CN111505705B - Microseism P wave first arrival pickup method and system based on capsule neural network - Google Patents
Microseism P wave first arrival pickup method and system based on capsule neural network Download PDFInfo
- Publication number
- CN111505705B CN111505705B CN202010057384.6A CN202010057384A CN111505705B CN 111505705 B CN111505705 B CN 111505705B CN 202010057384 A CN202010057384 A CN 202010057384A CN 111505705 B CN111505705 B CN 111505705B
- Authority
- CN
- China
- Prior art keywords
- micro
- training
- data
- arrival
- capsule
- 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
- 239000002775 capsule Substances 0.000 title claims abstract description 52
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 37
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000012549 training Methods 0.000 claims abstract description 86
- 238000001514 detection method Methods 0.000 claims abstract description 15
- 238000012545 processing Methods 0.000 claims abstract description 9
- 230000006870 function Effects 0.000 claims description 24
- 210000002569 neuron Anatomy 0.000 claims description 5
- 239000000284 extract Substances 0.000 claims description 3
- 238000000605 extraction Methods 0.000 abstract description 8
- 238000011156 evaluation Methods 0.000 abstract description 3
- 239000000523 sample Substances 0.000 description 10
- 238000010586 diagram Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- 230000004913 activation Effects 0.000 description 1
- 230000003416 augmentation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000005251 gamma ray Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000644 propagated effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000008685 targeting Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 230000004580 weight loss Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V1/00—Seismology; Seismic or acoustic prospecting or detecting
- G01V1/28—Processing seismic data, e.g. for interpretation or for event detection
- G01V1/288—Event detection in seismic signals, e.g. microseismics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V2210/00—Details of seismic processing or analysis
- G01V2210/60—Analysis
- G01V2210/67—Wave propagation modeling
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Biophysics (AREA)
- Health & Medical Sciences (AREA)
- Remote Sensing (AREA)
- Environmental & Geological Engineering (AREA)
- Emergency Management (AREA)
- Business, Economics & Management (AREA)
- Acoustics & Sound (AREA)
- Geology (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Geophysics (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention relates to the technical field of microseism data processing, in particular to a microseism P wave first-arrival picking method and system based on a capsule neural network; the method includes preparing a raw data set; making a data training set; selecting first arrival points of a part of sample signals of the original data set to be labeled to serve as labeled parts, and selecting the other part of sample signals to serve as unlabeled parts; inputting the data training set into a combined training model, and performing prediction evaluation on the micro seismic signal characteristics; performing target detection on the micro-seismic signal characteristics to obtain a primary arrival point of the micro-seismic signal; the system comprises a data acquisition module, a data training set making module, a data training module and an output module; according to the embodiment of the invention, the capsule neural network and the semi-supervised learning are combined, the RPN is used for detecting the micro-seismic signals, the first arrival point pickup of the micro-seismic signals is realized, the feature extraction accuracy of the micro-seismic signals is improved, and the accurate pickup of the P wave first arrival point is realized.
Description
Technical Field
The invention relates to the technical field of microseism data processing, in particular to a microseism P wave first-arrival picking method and system based on a capsule neural network.
Background
The method has important significance for stable and high yield of oil field development by effectively detecting the micro seismic signals. The effective signal energy of the microseism is weak, the signal-to-noise ratio is low, and the microseism is even completely submerged in noise. Although the conventional seismic data processing methods are numerous, if the conventional seismic data processing methods are directly applied to microseism data, a satisfactory effect cannot be obtained, and the quality and the accuracy of microseism monitoring are directly influenced. Therefore, finding an appropriate method to identify weak valid signals in the microseismic data is the key to microseismic data processing and interpretation.
The extraction method of the first arrival of the microseism P wave based on the AB algorithm combined with the Allen algorithm and the Bear algorithm is proposed in a paper of 'a method from the position of the microseism P wave to the pickup' in 2018. According to the method, on the basis of analyzing the characteristics of the micro-seismic signals, the time window average parameters and characteristic functions in the original energy ratio method are partially modified, and some new mathematical calculation methods are used in the energy ratio method, so that the judgment capability and the pickup effect of the algorithm on the micro-seismic signals with low signal-to-noise ratio are improved to a certain extent.
The method has the defects that although the signal identification rate is improved to a certain extent, the extraction accuracy of the micro-seismic signal features is not high, so that the P wave first arrival point pickup accuracy is only 73.5%.
Disclosure of Invention
In order to overcome the defects in the prior art, the embodiment of the invention provides a method and a system for picking up the first-arrival of the microseism P-wave based on a capsule neural network, and the accuracy rate of picking up the first-arrival points is further improved.
On one hand, the embodiment of the invention provides a microseism P wave first arrival picking method based on a capsule neural network, which comprises the following steps:
s1, preparing an original data set; adding random Gaussian noise into all the original data sets to enhance data set processing; all the original data sets comprise microseism forward signals with different main frequencies and different signal-to-noise ratios and actual seismic records;
s2, making a data training set; selecting first arrival points of a part of sample signals of the original data set to be labeled to serve as labeled parts, and selecting the other part of sample signals to serve as unlabeled parts;
s3, inputting the data training set into a combined training model, and predicting and evaluating the micro seismic signal characteristics; the combined training model consists of a capsule neural network and a semi-supervised learning model; performing supervised training on the labeled part, and performing unsupervised training on the unlabeled part;
s4, performing target detection on the micro seismic signal characteristics to obtain a first arrival point of the micro seismic signal; specifically, anchor points with different scales are set through an RPN network rule, candidate frames are extracted from a convolution characteristic layer, and end-to-end training is carried out.
On the other hand, the embodiment of the invention provides a micro-seismic P-wave first arrival pickup system based on a capsule neural network, which comprises the following components:
a data acquisition module for preparing an original data set; adding random Gaussian noise into all the original data sets to enhance data set processing; all the original data sets comprise microseism forward signals with different main frequencies and different signal-to-noise ratios and actual seismic records;
the data training set making module is used for making a data training set; selecting first arrival points of a part of sample signals of the original data set to be labeled to serve as labeled parts, and selecting the other part of sample signals to serve as unlabeled parts;
the data training module inputs the data training set into a combined training model and predicts and evaluates the micro-seismic signal characteristics; the combined training model consists of a capsule neural network and a semi-supervised learning model; performing supervised training on the labeled part, and performing unsupervised training on the unlabeled part;
the output module is used for carrying out target detection on the micro-seismic signal characteristics to obtain a first arrival point of the micro-seismic signal; specifically, anchor points with different scales are set through an RPN network rule, candidate frames are extracted from a convolution characteristic layer, and end-to-end training is carried out.
The embodiment of the invention provides a method and a system for picking up a microseism P wave first arrival based on a capsule neural network; through the combination of a capsule neural network and semi-supervised learning, the RPN is used for micro-seismic signal detection, and the first arrival point pickup of micro-seismic signals is realized; the micro-seismic signal feature extraction accuracy and the P wave first arrival point accurate pickup are improved.
Drawings
In order to more clearly illustrate the technical solution of the present invention, the drawings needed to be used in the technical description of the present invention will be briefly introduced below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without inventive labor.
FIG. 1 is a schematic flow chart of a microseism P-wave first-break pickup method based on a capsule neural network according to an embodiment of the present invention;
FIG. 2 is a sub-flow diagram of a microseism P-wave first-arrival picking method based on a capsule neural network according to an embodiment of the invention
FIG. 3 is a flowchart of a training method of a combination training model according to an embodiment of the present invention;
FIG. 4 is a graph of training results of the combined training model according to the embodiment of the present invention;
FIG. 5 is a flow chart of the RPN network target detection according to the present invention;
FIG. 6 shows the first arrival picking-up result of micro seismic signals according to an embodiment of the present invention;
FIG. 7 is a schematic structural diagram of a microseism P-wave first-arrival pickup system based on a capsule neural network according to an embodiment of the present invention;
reference numerals:
data acquisition module-1 data training set making module-2 data training module-3 output module-4
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without inventive step based on the embodiments of the present invention, are within the scope of protection of the present invention.
FIG. 1 is a schematic flow chart of a microseism P-wave first-break pickup method based on a capsule neural network according to an embodiment of the present invention; as shown in fig. 1, the method comprises the following steps:
s1, preparing an original data set; adding random Gaussian noise into all the original data sets to enhance data set processing; all the original data sets comprise microseism forward signals with different main frequencies and different signal-to-noise ratios and actual seismic records;
s2, making a data training set; selecting first arrival points of a part of sample signals of the original data set to be labeled to serve as labeled parts, and selecting the other part of sample signals to serve as unlabeled parts;
s3, inputting the data training set into a combined training model, and predicting and evaluating the micro seismic signal characteristics; the combined training model consists of a capsule neural network and a semi-supervised learning model; performing supervised training on the labeled part, and performing unsupervised training on the unlabeled part;
s4, performing target detection on the micro seismic signal characteristics to obtain a first arrival point of the micro seismic signal; specifically, anchor points with different scales are set through an RPN network rule, candidate frames are extracted from a convolution characteristic layer, and end-to-end training is carried out.
Specifically, a Capsule neural network (Capsule-net) and a semi-supervised learning (Temporal engineering) model form a combined training model, and micro seismic signal data samples form a training data set and are added into the combined training model for signal feature extraction; the capsule neural network is beneficial to extracting the characteristics of signals, makes up for the loss of a large amount of information when the system outputs, and effectively ensures the retention of characteristic information due to the isodegeneration of the capsule neural network; the semi-supervised learning can effectively avoid the over-fitting phenomenon in the data training process, reduce the number of labels of a training data set, improve the quality of the training data set and accurately extract the micro-seismic signal characteristics;
fig. 5 is a flow chart of target detection of an RPN network according to an embodiment of the present invention, and as shown in fig. 5, target detection is performed on the micro-seismic signal characteristics to obtain a first arrival point of the micro-seismic signal; anchor points with different scales are set through an RPN network rule to extract candidate frames in a convolution characteristic layer, and end-to-end training is carried out; and (3) the digital capsule pair signals in a vector neuron layer of the capsule neural network are used for generating region probes, the layer judges whether the anchor point belongs to the foreground or the background through softmax, and then the boundary box regression is utilized to correct the anchor point to obtain accurate probes. And collecting the characteristic spectrogram and the proposals to form an ROI layer, and enabling the ROI layer and the proposals to enter two parallel full-connection layers after passing through the full-connection layer, wherein one is the classification of the region, and the other is the regression of the frame. In the process of regression and classification, a learning parameter x, y respectively represents the abscissa and the ordinate h of the midpoint of the predefined frame, and w respectively represents the height and the width of the predefined frame. In the RPN network, the objective classification function:
objective score function:
classification loss function:
target detection loss function:
Loss＝L bbox +L score +L class
in the above formula, N obj Is the number of targets in the frame, λ bbox Is the weight loss of the part, K is the number of predefined frames, I k ∈{0,1}，I k Is the predefined box with the largest intersection ratio of the target boxes, X is composed of a four-dimensional vector (delta) x ,δ y ,δ w ,δ h ) The amount of correction of the constituent predefined boxes,
and representing the intersection ratio correction quantity of the predefined frame and the target regression frame. Gamma ray k A predetermined frame is represented that is a predetermined frame,
representing the intersection ratio of the pre-defined box and the target regression box.
P after softmax normalization in RNP networks c ∈[0,1]. lc 1 indicates that this predetermined box belongs to class c. FIG. 6 shows the first arrival picking-up result of micro seismic signals according to an embodiment of the present invention; as shown in FIG. 6, candidate boxes are extracted at the convolution feature layerThereby effectively detecting the initial point of the micro seismic signal.
The embodiment of the invention provides a microseism P wave first arrival picking method based on a capsule neural network; through the combination of a capsule neural network and semi-supervised learning, the RPN is used for micro-seismic signal detection, and the first arrival point pickup of micro-seismic signals is realized; the micro-seismic signal feature extraction accuracy and the P wave first arrival point accurate pickup are improved.
Further, fig. 2 is a sub-flow diagram of a microseism P-wave first arrival pickup method based on a capsule neural network according to an embodiment of the present invention; as shown in fig. 2, the step S3 specifically includes:
s31, inputting the data training set, performing convolution operation on all microseismic signals to obtain a convolution layer, and performing convolution operation on the convolution layer to obtain a vector neuron layer original capsule;
s32, transmitting the original capsule layer to a digital capsule layer by adopting a dynamic routing algorithm to obtain the micro-seismic signal characteristic vector;
s33, constructing a total loss function by weighting and summing the unsupervised loss function and the supervised loss function of the micro seismic signals through semi-supervised learning, and acquiring the micro seismic signal characteristics by minimizing the total loss function.
Specifically, fig. 3 is a flowchart of a training method of a combined training model according to an embodiment of the present invention; as shown in FIG. 3, the input is a processed microseismic signal x i . Performing convolution operation on the section of signal to obtain a convolution layer, and performing convolution operation on the convolution layer to obtain a vector neuron layer original capsule; the original capsule layer is propagated to the digital capsule layer by adopting a dynamic routing algorithm, and the specific content of the dynamic routing algorithm is not described herein; the operation parameter formula of the capsule neural network is as follows:
c is the coupling coefficient, S is the input of the capsule, u is the output of the previous layer of the capsule network, and W is the weight to be multiplied by each output. b is initialized to be 0, so that in the process of solving S by forward propagation, W is designed to be a random value, C can be obtained by initializing b to be 0, and S of the next layer can be obtained. Reducing calculation cost by using an activation function Squashing to obtain an extracted feature vector V j (ii) a The formula is as follows:
constructing a total loss function by weighting and summing the unsupervised loss function and the supervised loss function of the micro seismic signals through semi-supervised learning, and acquiring micro seismic signal characteristics through minimizing the total loss function; the total loss function is:
we can divide the evaluation results of the two branches into two different phases: the training set is first classified and then the network is subjected to different augmentation and loss training with the same input. For tagged data set y i Supervised training is performed, and then prediction evaluation is obtained by performing unsupervised training on the unlabeled part in the data set by using the prediction obtained by the just supervised training as the target of the unsupervised loss component.
After each training epoch, the feature vector V is extracted by updating j ←aV j +(1-a)v j Output the network v j Accumulated to the output V j Wherein a is setAnd (4) a resultant momentum term. Due to dropout regularization and random noise addition in the network, V contains a weighted average of the network aggregate output from previous training epochs, while recent epochs are weighted more heavily than distant epochs. To generate a training targeting V, we need to correct for start-up bias in V by dividing by a factor (1-at), resulting in
In this process, we can assign the unsupervised weight function w (t) for the first training period to be zero.
Training data for a total of N inputs, where x is represented for all training data i Where i e {1.. N }, the set of training indices containing labels is | L | ═ M, and for each i e L, there is a known correct label y i E {1.. C }, where C is the number of different classes and B is the mini-batch index set. FIG. 4 is a graph of training results of the combined training model according to the embodiment of the present invention; as shown in FIG. 4, the combined model extracts the micro-seismic signal characteristic curve graph, and the accuracy can reach 93%.
The embodiment of the invention provides a microseism P wave first arrival picking method based on a capsule neural network; through the combination of a capsule neural network and semi-supervised learning, the RPN is used for micro-seismic signal detection, and the first arrival point pickup of micro-seismic signals is realized; the micro-seismic signal feature extraction accuracy and the P wave first arrival point accurate pickup are improved.
Based on the above embodiments, fig. 7 is a schematic structural diagram of a microseismic P-wave first arrival pickup system based on a capsule neural network according to an embodiment of the present invention; as shown in fig. 7, includes:
a data acquisition module 1 for preparing an original data set; adding random Gaussian noise into all the original data sets to enhance data set processing; all the original data sets comprise microseism forward signals with different main frequencies and different signal-to-noise ratios and actual seismic records;
a data training set making module 2 for making a data training set; selecting first arrival points of a part of sample signals of the original data set to be labeled to serve as labeled parts, and selecting the other part of sample signals to serve as unlabeled parts;
the data training module 3 inputs the data training set into a combined training model to predict and evaluate the micro-seismic signal characteristics; the combined training model consists of a capsule neural network and a semi-supervised learning model; performing supervised training on the labeled part, and performing unsupervised training on the unlabeled part;
the output module 4 is used for carrying out target detection on the micro seismic signal characteristics to obtain a first arrival point of the micro seismic signal; specifically, anchor points with different scales are set through an RPN network rule, candidate frames are extracted from a convolution characteristic layer, and end-to-end training is carried out.
The embodiment of the invention provides a microseism P wave first-arrival picking system based on a capsule neural network, which executes the method; through the combination of a capsule neural network and semi-supervised learning, the RPN is used for micro-seismic signal detection, and the first arrival point pickup of micro-seismic signals is realized; the micro-seismic signal feature extraction accuracy and the P wave first arrival point accurate pickup are improved.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. Based on the understanding, the above technical solutions substantially or otherwise contributing to the prior art may be embodied in the form of a software product, which may be stored in a computer-readable storage medium, such as ROM/RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method according to the various embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (4)
1. A microseism P wave first arrival picking method based on a capsule neural network is characterized by comprising the following steps:
s1, preparing an original data set; adding random Gaussian noise into all the original data sets to enhance data set processing; all the original data sets comprise microseism forward signals with different main frequencies and different signal-to-noise ratios and actual seismic records;
s2, making a data training set; selecting first arrival points of a part of sample signals of the original data set to be labeled to serve as labeled parts, and selecting the other part of sample signals to serve as unlabeled parts;
s3, inputting the data training set into a combined training model, and extracting micro seismic signal characteristics; the combined training model consists of a capsule neural network and a semi-supervised learning model; performing supervised training on the labeled part, and performing unsupervised training on the unlabeled part;
s4, carrying out target detection on the micro seismic signal characteristics to obtain a first arrival point of the micro seismic signal; specifically, anchor points with different scales are set through an RPN network rule, candidate frames are extracted from a convolution characteristic layer, and end-to-end training is carried out.
2. The method for picking up the first arrival of the microseism P wave based on the capsule neural network as claimed in claim 1, wherein the step S3 specifically comprises:
s31, inputting the data training set, performing convolution operation on all microseismic signals to obtain a convolution layer, and performing convolution operation on the convolution layer to obtain a vector neuron layer original capsule;
s32, transmitting the original capsule layer to a digital capsule layer by adopting a dynamic routing algorithm to obtain the micro-seismic signal characteristic vector;
s33, constructing a total loss function by weighting and summing the unsupervised loss function and the supervised loss function of the micro seismic signals through semi-supervised learning, and acquiring the micro seismic signal characteristics by minimizing the total loss function.
3. A microseism P wave first arrival picking system based on a capsule neural network is characterized by comprising:
a data acquisition module for preparing an original data set; adding random Gaussian noise into all the original data sets to enhance data set processing; all the original data sets comprise microseism forward signals with different main frequencies and different signal-to-noise ratios and actual seismic records;
the data training set making module is used for making a data training set; selecting first arrival points of a part of sample signals of the original data set to be labeled to serve as labeled parts, and selecting the other part of sample signals to serve as unlabeled parts;
the data training module inputs the data training set into a combined training model and extracts the micro-seismic signal characteristics; the combined training model consists of a capsule neural network and a semi-supervised learning model; performing supervised training on the labeled part, and performing unsupervised training on the unlabeled part;
the output module is used for carrying out target detection on the micro-seismic signal characteristics to obtain a first arrival point of the micro-seismic signal; specifically, anchor points with different scales are set through an RPN network rule, candidate frames are extracted from a convolution characteristic layer, and end-to-end training is carried out.
4. The system for picking up P-wave first arrival of micro-earthquakes based on capsule neural network as claimed in claim 3, wherein the data training module comprises:
inputting the data training set, performing convolution operation on all microseismic signals to obtain a convolution layer, and performing convolution operation on the convolution layer to obtain a vector neuron layer original capsule;
transmitting the original capsule layer to a digital capsule layer by adopting a dynamic routing algorithm to obtain the micro-seismic signal characteristic vector;
and constructing a total loss function by weighting and summing the unsupervised loss function and the supervised loss function of the micro seismic signals through semi-supervised learning, and acquiring the micro seismic signal characteristics by minimizing the total loss function.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010057384.6A CN111505705B (en) | 2020-01-19 | 2020-01-19 | Microseism P wave first arrival pickup method and system based on capsule neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010057384.6A CN111505705B (en) | 2020-01-19 | 2020-01-19 | Microseism P wave first arrival pickup method and system based on capsule neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111505705A CN111505705A (en) | 2020-08-07 |
| CN111505705B true CN111505705B (en) | 2022-08-02 |
Family
ID=71871026
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010057384.6A Active CN111505705B (en) | 2020-01-19 | 2020-01-19 | Microseism P wave first arrival pickup method and system based on capsule neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111505705B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112257726B (en) * | 2020-10-29 | 2023-12-15 | 上海明略人工智能(集团)有限公司 | Target detection training method, system, electronic equipment and computer readable storage medium |
| CN113109869A (en) * | 2021-03-30 | 2021-07-13 | 成都理工大学 | Automatic picking method for first arrival of shale ultrasonic test waveform |
| CN113300788B (en) * | 2021-04-19 | 2023-04-21 | 嘉兴学院 | Blind receiver method and device based on Capsule network |
| CN113671564B (en) * | 2021-08-11 | 2023-07-21 | 西安石油大学 | NARX dynamic neural network-based microseism effective event automatic pickup method |
| CN118112647A (en) * | 2024-02-20 | 2024-05-31 | 中国地质科学院地质力学研究所 | Microseism P-wave first arrival pickup method based on mixed depth neural network |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2002010797A1 (en) * | 2000-07-31 | 2002-02-07 | Scientific Prediction, Inc. | System and method for optimizing a mining operation |
| CN107807387A (en) * | 2017-10-31 | 2018-03-16 | 中国科学技术大学 | Acquisition methods when seismic first break based on neutral net is walked |
| CN109063687A (en) * | 2018-08-29 | 2018-12-21 | 长江大学 | A kind of microseism P wave recognition methods and system based on depth convolutional neural networks |
| CN109613610A (en) * | 2019-01-17 | 2019-04-12 | 中南大学 | The automatic pick method of microseismic signals arrival time difference |
| CN110082821A (en) * | 2019-03-26 | 2019-08-02 | 长江大学 | A kind of no label frame microseism signal detecting method and device |
| CN110210296A (en) * | 2019-04-25 | 2019-09-06 | 长江大学 | A kind of microseism useful signal detection method of combination U-net network and DenseNet network |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140334260A1 (en) * | 2013-05-09 | 2014-11-13 | Schlumberger Technology Corporation | Neural Network Signal Processing of Microseismic Events |
| US11105942B2 (en) * | 2018-03-27 | 2021-08-31 | Schlumberger Technology Corporation | Generative adversarial network seismic data processor |
-
2020
- 2020-01-19 CN CN202010057384.6A patent/CN111505705B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2002010797A1 (en) * | 2000-07-31 | 2002-02-07 | Scientific Prediction, Inc. | System and method for optimizing a mining operation |
| CN107807387A (en) * | 2017-10-31 | 2018-03-16 | 中国科学技术大学 | Acquisition methods when seismic first break based on neutral net is walked |
| CN109063687A (en) * | 2018-08-29 | 2018-12-21 | 长江大学 | A kind of microseism P wave recognition methods and system based on depth convolutional neural networks |
| CN109613610A (en) * | 2019-01-17 | 2019-04-12 | 中南大学 | The automatic pick method of microseismic signals arrival time difference |
| CN110082821A (en) * | 2019-03-26 | 2019-08-02 | 长江大学 | A kind of no label frame microseism signal detecting method and device |
| CN110210296A (en) * | 2019-04-25 | 2019-09-06 | 长江大学 | A kind of microseism useful signal detection method of combination U-net network and DenseNet network |
Non-Patent Citations (2)
| Title |
|---|
| Research Developments and Prospects on Microseismic Source Location in Mines;Jiulong Cheng 等;《Engineering》;20180823;第4卷;第653-660页 * |
| 改进的神经网络级联相关算法及其在初至拾取中的应用;宋建国 等;《石油地球物理勘探》;20180228;第53卷(第1期);第8-16页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111505705A (en) | 2020-08-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111505705B (en) | Microseism P wave first arrival pickup method and system based on capsule neural network | |
| CN105224872B (en) | A kind of user's anomaly detection method based on neural network clustering | |
| CN112766379A (en) | Data equalization method based on deep learning multi-weight loss function | |
| CN110084610B (en) | Network transaction fraud detection system based on twin neural network | |
| CN113392931B (en) | Hyperspectral open set classification method based on self-supervision learning and multitask learning | |
| CN109948693A (en) | Expand and generate confrontation network hyperspectral image classification method based on super-pixel sample | |
| CN109919241B (en) | Hyperspectral unknown class target detection method based on probability model and deep learning | |
| CN114564982B (en) | Automatic identification method for radar signal modulation type | |
| Lara et al. | A deep learning approach for automatic recognition of seismo-volcanic events at the Cotopaxi volcano | |
| Wang et al. | Knowledge transfer for structural damage detection through re-weighted adversarial domain adaptation | |
| CN112700324A (en) | User loan default prediction method based on combination of Catboost and restricted Boltzmann machine | |
| Li et al. | Robust deep neural networks for road extraction from remote sensing images | |
| CN117292494A (en) | Signal identification method, system, computer equipment and medium for sound and vibration fusion | |
| CN112966728B (en) | Transaction monitoring method and device | |
| Payne et al. | Two-stage Metropolis-Hastings for tall data | |
| Son et al. | Partial convolutional LSTM for spatiotemporal prediction of incomplete data | |
| CN115375966A (en) | Image countermeasure sample generation method and system based on joint loss function | |
| CN111523258B (en) | Microseism effective signal first arrival pickup method and system based on MS-Net network | |
| Wang et al. | Sonar Objective Detection Based on Dilated Separable Densely Connected CNNs and Quantum‐Behaved PSO Algorithm | |
| Chen et al. | Combining random forest and graph wavenet for spatial-temporal data prediction | |
| Lv et al. | Iterative Sample Generation and Balance Approach for Improving Hyperspectral Remote Sensing Imagery Classification with Deep Learning Network | |
| CN118378178B (en) | Transformer fault identification method and system based on residual map convolution neural network | |
| CN112434716A (en) | Underwater target data amplification method and system based on conditional adversarial neural network | |
| Zhang et al. | Microseismic event detection based on multiscale detection convolutional neural network | |
| Liu et al. | Anomaly detection based on semi-supervised generative adversarial networks |
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 |