CN107463991A - A kind of Regional Landslide method for evaluating hazard based on slopes unit and machine learning - Google Patents
A kind of Regional Landslide method for evaluating hazard based on slopes unit and machine learning Download PDFInfo
- Publication number
- CN107463991A CN107463991A CN201710506290.0A CN201710506290A CN107463991A CN 107463991 A CN107463991 A CN 107463991A CN 201710506290 A CN201710506290 A CN 201710506290A CN 107463991 A CN107463991 A CN 107463991A
- Authority
- CN
- China
- Prior art keywords
- landslide
- hazard
- unit
- index
- evaluation
- 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.)
- Pending
Links
Classifications
-
- 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
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Molecular Biology (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Health & Medical Sciences (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of Regional Landslide method for evaluating hazard based on slopes unit and machine learning, comprise the following steps:Research area ramp unit to be evaluated is marked off according to digital elevation model first, and establishes the Landslide Hazard Assessment index system suitable for research area;Then the regularity of distribution of the evaluation index in the ramp unit to have come down is analyzed, and on this basis, builds the Landslide hazard grade classification table of each index;The models for hazard assessment of landslide based on LM BP neural networks is finally established, calculates the risk factor of each evaluation unit, and carry out Hazard rank division.Method provided by the present invention calculates accurately, and principle is reliable, and operating process is easy, can be generally applicable to the needs of Regional Landslide hazard assessment.
Description
Technical field
The present invention relates to landslide disaster risk assessment technology field, and in particular to one kind is based on slopes unit and machine learning
Regional Landslide method for evaluating hazard.
Background technology
Regional Landslide hazard assessment is to be directed to evaluation unit, with the influence factor (evaluation index) of development of landslide for base
Plinth, establish evaluation model and carry out hazard assessment.China's landslide disaster, which has, endangers big, widespread feature, serious threat
The life and property safety of the people, governs the economic development in the multiple area of landslide disaster.Therefore, endangered for Regional Landslide
Dangerous evaluation, personnel's injury and property loss caused by can effectively reducing geological disaster, and rational evaluation unit is this
The basis of all, evaluation unit are the base units of Regional Landslide hazard assessment, and research object, its size and border are straight
Connect influence evaluation result.The evaluation unit of Regional Landslide hazard assessment can be summarized as following 5 class:Homogeneous conditioning unit,
Domain unit, grid unit, region unit, ramp unit and geomorphic unit.As GIS is in fields such as Regional Landslide hazard assessments
Using increasing increasingly, the computer disposal of grid unit is relatively easy in addition, and many researchers select to select lattice during evaluation unit
Net unit is as evaluation unit.But the development on landslide be based on slope, this regular grid and Slope Failure
Mechanical mechanism is without any association, with geology, landform and ambient boundary also without any relation.Ramp unit is the basic of development of landslide
Unit, and in all kinds of influence factors, the development of stage of development of river or cheuch on landslide have highly important influence.Cause
This, using the ramp unit based on watershed and the division of ditch valley line as evaluation unit, can retain geology inside evaluation unit
The relative fullness of environment, the comprehensive effect for embodying all kinds of influence factors, makes evaluation result more realistic.
Regional Landslide Hazard Risk Assessment method has a lot, and these methods can be divided into three classes:Qualitative risk evaluation, semidefinite
Measure risk assessment, Quantitative Risk Evaluation.Main evaluation method has expert point rating method, index score method, analytic hierarchy process (AHP), decision-making
Tree analysis process, markov modeling, fuzzy comprehensive evaluation method etc., but these methods largely have very strong artificial subjectivity
Property, influence the accuracy and objectivity of evaluation result.Such as Application No. 201310357139.7《It is a kind of based on Analysis on Mechanism
Small watershed area landslide disaster fire risk district method》, Application No. 201310627906.1《A kind of Regional Landslide calamity source
Method of estimation》, Application No. 201610330203.6《Risk of landslip quantization method》Deng, the above method in terms of self study according to
Many deficiencies so be present, it calculates structure and still had some limitations.
The content of the invention
The present invention cut the relative fullness of geological environment, tradition inside ramp unit for traditional grid evaluation unit
Index point system weight determines the defects of link human factor is too high, there is provided according to research area's digital elevation model, has slided
Slope disaster data and geology, meteorology, remote sensing image data etc. are used for the method for Regional Landslide hazard assessment, and this method can
Realize calculating and the risk zoning of the risk factor to studying each evaluation unit in area.Using machine learning method, overcome well
Subjective factor excessive this shortcoming, especially BP neural network in prior art evaluation method, i.e. error back propagation
Multilayer feedforward formula network.Slopes unit is combined with machine learning method, and is applied in Regional Landslide hazard assessment, is belonged to
The research contents of multi-crossed disciplines.In recent years, the fast development of computer technology, information technology especially GIS technology is region
The hazard assessment on landslide provides great support.
To achieve the above object, technical scheme is as follows:
(1) hydrological analysis to research area DEM (digital elevation model), the evaluation unit in Research on partition area are passed through:
1) ArcGIS 10.2 raster symbol-base device instrument is utilized, obtains DEM maximum first, and is subtracted with maximum
The DATA values of DEM each grid cell, obtain inverting DEM;
2) hollow instrument is filled out using the hydrological analysis modules of ArcGIS 10.2, depression filling is carried out to positive and negative DEM respectively, obtained
To without depression DEM;
3) instrument is flowed to using the hydrological analysis modules of ArcGIS 10.2, water (flow) direction meter is carried out to positive and negative DEM respectively
Calculate, obtain flowing to grid;
4) instrument is calculated using the integrated flow of the hydrological analysis modules of ArcGIS 10.2, positive and negative DEM is confluxed respectively
Integrated flow calculates, and obtains flow grid;
5) ArcGIS 10.2 raster symbol-base device instrument is utilized, setting collection waters threshold value, extracts river network;
6) the Watershed instruments of the hydrological analysis modules of ArcGIS 10.2 are utilized, generate the positive and negative network of waterways, positive and negative collection current
Domain;
7) the basin face of catchmenting of generation merge with reversely basin face of catchmenting, then postmenstruation artificial cognition, change it is unreasonable
Unit, amending method include border amendment, fragment merge, crack filling etc., finally give by valley line and ridge line institute
The region of composition is ramp unit;
(2) by the research to the pregnant calamity condition of landslide disaster and formation mechenism, the finger of Regional Landslide hazard assessment is established
Mark system:
1) in regional differentiation, index primary and secondary is obvious, distinguishes independence between opinion scale, index, the principle of availability
Under establish assessment indicator system, the primary condition of development of landslide has:Geological conditions, geographic and geomorphic conditions, weather conditions, hydrology bar
Part, Vegetation condition, groundwater condition, distribution of faults and activity, earthquake, Human dried bloodstains;
2) according to environment body factor and the major class of triggering factors two, one-level assessment indicator system is established, respectively including landform
Landforms, ground mulching, precipitation, geological conditions;
3) under one-level assessment indicator system, further refinement, wherein, topography and geomorphology includes:Elevation, the gradient, slope aspect, height
Difference;Ground mulching includes vegetation-cover index, normalization aqua index;Precipitation includes average annual precipitation, precipitation year border coefficient of variation;
Geological conditions includes lithology, away from tomography distance;
4) be based on ArcGIS 10.2, the ZonalStatisticsAsTable instruments of utilization space analysis module, pass through by
Ramp unit figure layer and each indicatrix layer overlay analysis, obtain the quantized value of several Raw performances of each ramp unit, obtain
Study the Raw performance matrix A in area;
5) according to relatively independent principle between the index of one of assessment indicator system establishment principle, matrix A is imported into SPSS
21.0 do each index related analysis, obtain the correlation coefficient ρ of Raw performance between any two;
(3) analysis of distribution of the evaluation index in the ramp unit to have come down:
1) using GIS as platform, by the overlay analysis to landslide disaster point and ramp unit has occurred, obtain each evaluation and refer to
It is marked on the desired value of the ramp unit to have come down;
2) using SPSS as platform, statistical analysis is carried out to desired value of the evaluation index in the ramp unit to have come down,
Obtain the histogram frequency distribution diagram or directional spreding radar map of each evaluation index;
3) according to the histogram frequency distribution diagram of each evaluation index or directional spreding radar map, each evaluation index is marked off not
With Landslide hazard grade corresponding to section, and the risk factor monotonicity in each section is determined, establish the cunning of each index respectively
Slope Hazard rank division table;
(4) foundation of the Landslide Hazard Assessment model based on LM-BP neutral nets:
1) BP neural network is due to its good generalization ability and None-linear approximation ability, and model easy constructed by
To the application of many industry fields.At present, BP neural network is mainly used in the fields such as function approximation, pattern-recognition, especially letter
Number approaches field.But which kind of field no matter applied to, BP algorithm all has the defects of convergence rate is excessively slow;
2) BP neural network (abbreviation LM-BP neutral nets) biggest advantage based on LM algorithm optimizations is exactly local quick
Convergence, this has been saved the substantial amounts of time for the training of neutral net, especially the huge neutral net of training data.Meanwhile
The powerful global search of LM-BP neutral nets also preferably ensure that the neutral net after training can have good extensive energy
Power, that is, network extrapolability;
3) foundation of neutral net master sample:Master sample (including training sample and test sample) is referred to evaluation
Mark according to certain mathematical method and Landslide Hazards degree of danger with the basis of the functional relation of Landslide, predicting
Evaluation index grade scale considers foundation;
4) foundation of LM-BP neural network models:The prediction and evaluation of neutral net is realized on MATLAB, be it is most convenient,
Quick method, Neural Network Toolbox powerful MATLAB and largely on neutral net structure, study etc. function be
The realization of neutral net provides great support.Normalization of the step of LM-BP Establishment of Neural Model including data, wound
Build LM-BP neutral nets, training LM-BP neutral nets, test LM-BP neutral nets and neural network forecast;
(5) Regional Landslide hazard assessment:
1) on the basis of the LM-BP neutral nets established, by inputting master sample matrix, to LM-BP nerve nets
Network is repeatedly trained, until reaching default precision;
2) LM-BP neutral nets are tested, to verify the generalization ability of LM-BP neutral nets;
3) the good LM-BP neutral nets of generalization ability are preserved, landslide hazard angle value is carried out to each evaluation unit in research area
Prediction;
4) take equidistant method point four grades to divide Landslide hazard rank, landslide hazard angle value 0~
In the range of 0.25 corresponding Hazard rank be I grade, in the range of 0.25~0.5 corresponding Hazard rank be II grade, 0.5~
Corresponding Hazard rank is III grade in the range of 0.75, and corresponding Hazard rank is IV grade in the range of 0.75~1.0;
5) hazard assessment interpretation of result and checking:By statistical analysis, each grade hazardous area slope number, number are drawn
Percentage, area, area percentage statistical form, the distribution situation of each danger classes is drawn, by real with research area's disaster distribution
Whether the contrast of border situation, checking evaluation result meet the fact.
Compared with prior art, the beneficial effects of the invention are as follows:
Evaluation unit using ramp unit as Regional Landslide hazard assessment, remain geological environment inside evaluation unit
Relative fullness;By to evaluation index analysis of distribution, on the basis of LM-BP neural network theories, being managed using interpolation
By establishing LM-BP neutral net master sample matrixes, sample matrix is eliminated the reliance on other evaluation methods and is obtained, avoid
The influence of human factor during neural network model is built using the methods of expert point rating method, index method.
Brief description of the drawings
Fig. 1 is technical scheme line map.
Fig. 2 is ramp unit division flow chart.
Fig. 3 is LM-BP neural network algorithm flow charts.
Fig. 4 is research area's Landslide hazard division result figure.
Embodiment
The present invention is further described with reference to the accompanying drawings and examples.
Below in conjunction with the accompanying drawings 1 to 4, the preferred embodiments of the present invention are further described.
So that on blue Chengdu-Chongqing product oil oil pipeline K558-K642 mileages section (Guangyuan section), there may be the model on the slope of influence
Enclose to study area, Landslide Hazard Assessment is carried out to the research area with the inventive method.
Research area is located at Sichuan Province's Guangyuan City domestic, 15 ' -106 ° 4 ' of east longitude 105 °, 3 ' -32 ° 45 ' of north latitude 32 °, by north orientation
South is across 19 small towns in Chaotian District, Li Zhouqu, Zhao Huaqu, Qingchuan County, Jiange County county of totally three area two.Study the geology calamity that come down in area
Evil has more than 100 to locate, and part landslide is apart from pipeline less than 100m, safe operation, nearby residents and its people of these landslides to pipeline
The great threat of economic structure.
In order to carry out Landslide Hazard Assessment to research area, the technology path (such as Fig. 1) provided using the inventive method is right
Study area and carry out method validation, step is as follows:
The first step, evaluation unit classification.
Research area DEM (ASTER V2) is obtained, using ArcGIS 10.2 hydrological analysis module, is drawn according to ramp unit
Split flow divides evaluation unit, as shown in Figure 2.Research area marks off 315 ramp units, the wherein unreasonable list of manual amendment altogether
Member is on the platforms of ArcGIS 10.2, according to GF-1 remote sensing images, artificial cognition, changes irrational unit, amending method bag
Include border amendment, fragment merging, crack filling etc..
Second step, the foundation of Raw performance system.
According to selecting index principle, the formation mechenism of Landslide Hazards, Raw performance system, Raw performance system are established
In index (i.e. initial evaluation index) four classes can be classified as, be landforms, precipitation, ground mulching, the class of geology four respectively, such as the institute of table 1
Show.
The Landslide Hazard Assessment Raw performance of table 1
3rd step, the foundation of index system.
Based on ArcGIS 10.2, by by ramp unit figure layer and each indicatrix layer overlay analysis, obtaining each ramp unit
11 Raw performances quantized value, obtain study area totally 315 rows 11 row Raw performance matrix A315×11.According to evaluation index
Relatively independent principle between the index of one of Establishing principle, by A315×11Import SPSS21.0 and do each index related analysis, obtain
To the correlation coefficient ρ of 11 Raw performances between any two.What table 2 represented is that 11 counted based on Raw performance matrix are initially referred to
The coefficient correlation of mark between any two.
Correlation coefficient charts between the Landslide Hazard Assessment initial evaluation index of table 2
The coefficient correlation that NDVI and NDWI coefficient correlation is 0.983, Rain and Rain-Ratio is 0.953, is all exceeded
0.8, it is highly correlated.Generally, Rain and the coefficient correlation absolute value ratio of other indexs in addition to Rain-Ratio
Rain-Ratio and the coefficient correlation absolute value of other indexs in addition to Rain are smaller, i.e. Rain and other indexs
Correlation is lower, and Rain (average annual precipitation for many years) in Landslide Hazard Assessment using more extensive.NDVI with except NDWI it
The coefficient correlation absolute value of other outer indexs differs than NDWI with the coefficient correlation absolute value of other indexs in addition to NDVI
Less.But NDVI standard deviation be about 0.117, NDWI standard deviation be about 0.110, NDVI standard deviation be more than NDWI, and
NDVI and NDWI dimension and value section are identical, it is believed that this index of NDVI is than information content that NDWI this index provides
Greatly.Consider, final Landslide Hazard Assessment index deletes NDWI, Rain-Ratio from initial evaluation index system.
Therefore, Landslide Hazard Assessment index system includes NDVI, Aspect, Slope, Elevation, Rain altogether,
Height-Difference, Curvature, Distance, Lithology totally 9 evaluation indexes.
4th step, analysis of distribution of the evaluation index in the ramp unit to have come down.
Landslide point figure layer is overlapped with ramp unit face figure layer in the platforms of ArcGIS 10.2, calamity has been occurred
Evil ramp unit simultaneously exports its all attribute list, and inputs SPSS 21.0, and the frequency distribution for obtaining each evaluation index is straight
Side's figure or directional spreding radar map, carry out statistical analysis.
According to the histogram frequency distribution diagram of each evaluation index or directional spreding radar map, it is different to mark off each evaluation index
Landslide hazard grade corresponding to section, and the risk factor monotonicity in each section is determined, the landslide of each index is established respectively
Hazard rank divides table.Landslide hazard grade is divided into four ranks altogether, low dangerous (I) of generation of respectively coming down, in
Dangerous (II), high-risk (III), high dangerous (IV) four ranks.
5th step, the foundation of neutral net master sample.
As shown in figure 3, training sample construction step is as follows:
Ⅰ:Build empty matrix
In order that LM-BP neutral nets Fast Convergent, learning sample capacity of each Landslide hazard rank in training
200 are arranged to, therefore, the capacity of training sample is 800, and totally 9 evaluation indexes, build empty matrix X first800×9。
Ⅱ:Build input vector
According to the Landslide hazard grade classification table of evaluation index, according to the order of risk factor from low to high, in each area
It is interior to press section monotonicity interpolation, the sample vector of each evaluation index is built respectively, every 200 are a Hazard rank,
The sample vector length of each evaluation index is 800.
Ⅲ:Build output vector
Risk factor interzone is [0,1], and output vector is equidistant 800 value gained of interpolation between [0,1].
Ⅳ:Merge
The input vector of 9 evaluation indexes and output vector are merged into a matrix.
The standard exercise sample matrix for training LM-BP neutral nets, part of standards instruction are constructed according to above-mentioned steps
It is as shown in table 3 to practice sample matrix, and the standard testing sample moment for testing LM-BP neutral nets is constructed with reference to above-mentioned steps
Battle array, totally 20 test samples.
The part of standards training sample matrix of table 3
5th step, the foundation of LM-BP neural network models and the calculating of risk factor.
LM-BP neutral nets are completed on the platforms of MATLAB 2014, and it is as follows to complete step:
Ⅰ:The normalization of data
Data normalization is that data are limited in a segment limit section by certain mathematical method.Normalized one
As it is limited between [0,1], [0.1,0.9] or [- 1,1].Using the mapminmax functions that MATLAB is carried by sample
Each column vector of matrix is normalized between [0,1].
After normalization sample matrix all between [0,1], it is necessary to explanation, the output vector in sample matrix
Originally all between [0,1], therefore it need not be normalized again.
Ⅱ:Create LM-BP neutral nets
It is determined that the rational hidden layer number of plies and nodes, determine network structure.BP neural network structure is by input layer, hidden
Formed containing layer and output layer.The number of input layer takes the dimension of input vector;Node in hidden layer there is no what is quickly determined at present
Method, can only be by according to forefathers' empirical equation, calculating general node in hidden layer, then carries out constantly debugging, tentative calculation network,
The error sum of squares exported by contrasting network, obtains optimal node in hidden layer.
From Kolmogorov theorems, one 3 layers of neutral net can complete any n dimensions to m dimensions with arbitrary accuracy
Mapping.LM-BP neural network structures are arranged to 3 layers, respectively input layer, hidden layer and output layer.Debugged according to multiple, finally
The node in hidden layer for determining LM-BP neutral nets is 10.The transmission function of hidden layer and output layer is respectively tansig,
purelin。
Ⅲ:Train LM-BP neutral nets
Training LM-BP neutral nets are exactly the process of LM-BP neutral net self-teachings, are the mistakes for adjusting weights and threshold value
Journey, it is a crucial link of the Regional Landslide hazard assessment based on LM-BP neutral nets.The precision of network training will be direct
The precision of evaluation result is had influence on, the precision of LM-BP neural metwork trainings can refer to root-mean-square error MSE and make analysis.
The training function of network is trainlm, and training parameter is as follows:
Net.trainParam.show=60;
Net.trainParam.lr=0.5;
Net.trainParam.epochs=1000;
Net.trainParam.goal=1e-8;
Ⅳ:Test LM-BP neutral nets
Test LM-BP neutral nets are in order to sentence the generalization ability of the LM-BP neutral nets after knowing training, i.e., to unknown sample
The ability of this prediction.Only on the premise of the LM-BP neutral nets after ensureing to train have enough generalization abilities, it could carry out
Vth step, i.e., using neutral net.
In order to more accurately judge the generalization ability of the LM-BP neutral nets after training, 20 groups of test sample data are selected
Neutral net is tested.Network test is exactly that the Input parts of test matrix are input into LM-BP neutral nets Net to enter
Row emulation, the Output of simulation data and test matrix parts are contrasted, analyze whether its error meets the requirements, LM-BP god
It is as shown in table 4 through network test error.
The LM-BP neutral net test error tables of table 4
As shown in Table 4, the Error Absolute Value of 20 groups of test datas is both less than 0.02, meets Regional Landslide hazard assessment
It is required that LM-BP neutral net Net generalization abilities are good, therefore, network N et can come down to the research each evaluation unit in area
Dangerous angle value prediction.
Ⅴ:Prediction
LM-BP neutral nets Net that is qualified and preserving is emulated after prediction data is exactly input to training by prediction,
LM-BP neutral nets can export predicted value automatically.
Prediction data is exactly the data matrix Y of the 9 evaluation index values composition for 315 ramp units for studying area315×9, will
It is input to LM-BP neutral nets Net and emulated after being normalized, export the landslide hazard of 315 ramp units
Angle value.
6th step, hazard assessment interpretation of result and checking.
Four grades of equidistant method point are taken to divide Landslide hazard rank., will on the platforms of ArcGIS 10.2
The landslide hazard angle value of research 315, area ramp unit is connected to ramp unit figure layer, and landslide hazard angle value is in 0~0.25 scope
Hazard rank is I grade corresponding to interior, and corresponding Hazard rank is II grade in the range of 0.25~0.5, in the range of 0.5~0.75
Corresponding Hazard rank is III grade, and corresponding Hazard rank is IV grade in the range of 0.75~1.0, obtains studying area landslide danger
Dangerous division result figure, as shown in Figure 4.
By statistical analysis, each grade hazardous area slope number, number percentage, area, area percentage statistics are drawn
Table, as shown in table 5.
Each grade danger slope number of table 5 and area statistics
As shown in Table 5, sum is accounted in high-risk danger zone (III), high hazardous area (IV) slope number and area
70%, research area's Landslide is generally higher.The fact that this result with research area is Landslide Hazards multiple area phase
It coincide.
The above described is only a preferred embodiment of the present invention, any formal limitation not is made to the present invention, though
So the present invention is disclosed above with preferred embodiment, but is not limited to the present invention, any to be familiar with this professional technology people
Member, without departing from the scope of the present invention, when the technology contents using the disclosure above make a little change or modification
For the equivalent embodiment of equivalent variations, as long as being the content without departing from technical solution of the present invention, the technical spirit according to the present invention
Any simple modification, equivalent change and modification made to above example, in the range of still falling within technical solution of the present invention.
Claims (6)
1. a kind of Regional Landslide method for evaluating hazard based on slopes unit and machine learning, it is characterised in that including as follows
Step:
S1, the hydrological analysis by research area DEM being digital elevation model, the evaluation unit in Research on partition area;
S2, by the research to the pregnant calamity condition of landslide disaster and formation mechenism, establish the index body of Regional Landslide hazard assessment
System;
The analysis of distribution of S3, evaluation index in the ramp unit to have come down;
S4, Landslide Hazard Assessment model based on LM-BP neutral nets foundation;
S5, Regional Landslide hazard assessment.
2. a kind of Regional Landslide method for evaluating hazard based on slopes unit and machine learning according to claim 1,
Characterized in that, in the step S1, by the hydrological analysis to studying area DEM, the evaluation unit in Research on partition area, it is used
The softwares of ArcGIS 10.2 are analyzed, and are comprised the concrete steps that:
S1.1, the raster symbol-base device instrument using ArcGIS 10.2, obtain DEM maximum first, and are subtracted with maximum
The DATA values of DEM each grid cell, obtain inverting DEM;
S1.2, using the hydrological analysis modules of ArcGIS 10.2 hollow instrument is filled out, depression filling is carried out to positive and negative DEM respectively, obtained
Without depression DEM;
S1.3, using the hydrological analysis modules of ArcGIS 10.2 instrument is flowed to, water (flow) direction calculating is carried out to positive and negative DEM respectively,
Obtain flowing to grid;
S1.4, the integrated flow calculating instrument using the hydrological analysis modules of ArcGIS 10.2, conflux to positive and negative DEM respectively
Integrated flow calculates, and obtains flow grid;
S1.5, the raster symbol-base device instrument using ArcGIS 10.2, setting collection waters threshold value, extract river network;
S1.6, the Watershed instruments using the hydrological analysis modules of ArcGIS 10.2, generate the positive and negative network of waterways, positive and negative collection current
Domain;
S1.7, the basin face of catchmenting of generation merge with reversely basin face of catchmenting, then postmenstruation artificial cognition, change it is irrational
Unit, amending method include border amendment, fragment merge, crack filling etc., finally give by valley line and ridge line institute group
Into region be ramp unit.
3. a kind of Regional Landslide method for evaluating hazard based on slopes unit and machine learning according to claim 2,
Characterized in that, in the step S2, by the research to the pregnant calamity condition of landslide disaster and formation mechenism, Regional Landslide danger is established
The step of index system of dangerous evaluation is:
S2.1, in regional differentiation, index primary and secondary is obvious, distinguishes under independence between opinion scale, index, the principle of availability
Assessment indicator system is established, the primary condition of development of landslide has:Geological conditions, geographic and geomorphic conditions, weather conditions, hydrology bar
Part, Vegetation condition, groundwater condition, distribution of faults and activity, earthquake, Human dried bloodstains;
S2.2, according to environment body factor and the major class of triggering factors two, establish one-level assessment indicator system, respectively including landform
Looks, ground mulching, precipitation, geological conditions;
S2.3, under one-level assessment indicator system, further refinement, wherein, topography and geomorphology includes:Elevation, the gradient, slope aspect, height
Difference;Ground mulching includes vegetation-cover index, normalization aqua index;Precipitation includes average annual precipitation, precipitation year border coefficient of variation;
Geological conditions includes lithology, away from tomography distance;
S2.4, based on ArcGIS 10.2, the ZonalStatisticsAsTable instruments of utilization space analysis module, pass through by
Ramp unit figure layer and each indicatrix layer overlay analysis, obtain the quantized value of several Raw performances of each ramp unit, obtain
Study the Raw performance matrix A in area;
Relatively independent principle between S2.5, the index according to one of assessment indicator system establishment principle, SPSS is imported by matrix A
21.0 do each index related analysis, obtain the correlation coefficient ρ of Raw performance between any two.
4. a kind of Regional Landslide method for evaluating hazard based on slopes unit and machine learning according to claim 3,
Characterized in that, in the step S3, the step of analysis of distribution of the evaluation index in the ramp unit to have come down
For:
S3.1, using GIS as platform, by the overlay analysis to landslide disaster point and ramp unit has occurred, obtain each evaluation and refer to
It is marked on the desired value of the ramp unit to have come down;
S3.2, using SPSS as platform, to evaluation index the ramp unit to have come down desired value carry out statistical analysis, obtain
To the histogram frequency distribution diagram or directional spreding radar map of each evaluation index;
S3.3, histogram frequency distribution diagram or directional spreding radar map according to each evaluation index, mark off each evaluation index not
With Landslide hazard grade corresponding to section, and the risk factor monotonicity in each section is determined, establish the cunning of each index respectively
Slope Hazard rank division table.
5. a kind of Regional Landslide method for evaluating hazard based on slopes unit and machine learning according to claim 4,
Characterized in that, in the step S4, the Landslide Hazard Assessment model based on LM-BP neutral nets is the step of foundation:
S4.1, neutral net master sample foundation:Master sample, including training sample and test sample, referred to evaluation
Mark according to certain mathematical method and Landslide Hazards degree of danger with the basis of the functional relation of Landslide, predicting
Evaluation index grade scale considers foundation;
The structure of training sample includes:Build empty matrix, structure input vector, structure output vector, by several evaluation indexes
Input vector and output vector merge into a matrix;
The establishment step of S4.2, LM-BP neural network model is followed successively by:Sample data is normalized using matlab, creates LM-
BP neural network, training LM-BP neutral nets, test LM-BP neutral nets and neural network forecast.
6. a kind of Regional Landslide method for evaluating hazard based on slopes unit and machine learning according to claim 5,
It is characterized in that, in the step S5, the step of Regional Landslide hazard assessment:
S5.1, on the basis of the LM-BP neutral nets established, by inputting master sample matrix, to LM-BP neutral nets
Repeatedly trained, until reaching default precision;
S5.2, LM-BP neutral nets are tested, to verify the generalization ability of LM-BP neutral nets;
S5.3, the good LM-BP neutral nets of generalization ability are preserved, it is pre- to carry out landslide hazard angle value to each evaluation unit in research area
Survey;
S5.4, to take equidistant method by Landslide hazard partition of the level be four grades, and landslide hazard angle value is in 0~0.25 scope
Hazard rank is I grade corresponding to interior, and corresponding Hazard rank is II grade in the range of 0.25~0.5, in the range of 0.5~0.75
Corresponding Hazard rank is III grade, and corresponding Hazard rank is IV grade in the range of 0.75~1.0;
S5.5, hazard assessment interpretation of result and checking:By statistical analysis, each grade hazardous area slope number, number are drawn
Percentage, area, area percentage statistical form, the distribution situation of each danger classes is drawn, by real with research area's disaster distribution
Whether the contrast of border situation, checking evaluation result meet the fact.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710506290.0A CN107463991A (en) | 2017-06-28 | 2017-06-28 | A kind of Regional Landslide method for evaluating hazard based on slopes unit and machine learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710506290.0A CN107463991A (en) | 2017-06-28 | 2017-06-28 | A kind of Regional Landslide method for evaluating hazard based on slopes unit and machine learning |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107463991A true CN107463991A (en) | 2017-12-12 |
Family
ID=60546372
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710506290.0A Pending CN107463991A (en) | 2017-06-28 | 2017-06-28 | A kind of Regional Landslide method for evaluating hazard based on slopes unit and machine learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107463991A (en) |
Cited By (30)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108229726A (en) * | 2017-12-15 | 2018-06-29 | 同济大学 | A kind of shiver stone side slope catastrophe dangerous situation early alarming and forecasting method |
CN109163998A (en) * | 2018-08-14 | 2019-01-08 | 重庆大学 | A kind of bank slope bumps degree division methods acted on for analyzing normal erosion |
CN109359738A (en) * | 2018-10-19 | 2019-02-19 | 西南交通大学 | A kind of Landslide hazard appraisal procedure based on QPSO-BP neural network |
CN109409719A (en) * | 2018-10-17 | 2019-03-01 | 西南石油大学 | A kind of lower long-distance oil & gas pipeline vulnerability assessment method of landslide effect |
CN109584510A (en) * | 2018-11-30 | 2019-04-05 | 中国公路工程咨询集团有限公司 | A kind of road landslide of high slope disaster alarm method based on valuation functions training |
CN109726902A (en) * | 2018-12-19 | 2019-05-07 | 天津理工大学 | A kind of Slope Stability Evaluation method |
CN110532872A (en) * | 2019-07-24 | 2019-12-03 | 宁德市公路局 | A kind of landslide hierarchy system and method based on convolution supporting vector neural network |
CN110569554A (en) * | 2019-08-13 | 2019-12-13 | 成都垣景科技有限公司 | Landslide susceptibility evaluation method based on spatial logistic regression and geographic detector |
CN110807399A (en) * | 2019-10-29 | 2020-02-18 | 北京师范大学 | Single-category support vector machine-based collapse and slide hidden danger point detection method |
CN110866318A (en) * | 2019-12-30 | 2020-03-06 | 桂林理工大学 | Dangerous rock instability mode discrimination method and device based on deep learning technology |
CN111046517A (en) * | 2019-08-13 | 2020-04-21 | 成都垣景科技有限公司 | Local scale landslide vulnerability prediction method based on hierarchical Bayesian method |
CN111046783A (en) * | 2019-12-09 | 2020-04-21 | 山西能源学院 | Slope geological disaster boundary extraction method for improving watershed algorithm |
CN111415065A (en) * | 2020-02-26 | 2020-07-14 | 广州地理研究所 | Mountain disaster ecological damage risk evaluation method based on action process |
CN111858803A (en) * | 2020-07-06 | 2020-10-30 | 东华理工大学 | Landslide land disaster risk zoning map generation method |
CN112036424A (en) * | 2020-04-30 | 2020-12-04 | 自然资源部第一海洋研究所 | Submarine landslide hazard analysis method based on unsupervised machine learning |
CN112132369A (en) * | 2019-06-24 | 2020-12-25 | 中国石油化工股份有限公司 | Petroleum geology evaluation index assignment method |
CN113034499A (en) * | 2021-05-24 | 2021-06-25 | 西南交通大学 | Slope surface danger area prediction method, device, equipment and readable storage medium |
CN113283817A (en) * | 2021-07-20 | 2021-08-20 | 光谷技术有限公司 | Disaster assessment method and system |
CN113378396A (en) * | 2021-06-22 | 2021-09-10 | 中国科学院、水利部成都山地灾害与环境研究所 | Early identification method for hidden danger points of small watershed geological disaster |
CN113408497A (en) * | 2021-08-03 | 2021-09-17 | 广东省科学院广州地理研究所 | Space scale selection method and device for mass landslide |
CN113536659A (en) * | 2021-06-09 | 2021-10-22 | 上海交通大学 | Method, system and storage medium for rapidly predicting post-earthquake road disaster area |
CN113591700A (en) * | 2021-07-30 | 2021-11-02 | 兰州大学 | Prediction method for potential landslide and river blockage |
CN113780741A (en) * | 2021-08-11 | 2021-12-10 | 中国地质调查局武汉地质调查中心 | Landslide risk evaluation method and system based on slope characteristics and storage medium |
CN114066165A (en) * | 2021-10-20 | 2022-02-18 | 国网黑龙江省电力有限公司电力科学研究院 | Improved power transmission line high-order landslide risk evaluation system and method |
CN114662950A (en) * | 2022-03-31 | 2022-06-24 | 中南大学 | Slope unit thinning method suitable for regional earthquake landslide risk assessment |
CN115162376A (en) * | 2022-07-28 | 2022-10-11 | 河海大学 | Landslide danger zoning method for landslide process probability analysis based on GIS platform |
CN116050120A (en) * | 2023-01-06 | 2023-05-02 | 中国自然资源航空物探遥感中心 | Landslide hidden danger activity remote sensing evaluation modeling method, system and storage medium |
CN116341901A (en) * | 2023-03-07 | 2023-06-27 | 重庆大学 | Integrated evaluation method for landslide surface domain-monomer hazard early warning |
CN113642849B (en) * | 2021-07-19 | 2023-10-20 | 北京师范大学 | Geological disaster risk comprehensive evaluation method and device considering spatial distribution characteristics |
CN117493805A (en) * | 2023-11-04 | 2024-02-02 | 广东省核工业地质调查院 | Grading and value-taking method for slope unit in geological disaster evaluation process |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140229420A1 (en) * | 2006-10-17 | 2014-08-14 | Corelogic Solutions, Llc | Systems and methods for quantifying flood risk |
CN105225046A (en) * | 2015-09-30 | 2016-01-06 | 武汉工程大学 | A kind of Regional Landslide sensitivity assessment data sampling method |
CN106408120A (en) * | 2016-09-13 | 2017-02-15 | 江苏大学 | Local region landslide prediction device and local region landslide prediction method |
-
2017
- 2017-06-28 CN CN201710506290.0A patent/CN107463991A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140229420A1 (en) * | 2006-10-17 | 2014-08-14 | Corelogic Solutions, Llc | Systems and methods for quantifying flood risk |
CN105225046A (en) * | 2015-09-30 | 2016-01-06 | 武汉工程大学 | A kind of Regional Landslide sensitivity assessment data sampling method |
CN106408120A (en) * | 2016-09-13 | 2017-02-15 | 江苏大学 | Local region landslide prediction device and local region landslide prediction method |
Non-Patent Citations (3)
Title |
---|
宫清华 等: ""基于斜坡单元的浅层滑坡风险区划-以华南松岗河小流域为例"", 《安全与环境学报》 * |
谭立霞: ""GIS支持下基于支持向量机的滑坡灾害危险性评价研究"", 《中国优秀硕士学位论文全文数据库 基础科学辑》 * |
高彩云: ""基于智能算法的滑坡位移预测与危险性评价研究"", 《中国优秀博士学位论文全文数据库 基础科学辑》 * |
Cited By (45)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108229726A (en) * | 2017-12-15 | 2018-06-29 | 同济大学 | A kind of shiver stone side slope catastrophe dangerous situation early alarming and forecasting method |
CN109163998A (en) * | 2018-08-14 | 2019-01-08 | 重庆大学 | A kind of bank slope bumps degree division methods acted on for analyzing normal erosion |
CN109163998B (en) * | 2018-08-14 | 2021-06-08 | 重庆大学 | Bank slope concave-convex degree dividing method for analyzing flowing water erosion effect |
CN109409719A (en) * | 2018-10-17 | 2019-03-01 | 西南石油大学 | A kind of lower long-distance oil & gas pipeline vulnerability assessment method of landslide effect |
CN109409719B (en) * | 2018-10-17 | 2022-07-15 | 西南石油大学 | Method for evaluating vulnerability of long-distance oil and gas pipeline under landslide action |
CN109359738A (en) * | 2018-10-19 | 2019-02-19 | 西南交通大学 | A kind of Landslide hazard appraisal procedure based on QPSO-BP neural network |
CN109584510A (en) * | 2018-11-30 | 2019-04-05 | 中国公路工程咨询集团有限公司 | A kind of road landslide of high slope disaster alarm method based on valuation functions training |
CN109584510B (en) * | 2018-11-30 | 2021-02-02 | 中国公路工程咨询集团有限公司 | Road high slope landslide hazard early warning method based on evaluation function training |
CN109726902A (en) * | 2018-12-19 | 2019-05-07 | 天津理工大学 | A kind of Slope Stability Evaluation method |
CN112132369A (en) * | 2019-06-24 | 2020-12-25 | 中国石油化工股份有限公司 | Petroleum geology evaluation index assignment method |
CN110532872A (en) * | 2019-07-24 | 2019-12-03 | 宁德市公路局 | A kind of landslide hierarchy system and method based on convolution supporting vector neural network |
CN110569554A (en) * | 2019-08-13 | 2019-12-13 | 成都垣景科技有限公司 | Landslide susceptibility evaluation method based on spatial logistic regression and geographic detector |
CN110569554B (en) * | 2019-08-13 | 2020-11-10 | 成都垣景科技有限公司 | Landslide susceptibility evaluation method based on spatial logistic regression and geographic detector |
CN111046517A (en) * | 2019-08-13 | 2020-04-21 | 成都垣景科技有限公司 | Local scale landslide vulnerability prediction method based on hierarchical Bayesian method |
CN110807399A (en) * | 2019-10-29 | 2020-02-18 | 北京师范大学 | Single-category support vector machine-based collapse and slide hidden danger point detection method |
CN111046783A (en) * | 2019-12-09 | 2020-04-21 | 山西能源学院 | Slope geological disaster boundary extraction method for improving watershed algorithm |
CN111046783B (en) * | 2019-12-09 | 2023-06-30 | 山西能源学院 | Slope geological disaster boundary extraction method for improving watershed algorithm |
CN110866318A (en) * | 2019-12-30 | 2020-03-06 | 桂林理工大学 | Dangerous rock instability mode discrimination method and device based on deep learning technology |
CN111415065A (en) * | 2020-02-26 | 2020-07-14 | 广州地理研究所 | Mountain disaster ecological damage risk evaluation method based on action process |
CN111415065B (en) * | 2020-02-26 | 2023-05-12 | 广州地理研究所 | Mountain disaster ecological damage risk evaluation method based on action process |
CN112036424B (en) * | 2020-04-30 | 2024-04-09 | 自然资源部第一海洋研究所 | Submarine landslide risk analysis method based on unsupervised machine learning |
CN112036424A (en) * | 2020-04-30 | 2020-12-04 | 自然资源部第一海洋研究所 | Submarine landslide hazard analysis method based on unsupervised machine learning |
CN111858803B (en) * | 2020-07-06 | 2021-09-21 | 东华理工大学 | Landslide land disaster risk zoning map generation method |
CN111858803A (en) * | 2020-07-06 | 2020-10-30 | 东华理工大学 | Landslide land disaster risk zoning map generation method |
CN113034499B (en) * | 2021-05-24 | 2021-09-21 | 西南交通大学 | Slope surface danger area prediction method, device, equipment and readable storage medium |
CN113034499A (en) * | 2021-05-24 | 2021-06-25 | 西南交通大学 | Slope surface danger area prediction method, device, equipment and readable storage medium |
CN113536659A (en) * | 2021-06-09 | 2021-10-22 | 上海交通大学 | Method, system and storage medium for rapidly predicting post-earthquake road disaster area |
CN113378396A (en) * | 2021-06-22 | 2021-09-10 | 中国科学院、水利部成都山地灾害与环境研究所 | Early identification method for hidden danger points of small watershed geological disaster |
CN113378396B (en) * | 2021-06-22 | 2023-11-24 | 中国科学院、水利部成都山地灾害与环境研究所 | Early identification method for small-basin geological disaster hidden danger points |
CN113642849B (en) * | 2021-07-19 | 2023-10-20 | 北京师范大学 | Geological disaster risk comprehensive evaluation method and device considering spatial distribution characteristics |
CN113283817B (en) * | 2021-07-20 | 2021-09-21 | 光谷技术有限公司 | Disaster assessment method and system |
CN113283817A (en) * | 2021-07-20 | 2021-08-20 | 光谷技术有限公司 | Disaster assessment method and system |
CN113591700A (en) * | 2021-07-30 | 2021-11-02 | 兰州大学 | Prediction method for potential landslide and river blockage |
CN113408497A (en) * | 2021-08-03 | 2021-09-17 | 广东省科学院广州地理研究所 | Space scale selection method and device for mass landslide |
CN113408497B (en) * | 2021-08-03 | 2021-11-09 | 广东省科学院广州地理研究所 | Space scale selection method and device for mass landslide |
CN113780741A (en) * | 2021-08-11 | 2021-12-10 | 中国地质调查局武汉地质调查中心 | Landslide risk evaluation method and system based on slope characteristics and storage medium |
CN113780741B (en) * | 2021-08-11 | 2024-01-23 | 中国地质调查局武汉地质调查中心 | Landslide risk evaluation method, system and storage medium based on slope characteristics |
CN114066165A (en) * | 2021-10-20 | 2022-02-18 | 国网黑龙江省电力有限公司电力科学研究院 | Improved power transmission line high-order landslide risk evaluation system and method |
CN114662950A (en) * | 2022-03-31 | 2022-06-24 | 中南大学 | Slope unit thinning method suitable for regional earthquake landslide risk assessment |
CN115162376A (en) * | 2022-07-28 | 2022-10-11 | 河海大学 | Landslide danger zoning method for landslide process probability analysis based on GIS platform |
CN116050120A (en) * | 2023-01-06 | 2023-05-02 | 中国自然资源航空物探遥感中心 | Landslide hidden danger activity remote sensing evaluation modeling method, system and storage medium |
CN116050120B (en) * | 2023-01-06 | 2023-09-01 | 中国自然资源航空物探遥感中心 | Landslide hidden danger activity remote sensing evaluation modeling method, system and storage medium |
CN116341901A (en) * | 2023-03-07 | 2023-06-27 | 重庆大学 | Integrated evaluation method for landslide surface domain-monomer hazard early warning |
CN116341901B (en) * | 2023-03-07 | 2023-09-12 | 重庆大学 | Integrated evaluation method for landslide surface domain-monomer hazard early warning |
CN117493805A (en) * | 2023-11-04 | 2024-02-02 | 广东省核工业地质调查院 | Grading and value-taking method for slope unit in geological disaster evaluation process |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107463991A (en) | A kind of Regional Landslide method for evaluating hazard based on slopes unit and machine learning | |
CN104346538B (en) | Earthquake hazard assessment method based on three kinds of the condition of a disaster factor control | |
Quan et al. | A systematic review of GIS-based local climate zone mapping studies | |
Romano et al. | Multi-criteria decision analysis for land suitability mapping in a rural area of Southern Italy | |
CN107943880B (en) | Geological disaster susceptibility improvement and evaluation method based on analytic hierarchy process | |
CN104298841B (en) | A kind of Flood Forecasting Method and system based on historical data | |
CN109584510B (en) | Road high slope landslide hazard early warning method based on evaluation function training | |
Roy et al. | Predicting land cover change in a Mediterranean catchment at different time scales | |
Quan et al. | GIS-based landslide susceptibility mapping using analytic hierarchy process and artificial neural network in Jeju (Korea) | |
CN106803223B (en) | A kind of mountain flood risk evaluating method based on system polymorphic theory | |
CN103971184B (en) | Transmission line of electricity path generating method based on GIS Spatial GISs | |
CN106651211A (en) | Different-scale regional flood damage risk evaluation method | |
CN107220754B (en) | County scale mountain torrent disaster risk assessment method | |
CN108304536A (en) | A kind of geographical environmental simulation of the geographical environmental element of coupling and predicting platform | |
CN107315722A (en) | Based on gram in gold and the hydrographic(al) network Optimized model that is combined of information entropy theory | |
CN109829584A (en) | A kind of mountain torrents risk dynamic assessment method | |
CN104732448B (en) | A kind of power network infrastructure Vulnerability Assessment | |
CN115689293B (en) | Urban waterlogging toughness assessment method based on pressure-state-response framework | |
CN107832982A (en) | One kind fits trip's index calculation method based on big data tourism trip assessment models region | |
CN103413036A (en) | Continuous forest fire weather level forecasting model and application thereof | |
CN107016205A (en) | A kind of multi-model construction method of groundwater Numerical Simulation | |
Chen et al. | The application of the genetic adaptive neural network in landslide disaster assessment | |
CN113191582B (en) | Road torrential flood vulnerability evaluation method based on GIS and machine learning | |
CN114417724A (en) | Simulation method for land use evolution of mountain city | |
Jam et al. | Raster-based landslide susceptibility mapping using compensatory MADM methods |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171212 |
|
RJ01 | Rejection of invention patent application after publication |