CN104636357A - Positioning method based on artificial neural network and centroid algorithm - Google Patents
Positioning method based on artificial neural network and centroid algorithm Download PDFInfo
- Publication number
- CN104636357A CN104636357A CN201310553275.3A CN201310553275A CN104636357A CN 104636357 A CN104636357 A CN 104636357A CN 201310553275 A CN201310553275 A CN 201310553275A CN 104636357 A CN104636357 A CN 104636357A
- Authority
- CN
- China
- Prior art keywords
- side slope
- finite element
- slope
- artificial neural
- neural network
- 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
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a positioning method based on an artificial neural network and a centroid algorithm. The positioning method includes the following steps that finite element analysis is conducted on a side slope and physical parameters are collected, and a three-dimensional finite element model of the side slope is obtained; a computer is used for calculating the location of the center of gravity of the side slope for the finite element model; the finite element analysis is conducted on a side slope to be detected, side slope data, related to the current side slope parameters, in a simulation database of a database are called, a forecasting function is trained by using the artificial neural network, and a mature objective function is obtained; a finite element and physical parameters of the side slope to be detected are substituted into the mature objective function, and a forecasting result of a current sliding displacement sequence is obtained. Compared with a traditional analytical method, the positioning method is higher and more scientific in analytical precision and the motion of the side slope is forecasted very conveniently.
Description
Technical field
The present invention relates to a kind of localization method based on artificial neural network and centroid algorithm.
Background technology
Slope deforming is by the various factors such as precipitation, underground water table, internal stress change, and cause the deformation mechanism of different side slope to be not quite similar, its deformation process has complicacy, randomness and uncertainty, and the Deformation Prediction of side slope remains a difficult problem.
Side slope is due to contour structures complexity, usually there is irregular profile and profile, when mechanical analysis, be usually difficult to the center of gravity that can calculate edge slope structure, with regard to being difficult to, the prediction of corresponding globality is done, for road safety brings great hidden danger to the movement tendency of side slope.
The kind of present side slope is relatively fixing, and namely physical property such as density is relative fixing with information such as materials, and the development of measuring technique, also can know the 3D shape of side slope accurately.
Summary of the invention
The present invention is directed to the proposition of above problem, and a kind of localization method based on artificial neural network and centroid algorithm of development, there are following steps:
-finite element analysis is carried out to side slope and gathers physical parameter, obtain the three-dimensional finite element model of side slope; Use computing machine to be described finite element model, calculate the position of side slope center of gravity;
-repeat above-mentioned steps, set up side slope center of gravity database;
-finite element analysis is carried out to side slope to be detected, according to the quantity of finite element and the physical parameter of side slope, set up the anticipation function of the center of gravity of current side slope;
Slope data relevant to current slope parameter in simulated database in-calling data storehouse, the anticipation function described in the training of end user's artificial neural networks, obtains ripe objective function;
-finite element of described side slope to be detected and physical parameter are brought into the objective function of described maturation, obtain predicting the outcome of current slide displacement sequence.
The current side slope center of gravity calculated by anticipation function is predicted the outcome, adopts centroid algorithm to carry out modified result.
In described database, slope data at least comprises: side slope kind, side slope size, edge slope structure and side slope material.
For the finite element unique point of side slope to be measured, use the method for interpolation, be similar to the finite element of approximate side slope in database.
According to analyzing the finite element drawing side slope to be detected, transferring slope data similar with it in a database, forming similar data set, use the objective function described in set of metadata of similar data set pair to train.
Owing to have employed technique scheme, a kind of localization method based on artificial neural network and centroid algorithm provided by the invention, by carrying out a large amount of analytical tests to existing side slope, forming database, finite element analysis is carried out to side slope to be detected, form target analysis function, function end user artificial neural networks is trained, finally obtains the centre of gravity place of side slope, compared to traditional analytical approach, there is higher more scientific analysis precision, very convenient prediction side slope motion.
Accompanying drawing explanation
In order to the technical scheme of clearer explanation embodiments of the invention or prior art, introduce doing one to the accompanying drawing used required in embodiment or description of the prior art simply below, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skill in the art, under the prerequisite not paying creative work, other accompanying drawing can also be obtained according to these accompanying drawings.
Fig. 1 is process flow diagram of the present invention
Embodiment
For making the object of embodiments of the invention, technical scheme and advantage clearly, below in conjunction with the accompanying drawing in the embodiment of the present invention, clear complete description is carried out to the technical scheme in the embodiment of the present invention:
As shown in Figure 1: a kind of localization method based on artificial neural network and centroid algorithm, mainly comprises the steps:
First, finite element analysis carried out to side slope and gathers physical parameter, obtaining the three-dimensional finite element model of side slope; Use computing machine to be described finite element model, calculate the position of side slope center of gravity.
Then, repeat above-mentioned steps, set up side slope center of gravity database.
Secondly, finite element analysis is carried out to side slope to be detected, according to the quantity of finite element and the physical parameter of side slope, sets up the anticipation function of the center of gravity of current side slope;
Slope data relevant to current slope parameter in simulated database in calling data storehouse, the anticipation function described in the training of end user's artificial neural networks, obtains ripe objective function.
Finally, the finite element of described side slope to be detected and physical parameter are brought into the objective function of described maturation, obtain predicting the outcome of current slide displacement sequence.
In order to further increase the precision of measuring and calculating, as a preferably embodiment, the current side slope center of gravity calculated by anticipation function being predicted the outcome, adopting centroid algorithm to carry out modified result.
In order to increase the precision of prediction, ensureing when having some samples, ensureing precision of prediction.As a preferably embodiment, in described database, slope data at least comprises: side slope kind, side slope size, edge slope structure and side slope material.
Consider, the finite element of side slope in real process, may not be corresponding with the finite element in database, for the finite element unique point of side slope to be measured, uses the method for interpolation, is similar to the finite element of approximate side slope in database.
In order to reduce operand, as a preferably embodiment, according to analyzing the finite element drawing side slope to be detected, transferring slope data similar with it in a database, forming similar data set, use the objective function described in set of metadata of similar data set pair to train.
The above; be only the present invention's preferably embodiment; but protection scope of the present invention is not limited thereto; anyly be familiar with those skilled in the art in the technical scope that the present invention discloses; be equal to according to technical scheme of the present invention and inventive concept thereof and replace or change, all should be encompassed within protection scope of the present invention.
Claims (5)
1., based on a localization method for artificial neural network and centroid algorithm, there are following steps:
-finite element analysis is carried out to side slope and gathers physical parameter, obtain the three-dimensional finite element model of side slope; Use computing machine to be described finite element model, calculate the position of side slope center of gravity;
-repeat above-mentioned steps, set up side slope center of gravity database;
-finite element analysis is carried out to side slope to be detected, according to the quantity of finite element and the physical parameter of side slope, set up the anticipation function of the center of gravity of current side slope;
Slope data relevant to current slope parameter in simulated database in-calling data storehouse, the anticipation function described in the training of end user's artificial neural networks, obtains ripe objective function;
-finite element of described side slope to be detected and physical parameter are brought into the objective function of described maturation, obtain predicting the outcome of current slide displacement sequence.
2. a kind of localization method based on artificial neural network and centroid algorithm according to claim 1, is further characterized in that: predict the outcome for the current side slope center of gravity calculated by anticipation function, adopts centroid algorithm to carry out modified result.
3. a kind of localization method based on artificial neural network and centroid algorithm according to claim 1, is further characterized in that: in described database, slope data at least comprises: side slope kind, side slope size, edge slope structure and side slope material.
4. a kind of localization method based on artificial neural network and centroid algorithm according to claim 1, is further characterized in that: for the finite element unique point of side slope to be measured, uses the method for interpolation, is similar to the finite element of approximate side slope in database.
5. a kind of localization method based on artificial neural network and centroid algorithm according to claim 3, be further characterized in that: according to analyzing the finite element drawing side slope to be detected, transfer slope data similar with it in a database, form similar data set, use the objective function described in set of metadata of similar data set pair to train.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310553275.3A CN104636357A (en) | 2013-11-07 | 2013-11-07 | Positioning method based on artificial neural network and centroid algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310553275.3A CN104636357A (en) | 2013-11-07 | 2013-11-07 | Positioning method based on artificial neural network and centroid algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104636357A true CN104636357A (en) | 2015-05-20 |
Family
ID=53215128
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310553275.3A Pending CN104636357A (en) | 2013-11-07 | 2013-11-07 | Positioning method based on artificial neural network and centroid algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104636357A (en) |
-
2013
- 2013-11-07 CN CN201310553275.3A patent/CN104636357A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102708227B (en) | SPH (smoothed particle hydrodynamics) algorithm-based simulation method and simulation system of process of breaking dam by flood | |
| CN105550199B (en) | A kind of point polymerization and device based on multi-source map | |
| CN103942837B (en) | The direct building method of blade point cloud model cross section curve based on Successive linear programming | |
| CN107301648B (en) | Redundant point cloud removing method based on overlapping area boundary angle | |
| RU2012102394A (en) | METHOD FOR CALCULATING PHYSICAL VALUES, METHOD FOR NUMERICAL ANALYSIS, PROGRAM FOR CALCULATING PHYSICAL VALUES, PROGRAM FOR NUMERICAL ANALYSIS, DEVICE FOR CALCULATING PHYSICAL VALUES AND DEVICES FOR NUMERICAL ANALYSIS | |
| CN105160700B (en) | A kind of cross section curve reconstructing method for reconstructing three-dimensional model | |
| CN106612495B (en) | A kind of indoor orientation method and system based on propagation loss study | |
| CN105205864B (en) | Geologic structure face threedimensional model method for automatic modeling and system based on multi-source data | |
| CN107221028B (en) | A kind of geologic body closed surface three-dimensional rebuilding method based on seismic interpretation data | |
| CN104408760A (en) | Binocular-vision-based high-precision virtual assembling system algorithm | |
| CN108280852A (en) | A kind of door and window point cloud shape detecting method and system based on laser point cloud data | |
| GB2536123A (en) | Integrated oilfield asset modeling using multiple resolutions of reservoir detail | |
| CN107340534A (en) | A kind of observation system shot point Optimization Design based on dimensional topography gradient slope aspect | |
| CN108256218A (en) | A kind of subterranean communication tunnel fine modeling method based on actual measurement stringcourse data | |
| CN108287974A (en) | Coupling evaluation method towards land use change survey Cellular Automata Simulation precision | |
| CN105045958A (en) | Implementation system and method of GPS (Global Positioning System) elevation fitting on the basis of BP (Back Propagation) neural network | |
| CN104636510A (en) | Positioning method based on ant colony optimization algorithm and centroid algorithm | |
| CN104636508A (en) | Positioning method based on particle swarm optimization algorithm and centroiding algorithm | |
| CN104636575A (en) | Positioning method based on simulated annealing algorithm and centroiding algorithm | |
| CN104636350A (en) | Positioning method based on harmony search algorithm and centroiding algorithm | |
| CN104636357A (en) | Positioning method based on artificial neural network and centroid algorithm | |
| CN104636507A (en) | Positioning method based on genetic algorithm and centroiding algorithm | |
| Tan et al. | An integrated optimization algorithm for parameter structure identification in groundwater modeling | |
| CN104279956A (en) | Determination method for rock structural surface reference plane | |
| CN110717290B (en) | Method for drawing dam contour line based on three-time interpolation method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20150520 |