CN110109995B - Fully mechanized mining face multi-source heterogeneous data fusion method based on deep learning - Google Patents
Fully mechanized mining face multi-source heterogeneous data fusion method based on deep learning Download PDFInfo
- Publication number
- CN110109995B CN110109995B CN201910397221.XA CN201910397221A CN110109995B CN 110109995 B CN110109995 B CN 110109995B CN 201910397221 A CN201910397221 A CN 201910397221A CN 110109995 B CN110109995 B CN 110109995B
- Authority
- CN
- China
- Prior art keywords
- heterogeneous data
- source heterogeneous
- deep learning
- data
- fully mechanized
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/25—Integrating or interfacing systems involving database management systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/251—Fusion techniques of input or preprocessed data
Abstract
The invention discloses a fully mechanized coal mining face multi-source heterogeneous data fusion method based on deep learning, which is suitable for the technical field of intelligent mining of coal mines. According to multi-source heterogeneous data information of the fully mechanized mining face in the mining process, establishing a deep learning fusion model of the multi-source heterogeneous data of the fully mechanized mining face: establishing a multi-source heterogeneous data model of the fully mechanized coal mining face; unifying the spatial reference, the data attribute and the element relation of the multi-source heterogeneous data model; generating an irregular triangular net of a multi-source heterogeneous data model of the fully mechanized coal mining face; and extracting characteristic parameters of the irregular triangulation network, checking whether the iteration conditions of the deep learning algorithm are met, and establishing a deep learning fusion model of the multi-source heterogeneous data of the fully mechanized mining face. The method can quickly fuse various modes of heterogeneous data of the fully mechanized coal mining face in the mining process, and effectively improve the efficiency and accuracy of data fusion.
Description
Technical Field
The invention relates to a multi-source heterogeneous data fusion method for a fully mechanized coal mining face, in particular to a deep learning-based multi-source heterogeneous data fusion method for the fully mechanized coal mining face, which is suitable for the technical field of intelligent mining of coal mines.
Background
The intelligent coal mining is a leading-edge technology commonly pursued in the international coal mining field, is an important means for reducing casualties and guaranteeing safe production, and is an effective way for realizing safe, efficient and green mining of coal mines in China. In order to realize intelligent mining of coal mines, a large amount of multi-source heterogeneous data, such as groove wave detection data, laser scanner data, mine ground measurement data, ground penetrating radar data, three-dimensional seismic data, actual measurement production data in the mining process and the like, need to be acquired in the mining process of a fully mechanized mining face. These data acquisition methods, time and sensors used are different, and therefore, there is a problem that spatial references, data attributes, element relationships and the like are not uniform. However, the existing multi-source heterogeneous data processing method mostly adopts a 'warfare for each' processing mode, that is, databases measured by respective sensors are respectively established, and the databases are not related to each other, so that the corresponding relation between the operation object and the mining environment in the mining process cannot be accurately reflected.
Disclosure of Invention
The technical problem is as follows: aiming at the defects of the technology, the method for fusing various modes of heterogeneous data of the fully mechanized coal mining face in the mining process, which has the advantages of simple steps, high execution efficiency, and capability of providing accurate multi-source heterogeneous data fusion information for intelligent mining of a coal mine quickly and efficiently and saving manpower
The technical scheme is as follows: in order to achieve the purpose, the deep learning-based fully-mechanized working face multi-source heterogeneous data fusion method provided by the invention establishes a deep learning fusion model of fully-mechanized working face multi-source heterogeneous data according to multi-source heterogeneous data information of the fully-mechanized working face in a mining process, and specifically comprises the following steps:
according to multi-source heterogeneous data information of the fully mechanized mining face in the mining process, a deep learning fusion model of the multi-source heterogeneous data of the fully mechanized mining face is established; the method comprises the following specific steps:
a, acquiring multi-source heterogeneous data information in the mining process of a fully mechanized mining face, comprising: the working face channel wave detection data, the laser scanner data, the mine ground survey data, the ground penetrating radar data, the three-dimensional seismic data to carry out optimization processing to above-mentioned all multisource heterogeneous data information, include: removing noise points by adopting a chord height difference method, carrying out smooth filtering by adopting an average filtering method, carrying out data repairing by adopting cubic spline interpolation, and establishing a multi-source heterogeneous data model of the fully mechanized coal mining face;
b, converting multi-source heterogeneous data information into a fully mechanized working face multi-source heterogeneous data model by using a Morojinsky seven-parameter coordinate conversion model, and performing unification processing on a spatial reference, data attributes and element relations in the fully mechanized working face multi-source heterogeneous data model;
c, generating an irregular triangular net of the multi-source heterogeneous data model of the fully mechanized coal mining face by using a point-by-point insertion method;
d, extracting characteristic parameters of the irregular triangular net, wherein the characteristic parameters comprise the number of layers, the number of nodes and a training algebra of the triangular net, checking the number of layers, the number of nodes and the absolute value linear convergence of the training algebra of the triangular net so as to judge whether the iteration condition of the deep learning algorithm is met, and if the iteration condition is not met, realizing that the multisource heterogeneous data model of the fully mechanized mining face meets the iteration condition requirement of the deep learning algorithm by adjusting the number of layers, the number of nodes and the size of the training algebra;
e, if the iteration condition of the deep learning algorithm is met, performing feature training by using the deep learning training sample, then verifying the multi-source heterogeneous data model performance of the fully mechanized mining face by using the deep learning verification sample,
f, completing feature training on the deep learning training samples and completing model performance verification on the deep learning verification samples in the step e, and establishing a deep learning fusion model of the multi-source heterogeneous data of the fully mechanized mining face by using a general convolution neural network method.
The spatial reference unification processing process comprises the following steps: the obtained multi-source heterogeneous data is converted into a China geodetic coordinate system (CGCS2000) by adopting a Morojinskyqi parameter coordinate conversion model,
in the formula, XCGCS,YCGCS,ZCGCSIs the coordinate value, X, of the Chinese geodetic coordinate systemM,YM,ZMCoordinate values under a coordinate system of the multi-source heterogeneous data measurement sensor, wherein delta X, delta Y and delta Z are translation parameters between two coordinate origins, RX,RY,RZIs 3 coordinate axis rotation parameters, epsilon is a scale parameter;
the data attribute unification processing process comprises the following steps: adopting a multi-stage description method to carry out uniform processing on the attributes of the multi-source heterogeneous data, and setting the data attributes of the heterogeneous data A and B as follows: (mu.) aA,sA) And (mu)B,sB) Judging whether the data attributes of A and B satisfy the following formula:
where sim () is a data attribute consistency function, μ and s represent coordinate similarity and direction similarity, respectively, N is the total number of heterogeneous data, ω is1And ω2Is a weight parameter;
the element relationship unification processing process comprises the following steps: adopting a 9-intersection model and calculating the following topological relation equation of heterogeneous data A and B:
wherein R () is a data topological relation function, A0,And A-Respectively representing the internal topological relation, the boundary topological relation and the external topological relation of the data A, B0,And B-And respectively representing the internal topological relation, the boundary topological relation and the external topological relation of the data B.
The specific steps of the irregular triangulation network for generating the multi-source heterogeneous data model of the fully mechanized coal mining face by using the point-by-point insertion method are as follows:
c1 defining an initial polygon containing all the fully mechanized face multi-source heterogeneous data point sets;
c2 randomly selecting a point from the multi-source heterogeneous data point set of the fully mechanized mining face, inserting the point into the gravity center point of the initial polygon to establish an initial triangulation network;
c3, performing iterative computation until all data points are processed, namely, inserting a discrete sampling point in the initial triangulation network, finding out a triangle containing the point in the initial triangulation network, and connecting the point with three vertexes of the triangle to generate three new small triangles; then, optimizing the triangular net from inside to outside by using an LOP (local Optimization procedure) algorithm to ensure that the new small triangle meets the requirement of the irregular triangle;
c4 deletes the triangle containing the initial polygon points to form the final irregular triangulation network.
In the step d, the iteration condition of the deep learning algorithm needs to satisfy the following equation:
in the formula, P represents a combined probability distribution function of a deep learning hidden layer and an input layer, is the number h of layers of an irregular triangular net, v is the number of nodes of the layers of the irregular triangular net, sigm is a deep learning conditional probability excitation function, and wijRepresenting a weight matrix between the ith neuron of the deep learning input layer and the jth neuron of the hidden layer, ajBias of the jth neuron of the input layer, biTo hide the bias of the ith neuron of the layer, nv nhRepresenting the number of input layers and hidden layers, respectively.
The steps for verifying the performance of the sample verification model comprise:
establishing a training set S ═ x according to a multi-source heterogeneous data model of the fully mechanized mining face(1),x(2),…,x(N)And f, deep learning training is carried out: s(k)=f(x(k))=f(σ(wx(k)+ b)), wherein: n represents the total number of multi-source heterogeneous data, x represents the multi-source heterogeneous data to be trained, sigma represents a deep learning activation function, w represents a weight matrix from an input layer to a hidden layer, b represents the bias of the hidden layer, and s is a training characteristic expression of x;
using the formula: y is(k)=g(s(k))=g(σ(wTs(k)+bT) Carry out deep learning verification, where: w is aTA transposed matrix representing w, bTA verification characteristic expression which represents a transposed matrix of b and y is s;
ensuring that the minimum error value xi of the multi-source heterogeneous data fusion meets the following requirements in the deep learning training process:
has the advantages that: compared with the prior art, the invention has the following advantages:
1) the multi-source heterogeneous data fusion method for the fully-mechanized coal mining face based on the deep learning can rapidly fuse various heterogeneous data of the fully-mechanized coal mining face in the mining process, including channel wave detection data, laser scanner data, mine ground measurement data, ground penetrating radar data, three-dimensional seismic data and actual measurement production data in the mining process, and effectively improves the efficiency and accuracy of data fusion.
2) The fusion method provided by the invention can effectively fuse multi-source heterogeneous data of the fully mechanized mining face together, can effectively ensure the accuracy and reliability of the established model, and simultaneously provides accurate multi-source heterogeneous data for unmanned mining of the fully mechanized mining face.
Drawings
FIG. 1 is a flow chart of a fully mechanized coal mining face multi-source heterogeneous data fusion method based on deep learning.
FIG. 2(a) is a schematic diagram of an initial polygon definition for generating an irregular triangulation network of a multi-source heterogeneous data model by using a point-by-point interpolation method according to the present invention.
FIG. 2(b) is a schematic diagram of an initial triangulation network for generating an irregular triangulation network of a multi-source heterogeneous data model by using a point-by-point interpolation method according to the present invention.
Fig. 2(c) is a schematic diagram of the irregular triangulation network generated by the point-by-point insertion method according to the present invention after inserting a second point.
Fig. 2(d) is a schematic diagram of the irregular triangulation network generated by the point-by-point insertion method according to the present invention after a third point is inserted.
Fig. 2(e) is a schematic diagram after the iteration of generating the irregular triangulation network of the multi-source heterogeneous data model by using the point-by-point insertion method is completed.
Fig. 2(f) is a schematic diagram of an irregular triangulation network finally generated by generating an irregular triangulation network of a multi-source heterogeneous data model by using a point-by-point insertion method.
Detailed Description
The invention will be further described with reference to examples in the drawings to which:
as shown in fig. 1, the deep learning-based fully-mechanized working face multi-source heterogeneous data fusion method of the invention establishes a deep learning fusion model of fully-mechanized working face multi-source heterogeneous data according to multi-source heterogeneous data information of the fully-mechanized working face in a mining process, and specifically includes the following steps:
a, acquiring multi-source heterogeneous data information in the mining process of a fully mechanized mining face, comprising: the working face channel wave detection data, the laser scanner data, the mine ground survey data, the ground penetrating radar data, the three-dimensional seismic data to carry out optimization processing to above-mentioned all multisource heterogeneous data information, include: removing noise points by adopting a chord height difference method, carrying out smooth filtering by adopting an average filtering method, carrying out data repairing by adopting cubic spline interpolation, and establishing a multi-source heterogeneous data model of the fully mechanized coal mining face;
b, converting multi-source heterogeneous data information into a fully mechanized working face multi-source heterogeneous data model by using a Morojinsky seven-parameter coordinate conversion model, and performing unification processing on a spatial reference, data attributes and element relations in the fully mechanized working face multi-source heterogeneous data model;
the spatial reference unification processing process comprises the following steps: the obtained multi-source heterogeneous data is converted into a China geodetic coordinate system (CGCS2000) by adopting a Morojinskyqi parameter coordinate conversion model,
in the formula, XCGCS,YCGCS,ZCGCSIs the coordinate value, X, of the Chinese geodetic coordinate systemM,YM,ZMCoordinate values under a coordinate system of the multi-source heterogeneous data measurement sensor, wherein delta X, delta Y and delta Z are translation parameters between two coordinate origins, RX,RY,RZIs 3 coordinate axis rotation parameters, epsilon is a scale parameter;
the data attribute unification processing process comprises the following steps: adopting a multi-stage description method to carry out uniform processing on the attributes of the multi-source heterogeneous data, and setting the data attributes of the heterogeneous data A and B as follows: (mu.) aA,sA) And (mu)B,sB) Judging whether the data attributes of A and B satisfy the following formula:
where sim () is a data attribute consistency function, μ and s represent coordinate similarity and direction similarity, respectively, N is the total number of heterogeneous data, ω is1And ω2Is a weight parameter;
the element relationship unification processing process comprises the following steps: adopting a 9-intersection model and calculating the following topological relation equation of heterogeneous data A and B:
wherein R () is a data topological relation function, A0,And A-Respectively representing the internal topological relation, the boundary topological relation and the external topological relation of the data A, B0,And B-Respectively representing the internal topological relation, the boundary topological relation and the external topological relation of the data B;
as shown in fig. 2(a), an irregular triangulation network of a multi-source heterogeneous data model of the fully mechanized coal mining face is generated by using a point-by-point interpolation method, and the specific steps are as follows:
c1 defining an initial polygon containing all of the fully mechanized face multisource heterogeneous data point sets as shown in FIG. 2 (b);
c2 randomly selecting a point from the data point set defined by c1, inserting the point into the gravity center point of the initial polygon to establish an initial triangulation network;
c3, performing iterative computation until all data points are processed, as shown in fig. 2(c), inserting a second point, and as shown in fig. 2(d), inserting a third point, i.e., inserting a discrete sampling point inside the initial triangulation network, finding out the triangle containing the point in the initial triangulation network, and connecting the point with three vertices of the triangle to generate three new small triangles; then, the triangle network is optimized from inside to outside by using LOP (local Optimization procedure) algorithm to ensure that the new small triangle meets the requirement of the irregular triangle, wherein
c4 deleting the triangles containing the initial polygon points to form the final irregular triangulation network, as shown in FIG. 2 (f);
d, extracting characteristic parameters of the irregular triangular net, including the number of layers, the number of nodes and the training algebra of the triangular net, verifying whether the iteration condition of the deep learning algorithm is met, namely the number of layers, the number of nodes and the absolute value of the training algebra of the triangular net are linearly converged, and if the iteration condition is not met, realizing that the multi-source heterogeneous data model of the fully mechanized mining face meets the iteration condition requirement of the deep learning algorithm by adjusting the number of layers, the number of nodes and the size of the training algebra of the triangular net;
the iteration condition of the deep learning algorithm is to satisfy the following equation:
in the formula, P represents a combined probability distribution function of a deep learning hidden layer and an input layer, is the number h of layers of an irregular triangular net, v is the number of nodes of the layers of the irregular triangular net, sigm is a deep learning conditional probability excitation function, and wijRepresenting a weight matrix between the ith neuron of the deep learning input layer and the jth neuron of the hidden layer, ajBias of the jth neuron of the input layer, biTo hide the bias of the ith neuron of the layer, nv nhRespectively representing the number of input layers and hidden layers, and as shown in fig. 2(e), the schematic diagram of the triangulation network after iteration is completed;
e, if the iteration condition of the deep learning algorithm is met, performing feature training by using the deep learning training sample, and then verifying the multi-source heterogeneous data model performance of the fully mechanized mining face by using the deep learning verification sample, wherein the specific steps are as follows:
establishing a training set S ═ x according to a multi-source heterogeneous data model of the fully mechanized mining face(1),x(2),…,x(N)Get deep learning trainingRefining: s(k)=f(x(k))=f(σ(wx(k)+ b)), wherein: n represents the total number of multi-source heterogeneous data, x represents the multi-source heterogeneous data to be trained, sigma represents a deep learning activation function, w represents a weight matrix from an input layer to a hidden layer, b represents the bias of the hidden layer, and s is a training characteristic expression of x;
using the formula: y is(k)=g(s(k))=g(σ(wTs(k)+bT) Carry out deep learning verification, where: w is aTA transposed matrix representing w, bTA verification characteristic expression which represents a transposed matrix of b and y is s;
ensuring that the minimum error value xi of the multi-source heterogeneous data fusion meets the following requirements in the deep learning training process:
f, completing feature training on the deep learning training samples and completing model performance verification on the deep learning verification samples in the step e, and establishing a deep learning fusion model of the multi-source heterogeneous data of the fully mechanized mining face by using a general convolution neural network method.
Claims (4)
1. A fully mechanized mining face multi-source heterogeneous data fusion method based on deep learning is characterized in that: according to multi-source heterogeneous data information of the fully mechanized mining face in the mining process, a deep learning fusion model of the multi-source heterogeneous data of the fully mechanized mining face is established; the method comprises the following specific steps:
a, acquiring multi-source heterogeneous data information in the mining process of a fully mechanized mining face, comprising: the working face channel wave detection data, the laser scanner data, the mine ground survey data, the ground penetrating radar data, the three-dimensional seismic data to carry out optimization processing to above-mentioned all multisource heterogeneous data information, include: removing noise points by adopting a chord height difference method, carrying out smooth filtering by adopting an average filtering method, carrying out data repairing by adopting cubic spline interpolation, and establishing a multi-source heterogeneous data model of the fully mechanized coal mining face;
b, converting multi-source heterogeneous data information into a fully mechanized working face multi-source heterogeneous data model by using a Morojinsky seven-parameter coordinate conversion model, and performing unification processing on a spatial reference, data attributes and element relations in the fully mechanized working face multi-source heterogeneous data model;
c, generating an irregular triangular net of the multi-source heterogeneous data model of the fully mechanized coal mining face by using a point-by-point insertion method;
d, extracting characteristic parameters of the irregular triangular net, wherein the characteristic parameters comprise the number of layers, the number of nodes and a training algebra of the triangular net, checking the number of layers, the number of nodes and the absolute value linear convergence of the training algebra of the triangular net so as to judge whether the iteration condition of the deep learning algorithm is met, and if the iteration condition is not met, realizing that the multisource heterogeneous data model of the fully mechanized mining face meets the iteration condition requirement of the deep learning algorithm by adjusting the number of layers, the number of nodes and the size of the training algebra;
e, if the iteration condition of the deep learning algorithm is met, performing feature training by using the deep learning training sample, then verifying the multi-source heterogeneous data model performance of the fully mechanized mining face by using the deep learning verification sample,
f, completing feature training on the deep learning training samples and completing model performance verification on the deep learning verification samples in the step e, and establishing a deep learning fusion model of the multi-source heterogeneous data of the fully mechanized mining face by using a general convolutional neural network method;
the specific steps of the irregular triangulation network for generating the multi-source heterogeneous data model of the fully mechanized coal mining face by using the point-by-point insertion method are as follows:
c1 defining an initial polygon containing all the fully mechanized face multi-source heterogeneous data point sets;
c2 randomly selecting a point from the multi-source heterogeneous data point set of the fully mechanized mining face, inserting the point into the gravity center point of the initial polygon to establish an initial triangulation network;
c3, performing iterative computation until all data points are processed, namely, inserting a discrete sampling point in the initial triangulation network, finding out a triangle containing the point in the initial triangulation network, and connecting the point with three vertexes of the triangle to generate three new small triangles; then, optimizing the triangular net from inside to outside by using an LOP (local Optimization procedure) algorithm to ensure that the new small triangle meets the requirement of the irregular triangle;
c4 deletes the triangle containing the initial polygon points to form the final irregular triangulation network.
2. The deep learning-based multi-source heterogeneous data fusion method for the fully mechanized mining face according to claim 1, wherein the spatial reference unification process comprises: the obtained multi-source heterogeneous data is converted into a China geodetic coordinate system (CGCS2000) by adopting a Morojinskyqi parameter coordinate conversion model,
in the formula, XCGCS,YCGCS,ZCGCSIs the coordinate value, X, of the Chinese geodetic coordinate systemM,YM,ZMCoordinate values under a coordinate system of the multi-source heterogeneous data measurement sensor, wherein delta X, delta Y and delta Z are translation parameters between two coordinate origins, RX,RY,RZIs 3 coordinate axis rotation parameters, epsilon is a scale parameter;
the data attribute unification processing process comprises the following steps: adopting a multi-stage description method to carry out uniform processing on the attributes of the multi-source heterogeneous data, and setting the data attributes of the heterogeneous data A and B as follows: (mu.) aA,sA) And (mu)B,sB) Judging whether the data attributes of A and B satisfy the following formula:
where sim () is a data attribute consistency function, μ and s represent coordinate similarity and direction similarity, respectively, N is the total number of heterogeneous data, ω is1And ω2Is a weight parameter;
the element relationship unification processing process comprises the following steps: adopting a 9-intersection model and calculating the following topological relation equation of heterogeneous data A and B:
wherein R () is a data topological relation function, A0,And A-Respectively representing the internal topological relation, the boundary topological relation and the external topological relation of the data A, B0,And B-And respectively representing the internal topological relation, the boundary topological relation and the external topological relation of the data B.
3. The deep learning-based fully-mechanized working face multi-source heterogeneous data fusion method according to claim 1, wherein the iteration condition of the deep learning algorithm in step d satisfies the following equation:
in the formula, P represents a combined probability distribution function of a deep learning hidden layer and an input layer, is the number h of layers of an irregular triangular net, v is the number of nodes of the layers of the irregular triangular net, sigm is a deep learning conditional probability excitation function, and wijRepresenting a weight matrix between the ith neuron of the deep learning input layer and the jth neuron of the hidden layer, ajBias of the jth neuron of the input layer, biTo hide the bias of the ith neuron of the layer, nv nhRepresenting the number of input layers and hidden layers, respectively.
4. The deep learning-based fully mechanized mining face multi-source heterogeneous data fusion method according to claim 1, characterized in that the step of verifying the performance of the sample verification model is:
establishing a training set S ═ x according to a multi-source heterogeneous data model of the fully mechanized mining face(1),x(2),…,x(N)And f, deep learning training is carried out: s(k)=f(x(k))=f(σ(wx(k)+ b)), wherein: n represents the total number of multi-source heterogeneous data, x represents the multi-source heterogeneous data to be trained, sigma represents a deep learning activation function, w represents a weight matrix from an input layer to a hidden layer, b represents the bias of the hidden layer, and s is a training characteristic expression of x;
using the formula: y is(k)=g(s(k))=g(σ(wTs(k)+bT) Carry out deep learning verification, where: w is aTA transposed matrix representing w, bTA verification characteristic expression which represents a transposed matrix of b and y is s;
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910397221.XA CN110109995B (en) | 2019-05-14 | 2019-05-14 | Fully mechanized mining face multi-source heterogeneous data fusion method based on deep learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910397221.XA CN110109995B (en) | 2019-05-14 | 2019-05-14 | Fully mechanized mining face multi-source heterogeneous data fusion method based on deep learning |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110109995A CN110109995A (en) | 2019-08-09 |
CN110109995B true CN110109995B (en) | 2021-12-17 |
Family
ID=67489853
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910397221.XA Active CN110109995B (en) | 2019-05-14 | 2019-05-14 | Fully mechanized mining face multi-source heterogeneous data fusion method based on deep learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110109995B (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111736234B (en) * | 2020-07-27 | 2021-06-11 | 中国矿业大学 | Coal thickness prediction method based on ground penetrating radar and earthquake transmission channel waves |
CN112329302B (en) * | 2020-11-02 | 2021-06-15 | 中煤科工集团重庆研究院有限公司 | Simulation and three-dimensional visualization analysis system for underground dust production environment |
CN113746855A (en) * | 2021-09-09 | 2021-12-03 | 国网电子商务有限公司 | Data access method of energy industry cloud network and related equipment |
CN115774861B (en) * | 2022-12-22 | 2023-07-21 | 广东五度空间科技有限公司 | Natural resource multi-source heterogeneous data convergence fusion service system |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101906965B (en) * | 2010-07-14 | 2011-12-21 | 中国矿业大学(北京) | Technology for identifying and analyzing stratums in VR (Virtual Reality) geological environment |
CN102609986B (en) * | 2012-03-14 | 2014-05-14 | 中国矿业大学(北京) | Ore body modeling system in digital mine and modeling and detecting method of ore body modeling system |
US11120314B2 (en) * | 2017-04-07 | 2021-09-14 | Intel Corporation | Joint training of neural networks using multi scale hard example mining |
CN108647607A (en) * | 2018-04-28 | 2018-10-12 | 国网湖南省电力有限公司 | Objects recognition method for project of transmitting and converting electricity |
CN109063753B (en) * | 2018-07-18 | 2021-09-14 | 北方民族大学 | Three-dimensional point cloud model classification method based on convolutional neural network |
CN109547032B (en) * | 2018-10-12 | 2020-06-19 | 华南理工大学 | Confidence propagation LDPC decoding method based on deep learning |
CN109543828B (en) * | 2018-12-28 | 2020-10-23 | 中国石油大学(华东) | Water absorption profile prediction method based on small sample condition |
-
2019
- 2019-05-14 CN CN201910397221.XA patent/CN110109995B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN110109995A (en) | 2019-08-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110109995B (en) | Fully mechanized mining face multi-source heterogeneous data fusion method based on deep learning | |
Liu et al. | Construction of iso-contours, bisectors, and Voronoi diagrams on triangulated surfaces | |
AU2013317709B2 (en) | On-ground or near-ground discrete object detection method and system | |
CN112784403A (en) | Numerical simulation method for establishing jointed rock mass discrete element model based on point cloud data | |
CN112446844B (en) | Point cloud feature extraction and registration fusion method | |
CN112949407A (en) | Remote sensing image building vectorization method based on deep learning and point set optimization | |
CN111028238A (en) | Robot vision-based three-dimensional segmentation method and system for complex special-shaped curved surface | |
CN114693932A (en) | Large aircraft large component point cloud semantic segmentation method | |
CN114170446A (en) | Temperature and brightness characteristic extraction method based on deep fusion neural network | |
CN113989631A (en) | Infrared image target detection network compression method based on convolutional neural network | |
CN113838005A (en) | Intelligent rock fracture identification and three-dimensional reconstruction method and system based on dimension conversion | |
CN114474004B (en) | Error compensation planning control strategy for multi-factor coupling vehicle-mounted building robot | |
CN112634447B (en) | Outcrop stratum layering method, device, equipment and storage medium | |
CN115546116A (en) | Method and system for extracting and calculating spacing of discontinuous surface of fully-covered rock mass | |
CN115619953A (en) | Rugged terrain-oriented mobile robot terrain mapping method and system | |
De Geyter et al. | Automated training data creation for semantic segmentation of 3D point clouds | |
CN111445569B (en) | Sedimentary geological evolution dynamic simulation method | |
Yotsumata et al. | Quality improvement for airborne lidar data filtering based on deep learning method | |
CN112069445A (en) | 2D SLAM algorithm evaluation and quantification method | |
CN111260783B (en) | Ore body three-dimensional automatic modeling method based on K neighbor and Poisson curved surface | |
CN116049941B (en) | Method for extracting and analyzing multidimensional state of assembled ring truss structural member before assembly | |
CN115903814B (en) | Multi-robot optimal formation path planning method based on convex polygon tree | |
Ovchinnikov et al. | Creating 3D Models Using Segmented Point Clouds | |
CN111382799B (en) | Earthquake fault image processing method | |
CN117252863B (en) | Quick detection and analysis method for geographic information abnormal data |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |