dot product result between the second norm of the i-th enhanced node attribute temporal context semantic coding vector and the second norm of the node attribute temporal context semantic clustering center vector to obtain the i-th semantic correlation coefficient.

[0039] For example, according to the oil safety supply management method based on big data in an embodiment of the present application, a prediction result is determined based on the spatial clustering representation vector of the node time series attribute information, and the prediction result is used to represent the oil supply interruption risk level label, including:

[0040] The node time series attribute information space clustering representation vector is input into a classifier-based power outage risk predictor to obtain the prediction result.

[0041] The embodiment of the present application also provides a petroleum safety supply management system based on big data, which includes:

[0042] A data acquisition module, used to obtain a time series of node attribute data of multiple nodes related to oil supply security, wherein the node attribute data includes reserves, production, consumption, export, import, price, GDP, population, urbanization rate, passenger volume, freight volume and car ownership;

[0043] A time series analysis module, used for performing a time series analysis of node attribute data of each node to obtain a plurality of node attribute time series context semantic encoding vectors;

[0044] An enhanced clustering module, used for performing dual semantic enhancement and node spatiotemporal clustering representation on a plurality of the node attribute temporal context semantic encoding vectors to obtain a node temporal attribute information spatial clustering representation vector;

[0045] A prediction module, used to determine a prediction result based on the spatial clustering representation vector of the node time series attribute information, wherein the prediction result is used to represent a risk level label of oil supply interruption;

[0046] The production capacity adjustment module is used to adjust the oil production capacity release strategy based on the prediction results.

[0047] This application has at least the following technical effects:

[0048] Compared with the prior art, the present application obtains the time series of node attribute data of multiple nodes in the data preparation stage to provide an important data basis for subsequent time series analysis. At the same time, it combines big data and artificial intelligence technologies to perform time series context understanding on the time series of node attribute data of multiple nodes, and introduces a masked reverse clustering mechanism to aggregate the attribute features of each node to form a full-node clustering representation with spatiotemporal dynamic characteristics, so as to evaluate the level of oil supply interruption risk based on this, and can further release, increase or reduce production capacity accordingly according to the level of oil supply interruption risk to ensure stable and safe oil supply. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings of the embodiments of the present application are briefly introduced below. Obviously, the drawings described below only relate to some embodiments of the present application, and are not intended to limit the present application.

[0050] Figure 1 A schematic diagram of the application architecture of the oil safety supply management method based on big data in an embodiment of the present application is shown;

[0051] Figure 2 A flow chart of a method for managing safe oil supply based on big data in an embodiment of the present application is shown;

[0052] Figure 3 A flowchart of sub-step S530 of the oil safety supply management method based on big data in an embodiment of the present application is shown;

[0053] Figure 4 A schematic diagram of the structure of a petroleum safety supply management system based on big data in an embodiment of the present application is shown;

[0054] Figure 5 An application scenario diagram of the oil safety supply management method based on big data in an embodiment of the present application is shown; and

[0055] Figure 6 A schematic diagram of a storage medium according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0056] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.

[0057] The terms used in this specification are those common terms currently widely used in the art in consideration of the functions of the present application, but these terms may vary according to the intention of a person of ordinary skill in the art, precedents, or new technologies in the art. In addition, specific terms may be selected by the applicant, and in this case, their detailed meanings will be described in the detailed description of the present application. Therefore, the terms used in the specification should not be understood as simple names, but rather as a general description based on the meaning of the terms and the present application.

[0058] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0059] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. At the same time, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0060] Figure 1 A schematic diagram of the application architecture of the oil safety supply management method based on big data in an embodiment of the present application is shown, including a server 100 and a terminal device 200.

[0061] The terminal device 200 and the server 100 can be connected via the Internet to achieve mutual communication. Optionally, the above-mentioned Internet uses standard communication technology and / or protocol. The Internet is usually the Internet, but it can also be any network, including but not limited to any combination of a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a mobile, wired or wireless network, a private network or a virtual private network. In some embodiments, the data exchanged through the network can be represented by technologies and / or formats including Hyper Text Markup Language (HTML), Extensible Markup Language (XML), etc. In addition, conventional encryption technologies such as Secure Socket Layer (SSL), Transport Layer Security (TLS), Virtual Private Network (VPN), Internet Protocol Security (IPsec), etc. can be used to encrypt all or some links. In other embodiments, customized and / or dedicated data communication technologies can also be used to replace or supplement the above-mentioned data communication technologies.

[0062] The server 100 can provide various network services for the terminal device 200, wherein the server 100 can be a single server, a server cluster consisting of several servers, or a cloud computing center. Specifically, the server 100 may include a processor 110 (Center Processing Unit, CPU), a memory 120, an input device 130, and an output device 140, etc. The input device 130 may include a keyboard, a mouse, a touch screen, etc., and the output device 140 may include a display device, such as a liquid crystal display (Liquid Crystal Display, LCD), a cathode ray tube (Cathode Ray Tube, CRT), etc.

[0063] The memory 120 may include a read-only memory (ROM) and a random access memory (RAM), and provides the processor 110 with program instructions and data stored in the memory 120. In the embodiment of the present application, the memory 120 may be used to store the program of the oil safety supply management method based on big data in the embodiment of the present application.

[0064] The processor 110 calls the program instructions stored in the memory 120, and the processor 110 is used to execute the steps of any one of the big data-based oil safety supply management methods in the embodiments of the present application according to the obtained program instructions.

[0065] In addition, the application architecture diagram in the embodiment of the present application is intended to more clearly illustrate the technical solution in the embodiment of the present application, and does not constitute a limitation on the technical solution provided in the embodiment of the present application. Of course, for other application architectures and business applications, the technical solution provided in the embodiment of the present application is also applicable to similar problems.

[0066] The following is a non-restrictive explanation of the big data-based oil supply security management method provided according to at least one embodiment of the present application through several examples or embodiments. As described below, different features in these specific examples or embodiments can be combined with each other without conflicting with each other to obtain new examples or embodiments, and these new examples or embodiments also fall within the scope of protection of the present application.

[0067] In response to the above technical problems, the technical concept of the present application is: in the data preparation stage, the time series of node attribute data of multiple nodes are obtained to provide an important data basis for subsequent time series analysis, and at the same time, big data and artificial intelligence technologies are combined to perform time series context understanding on the time series of node attribute data of multiple nodes, and a masked reverse clustering mechanism is introduced to aggregate the attribute characteristics of each node to form a full-node clustering representation with spatiotemporal dynamic characteristics, so as to evaluate the level of oil supply interruption risk based on this, and further release, increase or reduce production capacity according to the level of oil supply interruption risk to ensure stable and safe oil supply.

[0068] Based on this, Figure 2 The flowchart of the oil safety supply management method based on big data in the embodiment of the present application is shown. For example, the oil safety supply management method based on big data can be executed by a server, which can be Figure 1 The server 100 shown in FIG. Figure 2As shown, according to the big data-based oil supply security management method of the embodiment of the present application, the steps include: S510, obtaining the time series of node attribute data of multiple nodes related to oil supply security, wherein the node attribute data include reserves, production, consumption, export volume, import volume, price, GDP, population, urbanization rate, passenger volume, freight volume and car ownership; S520, performing node attribute time series analysis on the time series of node attribute data of each node to obtain multiple node attribute time series context semantic coding vectors; S530, performing dual semantic enhancement and node spatiotemporal clustering representation on multiple node attribute time series context semantic coding vectors to obtain a node time series attribute information space clustering representation vector; S540, determining a prediction result based on the node time series attribute information space clustering representation vector, and the prediction result is used to represent the oil supply interruption risk level label; S550, adjusting the oil production capacity release strategy based on the prediction result.

[0069] Specifically, in the technical solution of the present application, first, a time series of node attribute data of multiple nodes related to oil supply security is obtained, wherein the node attribute data includes reserves, output, consumption, export volume, import volume, price, GDP, population, urbanization rate, passenger volume, freight volume and car ownership. More specifically, in an embodiment of the present application, multiple nodes may refer to different entities or geographical areas in the oil supply chain, and these nodes together constitute the oil supply network. For example, these nodes may be production nodes such as oil fields and oil wells, refining nodes such as oil refineries, storage nodes such as oil depots, etc. These nodes play different roles in the oil supply chain, and their attribute data (such as reserves, output, consumption, etc.) are crucial for assessing oil supply security. It should be understood that by analyzing the node attribute data of these nodes, the operation of the oil supply chain can be fully understood.

[0070] Then, each node attribute data in the time series of the node attribute data of the multiple nodes is arranged as a node attribute input vector according to the sample dimension, and then passed through a node attribute high-dimensional embedding encoder based on a fully connected layer to obtain a time series of node attribute embedding encoding vectors; and the time series of the node attribute embedding encoding vectors of each node is input into a time series encoder based on a bidirectional long short-term memory neural network model to obtain multiple node attribute temporal context semantic encoding vectors. That is, the node attribute data of different samples are integrated by arrangement, and the node attribute high-dimensional embedding encoder is constructed using a fully connected layer to map the original discrete data into a high-dimensional space to help the model distinguish different data points, thereby converting them into vector representations that can be read and processed by the network model; then, the time series encoder is constructed using a bidirectional long short-term memory neural network model to capture the complex temporal feature representations of the node attribute data of each sample.

[0071] Accordingly, in step S520, node attribute time series analysis is performed on the time series of node attribute data of each node to obtain multiple node attribute time series context semantic coding vectors, including: arranging each node attribute data in the time series of node attribute data of multiple nodes as a node attribute input vector according to the sample dimension and then passing it through a node attribute high-dimensional embedding encoder based on a fully connected layer to obtain a time series of node attribute embedding coding vectors; inputting the time series of node attribute embedding coding vectors of each node into a time series encoder based on a bidirectional long short-term memory neural network model to obtain multiple node attribute time series context semantic coding vectors.

[0072] It should be understood that there may be redundant information in the node attribute data, and different nodes may have different degrees of focus on the node attribute data of each sample. For example, changes in GDP are closely related to population and urbanization rate, and there may be semantic information with a high degree of repetition between the three. For another example, the production node is the starting point of the oil supply chain. In the node attribute data of the production node, attribute information such as reserves, output and export volume is more important for expressing the attribute characteristics of the node. That is, when describing and characterizing the node attribute characteristics of each node, different focuses and attention should be paid to the time series of the node attribute data of each node. However, these requirements are missing in the above-mentioned processing, which may cause important key information to be ignored. Therefore, in the technical solution of the present application, it is expected that a plurality of the node attribute temporal context semantic coding vectors are input into a saliency enhancement module based on the global context semantics of the node attributes to obtain a plurality of enhanced node attribute temporal context semantic coding vectors.

[0073] In particular, the saliency enhancement module based on the global context semantics of node attributes calculates the maximum value of each node attribute temporal context semantic coding vector in a plurality of the node attribute temporal context semantic coding vectors to represent the salient features in the plurality of the node attribute temporal context semantic coding vectors, and calculates the average value of each node attribute temporal context semantic coding vector in a plurality of the node attribute temporal context semantic coding vectors to represent the global features in the plurality of the node attribute temporal context semantic coding vectors, and then uses convolutional coding, activation and fusion processing to convert the salient features and the global features into weights, and uses the weights to guide the optimized expression of each of the node attribute temporal context semantic coding vectors.

[0074] Subsequently, multiple enhanced node attribute temporal context semantic coding vectors are input into a reverse clustering analysis network based on feature distribution mask to obtain a node temporal attribute information space cluster representation vector. Among them, the reverse clustering analysis network based on feature distribution mask first clusters multiple enhanced node attribute temporal context semantic coding vectors to learn the internal mechanism in the data, and extracts a representative cluster center vector, that is, the node attribute temporal context semantic cluster center vector. Then, the semantic correlation coefficient between each enhanced node attribute temporal context semantic coding vector and the node attribute temporal context semantic cluster center vector is calculated to quantify the similarity and correlation between the two; and the feature distributions in the multiple enhanced node attribute temporal context semantic coding vectors that differ greatly from the cluster center are emphasized by taking the inverse and normalizing, and the weight coefficients obtained after normalization are ensured to be within a reasonable numerical range. That is, these feature distributions that differ greatly from the cluster center may represent information in the node attribute data that is less important than the node attribute representation of the node. Then, the weighting step applies the normalized weight coefficient to the original coding vector. This process realizes the reverse screening of the feature distribution to filter out the feature information that is more likely to be interference components or noise distribution in the feature distribution. Then, the position difference between the multiple enhanced node attribute temporal context semantic coding vectors and the multiple reverse inhibition node attribute temporal context semantic coding vectors is calculated to filter out the feature information that is interference components or noise components in the feature distribution, and all node attribute features are aggregated by taking the average by position to form a more comprehensive and richer spatiotemporal node attribute feature representation.

[0075] Accordingly, in step S530, if Figure 3 As shown, dual semantic enhancement and node spatiotemporal clustering representation are performed on multiple node attribute temporal context semantic coding vectors to obtain node temporal attribute information spatial clustering representation vectors, including: S531, inputting multiple node attribute temporal context semantic coding vectors into a significance enhancement module based on node attribute global context semantics to obtain multiple enhanced node attribute temporal context semantic coding vectors; S532, inputting multiple enhanced node attribute temporal context semantic coding vectors into a feature distribution mask reverse clustering analysis network to obtain the node temporal attribute information spatial clustering representation vector.

[0076] Among them, in step S531, multiple node attribute temporal context semantic coding vectors are input into a significance enhancement module based on node attribute global context semantics to obtain multiple enhanced node attribute temporal context semantic coding vectors, including: calculating the maximum value of each node attribute temporal context semantic coding vector in the multiple node attribute temporal context semantic coding vectors to obtain a node attribute temporal context semantic salient feature vector composed of each said maximum value; calculating the average value of each node attribute temporal context semantic coding vector in the multiple node attribute temporal context semantic coding vectors to obtain a node attribute temporal context semantic global feature vector composed of each said average value; performing neighborhood coding and activation on the node attribute temporal context semantic salient feature vector and the node attribute temporal context semantic global feature vector to obtain a node attribute temporal context semantic salient feature context association activation feature vector and a node attribute temporal context semantic global feature vector. The following text associates the activation feature vector; after fusing the node attribute temporal context semantic saliency feature context association activation feature vector and the node attribute temporal context semantic global feature context association activation feature vector to obtain the node attribute temporal context semantic saliency-global feature context association activation feature vector, the node attribute temporal context semantic saliency-global feature context association activation feature vector is nonlinearly activated to obtain the node attribute temporal context semantic saliency-global feature context association weight feature vector; using the node attribute temporal context semantic saliency-global feature context association weight feature vector as the weight vector, performing positional dot multiplication on each of the multiple node attribute temporal context semantic coding vectors, and adding the sequence of vectors obtained by positional dot multiplication to the multiple node attribute temporal context semantic coding vectors element by element to obtain the multiple enhanced node attribute temporal context semantic coding vectors.

[0077] Specifically, the node attribute temporal context semantically significant feature vector and the node attribute temporal context semantically global feature vector are subjected to neighborhood encoding and activation to obtain the node attribute temporal context semantically significant feature context-associated activation feature vector and the node attribute temporal context semantically global feature context-associated activation feature vector, including: performing one-dimensional convolution encoding, non-linear activation and linear transformation on the node attribute temporal context semantically significant feature vector to obtain the node attribute temporal context semantically significant feature context-associated activation feature vector; performing one-dimensional convolution encoding, non-linear activation and linear transformation on the node attribute temporal context semantically global feature vector to obtain the node attribute temporal context semantically significant feature context-associated activation feature vector.

[0078] Specifically, the node attribute temporal context semantic saliency-global feature context associated activation feature vector is nonlinearly activated to obtain the node attribute temporal context semantic saliency-global feature context associated weight feature vector, including: after the node attribute temporal context semantic saliency-global feature context associated activation feature vector is respectively passed through the tanh activation function and the sigmoid activation function, the two obtained vectors are point multiplied to obtain the node attribute temporal context semantic saliency-global feature context associated weight feature vector.

[0079] In a specific example, the multiple node attribute temporal context semantic coding vectors are input into a saliency enhancement module based on node attribute global context semantics to obtain multiple enhanced node attribute temporal context semantic coding vectors, including: the multiple node attribute temporal context semantic coding vectors are input into the saliency enhancement module based on node attribute global context semantics to process with the following enhancement formula to obtain multiple enhanced node attribute temporal context semantic coding vectors; wherein the enhancement formula is:

[0080]

[0081]

[0082]

[0083]

[0084]

[0085] in, are multiple node attribute temporal context semantic encoding vectors, is the global feature vector of node attribute temporal context semantics, is the semantically significant feature vector of the node attribute temporal context, It means taking the average value, Indicates taking the maximum value, represents a one-dimensional convolution operation, express Activation function, and are the first weight matrix and the second weight matrix for linear transformation, and They are the node attribute temporal context semantic global feature context associated activation feature vector and the node attribute temporal context semantic salient feature context associated activation feature vector, express function, express Activation function, represents dot product, It is a plurality of the enhanced node attribute temporal context semantic encoding vectors.

[0086] Wherein, in step S532, a plurality of the enhanced node attribute temporal context semantic coding vectors are input into a feature distribution masked reverse clustering analysis network to obtain a spatial cluster representation vector of the node attribute temporal information, including: calculating a positional mean vector of a plurality of the enhanced node attribute temporal context semantic coding vectors to cluster the plurality of the enhanced node attribute temporal context semantic coding vectors to obtain a node attribute temporal context semantic clustering center vector; calculating a semantic correlation coefficient between each enhanced node attribute temporal context semantic coding vector in the plurality of the enhanced node attribute temporal context semantic coding vectors and the node attribute temporal context semantic clustering center vector to obtain a sequence of semantic correlation coefficients; taking the inverse of each semantic correlation coefficient in the sequence of semantic correlation coefficients to obtain a sequence of semantic anti-correlation coefficients; using S The igmoid function normalizes the sequence of semantic anti-correlation coefficients to obtain a sequence of semantic anti-correlation weight coefficients; using each semantic anti-correlation weight coefficient in the sequence of semantic anti-correlation weight coefficients as a weight, weighting each enhanced node attribute temporal context semantic coding vector in the multiple enhanced node attribute temporal context semantic coding vectors to obtain multiple reverse inhibition node attribute temporal context semantic coding vectors; calculating the positional difference between the multiple enhanced node attribute temporal context semantic coding vectors and the multiple reverse inhibition node attribute temporal context semantic coding vectors to obtain multiple clustering optimized node attribute temporal context semantic coding vectors; calculating the positional mean vector of the multiple clustering optimized node attribute temporal context semantic coding vectors to obtain the node temporal attribute information space cluster representation vector.

[0087] Specifically, the semantic correlation coefficients between each enhanced node attribute temporal context semantic coding vector in a plurality of the enhanced node attribute temporal context semantic coding vectors and the node attribute temporal context semantic clustering center vector are calculated to obtain a sequence of semantic correlation coefficients, including: dividing the matrix product result between the transposed vector of the i-th enhanced node attribute temporal context semantic coding vector and the node attribute temporal context semantic clustering center vector by the dot product result between the second norm of the i-th enhanced node attribute temporal context semantic coding vector and the second norm of the node attribute temporal context semantic clustering center vector to obtain the i-th semantic correlation coefficient.

[0088] In a specific example, a plurality of the enhanced node attribute temporal context semantic encoding vectors are input into a feature distribution masked reverse clustering analysis network to obtain a spatial clustering representation vector of the node temporal attribute information, including: inputting a plurality of the enhanced node attribute temporal context semantic encoding vectors into the feature distribution masked reverse clustering analysis network to process the vectors using the following clustering analysis formula to obtain a spatial clustering representation vector of the node temporal attribute information; wherein the clustering analysis formula is:

[0089]

[0090]

[0091]

[0092]

[0093]

[0094]

[0095]

[0096]

[0097] in, are multiple enhanced node attribute temporal context semantic encoding vectors, , , and are respectively the first enhanced node attribute temporal context semantic coding vector, the second enhanced node attribute temporal context semantic coding vector, the i-th enhanced node attribute temporal context semantic coding vector and the N-th enhanced node attribute temporal context semantic coding vector among the plurality of enhanced node attribute temporal context semantic coding vectors, The value of is the number of enhanced node attribute temporal context semantic coding vectors in the plurality of enhanced node attribute temporal context semantic coding vectors, is the node attribute temporal context semantic clustering center vector, represents the transpose of a vector, is the i-th semantic correlation coefficient, represents the two-norm of a vector, is the i-th semantic anti-correlation coefficient, is the i-th semantic anti-correlation weight coefficient, represents an exponential function with a natural constant as base, are multiple reverse inhibition node attribute temporal context semantic encoding vectors, , , and are respectively the first reverse inhibition node attribute temporal context semantic coding vector, the second reverse inhibition node attribute temporal context semantic coding vector, the i-th reverse inhibition node attribute temporal context semantic coding vector and the N-th reverse inhibition node attribute temporal context semantic coding vector among the plurality of reverse inhibition node attribute temporal context semantic coding vectors, are multiple clustering optimization node attribute temporal context semantic encoding vectors, , , and They are the first cluster optimization node attribute temporal context semantic coding vector, the second cluster optimization node attribute temporal context semantic coding vector, the i-th cluster optimization node attribute temporal context semantic coding vector and the N-th cluster optimization node attribute temporal context semantic coding vector, It is the spatial clustering representation vector of the node temporal attribute information.

[0098] Furthermore, the node time series attribute information space clustering representation vector is input into a classifier-based supply interruption risk predictor to obtain a prediction result, and the prediction result is used to represent the oil supply interruption risk level label; and based on the prediction result, the oil production capacity release strategy is adjusted. The specific definition and use of the oil supply interruption risk level label can be adjusted according to the actual oil safety strategy and risk management standards. These labels can provide clear information to decision makers, helping them to assess risks and take appropriate preventive or countermeasures.

[0099] Accordingly, in step S540, a prediction result is determined based on the node time series attribute information space cluster representation vector, and the prediction result is used to represent the oil supply interruption risk level label, including: inputting the node time series attribute information space cluster representation vector into a classifier-based supply interruption risk predictor to obtain the prediction result.

[0100] It should be understood that the role of the classifier is to use the given categories and known training data to learn classification rules and classifiers, and then classify (or predict) unknown data. Logistic regression, SVM, etc. are often used to solve binary classification problems. For multi-class classification problems, logistic regression or SVM can also be used, but multiple binary classifications are required to form a multi-classification, but this is prone to errors and inefficient. Commonly used multi-classification methods include the Softmax classification function.

[0101] In the technical solution of the present application, each of the multiple node attribute temporal context semantic coding vectors respectively represents the temporal context semantic aggregation representation of the high-dimensional embedding representation of the node attribute data of each node. Taking into account that the time series of the node attribute data of each node has source domain temporal distribution differences in the data source domain, this makes the multiple node attribute temporal context semantic coding vectors have relatively significant feature distribution inconsistency. Therefore, when the multiple enhanced node attribute temporal context semantic coding vectors are input into the feature distribution mask-based reverse clustering analysis network, the obtained node temporal attribute information space cluster representation vector has a causal association missing relative to the posterior class probability represented by the source semantic distribution, thereby affecting the accuracy of the classification result.

[0102] Based on this, in a preferred embodiment, when the applicant of the present application inputs the node timing attribute information space cluster representation vector into the outage risk predictor based on the classifier to obtain the prediction result, the node timing attribute information space cluster representation vector is optimized, specifically including the following steps: obtaining the node timing attribute information space cluster representation vector input into the outage risk predictor based on the classifier and obtaining the corresponding representations The corresponding oil supply interruption risk level labels label probability values; The label probability values ​​are subtracted from After dividing by one, calculate the square root of the sum of the squares and add One-half is added to obtain a level probability value; each eigenvalue of the node temporal attribute information space cluster representation vector is probabilized based on a probability activation function, such as a sigmoid function, to obtain a probabilistic node temporal attribute information space cluster representation vector; based on the comparison of each eigenvalue of the probabilistic node temporal attribute information space cluster representation vector with the level probability value, a node temporal attribute information space cluster cognitive symbol value is determined, wherein the node temporal attribute information space cluster cognitive symbol value is equal to one, zero and negative one in response to the eigenvalue of the probabilistic node temporal attribute information space cluster representation vector being greater than, equal to and less than the level probability value, respectively; the mean of all eigenvalues ​​of the probabilistic node temporal attribute information space cluster representation vector is calculated to obtain the overall phase shift value of the node temporal attribute information space cluster ; Multiply each eigenvalue of the probabilistic node temporal attribute information space clustering representation vector by the node temporal attribute information space clustering cognitive symbol value and the node temporal attribute information space clustering overall phase shift value to obtain a probabilistic node temporal attribute information space clustering cognitive feature vector and a probabilistic node temporal attribute information space clustering phase shift feature vector; Multiply the probabilistic node temporal attribute information space clustering cognitive feature vector and the probabilistic node temporal attribute information space clustering phase shift feature vector by the first weight and the second weight as hyperparameters respectively, and then calculate their difference to obtain a probabilistic node temporal attribute information space clustering differential representation vector; Take the absolute value of each eigenvalue of the probabilistic node temporal attribute information space clustering differential representation vector to obtain an optimized node temporal attribute information space clustering representation vector.

[0103] Specifically, in one example, optimizing the node temporal attribute information space cluster representation vector includes: optimizing the node temporal attribute information space cluster representation vector using the following optimization formula to obtain the optimized node temporal attribute information space cluster representation vector; wherein the optimization formula is:

[0104]

[0105]

[0106]

[0107] in, and are the i-th eigenvalue in the probabilistic node temporal attribute information space clustering representation vector and the optimized node temporal attribute information space clustering representation vector, respectively. is the grade probability value, is the j-th label probability value, is the spatial clustering cognitive symbol value of the node temporal attribute information, is the overall phase shift value of the spatial clustering of the node temporal attribute information, and are the first weight and the second weight respectively, Indicates taking the absolute value, Indicates the number of label probability values.

[0108] That is, in the above preferred embodiment, the level cognitive phase transformation response of the node temporal attribute information space cluster representation vector is obtained by comparing the probabilistic amplitude of the eigenvalue of the node temporal attribute information space cluster representation vector with the level probability representation, and the phase shift response of the eigenvalue of the node temporal attribute information space cluster representation vector relative to the level probability representation of the feature as a whole is subjected to a feature distribution sequence invariance transformation based on the level probability differential distribution expansion, so as to realize the causal constraint of the posterior class probability of the node temporal attribute information space cluster representation vector on its prior semantic feature distribution representation, so as to improve the accuracy of the prediction result obtained by the node temporal attribute information space cluster representation vector input into the classifier-based power outage risk predictor.

[0109] Based on the above embodiments, see Figure 4 As shown, it is a structural schematic diagram of a big data-based oil safety supply management system 800 in an embodiment of the present application. The big data-based oil safety supply management system 800 includes: a data acquisition module 810, which is used to obtain the time series of node attribute data of multiple nodes related to oil supply security, wherein the node attribute data includes reserves, production, consumption, export volume, import volume, price, GDP, population, urbanization rate, passenger volume, freight volume and car ownership; a time series analysis module 820, which is used to perform node attribute time series analysis on the time series of node attribute data of each node to obtain multiple node attribute time series context semantic encoding vectors; an enhanced clustering module 830, which is used to perform dual semantic enhancement and node spatiotemporal clustering representation on multiple node attribute time series context semantic encoding vectors to obtain node time series attribute information space clustering representation vector; a prediction module 840, which is used to determine the prediction result based on the node time series attribute information space clustering representation vector, and the prediction result is used to represent the oil supply interruption risk level label; a production capacity adjustment module 850, which is used to adjust the oil production capacity release strategy based on the prediction result.

[0110] The specific functions and operations of each module in the above-mentioned oil safety supply management system based on big data 800 have been referred to above. Figure 2 to Figure 3 It has been introduced in detail in the description of the big data-based oil security supply management method, and its repeated description will be omitted.

[0111] Figure 5 FIG. 1 is an application scenario diagram of the oil safety supply management method based on big data according to an embodiment of the present application. Figure 5 As shown, in this application scenario, first, the time series of node attribute data of multiple nodes related to oil supply security are obtained (for example, Figure 5 D), wherein the node attribute data includes reserves, production, consumption, export, import, price, GDP, population, urbanization rate, passenger volume, freight volume and car ownership, and then the time series of the node attribute data of each node is input into a server (for example, Figure 5 In S) shown in , the server is able to use the big data-based oil security supply management algorithm to process the time series of node attribute data of each of the nodes to obtain a prediction result for representing an oil supply interruption risk level label.

[0112] Based on the above embodiment, another exemplary embodiment of an electronic device is also provided in the embodiments of the present application. In some possible implementations, the electronic device in the embodiments of the present application may include a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor can implement the steps of the oil safety supply management method based on big data in the above embodiment when executing the program.

[0113] For example, in the case of electronic equipment Figure 1 Taking the server 100 in the example as an example, the processor in the electronic device is the processor 110 in the server 100, and the memory in the electronic device is the memory 120 in the server 100.

[0114] An embodiment of the present application also provides a computer-readable storage medium. Figure 6 1 shows a schematic diagram of a computer-readable storage medium 1000 according to an embodiment of the present application. Figure 6 As shown, the computer-readable storage medium 1000 stores computer-executable instructions 1001. When the computer-executable instructions 1001 are executed by the processor, the oil safety supply management method based on big data according to the embodiment of the present application described with reference to the above figures can be executed. The computer-readable storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. The volatile memory may include, for example, a random access memory (RAM) and / or a cache memory (cache), etc. The non-volatile memory may include, for example, a read-only memory (ROM), a hard disk, a flash memory, etc.

[0115] The embodiment of the present application also provides a computer program product or a computer program, which includes computer executable instructions, and the computer executable instructions are stored in a computer readable storage medium. The processor of the computer device reads the computer executable instructions from the computer readable storage medium, and the processor executes the computer executable instructions, so that the computer device executes the oil safety supply management method based on big data according to the embodiment of the present application.

[0116] Those skilled in the art will appreciate that the contents disclosed in this application may be subject to various variations and improvements. For example, the various devices or components described above may be implemented by hardware, or by software, firmware, or a combination of some or all of the three.

[0117] In addition, although the present application makes various references to certain units in the system according to embodiments of the present application, any number of different units can be used and run on the client and / or server. The units are only illustrative, and different aspects of the system and method can use different units.

[0118] Those skilled in the art will appreciate that all or part of the steps in the above method can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, such as a read-only memory, a disk or an optical disk. Optionally, all or part of the steps in the above embodiment can also be implemented using one or more integrated circuits. Accordingly, each module / unit in the above embodiment can be implemented in the form of hardware or in the form of software function modules. The present application is not limited to any particular form of combination of hardware and software.

[0119] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs. It should also be understood that terms such as those defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an idealized or extremely formal sense, unless explicitly defined as such herein.

[0120] The above is an explanation of the present application and should not be considered as a limitation thereof. Although several exemplary embodiments of the present application have been described, it will be readily understood by those skilled in the art that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present application. Therefore, all of these modifications are intended to be included within the scope of the present application as defined by the claims. It should be understood that the above is an explanation of the present application and should not be considered to be limited to the specific embodiments disclosed, and modifications to the disclosed embodiments and other embodiments are intended to be included within the scope of the appended claims. The present application is defined by the claims and their equivalents.

Claims

1. A method for managing oil supply security based on big data, characterized in that: include: Acquire a time series of node attribute data of a plurality of nodes related to oil supply security, wherein the node attribute data includes reserves, production, consumption, export, import, price, GDP, population, urbanization rate, passenger volume, freight volume, and vehicle ownership; The node attribute time series analysis is performed on the time series of the node attribute data of each node to obtain a plurality of node attribute time series context semantic encoding vectors, which include: Arranging each node attribute data in the time series of the node attribute data of the plurality of nodes into a node attribute input vector according to the sample dimension and then passing it through a node attribute high-dimensional embedding encoder based on a fully connected layer to obtain a time series of node attribute embedding encoding vectors; The time series of the node attribute embedding encoding vectors of each node is input into a time series encoder based on a bidirectional long short-term memory neural network model to obtain a plurality of the node attribute temporal context semantic encoding vectors; Performing dual semantic enhancement and node spatiotemporal clustering representation on the plurality of node attribute temporal context semantic encoding vectors to obtain a node temporal attribute information spatial clustering representation vector, which includes: Inputting the plurality of node attribute temporal context semantic coding vectors into a saliency enhancement module based on node attribute global context semantics to obtain a plurality of enhanced node attribute temporal context semantic coding vectors; Inputting a plurality of the enhanced node attribute temporal context semantic encoding vectors into a feature distribution mask-based reverse clustering analysis network to obtain a spatial clustering representation vector of the node temporal attribute information; Determine a prediction result based on the spatial clustering representation vector of the node time series attribute information, wherein the prediction result is used to represent a risk level label of oil supply interruption; Based on the forecast results, adjust the oil production capacity release strategy; The prediction result is determined based on the spatial clustering representation vector of the node time series attribute information, and the prediction result is used to represent the oil supply interruption risk level label, including: Inputting the node time series attribute information space clustering representation vector into a classifier-based power outage risk predictor to obtain the prediction result; When the node temporal attribute information space cluster representation vector is input into the classifier-based supply outage risk predictor to obtain a prediction result, the node temporal attribute information space cluster representation vector is optimized, specifically including the following steps: obtaining n label probability values ​​corresponding to n oil supply outage risk level labels obtained by inputting the node temporal attribute information space cluster representation vector into the classifier-based supply outage risk predictor; subtracting one-n from each of the n label probability values, calculating the square root of the sum of their squares, and adding them to one-n to obtain a level probability value, where n represents the number of label probability values; probabilizing each eigenvalue of the node temporal attribute information space cluster representation vector based on a probability activation function to obtain a probabilistic node temporal attribute information space cluster representation vector; determining a node temporal attribute information space cluster cognitive symbol value based on a comparison between each eigenvalue of the probabilistic node temporal attribute information space cluster representation vector and the level probability value, wherein the node temporal attribute information space cluster cognitive symbol values ​​respectively respond to the probabilistic node temporal attribute information space cluster representation vector. The eigenvalue of the space clustering representation vector is greater than, equal to and less than the level probability value and equal to one, zero and negative one; the mean of all eigenvalues ​​of the probabilistic node temporal attribute information space clustering representation vector is calculated to obtain the overall phase shift value of the node temporal attribute information space clustering; each eigenvalue of the probabilistic node temporal attribute information space clustering representation vector is multiplied by the node temporal attribute information space clustering cognitive symbol value and the node temporal attribute information space clustering overall phase shift value to obtain the probabilistic node temporal attribute information space clustering cognitive feature vector and the probabilistic node temporal attribute information space clustering phase shift feature vector; the probabilistic node temporal attribute information space clustering cognitive feature vector and the probabilistic node temporal attribute information space clustering phase shift feature vector are multiplied by the first weight and the second weight as hyperparameters respectively, and then their difference is calculated to obtain the probabilistic node temporal attribute information space clustering differential representation vector; the absolute value of each eigenvalue of the probabilistic node temporal attribute information space clustering differential representation vector is taken to obtain the optimized node temporal attribute information space clustering representation vector.

2. The oil safety supply management method based on big data according to claim 1 is characterized in that: Inputting the plurality of node attribute temporal context semantic coding vectors into a saliency enhancement module based on node attribute global context semantics to obtain a plurality of enhanced node attribute temporal context semantic coding vectors, including: Calculating the maximum value of each node attribute temporal context semantic coding vector among the plurality of node attribute temporal context semantic coding vectors to obtain a node attribute temporal context semantic salient feature vector composed of each of the maximum values; Calculating the average value of each node attribute temporal context semantic coding vector in the plurality of node attribute temporal context semantic coding vectors to obtain a node attribute temporal context semantic global feature vector composed of each of the average values; Performing neighborhood encoding and activation on the node attribute temporal context semantic salient feature vector and the node attribute temporal context semantic global feature vector to obtain a node attribute temporal context semantic salient feature context associated activation feature vector and a node attribute temporal context semantic global feature context associated activation feature vector; After fusing the node attribute temporal context semantic saliency feature context association activation feature vector and the node attribute temporal context semantic global feature context association activation feature vector to obtain a node attribute temporal context semantic saliency-global feature context association activation feature vector, nonlinearly activate the node attribute temporal context semantic saliency-global feature context association activation feature vector to obtain a node attribute temporal context semantic saliency-global feature context association weight feature vector; Taking the node attribute temporal context semantic saliency-global feature context associated weight feature vector as the weight vector, perform positional dot multiplication on each of the multiple node attribute temporal context semantic coding vectors, and add the sequence of vectors obtained by positional dot multiplication to the multiple node attribute temporal context semantic coding vectors element by element to obtain the multiple enhanced node attribute temporal context semantic coding vectors.

3. The oil safety supply management method based on big data according to claim 2 is characterized in that: Neighborhood encoding and activation are performed on the node attribute temporal context semantic salient feature vector and the node attribute temporal context semantic global feature vector to obtain the node attribute temporal context semantic salient feature context association activation feature vector and the node attribute temporal context semantic global feature context association Activation feature vector, including: Performing one-dimensional convolution coding, nonlinear activation and linear transformation on the node attribute temporal context semantically significant feature vector to obtain the node attribute temporal context semantically significant feature context-related activation feature vector; One-dimensional convolution coding, nonlinear activation and linear transformation are performed on the node attribute temporal context semantic global feature vector to obtain the node attribute temporal context semantic global feature context association activation feature vector.

4. The oil safety supply management method based on big data according to claim 3 is characterized in that: The node attribute temporal context semantic saliency-global feature context associated activation feature vector is nonlinearly activated to obtain a node attribute temporal context semantic saliency-global feature context associated weight feature vector, including: After the node attribute temporal context semantic saliency-global feature context associated activation feature vector is respectively passed through the tanh activation function and the sigmoid activation function, the two obtained vectors are point multiplied to obtain the node attribute temporal context semantic saliency-global feature context associated weight feature vector.

5. The oil safety supply management method based on big data according to claim 4 is characterized in that: Inputting a plurality of the enhanced node attribute temporal context semantic encoding vectors into a feature distribution mask-based reverse clustering analysis network to obtain the node temporal attribute information spatial clustering representation vector, including: Calculating the positional mean vector of the plurality of enhanced node attribute temporal context semantic coding vectors to cluster the plurality of enhanced node attribute temporal context semantic coding vectors to obtain a node attribute temporal context semantic clustering center vector; Calculating the semantic correlation coefficient between each enhanced node attribute temporal context semantic coding vector in the plurality of enhanced node attribute temporal context semantic coding vectors and the node attribute temporal context semantic clustering center vector to obtain a sequence of semantic correlation coefficients; Taking the reciprocal of each semantic correlation coefficient in the sequence of semantic correlation coefficients to obtain a sequence of semantic anti-correlation coefficients; Using a Sigmoid function to normalize the sequence of semantic anti-correlation coefficients to obtain a sequence of semantic anti-correlation weight coefficients; Using each semantic anti-correlation weight coefficient in the sequence of the semantic anti-correlation weight coefficients as a weight, weighting each enhanced node attribute temporal context semantic coding vector in the plurality of enhanced node attribute temporal context semantic coding vectors to obtain a plurality of reverse inhibition node attribute temporal context semantic coding vectors; Calculating positional differences between a plurality of the enhanced node attribute temporal context semantic coding vectors and a plurality of the reverse inhibited node attribute temporal context semantic coding vectors to obtain a plurality of clustering optimized node attribute temporal context semantic coding vectors; The position-based mean vectors of a plurality of the cluster-optimized node attribute temporal context semantic encoding vectors are calculated to obtain the node temporal attribute information space cluster representation vector.

6. The oil safety supply management method based on big data according to claim 5 is characterized in that: Calculating the semantic correlation coefficient between each enhanced node attribute temporal context semantic coding vector in the plurality of enhanced node attribute temporal context semantic coding vectors and the node attribute temporal context semantic clustering center vector to obtain a sequence of semantic correlation coefficients, including: The matrix product result between the transposed vector of the i-th enhanced node attribute temporal context semantic coding vector and the node attribute temporal context semantic clustering center vector is divided by the dot product result between the second norm of the i-th enhanced node attribute temporal context semantic coding vector and the second norm of the node attribute temporal context semantic clustering center vector to obtain the i-th semantic correlation coefficient.

7. A petroleum safety supply management system based on big data, used to execute the petroleum safety supply management method based on big data as claimed in claim 1, characterized in that: include: A data acquisition module, used to obtain a time series of node attribute data of multiple nodes related to oil supply security, wherein the node attribute data includes reserves, production, consumption, export, import, price, GDP, population, urbanization rate, passenger volume, freight volume and car ownership; A time series analysis module, used for performing a time series analysis of node attribute data of each node to obtain a plurality of node attribute time series context semantic encoding vectors; An enhanced clustering module, used for performing dual semantic enhancement and node spatiotemporal clustering representation on a plurality of the node attribute temporal context semantic encoding vectors to obtain a node temporal attribute information spatial clustering representation vector; A prediction module, used to determine a prediction result based on the spatial clustering representation vector of the node time series attribute information, wherein the prediction result is used to represent a risk level label of oil supply interruption; The production capacity adjustment module is used to adjust the oil production capacity release strategy based on the prediction results.

Images