CN113791444A - Earthquake motion record selection method based on cluster analysis - Google Patents
Earthquake motion record selection method based on cluster analysis Download PDFInfo
- Publication number
- CN113791444A CN113791444A CN202111079292.9A CN202111079292A CN113791444A CN 113791444 A CN113791444 A CN 113791444A CN 202111079292 A CN202111079292 A CN 202111079292A CN 113791444 A CN113791444 A CN 113791444A
- Authority
- CN
- China
- Prior art keywords
- seismic
- clustering
- earthquake
- record
- algorithm
- 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.)
- Withdrawn
Links
- 238000007621 cluster analysis Methods 0.000 title claims abstract description 25
- 238000010187 selection method Methods 0.000 title claims abstract description 10
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 38
- 238000004458 analytical method Methods 0.000 claims abstract description 30
- 238000001228 spectrum Methods 0.000 claims abstract description 15
- 238000002922 simulated annealing Methods 0.000 claims abstract description 13
- 239000002689 soil Substances 0.000 claims abstract description 8
- 238000012804 iterative process Methods 0.000 claims abstract description 6
- 238000004364 calculation method Methods 0.000 claims abstract description 4
- 238000000034 method Methods 0.000 claims description 14
- 238000012937 correction Methods 0.000 claims description 4
- 238000001914 filtration Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 claims description 4
- 230000008859 change Effects 0.000 claims description 2
- 150000001875 compounds Chemical class 0.000 claims description 2
- 230000010355 oscillation Effects 0.000 abstract description 2
- 230000002068 genetic effect Effects 0.000 description 14
- 230000001133 acceleration Effects 0.000 description 6
- 230000004044 response Effects 0.000 description 6
- 230000006870 function Effects 0.000 description 4
- 230000007547 defect Effects 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 238000013459 approach Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 230000008303 genetic mechanism Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000003321 amplification Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002457 bidirectional effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000002028 premature Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 239000000126 substance Substances 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
- G01V1/24—Recording seismic data
-
- 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
-
- 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
- G01V1/30—Analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Remote Sensing (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Biophysics (AREA)
- Evolutionary Biology (AREA)
- Geophysics (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Health & Medical Sciences (AREA)
- Geology (AREA)
- Acoustics & Sound (AREA)
- Environmental & Geological Engineering (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Probability & Statistics with Applications (AREA)
- Physiology (AREA)
- Genetics & Genomics (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention relates to a seismic oscillation record selection method based on cluster analysis, which comprises the following steps: step 1: selecting earthquake motion records based on a strong earthquake database; step 2: selecting the range of the seismic level and the seismic distance strip, and giving specific seismic information of each seismic record; and step 3: according to a fuzzy C-means clustering algorithm, taking the mean value of the magnitude and the epicenter distance of each earthquake motion record as a clustering center, repeatedly modifying the clustering center and the data membership degree through an iterative process, and giving a preliminary trial calculation clustering analysis result; and 4, step 4: introducing a simulated annealing algorithm to converge to global optimal clustering analysis; and 5: and on the basis of clustering analysis according to the magnitude and the epicenter distance, combining the soil layer shear wave velocity and the thickness of the soil covering layer of the field to give reaction spectrum curves of each group of earthquake motion records under different field types. The invention realizes the self-adaptive classification of the earthquake motion record database so as to provide technical support for the subsequent power time-course analysis.
Description
Technical Field
The invention belongs to the technical field of seismic engineering and disaster prevention and reduction engineering, and particularly relates to a method for selecting seismic motion records based on cluster analysis.
Background
In the field of seismic engineering, with the continuous development of seismic observation technology and the continuous increase of observation stations, global seismic motion record databases are more and more, and how to reasonably select seismic motion records becomes a key of performance-based seismic design. Because the earthquake occurs randomly in time and space, with the gradual enrichment of a large amount of earthquake source information, field information, fault types, earthquake occurrence time and the like, in order to explore a physical mechanism of earthquake occurrence and propagation and further dig the correlation among earthquake motion record data deeply, a big data method is needed to select the earthquake motion records.
The existing classification method mainly depends on the earthquake magnitude, the distance and the site condition to carry out primary selection on the earthquake motion records, the boundary between various types needs to be given in advance according to the expert experience, and the characteristics of the earthquake motion records can not be objectively reflected.
Disclosure of Invention
In order to overcome the defect that the traditional primary selection method only can consider fixed earthquake magnitude and distance grouping, the invention provides a method for primarily selecting earthquake magnitude records according to the earthquake magnitude and distance and considering field categories by carrying out cluster analysis on the earthquake magnitude records based on a global clustering algorithm in machine learning, deeply excavating the statistical properties of the earthquake magnitude and distance of the earthquake magnitude records and considering the influences of different field categories and different cluster groupings on earthquake magnitude spectrum characteristics.
In order to achieve the purpose, the invention is realized by the following technical scheme:
the invention relates to a seismic oscillation record selection method based on cluster analysis, which comprises the following steps:
and step 5, on the basis of carrying out cluster analysis according to the magnitude of earthquake and the distance between the epicenter, combining the shear wave velocity of the soil layer and the thickness of the covering soil layer of the field, giving out the clustering grouping condition of each group of earthquake motion records under the condition of different field types, and fully considering the influence of different field types and different clustering groupings on the earthquake motion spectrum characteristics.
Further, in step 1, the seismic motion recording needs to be subjected to reasonable baseline shift correction and filtering processing.
Further, step 2 specifically includes: the seismic record database contains specific seismic information such as: the earthquake name, station, earthquake magnitude, earthquake center distance, fault type, site condition and the like, and provides abundant data for subsequent research;
and selecting earthquake motion records in a wider range of the earthquake magnitude M-epicenter distance R strip to avoid obvious tendencies of results.
Further, in step 2, the band ranges are: 5< M <7.9, 10km < R <60 km.
Further, in step 3, the fuzzy C-means clustering algorithm measures the correlation size among data points by using a number between [0,1], namely the membership degree, then determines the clustering degree of each given data point according to the membership degree of the given data point, clusters the earthquake motion records into four groups, and obtains a preliminary trial clustering analysis result by repeatedly modifying the clustering center and the data membership degree.
Further, in step 4, the cluster analysis of the genetic simulated annealing algorithm is that the energy of the substance gradually approaches to a lower state along with the reduction of the temperature, and finally reaches a certain balance, the fuzzy cluster analysis is combined to obtain the optimal solution, namely the energy lowest state, the simulated annealing algorithm is combined with the genetic algorithm to amplify the individual fitness difference with similar fitness, so that the advantages of excellent individuals are more obvious, and the optimization approaches to the optimal solution by increasing the population size and the maximum genetic algebra.
Further, step 5 comprises: and classifying the fields according to the anti-seismic standard, giving a corresponding relation between the field type and the shear wave velocity, and giving a response spectrum curve of each field type and different cluster grouping earthquake motion records based on the clustering analysis result and considering the influence of different field types on the earthquake motion spectrum characteristics.
The invention has the beneficial effects that: the invention relates to a seismic record selection method based on cluster analysis, which considers the influences of earthquake magnitude, distance and field category, uses a cluster analysis method in machine learning, takes the mean value of the earthquake magnitude and the earthquake center distance as a cluster center, repeatedly modifies the cluster center and the data membership degree through iteration, and gives a preliminary trial calculation cluster analysis result; on the basis, the global optimal clustering classification is obtained more effectively and more quickly based on the clustering analysis of the genetic simulated annealing algorithm, and finally, the influences of different field categories and different clustering groups on the seismic frequency spectrum characteristics are given.
The earthquake motion records are classified according to the distance or the similarity, the correlation rule between different earthquake motion records is given based on the magnitude of earthquake and the distance of earthquake, and the influence of different field categories on the earthquake motion frequency spectrum characteristic is fully considered, so that the dynamic analysis result has good statistical properties, and a theoretical basis is provided for subsequent exploration.
Drawings
FIG. 1 is a flow chart of a seismic motion selection method based on cluster analysis according to the present invention.
FIG. 2 is the result of a seismic record clustering analysis based on the fuzzy C-means algorithm.
FIG. 3 is the result of a seismic record clustering analysis based on a genetic simulated annealing algorithm.
FIG. 4 is a comparison of seismic dynamic acceleration response spectral means based on different cluster groupings.
FIG. 5 is a seismic acceleration response spectrum mean comparison based on different field classes.
Detailed Description
In the following description, for purposes of explanation, numerous implementation details are set forth in order to provide a thorough understanding of the embodiments of the invention. It should be understood, however, that these implementation details are not to be interpreted as limiting the invention. That is, in some embodiments of the invention, such implementation details are not necessary. In addition, some conventional structures and components are shown in simplified schematic form in the drawings.
Referring to fig. 1, the embodiment provides a method for selecting a seismic record based on cluster analysis, which includes the following steps:
step S1, selecting earthquake motion records based on the strong earthquake database, and carrying out reasonable baseline offset correction and filtering processing;
s2, selecting a wider range of the seismic magnitude and the seismic center distance strip, and giving concrete seismic information of each seismic record;
step S3, according to a fuzzy C-means clustering algorithm, taking the mean value of the earthquake magnitude and the earthquake center distance of each earthquake motion record as a clustering center, repeatedly modifying the clustering center and the data membership degree through an iterative process, and giving a preliminary trial calculation clustering analysis result;
s4, searching an optimal solution by simulating the selection of the nature and the genetic mechanism, and introducing a simulated annealing algorithm to more effectively and more quickly converge to global optimal cluster analysis in order to prevent the later-stage fitness of the genetic algorithm from tending to be consistent, so that the seismic motion record cluster analysis result has stronger robustness, global convergence, implicit parallelism and wide adaptability;
and step S5, on the basis of cluster analysis according to the magnitude of earthquake and the distance between earthquakes, combining the shear wave velocity of the soil layer and the thickness of the covering soil layer of the field to give the clustering grouping condition of each group of earthquake motion records under different field types, and fully considering the influence of different field types and different clustering groupings on the earthquake motion frequency spectrum characteristics.
The seismic record selection method provided by the embodiment can overcome the defect that only fixed earthquake magnitude and distance grouping are considered in the prior art, the influence of seismic record data on clustering results is deeply mined based on a global clustering algorithm in machine learning, and classification is performed according to the correlation degree between different seismic records, so that the dynamic analysis result has good statistical properties.
The method comprises the following specific steps:
1. seismic motion selection based on a strong seismic database.
(1) Selecting 2157 groups of horizontal bidirectional seismic records in a database, striving for a wider magnitude M and a wider epicenter distance R, wherein the value range of the magnitude M-the epicenter distance R of the seismic record database of the primarily selected American Pacific seismic engineering research center is as follows: 5< M <7.9, 10km < R <60 km; the selected earthquake motion has detailed earthquake name, station information, field information, fault type and the like, and can be used for subsequent analysis.
2. And (4) carrying out clustering analysis based on a fuzzy C-means algorithm.
(1) The method comprises the steps of adopting a C-means clustering algorithm, dividing earthquake motion records into different categories through an iterative process, selecting a central point in a data set to have randomness, and introducing a fuzzy set concept into the selected data set when the earthquake motion records can not be strictly divided into a certain category, wherein the characteristic is not that the selected data set has randomness, and finally determining a certain clustering grouping degree according to the membership degree of a data point.
(2) Let sample space X ═ X1,x2,…,xn}TDividing the earthquake motion data into four classes, wherein each clustering center is { c1,c2,c3,c4}, element uijRepresenting the degree of membership that the ith data point belongs to the jth class, such that the objective function of euclidean distance achieves the smallest function value:
(3) and (3) giving a clustering center and a data membership:
(4) repeatedly modifying the clustering center and the data membership according to the formulab(t+1)-JbAnd (t) | < 1e-6, the algorithm converges to obtain the membership of each clustering center and each sample, the fuzzy clustering division is completed, the algorithm is ended, otherwise, iteration is performed until the maximum iteration frequency reaches 20 times, and the algorithm is ended. FIG. 2 shows the result of the seismic record clustering analysis based on the fuzzy C-means algorithm.
3. And (4) carrying out clustering analysis based on a genetic simulated annealing algorithm.
If the initial value is not properly selected, the clustering analysis result can be converged to a local optimal solution, a genetic algorithm optimizing solution is introduced on the basis, the solution space of the clustering analysis is encoded, a fitness function is constructed, and the value of each control parameter is determined through selection, intersection and variation of a genetic operator. However, the genetic algorithm is easy to generate the premature convergence problem, the fitness tends to be consistent in the later stage of the genetic algorithm, and a simulated annealing algorithm needs to be introduced to keep good individuals and maintain the diversity of the population.
The simulated annealing algorithm is combined with the genetic algorithm, when the temperature is continuously reduced, the individual fitness difference of fitness is amplified, so that the advantages of excellent individuals are more obvious, fuzzy cluster analysis is combined on the basis, wherein the change of energy is an objective function, the obtained optimal solution is the energy lowest state, and the seismic motion record cluster analysis result based on the genetic simulated annealing algorithm is shown in figure 3.
4. And (4) primarily selecting the seismic motion record of the clustering group.
The earthquake motion records are divided into four groups according to the magnitude of the earthquake magnitude and the magnitude of the epicenter distance, the earthquake motion acceleration time course is normalized, and the acceleration response spectrum is obtained, namely the dynamic amplification coefficient curve. In order to fully consider the influence of different clustering groups on the seismic motion spectrum characteristics, fig. 4 shows the seismic motion acceleration response spectrum mean comparison result of different clustering groups.
5. And (4) primarily selecting the seismic motion record in consideration of the field category.
The field classification is performed according to the seismic norm according to the equivalent shear wave velocity of the underground 30m depth, and the clustering grouping result based on the field category is given by considering each clustering grouping result as shown in table 1.
TABLE 1 field Category based earthquake motion record clustering grouping results
At least 11 earthquake motion records are selected for analysis during FEMA P-1050 specified time-course analysis, and the influence of the fourth field type on earthquake motion frequency spectrum characteristics cannot be fully considered due to too few earthquake motion records of the fourth field type, so that the earthquake motion records of the type are not continuously analyzed subsequently. FIG. 5 shows the result of comparing the mean of the seismic acceleration response spectra based on different field categories.
The method overcomes the defect that only fixed earthquake magnitude and earthquake-centre distance grouping are considered in the prior art, can objectively reflect the characteristics of earthquake motion records, deeply excavates the influence of the earthquake motion records on clustering analysis results, realizes the self-adaptive classification of the earthquake motion record database, and provides technical support for the subsequent power time-course analysis.
The above description is only an embodiment of the present invention, and is not intended to limit the present invention. Various modifications and alterations to this invention will become apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.
Claims (6)
1. A seismic record selection method based on cluster analysis is characterized by comprising the following steps: the seismic record selection method comprises the following steps:
step 1: selecting earthquake motion records based on a strong earthquake database;
step 2: selecting the range of the seismic level and the seismic distance strip, and giving specific seismic information of each seismic record;
and step 3: according to a fuzzy C-means clustering algorithm, taking the mean value of the magnitude and the epicenter distance of each earthquake motion record as a clustering center, repeatedly modifying the clustering center and the data membership degree through an iterative process, and giving a preliminary trial calculation clustering analysis result;
and 4, step 4: introducing a simulated annealing algorithm to converge to global optimal clustering analysis;
and 5: and on the basis of clustering analysis according to the magnitude and the epicenter distance, combining the soil layer shear wave velocity and the thickness of the soil covering layer of the field to give reaction spectrum curves of each group of earthquake motion records under different field types.
2. The method of selecting seismic motion record based on cluster analysis according to claim 1, wherein: the step 3 specifically comprises the following steps:
step 3-1: adopting a C-means clustering algorithm, dividing the seismic motion records into different categories through an iterative process, introducing a fuzzy set concept into the seismic motion records on the basis, and finally determining a certain clustering grouping degree according to the membership degree of data points;
step 3-2: let sample space X ═ X1,x2,…,xn}TDividing the earthquake motion data into four classes, wherein each clustering center is { c1,c2,c3,c4}, element uijRepresenting the degree of membership that the ith data point belongs to the jth class, such that the objective function of euclidean distance achieves the smallest function value:
step 3-3: and (3) giving a clustering center and a data membership:
step 3-4: repeatedly modifying the clustering center and the data membership according to the formulab(t+1)-JbAnd (t) | < 1e-6, the algorithm converges to obtain the membership of each clustering center and each sample, the fuzzy clustering division is completed, the algorithm is ended, otherwise, iteration is performed until the maximum iteration frequency reaches 20 times, and the algorithm is ended.
3. The method of selecting seismic motion record based on cluster analysis according to claim 1, wherein: the step 4 of introducing a simulated annealing algorithm to converge to the global optimal clustering analysis specifically comprises the following steps: when the temperature is continuously reduced, the individual fitness difference of the fitness is amplified, fuzzy clustering analysis is combined on the basis, wherein the change of energy is an objective function, and the obtained optimal solution is the energy lowest state.
4. The method of selecting seismic motion record based on cluster analysis according to claim 1, wherein: the seismic record in step 1 needs to be subjected to baseline shift correction and filtering processing.
5. The method of selecting seismic motion record based on cluster analysis according to claim 1, wherein: in step 2, the band range of magnitude M is: 5< M <7.9, epicenter R-banding range: 10km < R <60 km.
6. The method of selecting seismic motion record based on cluster analysis according to claim 1, wherein: in the step 2, the specific seismic information of the seismic record comprises an earthquake name, a station, an earthquake magnitude, an earthquake center distance, a fault type and a field condition.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111079292.9A CN113791444A (en) | 2021-09-15 | 2021-09-15 | Earthquake motion record selection method based on cluster analysis |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111079292.9A CN113791444A (en) | 2021-09-15 | 2021-09-15 | Earthquake motion record selection method based on cluster analysis |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113791444A true CN113791444A (en) | 2021-12-14 |
Family
ID=79183387
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111079292.9A Withdrawn CN113791444A (en) | 2021-09-15 | 2021-09-15 | Earthquake motion record selection method based on cluster analysis |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113791444A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114443883A (en) * | 2022-02-10 | 2022-05-06 | 北京永利信达科技有限公司 | Data processing method, system and medium based on big data and cloud computing |
CN114779329A (en) * | 2022-04-24 | 2022-07-22 | 福州大学 | Near-fault sea area seismic oscillation response spectrum calibration method |
CN116630676A (en) * | 2022-09-01 | 2023-08-22 | 中国地震局地球物理研究所 | Large-scale-range field classification processing method and device and electronic equipment |
-
2021
- 2021-09-15 CN CN202111079292.9A patent/CN113791444A/en not_active Withdrawn
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114443883A (en) * | 2022-02-10 | 2022-05-06 | 北京永利信达科技有限公司 | Data processing method, system and medium based on big data and cloud computing |
CN114779329A (en) * | 2022-04-24 | 2022-07-22 | 福州大学 | Near-fault sea area seismic oscillation response spectrum calibration method |
CN116630676A (en) * | 2022-09-01 | 2023-08-22 | 中国地震局地球物理研究所 | Large-scale-range field classification processing method and device and electronic equipment |
CN116630676B (en) * | 2022-09-01 | 2024-02-09 | 中国地震局地球物理研究所 | Large-scale-range field classification processing method and device and electronic equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN113791444A (en) | Earthquake motion record selection method based on cluster analysis | |
Sambridge et al. | Earthquake hypocenter location using genetic algorithms | |
US11215719B2 (en) | Performance-level seismic motion hazard analysis method based on three-layer dataset neural network | |
US20210026031A1 (en) | Dolomite reservoir prediction method and system based on well and seismic combination, and storage medium | |
CN101266299B (en) | Method for forecasting oil gas utilizing earthquake data object constructional features | |
CN108732620B (en) | Multi-wave earthquake oil and gas reservoir prediction method under unsupervised and supervised learning | |
CN108897975A (en) | Coalbed gas logging air content prediction technique based on deepness belief network | |
CN110344824B (en) | Acoustic curve generation method based on random forest regression | |
CN106569272B (en) | A kind of seismic attribute fusion method that dimension is risen based on data character space | |
CN106056577B (en) | SAR image change detection based on MDS-SRM Mixed cascading | |
CN113343574A (en) | Mishrif group lithology logging identification method based on neural network | |
CN115130375A (en) | Rock burst intensity prediction method | |
CN108280289A (en) | Bump danger classes prediction technique based on local weighted C4.5 algorithms | |
CN117272841B (en) | Shale gas dessert prediction method based on hybrid neural network | |
CN111611422A (en) | SVC-based method and system for automatically generating qualitative graph in earthquake disaster risk assessment | |
CN110009024A (en) | A kind of data classification method based on ID3 algorithm | |
CN115545543A (en) | Multi-disaster integrated natural disaster risk assessment and analysis system | |
Cho et al. | Development of locally specified soil stratification method with CPT data based on machine learning techniques | |
CN112434878A (en) | Cascade sample equalization-based seismic fluid prediction method | |
Whitley et al. | Comparing the niches of CMA-ES, CHC and pattern search using diverse benchmarks | |
CN110488350B (en) | Seismic inversion big data generation method based on convolutional neural network | |
CN113361619B (en) | Convolutional neural network lithology identification method based on rough set attribute reduction | |
Sinaga et al. | Spatial Variation in Infant Mortality with Geographically Weighted Poisson Regression (GWPR) Approach | |
Alimoradi et al. | Ga-based selection and scaling of strong ground motion records for structural design | |
CN116070148A (en) | Oil source classification method and system based on deep neural network model |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20211214 |
|
WW01 | Invention patent application withdrawn after publication |