CN110378070A - Based on PSO-SVR and the united Prediction of Displacement in Landslide method of DES - Google Patents
Based on PSO-SVR and the united Prediction of Displacement in Landslide method of DES Download PDFInfo
- Publication number
- CN110378070A CN110378070A CN201910715327.XA CN201910715327A CN110378070A CN 110378070 A CN110378070 A CN 110378070A CN 201910715327 A CN201910715327 A CN 201910715327A CN 110378070 A CN110378070 A CN 110378070A
- Authority
- CN
- China
- Prior art keywords
- displacement
- svr
- landslide
- value
- time series
- 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
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B21/00—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant
- G01B21/02—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant for measuring length, width, or thickness
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B21/00—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant
- G01B21/32—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant for measuring the deformation in a solid
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/20—Ensemble learning
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Pit Excavations, Shoring, Fill Or Stabilisation Of Slopes (AREA)
- Testing Or Calibration Of Command Recording Devices (AREA)
Abstract
The invention discloses one kind based on PSO-SVR and the united Prediction of Displacement in Landslide method of DES, it includes the displacement monitoring time series and multiple initial effects factors for obtaining monitoring point for displacement at landslide, it is two components: periodic term and trend term by displacement monitoring Time Series using the random noise of Wavelet noise-eliminating method removal displacement monitoring time series, and using HP filter.On this basis, using water level and rainfall as impact factor, landslide displacement main affecting factors feature is extracted using Principal Component Analysis, the hybrid predicting Optimized model for establishing a kind of joint population optimizing support vector regression (PSO-SVR) and double exponential smoothings (DES) realizes the displacement prediction on landslide by constructing periodic term and trend term training sample component respectively.Landslide total displacement predicted value finally is can be obtained into trend term and season displacement prediction value superposition.
Description
Technical field
The present invention relates to Geological Hazards Monitoring fields, and in particular to one kind is based on PSO-SVR and the united landslide displacement of DES
Prediction technique.
Background technique
It comes down as a kind of common geological disaster, usually by a variety of uncertain factors, such as geological conditions, landforms, the hydrology
The collective effects such as geology and physical factor and mankind's activity cause.Landslide Deformation forecast is always the important of landslide early-warning and predicting
Research direction and research hotspot.Landslide Deformation develops to be influenced by Seasonal (such as water level of yangtze river scheduling, periodical heavy rainfall)
When, accumulative displacement curve typically exhibits very strong Nonlinear Dynamical Characteristics, such as sudden transformation.Conventional method is to such cunning
When slope carries out prediction, easily the deformation behaviour of step evolution is mistakenly considered to come down to have entered to face the sliding stage, causes to judge by accident.So
Say that traditional analysis method is difficult to make accurate evaluation to the stability on landslide.
Summary of the invention
It is provided by the invention pre- based on the united landslide displacement of PSO-SVR and DES for above-mentioned deficiency in the prior art
Survey method can accurately predict the displacement on landslide.
In order to achieve the above object of the invention, the technical solution adopted by the present invention are as follows:
One kind is provided based on PSO-SVR and the united Prediction of Displacement in Landslide method of DES comprising:
Obtain the displacement monitoring time series and multiple initial effects factors of monitoring point for displacement at landslide；
Displacement monitoring time sequence is denoised using Wavelet noise-eliminating method, and is normalized, is used later
Displacement monitoring Time Series are that periodic term displacement and trend term are displaced by HP filter；
DES algorithm anticipation trend item shift value is passed through using the displacement monitoring time series of set period of time；
Characteristics extraction is carried out to multiple initial effects factors using principal component analytical method, and chooses and is greater than given threshold
The characteristic value character pair vector sum initial effects factor constitute main affecting factors；
According to particle swarm optimization algorithm, the ginseng of optimal SVR Radial basis kernel function is obtained by the speed of more new particle
Number γ and penalty factor；
Anti-normalization processing is carried out to trend term predicted value and periodic term predicted value, is added obtains final prediction of coming down later
Displacement.
The invention has the benefit that this programme extracts master by the linear relationship between principal component analysis impact factor
Composition characteristics, reduce data redundancy to it is related, improve the quality of data of impact factor, use double smoothing anticipation trend item
Displacement can eliminate the unstability of fitting result.
Detailed description of the invention
Fig. 1 is the flow chart based on PSO-SVR Yu the united Prediction of Displacement in Landslide method of DES.
Fig. 2 is DES trend term prediction result.
Fig. 3 is PSO-SVR periodic term prediction result.
Fig. 4 is landslide accumulative displacement prediction result.
Specific embodiment
A specific embodiment of the invention is described below, in order to facilitate understanding by those skilled in the art this hair
It is bright, it should be apparent that the present invention is not limited to the ranges of specific embodiment, for those skilled in the art,
As long as various change is in the spirit and scope of the present invention that the attached claims limit and determine, these variations are aobvious and easy
See, all are using the innovation and creation of present inventive concept in the column of protection.
The flow chart based on PSO-SVR Yu the united Prediction of Displacement in Landslide method of DES is shown with reference to Fig. 1, Fig. 1；Such as Fig. 1
Shown, this method 100 includes step 101 to step 106.
In a step 101, the displacement monitoring time series and multiple initial effects factors of monitoring point for displacement at landslide are obtained；
The plurality of initial effects factor includes the reservoir level of the moon, Reservoir Water Level amplitude, reservoir level where displacement monitoring time series
The accumulated rainfall of rate of change, rainfall and displacement monitoring time series the first two months.
In a step 102, displacement monitoring time sequence is denoised using Wavelet noise-eliminating method, and place is normalized
Reason uses HP filter (Hodrick-Prescott filter) by displacement monitoring Time Series for periodic term displacement later
It is displaced with trend term；
After this programme passes through Wavelet Denoising Method, the random noise in displacement monitoring time series can be removed, is used for improving
The accuracy of Prediction of Displacement in Landslide initial data；Time series is considered as the superposition of different frequency ingredient, and HP filter will become
Change the smooth sequence in indefinite time series data with certain variation tendency to separate, time series is divided into periodicity
Fluctuate data and trend factor data.
This programme carries out the decomposition of time series by HP filter, settles at one go, does not need to introduce multiple methods independent
It is handled, decomposition step is simplified, while avoiding the error introduced in multiple method treatment processes.
In step 103, DES algorithm (double smoothing is passed through using the displacement monitoring time series of set period of time
Method) anticipation trend item shift value；Set period of time herein is that the displacement monitoring time series of monitoring point for displacement at landslide is former
A month data, for example the displacement monitoring time series obtained is August, then set period of time is assumed to be 5 months, then setting
The displacement monitoring time series for section of fixing time is the data in 3,4,5,6 and July.
When implementation, this programme preferably uses the displacement monitoring time series of set period of time to pass through DES algorithm anticipation trend
Item shift value further comprises:
History before acquisition displacement monitoring time series in set period of time is displaced monitoring time sequence；
Using history displacement monitoring time sequence calculate anticipation trend item displacement previous moment single exponential smoothing value and
Double smoothing value:
Wherein, α and β is coefficent of exponential smoothing；WithThe respectively primary and secondary exponential smoothing value of t moment；WithThe respectively primary and secondary exponential smoothing value at t-1 moment；xtFor the landslide displacement time series of t moment；
According to trend term displacement and its single exponential smoothing value and double smoothing value of previous moment, trend term is calculated
It is displaced corresponding trend term predicted value:
Wherein, atAnd btRespectively intermediate parameters；T is period, F1+TFor the predicted value after trend term displacement when cycle T.
This programme calculates its primary and two by the primary and secondary exponential smoothing value of historical time sequence last moment
Secondary exponential smoothing value, and using its primary and secondary exponential smoothing value as the primary and secondary of update subsequent time time series
The basic data of exponential smoothing value, and so on, until calculate the time series of displacement monitoring time series previous moment
Primary and secondary exponential smoothing value, and primary and secondary exponential smoothing value of displacement monitoring time series is calculated with this, and use
The smooth value calculates trend term predicted value.
This programme carries out trend term predictor calculation using DES algorithm, when carrying out smooth value update, there is sample (to go through
History time series) the advantages that required amount is less, calculating is simple, adaptability is relatively strong and result is more stable.
At step 104, feature extraction is carried out to multiple initial effects factors using principal component analytical method, and chosen big
Main affecting factors are constituted in the characteristic value character pair vector sum initial effects factor of given threshold.
This programme is mutual landslide displacement impact factor (come down nearby river level, moon rainfall) by principal component analysis
Between complex relationship carry out simplifying processing, data redundancy can be solved under the premise of retaining original variable information, elimination variable
Between correlativity.
In one embodiment of the invention, characteristic value is carried out to multiple initial effects factors using principal component analytical method
It extracts, and chooses the characteristic value character pair vector sum initial effects factor composition number of principal components for being greater than given threshold according to further
Include:
Average value is gone to handle each initial effects factor, data press row processing composition matrix by treated later；This
Average value in step refers to the average value obtained after all initial effects factors additions divided by initial effects factor sum.
The covariance matrix of calculating matrix, and using Eigenvalues Decomposition method ask covariance matrix characteristic value and feature to
Amount；
Characteristic value is sorted from large to small, and choose be greater than the corresponding feature vector of characteristic value of given threshold as row to
Measure constitutive characteristic vector matrix；
Main affecting factors Y, Y=PX are obtained using matrix X and eigenvectors matrix P.
In step 105, it is displaced according to main affecting factors and periodic term, using particle swarm algorithm and SVR regression forecasting
The periodic term predicted value of model calculating cycle item displacement；
In an implementation of the invention in example, it is displaced according to main affecting factors and periodic term, using particle swarm algorithm
Periodic term predicted value with the displacement of SVR regressive prediction model calculating cycle item further comprises:
According to particle swarm optimization algorithm, the ginseng of optimal SVR Radial basis kernel function is obtained by the speed of more new particle
Number γ and penalty factor:
vi=w × vi+c1×rand()×(pbesti-xi)+c2×rand()×(gbesti-xi), xi=xi+vi
Wherein, viFor the present speed of main affecting factors, xiFor main affecting factors current location, w is inertial factor,
Rand () is randomly generated test problems, and generation value is between (0-1), pbestiAnd gbestiRespectively the optimal position of history and it is current most
Excellent position, c1And c2For Studying factors；The parameter γ and penalty factor of SVR Radial basis kernel function are particle (main affecting factors)
Optimal location.
According to the parameter and penalty factor of Radial basis kernel function, using the displacement of SVR regressive prediction model calculating cycle item
Periodic term predicted value, SVR regressive prediction model are as follows:
Wherein, F is periodic term predicted value；σiFor i-th of dual variable；For i-th of positive real number dual variable；C is to punish
Penalty factor；K(xi,xj) it is Radial basis kernel function；B is amount of bias；γ is the parameter of Radial basis kernel function；xiFor and xjBe with
Machine sample point；L is no appearance or manner suggestive of abject poverty；γ is the parameter of SVR Radial basis kernel function.
In step 106, anti-normalization processing is carried out to trend term predicted value and periodic term predicted value, is added obtains later
Come down final predictive displacement.
Below by taking the landslide of plain boiled water river as an example, Prediction of Displacement in Landslide effect is illustrated using the method that this programme provides:
Select in July, 2003 to the observation data of 86 groups of plain boiled water rivers landslide professional monitoring point ZG118 between in August, 2010 be to instruct
Practice sample (to be mainly used for calculating the time series of displacement monitoring time series previous moment during trend term predictor calculation
Primary and secondary exponential smoothing value), and (these tests are predicted using in September, 2010 to 30 groups of numbers between 2 months 2013 as test sample
Landslide displacement at sample), the Landslide Prediction displacement of each test sample is calculated.
When carrying out characteristics extraction to multiple initial effects factors using principal component analytical method, if characteristic value is smaller,
Then mean that the explanation dynamics of this principal component is less than the average explanation dynamics of original variable, this programme selects characteristic value to be greater than 1
Characteristic value；Principal component analysis is carried out to original displacement monitoring data and initial effects variable factors, obtains 2 principal components, is tied
Fruit such as table 1.One or two principal component reaches 82.37% to the accumulative variance contribution ratio of initial data, shows the two principal component packets
The most information of initial data is contained.
The population variance that table 1 is explained
In trend term displacement prediction, this programme smoothing factor α is chosen for the trend term prediction of 0.99,30 groups of test samples
Value is shown in Fig. 2.It when periodic term displacement prediction, is programmed using MATLAB and realizes PSO parameter optimization, obtain SVR Radial basis kernel function ginseng
Number γ=0.65938, C=2.1656.According to optimal model parameters, the periodic term for calculating 30 groups of test samples of test sample is pre-
Measured value is shown in Fig. 3；The trend term predicted value of test sample and periodic term predicted value are subjected to anti-normalization processing, addition is always tired out
Displacement result is counted, as shown in Figure 4.
Wherein, the root-mean-square error of plain boiled water river Landslide Prediction result is 3.4301mm, related coefficient 0.99846, prediction
Precision is 99.99195%, it was confirmed that this programme provide method predictive ability and reliability, surface its be more suitable it is non-linear
The prediction of landslide displacement.
Claims (5)
1. based on PSO-SVR and the united Prediction of Displacement in Landslide method of DES characterized by comprising
Obtain the displacement monitoring time series and multiple initial effects factors of monitoring point for displacement at landslide；
Displacement monitoring time sequence is denoised using Wavelet noise-eliminating method, and is normalized, is filtered later using HP
Displacement monitoring Time Series are that periodic term displacement and trend term are displaced by wave device；
DES algorithm anticipation trend item shift value is passed through using the displacement monitoring time series of set period of time；
Feature extraction is carried out to multiple initial effects factors using principal component analytical method, and chooses the feature for being greater than given threshold
It is worth the character pair vector sum initial effects factor and constitutes main affecting factors；
According to particle swarm optimization algorithm, obtained by the speed of more new particle optimal SVR Radial basis kernel function parameter and
Penalty factor；
Anti-normalization processing is carried out to trend term predicted value and periodic term predicted value, is added obtains the final prediction bits that come down later
It moves.
2. according to claim 1 be based on PSO-SVR and the united Prediction of Displacement in Landslide method of DES, which is characterized in that institute
It states and further comprises by DES algorithm anticipation trend item shift value using the displacement monitoring time series of set period of time:
History before acquisition displacement monitoring time series in set period of time is displaced monitoring time sequence；
The anticipation trend item displacement single exponential smoothing value of previous moment and secondary is calculated using history displacement monitoring time sequence
Exponential smoothing value:
Wherein, α and β is coefficent of exponential smoothing；WithThe respectively primary and secondary exponential smoothing value of t moment；WithThe respectively primary and secondary exponential smoothing value at t-1 moment；xtFor the landslide displacement time series of t moment；
According to trend term displacement and its single exponential smoothing value and double smoothing value of previous moment, trend term displacement is calculated
Corresponding trend term predicted value:
Wherein, atAnd btRespectively intermediate parameters；T is period, F1+TFor the predicted value after trend term displacement when cycle T.
3. according to claim 1 be based on PSO-SVR and the united Prediction of Displacement in Landslide method of DES, which is characterized in that more
A initial effects factor includes the reservoir level of the moon where displacement monitoring time series, Reservoir Water Level amplitude, Reservoir Water Level speed
The accumulated rainfall of rate, rainfall and displacement monitoring time series the first two months.
4. according to claim 3 be based on PSO-SVR and the united Prediction of Displacement in Landslide method of DES, which is characterized in that institute
It states and feature extraction is carried out to multiple initial effects factors using principal component analytical method, and choose the characteristic value for being greater than given threshold
The character pair vector sum initial effects factor constitutes number of principal components evidence:
Average value is gone to handle each initial effects factor, data press row processing composition matrix by treated later；
The covariance matrix of calculating matrix, and use Eigenvalues Decomposition method seeks the eigen vector of covariance matrix；
Characteristic value is sorted from large to small, and chooses the corresponding feature vector of characteristic value for being greater than given threshold as row vector structure
At eigenvectors matrix；
Main affecting factors Y, Y=PX are obtained using matrix X and eigenvectors matrix P.
5. according to claim 1 be based on PSO-SVR and the united Prediction of Displacement in Landslide method of DES, which is characterized in that institute
It states according to main affecting factors and periodic term displacement, is displaced using particle swarm algorithm and SVR regressive prediction model calculating cycle item
Periodic term predicted value further comprise:
According to particle swarm optimization algorithm, obtained by the speed of more new particle optimal SVR Radial basis kernel function parameter and
Penalty factor:
vi=w × vi+c1×rand()×(pbesti-xi)+c2×rand()×(gbesti-xi), xi=xi+vi
Wherein, viFor the present speed of main affecting factors, xiFor main affecting factors current location, w is inertial factor, rand
() is randomly generated test problems, and generation value is between (0-1), pbestiAnd gbestiThe respectively optimal position of history and current optimal
Position, c1And c2For Studying factors；
According to the parameter and penalty factor of optimal SVR Radial basis kernel function, using SVR regressive prediction model calculating cycle item position
The periodic term predicted value of shifting, the SVR regressive prediction model are as follows:
Wherein, F is periodic term predicted value；σiFor i-th of dual variable；For i-th of positive real number dual variable；C be punishment because
Son；K(xi,xj) it is Radial basis kernel function；B is amount of bias；γ is the parameter of Radial basis kernel function；xiFor and xjIt is with press proof
This point；L is no appearance or manner suggestive of abject poverty；γ is the parameter of SVR Radial basis kernel function.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910715327.XA CN110378070A (en) | 2019-08-05 | 2019-08-05 | Based on PSO-SVR and the united Prediction of Displacement in Landslide method of DES |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910715327.XA CN110378070A (en) | 2019-08-05 | 2019-08-05 | Based on PSO-SVR and the united Prediction of Displacement in Landslide method of DES |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110378070A true CN110378070A (en) | 2019-10-25 |
Family
ID=68257936
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910715327.XA Pending CN110378070A (en) | 2019-08-05 | 2019-08-05 | Based on PSO-SVR and the united Prediction of Displacement in Landslide method of DES |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110378070A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111241633A (en) * | 2020-01-20 | 2020-06-05 | 中国人民解放军国防科技大学 | Chopper residual life prediction method based on principal component analysis and double-exponential model |
| CN111337923A (en) * | 2020-04-10 | 2020-06-26 | 中国水利水电第四工程局有限公司 | Method for establishing landslide deformation time course model through time sequence InSAR data |
| CN111695487A (en) * | 2020-06-09 | 2020-09-22 | 河海大学 | Hydrodynamic landslide displacement prediction method of hybrid intelligent model |
| CN111914395A (en) * | 2020-06-30 | 2020-11-10 | 河海大学 | High arch dam valley amplitude deformation prediction analysis method based on ARIMA-GC-SVR |
| CN112149285A (en) * | 2020-08-31 | 2020-12-29 | 中国地质大学（武汉） | Landslide prediction method based on optimization parameter selection |
| CN112270400A (en) * | 2020-10-16 | 2021-01-26 | 西安工程大学 | Landslide displacement dynamic prediction method based on multiple influence factors |
| CN112668606A (en) * | 2020-12-02 | 2021-04-16 | 北京理工大学 | Step type landslide displacement prediction method based on gradient elevator and quadratic programming |
| CN112862069A (en) * | 2021-01-21 | 2021-05-28 | 西北大学 | Landslide displacement prediction method based on SVR-LSTM mixed deep learning |
| CN113312845A (en) * | 2021-05-31 | 2021-08-27 | 中国水利水电科学研究院 | Pressure measuring pipe water level prediction method of earth and rockfill dam infiltration line based on PSO-SVR |
| CN113627066A (en) * | 2021-08-03 | 2021-11-09 | 成都理工大学 | Displacement prediction method for reservoir bank landslide |
| CN114001703A (en) * | 2021-10-09 | 2022-02-01 | 四川轻化工大学 | Landslide deformation data real-time filtering method |
| CN114511150A (en) * | 2022-02-16 | 2022-05-17 | 成都理工大学 | Landslide displacement space-time prediction method based on deep learning |
| CN118503657A (en) * | 2024-07-18 | 2024-08-16 | 北京师范大学珠海校区 | Landslide risk dynamic prediction method, landslide risk dynamic prediction device, landslide risk dynamic prediction medium and electronic equipment |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105239608A (en) * | 2015-09-28 | 2016-01-13 | 中国地质大学（武汉） | Landslide displacement prediction method based on wavelet transform-rough set-support vector regression (WT-RS-SVR) combination |
| CN106845544A (en) * | 2017-01-17 | 2017-06-13 | 西北农林科技大学 | A kind of stripe rust of wheat Forecasting Methodology based on population Yu SVMs |
| CN109992847A (en) * | 2019-03-14 | 2019-07-09 | 桂林电子科技大学 | A kind of Prediction of Displacement in Landslide method of hybrid machine learning model |
-
2019
- 2019-08-05 CN CN201910715327.XA patent/CN110378070A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105239608A (en) * | 2015-09-28 | 2016-01-13 | 中国地质大学（武汉） | Landslide displacement prediction method based on wavelet transform-rough set-support vector regression (WT-RS-SVR) combination |
| CN106845544A (en) * | 2017-01-17 | 2017-06-13 | 西北农林科技大学 | A kind of stripe rust of wheat Forecasting Methodology based on population Yu SVMs |
| CN109992847A (en) * | 2019-03-14 | 2019-07-09 | 桂林电子科技大学 | A kind of Prediction of Displacement in Landslide method of hybrid machine learning model |
Non-Patent Citations (1)
| Title |
|---|
| 彭令 等: "基于核主成分分析和粒子群优化支持向量机的滑坡位移预测", 《武汉大学学报.信息科学版》 * |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111241633A (en) * | 2020-01-20 | 2020-06-05 | 中国人民解放军国防科技大学 | Chopper residual life prediction method based on principal component analysis and double-exponential model |
| CN111241633B (en) * | 2020-01-20 | 2024-05-31 | 中国人民解放军国防科技大学 | Chopper residual life prediction method based on principal component analysis and double-index model |
| CN111337923A (en) * | 2020-04-10 | 2020-06-26 | 中国水利水电第四工程局有限公司 | Method for establishing landslide deformation time course model through time sequence InSAR data |
| CN111695487A (en) * | 2020-06-09 | 2020-09-22 | 河海大学 | Hydrodynamic landslide displacement prediction method of hybrid intelligent model |
| CN111914395A (en) * | 2020-06-30 | 2020-11-10 | 河海大学 | High arch dam valley amplitude deformation prediction analysis method based on ARIMA-GC-SVR |
| CN111914395B (en) * | 2020-06-30 | 2022-11-08 | 河海大学 | High arch dam valley amplitude deformation prediction analysis method based on ARIMA-GC-SVR |
| CN112149285A (en) * | 2020-08-31 | 2020-12-29 | 中国地质大学（武汉） | Landslide prediction method based on optimization parameter selection |
| CN112270400A (en) * | 2020-10-16 | 2021-01-26 | 西安工程大学 | Landslide displacement dynamic prediction method based on multiple influence factors |
| CN112668606B (en) * | 2020-12-02 | 2022-07-08 | 北京理工大学 | Step type landslide displacement prediction method based on gradient elevator and quadratic programming |
| CN112668606A (en) * | 2020-12-02 | 2021-04-16 | 北京理工大学 | Step type landslide displacement prediction method based on gradient elevator and quadratic programming |
| CN112862069A (en) * | 2021-01-21 | 2021-05-28 | 西北大学 | Landslide displacement prediction method based on SVR-LSTM mixed deep learning |
| CN112862069B (en) * | 2021-01-21 | 2023-09-05 | 西北大学 | Landslide Displacement Prediction Method Based on SVR-LSTM Hybrid Deep Learning |
| CN113312845A (en) * | 2021-05-31 | 2021-08-27 | 中国水利水电科学研究院 | Pressure measuring pipe water level prediction method of earth and rockfill dam infiltration line based on PSO-SVR |
| CN113627066A (en) * | 2021-08-03 | 2021-11-09 | 成都理工大学 | Displacement prediction method for reservoir bank landslide |
| CN114001703A (en) * | 2021-10-09 | 2022-02-01 | 四川轻化工大学 | Landslide deformation data real-time filtering method |
| CN114001703B (en) * | 2021-10-09 | 2023-07-28 | 四川轻化工大学 | Landslide deformation data real-time filtering method |
| CN114511150A (en) * | 2022-02-16 | 2022-05-17 | 成都理工大学 | Landslide displacement space-time prediction method based on deep learning |
| CN114511150B (en) * | 2022-02-16 | 2022-09-13 | 成都理工大学 | Landslide displacement space-time prediction method based on deep learning |
| CN118503657A (en) * | 2024-07-18 | 2024-08-16 | 北京师范大学珠海校区 | Landslide risk dynamic prediction method, landslide risk dynamic prediction device, landslide risk dynamic prediction medium and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110378070A (en) | Based on PSO-SVR and the united Prediction of Displacement in Landslide method of DES | |
| CN108399248A (en) | A kind of time series data prediction technique, device and equipment | |
| CN105139093A (en) | Method for forecasting flood based on Boosting algorithm and support vector machine | |
| CN107505837A (en) | A kind of semi-supervised neural network model and the soft-measuring modeling method based on the model | |
| CN106909756A (en) | A kind of rolling bearing method for predicting residual useful life | |
| CN104182474A (en) | Method for recognizing pre-churn users | |
| CN106529185B (en) | A kind of combination forecasting method and system of ancient building displacement | |
| CN111783516A (en) | Ploughing quality natural grade evaluation method based on deep learning | |
| CN108053094A (en) | A kind of weight grade evaluation method and system | |
| CN112668606B (en) | Step type landslide displacement prediction method based on gradient elevator and quadratic programming | |
| CN103942433A (en) | Building settlement prediction method based on historical data analysis | |
| CN112288137A (en) | LSTM short-term load prediction method and device considering electricity price and Attention mechanism | |
| CN106706508A (en) | AHP (analytic hierarchy process)-based metal material seawater corrosion sensitivity evaluation method | |
| CN108830405B (en) | Real-time power load prediction system and method based on multi-index dynamic matching | |
| Prakash et al. | Improved higher lead time river flow forecasts using sequential neural network with error updating | |
| CN115640888A (en) | Yield prediction method of decreasing function embedded threshold sequence network | |
| CN106647285A (en) | Catalyst activity detecting method based on soft measurement technology | |
| Harkushenko et al. | Analysis of economic and mathematical models of information and communication technology (ICT) effect on the production output: Does the solow paradox exist? | |
| CN110322351A (en) | Multi-source driving quantization investment model under Depth Stratification strategy | |
| Prastyo et al. | Survival analysis of companies’ delisting time in Indonesian stock exchange using Bayesian multiple-period logit approach | |
| CN117370759A (en) | Gas consumption prediction method based on artificial intelligence | |
| CN117035155A (en) | Water quality prediction method | |
| CN115511341B (en) | Method and device for evaluating time-varying failure probability of reservoir bank slope | |
| CN116757545A (en) | Multi-stage manufacturing system quality prediction method based on multi-task deep learning | |
| CN116298190A (en) | Screening and determining method for geochemical investigation index elements |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191025 |