Regional Geological Data Fusion Visualization Monitoring System and Method Based on Neural Network Learning

Abstract

The present invention belongs to the technical field of geological data monitoring, and discloses a regional geological data fusion visualization monitoring system and method based on neural network learning. The method includes: constructing a regional geological node monitoring neural network learning model and a regional geological surface node monitoring neural network learning model and fusing them to complete the construction of a regional geological three-dimensional model based on neural network learning; performing node importance identification based on the adaptive adjustment method of the regional geological three-dimensional space expression vector value; performing dynamic node importance arrangement based on the adaptive threshold of the regional geological three-dimensional space expression vector; and realizing regional geological data fusion visualization monitoring based on the adaptive value of the regional geological three-dimensional space expression vector. The present invention further improves the accuracy of regional geological data fusion visualization monitoring and the accuracy of identifying geological anomaly areas compared with the prior art.

Description

Technical Field

[0001] The present invention belongs to the technical field of geological data monitoring, and particularly relates to a regional geological data fusion visualization monitoring system and method based on neural network learning. Background Art

[0002] Due to the complexity and uncertainty of geological phenomena, geologists consider various geological factors and complex conditions for formation identification and simulation. Within a certain range, multiple possible realizations are established to generate the optimal, most reasonable, and most geologically regular model design scheme, revealing the internal structure of geological bodies, the complex variation law of space, and the distribution characteristics of attribute parameters, and conducting comprehensive integrated and three-dimensional quantitative research and management, directly reflecting factors such as uncertainty in geological phenomena. The three-dimensional geological identification technology will provide a solid support for the research and development work of geologists, improve the ability to analyze geological bodies, including predicting borehole positions, automatically detecting fault attitudes, etc., and by coupling multi-source data in the study area, improving the method of spatial information processing, providing a unified three-dimensional display channel to support the all-round and integrated display of exploration data in the same coordinate system. Improving and perfecting the integration or coupling method of spatial geological data, the data expression form, and the model description of spatial geological geometric forms has become an urgent need for geologists. With the rapid development of technologies such as exploration geophysics and remote sensing image processing, it has made it possible to couple three-dimensional spatial multi-source data of geological data, photogrammetry data, remote sensing data, and geophysical data, providing a prerequisite and laying a foundation for quickly and accurately establishing a three-dimensional complex geological model, and broadening the application scope of resource analysis and evaluation for complex geological bodies including thrust faults and overturned folds. The geological model based on multi-source data and its information processing system can also provide important economic evolution bases, such as investment priority strategies, development plans, future exploration budgets, etc.

[0003] To solve the above problems, the prior art conducts formation identification and analysis technology in a VR geological environment, and discloses: geological data preprocessing, three-dimensional formation model identification and construction, and spatial visualization analysis and correction; each of the above steps requires corresponding verification and detection according to the "spatial information quality detection model"; with the development of machine learning, attempts are actively made to automatically execute operations such as identifying or classifying images through a computer system. In particular, attempts have been made to use neural networks, one of machine learning methods (for example, the deep learning method using a Convolution Neural Network (CNN)), to automate various classifications or judgments performed by humans, and by using deep learning of neural networks in it to read regional geological data images, thereby judging the state or development trend of a specific regional geology.

[0004] In existing multi-instance learning, since one instance is extracted from one instance package for learning, there is a problem that a large number of instance packages are required to improve the performance of the neural network. For example, in order to utilize existing multi-instance learning in a neural network for detecting regional geological evolution, a large number of whole-fault-images marked with signs related to whether there is evolution are needed.

[0005] As described above, in multi-instance learning, since the neural network before the completion of learning is used in the step of extracting learning data instances, there is a problem that the possibility of extracting wrong instances becomes high in the case of extracting multiple data instances from one instance package.

[0006] Through the above analysis, the problems and defects of the existing technology are as follows: The existing technology still needs to be further improved in the accuracy of regional geological data fusion visualization monitoring and the accuracy of identifying geological anomaly areas. Summary of the Invention

[0007] To overcome the problems in the related technology, the disclosed embodiments of the present invention provide a regional geological data fusion visualization monitoring system and method based on neural network learning.

[0008] The technical solution is as follows: A regional geological data fusion visualization monitoring method based on neural network learning includes:

[0009] S1, constructing a node monitoring neural network learning model for regional geology and surface layer, and performing fusion to complete the construction of a three-dimensional model of regional geology based on neural network learning;

[0010] S2, for the constructed three-dimensional model of regional geology based on neural network learning, performing node importance identification based on the method of adaptive adjustment of vector values for three-dimensional spatial expression of regional geology;

[0011] S3, dynamically arranging node importance based on the threshold adaptability of three-dimensional spatial expression vectors of regional geology;

[0012] S4, according to the dynamic arrangement of node importance, realizing the regional geological data fusion visualization monitoring based on the adaptability of three-dimensional spatial expression vector values of regional geology.

[0013] In step S1, the node monitoring neural network learning model is composed of a regional geology monitoring terminal, a monitoring terminal server, and a radio access network controller;

[0014] The regional geology monitoring terminal is wirelessly connected to Alibaba Cloud;

[0015] The monitoring terminal server has a synthesis function and is used to provide services to the regional geological data fusion visualization monitoring center;

[0016] The radio access network controller is connected to the cluster and generates drilling data and surface data units for collecting all server information;

[0017] The regional geological data fusion visualization monitoring center randomly traverses any regional geological monitoring terminal node of the regional geological three-dimensional space expression vector to request to query any content of the regional geological fusion monitoring. If the transmission time between any nodes in the cluster is less than the transmission time to Alibaba Cloud, the regional geological data fusion visualization monitoring center obtains a response.

[0018] Further, the responses obtained by the regional geological data fusion visualization monitoring center include:

[0019] (1) Traverse the monitoring terminal server of the regional geological three-dimensional space expression vector, synthesize the content that the regional geological data fusion visualization monitoring center requests to query the regional geological fusion monitoring, and the server responds to the request of the regional geological data fusion visualization monitoring center to query the regional geological fusion monitoring; check whether a certain synthesis of the traversed monitoring terminal server of the regional geological three-dimensional space expression vector synthesizes the content;

[0020] (2) If the content that the regional geological data fusion visualization monitoring center requests to query the regional geological fusion monitoring is synthesized, the synthesis sends the content to the regional geological data fusion visualization monitoring center, traverses the regional geological three-dimensional space expression vector and responds to the request of the regional geological data fusion visualization monitoring center to query the regional geological fusion monitoring of the local monitoring terminal server; determine whether other monitoring terminal servers synthesize the content;

[0021] (3) If no content is synthesized by all the regional geological monitoring terminal nodes in the search neural network layer, then the request to query the regional geological fusion monitoring by traversing the regional geological three-dimensional space expression vector will be forwarded to Alibaba Cloud;

[0022] (4) The request of the regional geological data fusion visualization monitoring center to query the regional geological fusion monitoring is responded in Alibaba Cloud. Alibaba Cloud sends the content of the request to query the regional geological fusion monitoring to the neural network layer head server, then to the server that traverses the regional geological three-dimensional space expression vector of the regional geological data fusion visualization monitoring center, and finally to the regional geological data fusion visualization monitoring center.

[0023] In step S2, the identification of node importance based on the adaptive adjustment method of the regional geological three-dimensional space expression vector value includes:

[0024] Let the neural network be H=(E,F), where E is the set of network nodes (E={1,2…E}); F is the set of edges in the network, F={1,2…F}; F ij is the edge between node i and node j, and the domain width of F ij is denoted by u ijIndicates that when there is an edge between node i and node j, then u ij > 0, otherwise u ij = 0;

[0025] S201, Node synthesis space, ca i represents the available resources of the node. The more available synthesis resources, the more important the node;

[0026] S202, The sum of the adjacency domain widths of the node. The expression for the sum of the adjacency domain widths of the node is:

[0027]

[0028] In the formula, N(i) is the set of neighbor nodes of node i in the network. The larger the sum of the adjacency domain widths of the node, the more important the node; n is the cumulative number of adjacency domain widths, g i is the adjacency domain width value of node i, and j is the jth neighbor node;

[0029] S203, The number of times of traversing the vector nodes in the geological three-dimensional space of the traversal area. The number of times of traversing the vector nodes in the geological three-dimensional space of the node depends on the spatial expression parameters and the degree of attention of the synthesized content. The more spatial expression parameters, the more times of traversing the vector nodes in the geological three-dimensional space of the node, the more the content is concerned, and the more times of traversing the vector nodes in the geological three-dimensional space of the traversal area; The spatial expression parameters include: core, geological boundary, stratum, fault, fold, original section, auxiliary section diagram, DTM / DEM data;

[0030] S204, The kernel degree center of the vector in the geological three-dimensional space of the traversal area. By introducing the number of times of traversing the vector nodes in the geological three-dimensional space of the node, the kernel degree centrality of the vector in the geological three-dimensional space of the traversal area is constructed, and this is used as the synthesis evolution standard for determining the network synthesis strategy. The synthesis position is determined by the position of the node and the situation of the vector in the geological three-dimensional space of the traversal area; The kernel degree centrality rate of the vector in the geological three-dimensional space of the node is the sum of the kernel degrees of the vectors in the geological three-dimensional space of all neighbor nodes and the number of times of traversing the vector nodes in the geological three-dimensional space of this node;

[0031] S205, The stability rate of the vector in the geological three-dimensional space of the traversal area.

[0032] In step S205, the stability rate of the vector in the geological three-dimensional space of the traversal area, the expression is:

[0033]

[0034] In the formula, is the vector stability rate The larger the value, the more stable the vector representation of the geological three-dimensional space in the traversal area of the node. The request for the vector representation of the geological three-dimensional space in the traversal area of the regional geological data fusion visualization monitoring center is easier to meet, and the importance of the node is greater; l is the Boltzmann constant, representing the property of the system itself, l = 1 / ln(n); e is the neighbor node, ko j is the kernel degree of the vector representation of the geological three-dimensional space in the traversal area of node j, ko e is the kernel degree of the vector representation of the geological three-dimensional space in the traversal area of the e-th neighbor node.

[0035] In step S203, the number of times of traversing the nodes of the vector representation of the geological three-dimensional space in the traversal area includes:

[0036] S2031, the significance of the vector representation of the geological three-dimensional space in the traversal area. If the attention received by the parameters of the vector representation of the geological three-dimensional space in the traversal area satisfies the Zipf distribution, the significance expression of the parameter k of the vector representation of the geological three-dimensional space in the traversal area ranked α is:

[0037]

[0038] In the formula, is the significance value of the parameter k of the vector representation of the geological three-dimensional space in the traversal area ranked α, α is the ranking value of the parameters of the vector representation of the geological three-dimensional space in the traversal area, θ is the number of parameters of the vector representation of the geological three-dimensional space in the traversal area that receive attention, μ is the skewness coefficient of the Zipf distribution, the larger μ is, the easier it is for the parameters of the vector representation of the geological three-dimensional space with high significance to be traversed; num is the total number of parameters of the vector representation of the geological three-dimensional space in the traversal area;

[0039] S2032, the attention of the regional geological monitoring terminal to the regional geological data fusion visualization monitoring center. By analyzing the long-term records of the vector representation of the geological three-dimensional space in the traversal area of the regional geological data fusion visualization monitoring center, the stable attention trend of the regional geological data fusion visualization monitoring center is obtained. Define the long-term attention of the regional geological data fusion visualization monitoring center to the regional geological monitoring terminal i of the vector representation of the geological three-dimensional space in the traversal area as The expression is:

[0040]

[0041] In the formula, is the statistical vector value of the current regional geological monitoring terminal node i; o long (ΔL long ) is the current statistical vector value of the vector representation of the geological three-dimensional space in the traversal area of all regional geological data fusion visualization monitoring centers within the neural network layer;

[0042] Define the concern of the regional geological data fusion visualization monitoring center traversing the vector node i of the regional geological three-dimensional space expression within the most recent period as short-term concern. The expression is:

[0043]

[0044] In the formula, is the statistical traversal regional geological three-dimensional space expression vector of the terminal node i within the most recent period, o short (ΔL short ) is the statistical traffic of all regional geological data fusion visualization monitoring centers in the neural network layer traversing the regional geological three-dimensional space expression vector within the most recent period;

[0045] The current concern trend of the regional geological data fusion visualization monitoring center depends on the long-term concern and short-term concern of the regional geological data fusion visualization monitoring center. The potential concern of the regional geological data fusion visualization monitoring center traversing the regional geological monitoring terminal node i of the regional geological three-dimensional space expression vector is defined as The expression is:

[0046]

[0047] In the formula, l1 is the influence proportion of the long-term concern on the current concern of the regional geological data fusion visualization monitoring center, and l2 is the influence proportion of the short-term concern on the current concern of the regional geological data fusion visualization monitoring center. Considering that the recent influence is greater, l2 is greater than l1;

[0048] S2033, regional geological data fusion visualization monitoring center, potential traversal regional geological three-dimensional space expression vector trend. The trend of the content of the regional geological data fusion visualization monitoring center traversing the regional geological three-dimensional space expression vector is not only affected by the concern trend, but also closely related to the degree of attention of the regional geological three-dimensional space expression parameters, that is, the regional geological data fusion visualization monitoring center always tends to request to query the significant content of the regional geological fusion they are concerned about; the request query regional geological fusion monitoring probability of content k in terminal node i is:

[0049]

[0050] In the formula, v is the number of concerned trend quantities of the regional geological three-dimensional space expression parameters;

[0051] Let the total number of regional geological data fusion visualization monitoring center traversing the regional geological three-dimensional space expression vector in the previous synthesis period be o ave, the regional geological data fusion visualization monitoring center traverses the potential traversal area geological three-dimensional space expression vector of content k in the regional geological three-dimensional space expression vector regional geological monitoring terminal node i It is:

[0052]

[0053] In step S204, traversing the regional geological three-dimensional space expression vector core degree center rate includes:

[0054] (A) Delete all nodes and edges with a connectivity of 1, and record the number of times the node is traversed in the regional geological three-dimensional space expression vector; if there are still nodes with a connectivity of 1, continue the above process, and mark the core degree of these deleted nodes as 1. The traversed regional geological three-dimensional space expression vector core degree is the number of times the node is traversed in the regional geological three-dimensional space expression vector plus 1. If the number of times the node j is traversed in the regional geological three-dimensional space expression vector is o j , then the traversed regional geological three-dimensional space expression vector core degree of node j is ko j , the expression is:

[0055] ko j = o j + 1

[0056] (B) For nodes with a connectivity of 2, mark the traversed regional geological three-dimensional space expression vector core degree of these nodes as the number of times the regional geological three-dimensional space expression vector is traversed + 2;

[0057] (C) Loop through the above process until all nodes are deleted and obtain the corresponding traversed regional geological three-dimensional space expression vector core degree; the number next to the node represents the number of times the regional geological three-dimensional space expression vector is traversed.

[0058] In step S3, based on the dynamic node importance arrangement adaptive to the regional geological three-dimensional space expression vector threshold, it includes:

[0059] S301, Evolution model. Suppose there are n nodes to be arranged in importance, and each node has 4 evaluation indicators. The value of the a-th evaluation indicator of the terminal node i is c ia , the evolution matrix C composed of all network nodes is expressed as:

[0060]

[0061] S302, Nonlinear programming evolution matrix. Since the dimensions of each indicator are different and there are differences in the order of magnitude, in order to eliminate the dimensional differences between the indicators, the index should be standardized, which is expressed as:

[0062]

[0063] where b ia is the normalized value of the ath evaluation index of the terminal node i;

[0064] The standard normalization matrix B is expressed as:

[0065]

[0066] S303. Calculate the vector threshold based on the exponential entropy for the three-dimensional spatial expression of regional geology;

[0067] The exponential entropy value w a is calculated by the following formula:

[0068]

[0069] Calculate the information entropy redundancy RR a , as shown in the following formula:

[0070] RR a = 1 - w a

[0071] Calculate the vector threshold u for the exponential-based three-dimensional spatial expression of regional geology a as shown in the following formula:

[0072]

[0073] The vector threshold matrix U for the exponential-based three-dimensional spatial expression of regional geology is obtained, as shown in the following formula:

[0074] U = [u1 u2 u3 u4]

[0075] where u1 is the first vector threshold for the three-dimensional spatial expression of regional geology, u2 is the second vector threshold for the three-dimensional spatial expression of regional geology, u3 is the third vector threshold for the three-dimensional spatial expression of regional geology, and u4 is the fourth vector threshold for the three-dimensional spatial expression of regional geology;

[0076] The weighted normalization evolution matrix Y is expressed as:

[0077]

[0078] The weighted attribute value K of the terminal node i i is expressed as:

[0079]

[0080] where Y ia is the weighted normalization evolution value of the ath evaluation index of the terminal node i;

[0081] Due to the collaborative synthesis among nodes in the neural network, adjacent nodes also contribute to the synthesis importance degree of the target node;

[0082] S304. Importance degree calculation: Based on the importance degree evaluation matrix, sum the attribute value of the terminal node i and the importance degree contributions of all adjacent nodes to obtain the importance degree ξ of the terminal node i i , and the expression is:

[0083]

[0084] In the formula, K i is the weighted attribute value of the terminal node i, θ i is the contribution distribution parameter, ξ i is the importance degree of the terminal node i after comprehensive evaluation, and ξ i reflects the value of the terminal node i in the network;

[0085] The importance degree evaluation matrix fully considers the node position, the geological three-dimensional spatial expression vector frequency in the traversed area, the synthesis space, and the available domain width, and more comprehensively reflects the importance degree of the node around the goal of synthesis value.

[0086] In step S4, implement the visualization monitoring of regional geological data fusion based on the adaptability of the regional geological three-dimensional spatial expression vector value, including:

[0087] S401. Node synthesis update rate R(i), and the expression is:

[0088]

[0089] In the formula, M is the number of regional geological three-dimensional spatial expression parameters updated from the terminal node i within a unit time, K(Ro m ) is the size of the regional geological three-dimensional spatial expression parameters updated from the terminal node i, and I(i) is the synthesis space size of the terminal node i;

[0090] S402. Node synthesis value proportion degree J(i), including the traversed regional geological three-dimensional spatial expression vector stability kernel degree center rate and the synthesis update rate, and the expression is:

[0091]

[0092] If the node synthesis update rate R(i) = 0, it means that the node synthesis space is not full, or there is no new input of regional geological three-dimensional spatial expression parameters.

[0093] Another object of the present invention is to provide a regional geological data fusion visualization monitoring system based on neural network learning. This system is implemented by the above-mentioned regional geological data fusion visualization monitoring method based on neural network learning, and the system includes:

[0094] A regional geological three-dimensional model construction module based on neural network learning, which is used to construct a node monitoring neural network learning model for regional geology and the surface layer, and perform fusion to complete the construction of a regional geological three-dimensional model based on neural network learning;

[0095] A node importance recognition module, which is used to recognize the importance of nodes in the constructed regional geological three-dimensional model based on neural network learning by means of an adaptive adjustment method of the vector value of the regional geological three-dimensional space expression;

[0096] A dynamic node importance arrangement module, which is used for dynamic node importance arrangement based on the threshold of the vector value of the regional geological three-dimensional space expression;

[0097] A regional geological data fusion visualization monitoring module, which is used to realize the regional geological data fusion visualization monitoring based on the adaptive vector value of the regional geological three-dimensional space expression according to the dynamic node importance arrangement

[0098] Combining all the above technical solutions, the beneficial effects of the present invention are as follows: The accuracy of the regional geological data fusion visualization monitoring and the accuracy of identifying geological anomaly areas of the present invention are further improved compared with the prior art. By adaptively weighting various features, the present invention more accurately selects high-speed synthetic regional geological monitoring terminal nodes, improves the synthetic hit rate and the response efficiency of the regional geological data fusion visualization monitoring center to request and query regional geological fusion monitoring. The present invention adopts multiple network models and simulates the algorithm from multiple angles. The results show that compared with the existing mechanism, this solution can obtain better traversal regional geological three-dimensional space expression vector synthesis efficiency in a more complex network access environment.

[0099] Taking a distributed neural network cluster as the object, the synthesis position of the content is determined by factors such as the node synthesis size, adjacent domain width, traversal regional geological three-dimensional space expression vector kernel degree center rate, and traversal regional geological three-dimensional space expression vector stability. This method uses the method of information entropy to allocate adaptive regional geological three-dimensional space expression vector values, and combines the node update rate to obtain the node with the largest synthesis value. Compared with other related solutions, this method has great advantages in the recognition accuracy and synthesis efficiency of synthesizing important nodes.

[0100] The neural network structure of the present invention takes into account the characteristics that the regional geological monitoring terminal nodes can traverse the vector of the three-dimensional spatial expression of regional geology and forward the request for traversing the vector of the three-dimensional spatial expression of regional geology to query the regional geological fusion monitoring, and calculates the characteristics related to node synthesis, including the synthesis space, adjacent domain width, core centrality of the traversed vector of the three-dimensional spatial expression of regional geology, and balance of the traversed vector of the three-dimensional spatial expression of regional geology, providing a basis for considering the importance of node synthesis.

[0101] Regarding the influence of multiple eigenvalues on the node synthesis value, the present invention adaptively allocates the threshold of each feature based on the vector of the three-dimensional spatial expression of regional geology by using information entropy. This helps to determine the importance of nodes flexibly and accurately; by calculating the update rate, frequent replacement of the content in the nodes is avoided, and the monitoring efficiency of the system is improved. The present invention conducts multi-angle experiments under a multi-network model. The experimental results show that compared with the existing algorithms, this algorithm can more accurately identify the importance of nodes and can more effectively improve the synthesis efficiency and the response rate of traversing the vector of the three-dimensional spatial expression of regional geology. BRIEF DESCRIPTION OF THE DRAWINGS

[0102] The drawings herein are incorporated into the specification and form a part of the specification, showing embodiments consistent with the present disclosure, and are used together with the specification to explain the principles of the present disclosure;

[0103] Figure 1 is a flowchart of a method for visual monitoring of regional geological data fusion based on neural network learning provided by an embodiment of the present invention;

[0104] Figure 2 is a schematic diagram of a system for visual monitoring of regional geological data fusion based on neural network learning provided by an embodiment of the present invention;

[0105] In the figure: 1. Module for constructing a three-dimensional model of regional geology based on neural network learning; 2. Module for identifying node importance; 3. Module for dynamically arranging node importance; 4. Module for visual monitoring of regional geological data fusion. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0106] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following detailed description of the specific embodiments of the present invention is provided in conjunction with the accompanying drawings. Many specific details are set forth in the following description to fully understand the present invention. However, the present invention can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0107] Embodiment 1, as Figure 1 shown, the method for visual monitoring of regional geological data fusion based on neural network learning provided by an embodiment of the present invention includes:

[0108] S1, construct a node monitoring neural network learning model for regional geology and the surface layer, and perform fusion to complete the construction of a three-dimensional model of regional geology based on neural network learning;

[0109] S2, for the constructed three-dimensional model of regional geology based on neural network learning, perform node importance recognition based on the adaptive adjustment method of vector values for three-dimensional spatial expression of regional geology;

[0110] S3, perform dynamic node importance arrangement based on the adaptive threshold of vector values for three-dimensional spatial expression of regional geology;

[0111] S4, according to the dynamic node importance arrangement, realize the visual monitoring of regional geology data fusion based on the adaptive vector values for three-dimensional spatial expression of regional geology.

[0112] Exemplarily, in step S1, the completion of the construction of the three-dimensional model of regional geology based on neural network learning includes:

[0113] Obtain the borehole data of regional geology, and classify the borehole data according to the category number and data volume of the boreholes; in the same category of borehole data, select the standard borehole data in this category, calculate the similarities and differences between other borehole data in this category and the standard borehole data, and generate the standard borehole data and the similarities and differences in a preset format in the server;

[0114] Obtain the surface layer data of regional geology, and the surface layer data is depth data and shooting data;

[0115] Generate borehole data and surface layer data, which includes a first synthesizing device and a second synthesizing device;

[0116] Among them, the first synthesizing device includes a synthesis processing module, a synthesizer, and a generator. Specifically:

[0117] The synthesis processing module is used to set the same category of borehole data as a small file package write operation into the synthesizer;

[0118] The synthesizer is used to synthesize the small file packages in memory, and then the small file packages will perform corresponding small file package merging operations according to different queries. After merging into a large borehole data file package, it is transmitted to the generator;

[0119] The generator is used to generate a large borehole data file package;

[0120] Among them, the second synthesizing device also includes a synthesis processing module, a synthesizer, and a generator. Specifically:

[0121] The synthesis processing module is used to set the depth data and shooting data as multiple small file package write operations into the synthesizer according to the preset file volume;

[0122] A synthesizer, which is used to synthesize multiple small file packages in memory respectively. After that, the multiple small file packages will perform corresponding small file package merging operations according to different queries. After being merged into a large surface data file package, it will be transmitted to a generator;

[0123] A generator, which is used to generate a large surface data file package;

[0124] In the process of completing the construction of a regional geological three-dimensional model based on neural network learning, the following are also constructed:

[0125] A geological database, which is used to transcribe the borehole data and surface data generated from a server;

[0126] A query module, which is used to query the data in the process of constructing a regional geological three-dimensional model based on neural network learning.

[0127] An interpolation module, which is used to perform interpolation processing on the discretely preprocessed borehole data obtained by sparse sampling in the geological database by using spatial interpolation and fitting algorithms;

[0128] A spatial data field construction module, which is used to judge and predict the distribution trend of geological body information and construct a three-dimensional regular spatial data field.

[0129] In the process of completing the construction of a regional geological three-dimensional model based on neural network learning, a regional geological modeling module is also constructed, which includes:

[0130] An extraction module, which is used to perform isosurface extraction on the geological bodies in the three-dimensional regular data field based on an implicit algorithm to obtain a geological ore body isosurface composed of countless small triangular facets;

[0131] A modeling module, which is used to convert multiple modeling points based on the geological ore body isosurface and construct a regional geological node monitoring neural network learning model based on the modeling points.

[0132] The geological surface modeling module also includes:

[0133] A conversion module, which is used to obtain the depth data and shooting data of the regional geological surface, convert them to obtain multiple surface modeling points, and determine the positions of the regional geological surface and the surface modeling points at the same time;

[0134] A surface modeling module, which is used to construct a regional geological surface node monitoring neural network learning model based on the positions of the regional geological surface and the surface modeling points.

[0135] When the surface data acquisition module acquires shooting data, it shoots videos at a first frame rate and a first resolution, shoots images at one-tenth of the first frame rate and a second resolution greater than the first resolution, and the images are shot within the same cycle as the videos.

[0136] In the construction of a 3D regional geological model based on neural network learning, a 3D model construction module is also constructed to fuse the constructed regional geological node monitoring neural network learning model and the constructed regional geological surface node monitoring neural network learning model, thereby completing the construction of the 3D regional geological model based on neural network learning.

[0137] In step S1, the neural network consists of a regional geological monitoring terminal, a monitoring terminal server, and a radio access network controller; the regional geological monitoring terminal is wirelessly connected and is connected to Alibaba Cloud through a specific regional geological monitoring terminal; the monitoring terminal server has a synthesis function and is used to provide services to the regional geological data fusion visualization monitoring center; the radio access network controller is connected to the generated borehole data and surface data units in the cluster and is used to collect all server information.

[0138] The regional geological data fusion visualization monitoring center can randomly traverse any regional geological monitoring terminal node in the 3D regional geological space expression vector to request the query of any content of the regional geological fusion monitoring. Assuming that the transmission time between any nodes in the cluster is less than the transmission time to Alibaba Cloud, the regional geological data fusion visualization monitoring center will obtain a response.

[0139] Exemplarily, the methods for the regional geological data fusion visualization monitoring center to obtain a response include:

[0140] (1) If the monitoring terminal server that traverses the 3D regional geological space expression vector synthesizes the content requested by the regional geological data fusion visualization monitoring center for querying the regional geological fusion monitoring, the server directly responds to the request of the regional geological data fusion visualization monitoring center for querying the regional geological fusion monitoring; check whether a certain synthesis of the monitoring terminal server that traverses the 3D regional geological space expression vector synthesizes the content.

[0141] (2) If the synthesis of the content requested by the regional geological data fusion visualization monitoring center for querying the regional geological fusion monitoring, the synthesis sends the content to the local monitoring terminal server that traverses the 3D regional geological space expression vector and responds to the request of the regional geological data fusion visualization monitoring center for querying the regional geological fusion monitoring; determine whether other monitoring terminal servers synthesize the content.

[0142] (3) If all regional geological monitoring terminal nodes in the neural network layer are searched and no synthesis content is found, the request for querying the regional geological fusion monitoring by traversing the 3D regional geological space expression vector will be forwarded to Alibaba Cloud.

[0143] (4) The request for querying the regional geological data fusion visualization monitoring center for regional geological fusion monitoring is responded in Alibaba Cloud. Alibaba Cloud sends the content of the request for querying regional geological fusion monitoring to the neural network layer head server, and then sends it to the regional geological data fusion visualization monitoring center to traverse the regional geological three-dimensional space expression vector server, and finally sends it to the regional geological data fusion visualization monitoring center.

[0144] Exemplarily, in step S2, the identification of the node importance degree based on the regional geological three-dimensional space expression vector value adaptive adjustment method includes:

[0145] Assume that the neural network is H=(E,F), where E is the set of network nodes, E={1,2…E}; F is the set of edges in the network, F={1,2…F}; F ij is the edge between node i and node j, and the domain width of F ij is represented by u ij . When there is an edge between node i and node j, then u ij >0, otherwise u ij =0;

[0146] S201, the node synthesis space, ca i represents the available resources of the node. The more available synthesis resources, the more important the node;

[0147] S202, the sum of the adjacent domain widths of the node. The expression for the sum of the adjacent domain widths of the node is:

[0148]

[0149] In the formula, N(i) is the set of neighbor nodes of node i in the network. The larger the sum of the adjacent domain widths of the node, the more important the node; n is the cumulative number of adjacent domain widths, g i is the adjacent domain width value of node i, and j is the jth neighbor node;

[0150] S203, the number of times of traversing the regional geological three-dimensional space expression vector nodes. The number of times of traversing the regional geological three-dimensional space expression vector nodes of the node depends on the spatial expression parameters and the degree of attention of the synthesis content. The more spatial expression parameters, the more times of traversing the regional geological three-dimensional space expression vector nodes of the node, the more the content is concerned, and the more times of traversing the regional geological three-dimensional space expression vector nodes; the spatial expression parameters include: core, geological boundary, stratum, fault, fold, original section, auxiliary section drawing, DTM / DEM data;

[0151] S204. Traverse the vector kernel degree center of the three-dimensional spatial expression of regional geology. By introducing the number of times of the three-dimensional spatial expression vector of the traversal area geology for nodes, construct the vector kernel degree centrality of the three-dimensional spatial expression of the traversal area geology, and use this as the synthesis evolution criterion for determining the network synthesis strategy. Determine the synthesis position based on the position of the node and the situation of the three-dimensional spatial expression vector of the traversal area geology; the vector kernel degree centrality rate of a node is the sum of the vector kernel degrees of the three-dimensional spatial expressions of all neighbor nodes of the traversal area geology and the number of times of the three-dimensional spatial expression vector of the traversal area geology of this node.

[0152] S205. Vector stability rate of the three-dimensional spatial expression of the traversal area geology.

[0153] Exemplarily, in step S205, the vector stability rate of the three-dimensional spatial expression of the traversal area geology has the following expression:

[0154]

[0155] In the formula, is the vector stability rate. The larger the value of, the more stable the three-dimensional spatial expression vector of the traversal area geology for the node, the easier it is to satisfy the query of the three-dimensional spatial expression vector request of the traversal area geology for the regional geology data fusion visualization monitoring center, and the greater the importance of the node; l is the Boltzmann constant, representing the property of the system itself, l = 1 / ln(n); e is the neighbor node, ko j is the vector kernel degree of the three-dimensional spatial expression of the traversal area geology of node j, ko e is the vector kernel degree of the three-dimensional spatial expression of the traversal area geology of the e-th neighbor node.

[0156] Exemplarily, in step S203, the number of times of the three-dimensional spatial expression vector nodes of the traversal area geology includes:

[0157] S2031. Significance of the three-dimensional spatial expression of regional geology. If the three-dimensional spatial expression parameters of regional geology are subject to the Zipf distribution, the significance expression of the three-dimensional spatial expression parameter k ranked α is:

[0158]

[0159] In the formula, is the significance value of the three-dimensional spatial expression parameter k ranked α, α is the ranking value of the three-dimensional spatial expression parameter, θ is the number of three-dimensional spatial expression parameters with attention, μ is the skewness coefficient of the Zipf distribution, the larger μ is, the easier it is for the three-dimensional spatial expression parameters with high significance to be traversed; num is the total number of three-dimensional spatial expression parameters.

[0160] S2032, the attention of the regional geological monitoring terminal to the regional geological data fusion visualization monitoring center is obtained by analyzing the long-term traversal regional geological three-dimensional spatial expression vector records of the regional geological data fusion visualization monitoring center, and a stable attention trend of the regional geological data fusion visualization monitoring center is obtained. The long-term attention of the regional geological data fusion visualization monitoring center to the traversal regional geological three-dimensional spatial expression vector regional geological monitoring terminal i is defined as The expression is:

[0161]

[0162] In the formula, is the statistical vector value of the current regional geological monitoring terminal node i; o log (ΔL long ) is the current statistical traversal regional geological three-dimensional spatial expression vector value of all regional geological data fusion visualization monitoring centers within the neural network layer;

[0163] Define the attention of the regional geological data fusion visualization monitoring center to the traversal regional geological three-dimensional spatial expression vector node i in the recent time as short-term attention The expression is:

[0164]

[0165] In the formula, is the statistical traversal regional geological three-dimensional spatial expression vector of the terminal node i in the recent period, o short (ΔL short ) is the statistical flow of the traversal regional geological three-dimensional spatial expression vector of all regional geological data fusion visualization monitoring centers in the neural network layer in the recent period;

[0166] The current attention trend of the regional geological data fusion visualization monitoring center depends on the long-term attention and short-term attention of the regional geological data fusion visualization monitoring center. The potential attention of the regional geological data fusion visualization monitoring center to the traversal regional geological three-dimensional spatial expression vector regional geological monitoring terminal node i is defined as The expression is:

[0167]

[0168] In the formula, l1 is the influence ratio of the long-term attention on the current attention of the regional geological data fusion visualization monitoring center, and l2 is the influence ratio of the short-term attention on the current attention of the regional geological data fusion visualization monitoring center. Considering that the recent influence is greater, l2 is greater than l1;

[0169] S2033, Regional Geological Data Fusion Visualization Monitoring Center, Potential Traversal Trend of Vector Representation of Regional Geological Three-Dimensional Space. The trend of the Regional Geological Data Fusion Visualization Monitoring Center traversing the vector content of the regional geological three-dimensional space expression is not only affected by the attention trend, but also closely related to the degree of attention of the regional geological three-dimensional space expression parameters. That is, the Regional Geological Data Fusion Visualization Monitoring Center always tends to request to query the regional geological fusion monitoring of the significant content they are concerned about; the request query probability of content k in terminal node i for regional geological fusion monitoring is:

[0170]

[0171] In the formula, v is the number of attention trends of the regional geological three-dimensional space expression parameters;

[0172] Let the total number of vectors representing the traversal of the regional geological three-dimensional space by the Regional Geological Data Fusion Visualization Monitoring Center in the previous synthesis cycle be o ave , then the potential vector representing the traversal of the regional geological three-dimensional space of the regional geological monitoring terminal node i for the vector representing the traversal of the regional geological three-dimensional space by the Regional Geological Data Fusion Visualization Monitoring Center is:

[0173]

[0174] Exemplarily, in step S204, the traversal of the central rate of the vector representing the regional geological three-dimensional space includes:

[0175] (A) Delete all nodes and edges with a connectivity of 1, and record the number of times the vector representing the traversal of the regional geological three-dimensional space of the nodes; if there are still nodes with a connectivity of 1, continue the above process, and mark the core degree of these deleted nodes as 1. The core degree of the vector representing the traversal of the regional geological three-dimensional space is the number of times the vector representing the traversal of the regional geological three-dimensional space of the node plus 1. If the number of times the vector representing the traversal of the regional geological three-dimensional space of node j is o j , then the core degree of the vector representing the traversal of the regional geological three-dimensional space of node j is ko j , and the expression is:

[0176] ko j = o j + 1

[0177] (B) For nodes with a connectivity of 2, mark the core degree of the vector representing the traversal of the regional geological three-dimensional space of these nodes as the number of times the vector representing the traversal of the regional geological three-dimensional space + 2;

[0178] (C) The above process is executed cyclically until all nodes are deleted, and the vector kernel degree of the three-dimensional spatial expression of the traversed regional geology is obtained; the number next to the node represents the number of the vector of the three-dimensional spatial expression of the traversed regional geology.

[0179] Exemplarily, in step S3, the dynamic node importance arrangement method based on the adaptive threshold of the three-dimensional spatial expression vector of the regional geology uses the information entropy theory to adaptively weight each index according to the change of the network index parameters; according to the information entropy theory, the higher the disorder of the index set, the greater the amount of information provided by the index, and the higher the threshold of the three-dimensional spatial expression vector of the regional geology for the comprehensive evaluation index.

[0180] The dynamic node importance arrangement method based on the adaptive threshold of the three-dimensional spatial expression vector of the regional geology specifically includes:

[0181] S301, Evolution model. There are n nodes to be arranged in importance, and each node has 4 evaluation indexes. The value of the a-th evaluation index of the terminal node i is c ia , and the evolution matrix C composed of all network nodes is expressed as:

[0182]

[0183] S302, Nonlinear programming evolution matrix. Since the dimensions of each index are different and there are differences in the order of magnitude, in order to eliminate the dimensional differences between the indexes, the index should be standardized, which is expressed as:

[0184]

[0185] In the formula, b ia is the standardized value of the a-th evaluation index of the terminal node i;

[0186] The standard normalization matrix B is expressed as:

[0187]

[0188] S303, Calculate the threshold based on the three-dimensional spatial expression vector of the regional geology according to the exponential entropy;

[0189] The exponential entropy value w a is calculated by the formula:

[0190]

[0191] Calculate the information entropy redundancy RR a , as shown in the following formula:

[0192] RR a =1 - w a

[0193] The calculation index is based on the vector threshold u of the three-dimensional spatial expression of regional geology a As shown in the following formula:

[0194]

[0195] The index is obtained based on the vector threshold matrix U of the three-dimensional spatial expression of regional geology, as shown in the following formula:

[0196] U = [u1 u2 u3 u4]

[0197] In the formula, u1 is the first vector threshold of the three-dimensional spatial expression of regional geology, u2 is the second vector threshold of the three-dimensional spatial expression of regional geology, u3 is the third vector threshold of the three-dimensional spatial expression of regional geology, and u4 is the fourth vector threshold of the three-dimensional spatial expression of regional geology;

[0198] The weighted normalization evolution matrix Y is expressed as:

[0199]

[0200] The weighted attribute value K of the terminal node i i is expressed as:

[0201]

[0202] In the formula, Y ia is the weighted normalization evolution value of the a-th evaluation index of the terminal node i;

[0203] Since the nodes in the neural network achieve collaborative synthesis, adjacent nodes will also contribute to the synthesis importance of the target node;

[0204] S304, importance calculation, based on the importance evaluation matrix, sum the attribute value of the terminal node i and the importance contributions of all adjacent nodes to obtain the importance ξ of the terminal node i i , and the expression is:

[0205]

[0206] In the formula, K i is the weighted attribute value of the terminal node i, θ i is the contribution distribution parameter, ξ i is the importance of the terminal node i after comprehensive evaluation, ξ i reflects the value of the terminal node i in the network;

[0207] The importance evaluation matrix fully considers the node position, the frequency of traversing the vector of the three-dimensional spatial expression of regional geology, the synthesis space, and the available domain width, and more comprehensively reflects the importance of the node around the goal of synthesis value.

[0208] Exemplarily, to implement the visualization monitoring of regional geological data fusion based on the adaptive vector value of the three-dimensional spatial expression of the region geology, it includes:

[0209] S401. The node synthesis update rate R(i), and the expression is:

[0210]

[0211] In the formula, M is the number of updated regional geological three-dimensional spatial expression parameters from the terminal node i within a unit time, K(Ro m ) is the size of the updated regional geological three-dimensional spatial expression parameters from the terminal node i, and I(i) is the synthesis space size of the terminal node i;

[0212] S402. The node synthesis value proportion degree J(i), which includes traversing the stability kernel degree center rate and the synthesis update rate of the regional geological three-dimensional spatial expression vector, and the expression is:

[0213]

[0214] If the node synthesis update rate R(i) = 0, it means that the node synthesis space is not full, or there is no new input of regional geological three-dimensional spatial expression parameters.

[0215] Among them, the parameters involved in the present invention are shown in Table 1;

[0216] Table 1 Parameter Table

[0217]

[0218] Example 2, as Figure 2 shown, the embodiment of the present invention provides a visualization monitoring system for regional geological data fusion based on neural network learning, including:

[0219] The regional geological three-dimensional model construction module 1 based on neural network learning is used to construct a node monitoring neural network learning model for the regional geology and its surface layer, and perform fusion to complete the construction of the regional geological three-dimensional model based on neural network learning;

[0220] The node importance recognition module 2 is used to recognize the importance of nodes of the constructed regional geological three-dimensional model based on the adaptive adjustment method of the regional geological three-dimensional spatial expression vector value;

[0221] The dynamic node importance arrangement module 3 is used for the dynamic node importance arrangement based on the adaptive threshold of the regional geological three-dimensional spatial expression vector;

[0222] The regional geological data fusion visualization monitoring module 4 is used to achieve the regional geological data fusion visualization monitoring based on the adaptive vector value of the regional geological three-dimensional space expression according to the arrangement of the dynamic node importance.

[0223] In the above embodiments, the descriptions of the respective embodiments have their own emphases. For the parts not detailed or recorded in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.

[0224] Regarding the information interaction, execution process, etc. between the above-mentioned devices / units, since they are based on the same concept as the method embodiments of the present invention, for their specific functions and the technical effects brought, reference may be specifically made to the method embodiment part, and details are not elaborated herein.

[0225] Those skilled in the art can clearly understand that, for the convenience and conciseness of description, only the above division of each functional unit and module is used as an example. In actual applications, the above functions can be allocated to different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiments can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated units can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of each functional unit and module are only for the convenience of mutual distinction and do not limit the protection scope of the present invention. The specific working process of the units and modules in the above system can refer to the corresponding process in the foregoing method embodiments.

[0226] The embodiment of the present invention also provides a computer device, which includes: at least one processor, a generator, and a computer program generated in the generator and executable on the at least one processor. When the processor executes the computer program, the steps in any of the above method embodiments are implemented.

[0227] The embodiment of the present invention also provides a computer-readable generation medium, on which a computer program is generated. When the computer program is executed by a processor, the steps in each of the above method embodiments can be implemented.

[0228] The embodiment of the present invention also provides an information data processing terminal, which is used to provide an input interface for the regional geological data fusion visualization monitoring center when executed on an electronic device to implement the steps in each of the above method embodiments. The information data processing terminal is not limited to mobile phones, computers, and switches.

[0229] An embodiment of the present invention further provides a server, which is used to provide an input interface for a regional geological data fusion visualization monitoring center when implemented on an electronic device, so as to implement the steps in the above method embodiments.

[0230] An embodiment of the present invention provides a computer program product, which, when running on an electronic device, enables the electronic device to implement the steps in the above method embodiments when executed.

[0231] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be generated in a computer-readable generation medium. Based on such an understanding, to implement all or part of the processes in the above method embodiments of the present application, a computer program can be used to instruct relevant hardware to complete. The computer program can be generated in a computer-readable generation medium. When the computer program is executed by a processor, the steps in the above method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file or some intermediate form, etc. The computer-readable medium can at least include: any entity or device that can carry the computer program code to the photographing device / terminal device, recording medium, computer generator, read-only generator (Read-Only Memory, ROM), random access generator (Wireless Access Network domAccess Memory, RAM), electrical carrier signal, telecommunication signal, and software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk or an optical disc, etc.

[0232] To verify the performance of the regional geological data fusion visualization monitoring of the present invention, this algorithm was compared with other node importance evaluation algorithms, the number of traversal regional geological three-dimensional space expression vector requests to query the regional geological fusion monitoring response times and the synthesis efficiency. The experiment conducted importance evaluation from two perspectives. One was to calculate the importance and level of nodes according to the topology (10 nodes), and the other was to construct a new topology (14 nodes) from the perspective of network efficiency to rank the nodes in terms of importance. This experiment simulated the implementation of a high-speed synthesis network through the NS3 simulator and implemented the regional geological data fusion visualization monitoring proposed by the present invention through coding. The simulation data was imported into Matlab for processing.

[0233] Taking a small-sample network as an example, various core node verification methods were used to evaluate and rank the node importance respectively. The results show that many algorithms cannot accurately distinguish the relative importance of nodes. Therefore, it is very important to adopt a more accurate method, and nodes n1 and n2 are used more precisely.

[0234] Comparison of network access efficiency. The network traversal area geological three-dimensional spatial expression vector efficiency (NAE) means that all the edges connected to a node are deleted at the same time, which may lead to an increase in the shortest traversal area geological three-dimensional spatial expression vector path between nodes, thus increasing the total length of the paths of the entire network to satisfy all traversal area geological three-dimensional spatial expression vectors. NAE reflects:

[0235]

[0236] In the formula, V is the shortest path between the terminal node i and the index node j, and Q is the number of nodes in the network that potentially traverse the area geological three-dimensional spatial expression vector.

[0237] Suppose the network has 14 nodes and 28 edges, the traversal area geological three-dimensional spatial expression vector times of each node are random, and importance assessment has been carried out for all nodes. Table 2 shows the node importance arrangements of different importance arrangement methods in the 14-node network. It can be seen that different methods have different node importance arrangements in the same network. Therefore, it is necessary to use the NAE performance index to more accurately analyze the importance of nodes in the network.

[0238] Comparison of the ability to satisfy the traversal area geological three-dimensional spatial expression vector request query area geological fusion monitoring. In order to more accurately verify the effectiveness of this method, the experiment used a neural network and compared the effects of different methods on selecting different nodes for traversal area geological three-dimensional spatial expression vector request query area geological fusion monitoring using the independent cascade model.

[0239] In order to analyze the differences between the method proposed in the present invention and other methods in evaluating the ability of nodes to respond to traversal area geological three-dimensional spatial expression vector request query area geological fusion monitoring, comparisons were made with existing methods respectively. Since the same node satisfies the same number of traversal area geological three-dimensional spatial expression vector request query area geological fusion monitoring in the same round, three nodes with good evaluability and incomplete consistency were used as source nodes.

[0240] The differences in the efficiency of the first three nodes of each evaluation method in satisfying the traversal area geological three-dimensional spatial expression vector request query area geological fusion monitoring. It can be learned that compared with other existing methods, the method of the present invention is superior to other methods in satisfying the traversal area geological three-dimensional spatial expression vector request query area geological fusion monitoring and can satisfy more traversal area geological three-dimensional spatial expression vector request query area geological fusion monitoring in the same round. The analysis shows that due to the introduction of the concept of access core, comprehensively considering the connectivity, access frequency and access diversity of nodes, it can better satisfy the access request query area monitoring.

[0241] The present invention takes into account the vector frequency of the three-dimensional spatial expression of the traversed area geology of the request query area geological fusion monitoring node, and the discovered "important" node is the node with the least comprehensive hop count. It can find the "important" node more accurately than Betw.

[0242] The total number of synthesized replacements for the content replaced by nodes. This parameter mainly includes the number of syntheses and the total number of syntheses in the simulation time. Through this parameter, the load situation of the synthesis system can be analyzed, and the synthesis position can be determined by considering the importance and replacement frequency of the nodes, so as to make the load of each node more balanced and effective.

[0243] This method takes the need to meet the requirements of the regional geological data fusion visualization monitoring center as the guiding research object, and adaptively assigns a threshold based on the three-dimensional spatial expression vector of the regional geology to various factors affecting the importance of node synthesis, and can more effectively meet the request query area geological fusion monitoring of the regional geological data fusion visualization monitoring center than other synthesis evolution methods.

[0244] As described above, only the relatively optimal specific implementation manner of the present invention is provided, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, any modifications, equivalent replacements, and improvements made within the spirit and principle of the present invention should be covered within the protection scope of the present invention.

Claims

1. A regional geological data fusion visualization monitoring method based on neural network learning, characterized in that: The method includes: S1, constructing a neural network learning model for regional geology and surface node monitoring, and integrating it to complete the construction of a regional geological three-dimensional model based on neural network learning; S2, for the constructed regional geological three-dimensional model based on neural network learning, node importance identification is performed based on the adaptive adjustment method of the regional geological three-dimensional spatial expression vector value; S3, dynamic node importance arrangement based on adaptive threshold of regional geological three-dimensional spatial expression vector; S4, according to the dynamic node importance arrangement, realizes the regional geological data fusion visualization monitoring based on the adaptive regional geological three-dimensional spatial expression vector value; In step S2, node importance identification based on the adaptive adjustment method of regional geological three-dimensional spatial expression vector value includes: Assume the neural network is , is the set of network nodes, ; is the edge set in the network, ; For Node and nodes The edge between Domain width Indicates that when the node and nodes When there is an edge between them, ,otherwise ; S201, node synthesis space, Represents the available resources of the node; S202, the sum of the width of the adjacent domain of the node. The expression of the sum of the width of the adjacent domain of the node is: ； In the formula, Nodes in the network The set of neighbor nodes of is the cumulative number of adjacent domain widths, For Node Adjacent domain width, For the neighbor nodes; S203, the number of times of traversing the regional geological three-dimensional spatial expression vector nodes, the number of times of traversing the regional geological three-dimensional spatial expression vector nodes for the nodes depends on the spatial expression parameters of the synthetic content and the degree of attention received, the more spatial expression parameters there are, the more times of traversing the regional geological three-dimensional spatial expression vector nodes for the nodes, the more attention received by the content, the more times of traversing the regional geological three-dimensional spatial expression vector nodes; the spatial expression parameters include: rock core, geological boundary, stratum, fault, fold, original section, auxiliary section, DTM / DEM data; S204, the nuclearity center of the traversed regional geological three-dimensional space expression vector is constructed by introducing the number of traversed regional geological three-dimensional space expression vectors of the node to construct the nuclearity centrality of the traversed regional geological three-dimensional space expression vector, and this is used as the synthesis evolution standard for determining the network synthesis strategy, and the synthesis position is determined by the position of the node and the traversed regional geological three-dimensional space expression vector situation; the nuclearity centrality of the traversed regional geological three-dimensional space expression vector of the node is the nuclearity of the traversed regional geological three-dimensional space expression vector of all neighboring nodes and the number of traversed regional geological three-dimensional space expression vectors of the node; S205, traversing the regional geological three-dimensional space to express the vector stability rate.

2. The regional geological data fusion visualization monitoring method based on neural network learning according to claim 1 is characterized in that: In step S1, the node monitoring neural network learning model is composed of a regional geological monitoring terminal, a monitoring terminal server and a wireless access network controller; The regional geological monitoring terminal is wirelessly connected to Alibaba Cloud; The monitoring terminal server is used to provide services to the regional geological data fusion visualization monitoring center; The wireless access network controller is connected to the cluster to generate drilling data and surface data units for collecting all server information; The regional geological data fusion visualization monitoring center randomly traverses any regional geological monitoring terminal node of the regional geological three-dimensional space expression vector to request to query any content of regional geological fusion monitoring. If the transmission time between any nodes in the cluster is less than the transmission time to Alibaba Cloud, the regional geological data fusion visualization monitoring center obtains a response.

3. The regional geological data fusion visualization monitoring method based on neural network learning according to claim 2 is characterized in that: The regional geological data fusion visualization monitoring center received responses, including: (1) Traverse the monitoring terminal server of the regional geological three-dimensional space expression vector, synthesize the regional geological data fusion visualization monitoring center to request to query the content of regional geological fusion monitoring, and the server responds to the request of the regional geological data fusion visualization monitoring center to query the regional geological fusion monitoring; check whether a synthesis of the monitoring terminal server traversing the regional geological three-dimensional space expression vector is synthesized content; (2) If the synthesized regional geological data fusion visualization monitoring center requests to query the regional geological fusion monitoring content, the synthesized content is sent to the regional geological data fusion visualization monitoring center, the regional geological three-dimensional space expression vector is traversed, and the local monitoring terminal server responds to the request of the regional geological data fusion visualization monitoring center to query the regional geological fusion monitoring; determine whether other monitoring terminal servers synthesize the content; (3) If all regional geological monitoring terminal nodes in the search neural network layer do not have synthetic content, the request to query regional geological fusion monitoring by traversing the regional geological three-dimensional spatial expression vector will be forwarded to Alibaba Cloud; (4) The request of the Regional Geological Data Fusion Visualization Monitoring Center to query the regional geological fusion monitoring is responded in Alibaba Cloud. Alibaba Cloud sends the request to query the regional geological fusion monitoring content to the neural network layer head server, and then sends it to the Regional Geological Data Fusion Visualization Monitoring Center to traverse the regional geological three-dimensional space expression vector server, and finally sends it to the Regional Geological Data Fusion Visualization Monitoring Center.

4. The regional geological data fusion visualization monitoring method based on neural network learning according to claim 1 is characterized in that: In step S205, the regional geological three-dimensional space is traversed to express the vector stability rate, and the expression is: ； In the formula, Vector stability rate The larger the value is, the more stable the traversal regional geological three-dimensional space expression vector of the node is, the easier it is to satisfy the traversal regional geological three-dimensional space expression vector request query regional geological fusion monitoring of the regional geological data fusion visualization monitoring center, and the greater the importance of the node is. is the Boltzmann constant, which represents the properties of the system itself, ; is the neighbor node, For Node The vector kernel of the traversal regional geological three-dimensional space expression, For the The vector kernel of the traversed regional geological three-dimensional space is expressed by the neighbor nodes.

5. The regional geological data fusion visualization monitoring method based on neural network learning according to claim 1 is characterized in that: In step S203, the number of times the regional geological three-dimensional space expression vector nodes are traversed includes: S2031, regional geological three-dimensional spatial expression vector significance, regional geological three-dimensional spatial expression parameter attention degree meets Distribution, ranking Regional geological three-dimensional spatial expression parameters The significance expression of is: ； In the formula, For ranking Regional geological three-dimensional spatial expression parameters The significance value of It is the ranking value of regional geological three-dimensional spatial expression parameters. is the number of parameters expressing the regional geological three-dimensional space that has attracted attention, for The skewness coefficient of the distribution, The larger it is, the easier it is to traverse the regional geological three-dimensional spatial expression parameters with high significance; is the total number of regional geological three-dimensional spatial expression parameters; S2032, regional geological monitoring terminal regional geological data fusion visualization monitoring center attention, by analyzing the long-term traversal regional geological three-dimensional space expression vector records of the regional geological data fusion visualization monitoring center, obtain the stable attention trend of the regional geological data fusion visualization monitoring center, define the regional geological data fusion visualization monitoring center's attention to the regional geological monitoring terminal traversal regional geological three-dimensional space expression vector Long-term attention is , the expression is: ； In the formula, It is the current regional geological monitoring terminal node The statistical vector value of ; The vector value of the three-dimensional space expression of regional geology for the current statistical traversal of the regional geological data fusion visualization monitoring center in the neural network layer; The regional geological data is integrated into the visualization monitoring center to traverse the regional geological three-dimensional space expression vector nodes in the nearest time. Short-term attention , the expression is: ； In the formula, The terminal node in the recent period The statistical traversal of the regional geological three-dimensional space expression vector, The statistical flow of the expression vector of the regional geological three-dimensional space traversed by the visualization monitoring center for all regional geological data in the neural network layer in the recent period; The current focus of the regional geological data fusion visualization monitoring center depends on the long-term and short-term focus of the regional geological data fusion visualization monitoring center. The regional geological data fusion visualization monitoring center traverses the regional geological three-dimensional space expression vector regional geological monitoring terminal node Potential concerns are defined as , the expression is: ； In the formula, For the long-term concern about the impact of the current focus on the regional geological data fusion visualization monitoring center, For the short-term focus on the impact of the current focus of the Regional Geological Data Fusion Visualization Monitoring Center, considering that the recent impact is greater, Greater than ; S2033, regional geological data fusion visualization monitoring center, potential traversal of regional geological three-dimensional space expression vector trend, regional geological data fusion visualization monitoring center tends to request query regional geological fusion monitoring of their significant content, terminal node Content Request to query regional geological fusion monitoring probability for: ； In the formula, The number of interesting trends in the three-dimensional spatial expression parameters of regional geology; Suppose the total number of regional geological three-dimensional spatial expression vectors traversed by the regional geological data fusion visualization monitoring center in the previous synthesis cycle is , then the regional geological data fusion visualization monitoring center traverses the regional geological three-dimensional space expression vector regional geological monitoring terminal node Content The potential traversal regional geological three-dimensional spatial expression vector for: 。 6. The regional geological data fusion visualization monitoring method based on neural network learning according to claim 1 is characterized in that: In step S204, the regional geological three-dimensional space is traversed to express the vector kernel central rate, including: (A) Delete all nodes and edges with a connectivity of 1, and record the number of times the node traverses the regional geological three-dimensional space expression vector; if there are still nodes with a connectivity of 1, continue the above process and mark the coreness of these deleted nodes as 1. The coreness of the traversed regional geological three-dimensional space expression vector is the number of traversed regional geological three-dimensional space expression vectors of the node plus 1. If the node The number of vectors expressing the traversed regional geological three-dimensional space is , then the node The vector kernel of the traversal regional geological three-dimensional space expression is , the expression is: ； (B) For nodes with a connectivity of 2, the coreness of the traversed regional geological three-dimensional space expression vectors of these nodes is marked as the number of traversed regional geological three-dimensional space expression vectors + 2; (C) The above process is executed repeatedly until all nodes are deleted and the corresponding traversed regional geological three-dimensional space expression vector coreness is obtained; the number next to the node indicates the number of traversed regional geological three-dimensional space expression vectors.

7. The regional geological data fusion visualization monitoring method based on neural network learning according to claim 1 is characterized in that: In step S3, the dynamic node importance arrangement based on the adaptive threshold of the regional geological three-dimensional spatial expression vector includes: S301, the evolution model, has The nodes to be ranked in importance, each node has 4 evaluation indicators, the terminal node No. The value of the evaluation index is , the evolution matrix composed of all network nodes It is expressed as: ； S302, nonlinear programming evolution matrix, standardizes the index and eliminates the dimension difference between the indicators, expressed as: ； In the formula, For terminal nodes No. The standardized value of the evaluation index; Standard normalized matrix It is expressed as: ； S303, calculating a vector threshold based on the regional geological three-dimensional spatial expression according to the exponential entropy; Exponential Entropy The calculation formula is: ； Calculating information entropy redundancy , the expression is: ； Calculation index based on regional geological three-dimensional spatial expression vector threshold , the expression is: ； Index based on regional geological three-dimensional spatial expression vector threshold matrix is obtained, the expression is: ； In the formula, They are the vector thresholds for the geological three-dimensional space expression of the 1st, 2nd, 3rd and 4th regions respectively; Weighted Normalized Evolution Matrix It is expressed as: ； Endpoints The weighted attribute value of It is expressed as: ； In the formula, For terminal nodes No. The weighted normalized evolution value of the evaluation index; S304, importance calculation, based on the importance evaluation matrix, the terminal nodes The attribute value of the node and the importance contribution of all adjacent nodes are summed to obtain the terminal node Importance , the expression is: ； In the formula, For terminal nodes The weighted attribute value of Assign parameters to contributions, Terminal nodes after comprehensive evaluation The importance of Reflects the terminal node Value in the network.

8. The regional geological data fusion visualization monitoring method based on neural network learning according to claim 7 is characterized in that: In step S4, regional geological data fusion visualization monitoring based on regional geological three-dimensional spatial expression vector value adaptation is realized, including: S401, node synthesis update rate , the expression is: ； In the formula, is the time from the terminal node Update the number of regional geological three-dimensional spatial expression parameters, From the terminal node The size of the updated regional geological three-dimensional spatial expression parameters, For terminal nodes The size of the synthesis space; S402, node synthesis value ratio , traversing the regional geological three-dimensional space to express the vector stability core center rate and synthetic update rate, the expression is: ； If the node synthesis update rate , which means that the node synthesis space is not full, or there is no new regional geological three-dimensional spatial expression parameter input.

9. A regional geological data fusion visualization monitoring system based on neural network learning, characterized in that: The system is implemented by the regional geological data fusion visualization monitoring method based on neural network learning according to any one of claims 1 to 8, and the system includes: A regional geological three-dimensional model construction module (1) based on neural network learning is used to construct a regional geological and surface node monitoring neural network learning model, and to integrate and complete the construction of a regional geological three-dimensional model based on neural network learning; A node importance identification module (2) is used to identify the node importance of the regional geological three-dimensional model constructed based on neural network learning, based on the regional geological three-dimensional space expression vector value adaptive adjustment method; A dynamic node importance ranking module (3) is used for dynamic node importance ranking based on adaptive threshold of regional geological three-dimensional spatial expression vector; The regional geological data fusion visualization monitoring module (4) is used to realize regional geological data fusion visualization monitoring based on the adaptive regional geological three-dimensional spatial expression vector value according to the dynamic node importance arrangement.

Images