Support vector machine for 3D modelling from sparse geological information of various origins

Tan

Sci. China Ser. D-Earth Sci.

et al. 2009

Geological body structure is the product of the geological evolution in the time dimension, which is presented in 3D configuration in the natural world. However, many geologists still record and process their geological data using the 2D or 1D pattern, which results in the loss of a large quantity of spatial data. One of the reasons is that the current methods have limitations on how to express underground geological objects. To analyze and interpret geological models, we present a layer data model to organize different kinds of geological datasets. The data model implemented the unification expression and storage of geological data and geometric models. In addition, it is a method for visualizing large-scaled geological datasets through building multi-resolution geological models rapidly, which can meet the demand of the operation, analysis, and interpretation of 3D geological objects. It proves that our methodology is competent for 3D modeling and self-adaptive visualization of large geological objects and it is a good way to solve the problem of integration and share of geological spatial data.geological object, geological model, network transmission, visualization Geological objects are composed of different kinds of data, such as borehole data, topography data, rock and soil data, geophysical and hydrological data. Geological and borehole data, in particular, provide useful information about the earth surface and underground. Therefore, they are used in many fields, such as city construction, resource exploration, environmental problems, and transportation. The related technical improvement provides new opportunities for analysis of geological data. The geological models describe the observations in different geological conditions through interpolation and interpretation. At present, it lacks a general model of geological data, which results in much time wasted on rebuilding data structure. Sometimes, geologists recognize visualization relevance by log or overlay of paper plot. This method surely can figure out reliable results, but it is time-consuming and has a heavy load. Especially, man-made error affects modeling precision in the process of study different geological spatial patterns by browsing tables.With the development of computer software, the techniques of interactive demonstration and visualization help overcome the shortcomings of the tables and plots. To manage and store large geological datasets, we propose the node-layer data model called NVDM. Geological object visualization is based on the tri-prism unit. First, preprocess the original borehole data and interpolate the strata into regular grids. Then, connect the grids

Section: Related Workmentioning

confidence: 99%

A methodology for 3D modeling and visualization of geological objects

Tan

Sci. China Ser. D-Earth Sci.

et al. 2009

“…In global methods, the interpolation depends on all sample points, while the local interpolation methods depend only on the data in its neighborhood [4]. Additionally, artificial neural networks [14], fuzzy methodology [15,16], evolutionary algorithms [17], support vector machines [18], and fractal models [19] have also been used to estimate the grade of the mineralized deposits.…”

Section: Introductionmentioning

confidence: 99%

Grade Distribution Modeling within the Bauxite Seams of the Wachangping Mine, China, Using a Multi-Step Interpolation Algorithm

Wang

2017

Minerals

Abstract:Mineral reserve estimation and mining design depend on a precise modeling of the mineralized deposit. A multi-step interpolation algorithm, including 1D biharmonic spline estimator for interpolating floor altitudes, 2D nearest neighbor, linear, natural neighbor, cubic, biharmonic spline, inverse distance weighted, simple kriging, and ordinary kriging interpolations for grade distribution on the two vertical sections at roadways, and 3D linear interpolation for grade distribution between sections, was proposed to build a 3D grade distribution model of the mineralized seam in a longwall mining panel with a U-shaped layout having two roadways at both sides. Compared to field data from exploratory boreholes, this multi-step interpolation using a natural neighbor method shows an optimal stability and a minimal difference between interpolation and field data. Using this method, the 97,576 m 3 of bauxite, in which the mass fraction of Al 2 O 3 (W a ) and the mass ratio of Al 2 O 3 to SiO 2 (W a/s ) are 61.68% and 27.72, respectively, was delimited from the 189,260 m 3 mineralized deposit in the 1102 longwall mining panel in the Wachangping mine, Southwest China. The mean absolute errors, the root mean squared errors and the relative standard deviations of errors between interpolated data and exploratory grade data at six boreholes are 2.544, 2.674, and 32.37% of W a ; and 1.761, 1.974, and 67.37% of W a/s , respectively. The proposed method can be used for characterizing the grade distribution in a mineralized seam between two roadways at both sides of a longwall mining panel.

2009 International Conference on Intelligent Networking and Collaborative Systems

“…Namely, the algorithm that is generated while the machine is being trained is readily applicable to a much wider area than the training set encompasses. Hence, a proper reconstruction of the final model is possible with sparse geo-inputs [7].…”

Section: Introductionmentioning

confidence: 99%

“…In fact, there is generally limited number of studies which addresses methodology adhered to herein, such as the case studies of landslide susceptibility in Hong Kong area [6], slope stability assessment by [9], or general geotechnical 3D modeling capabilities with machine learning [7]. The presented study gains gravity for its practical contribution in both, local and global scales.…”

Section: Introductionmentioning

confidence: 99%

Landslide Susceptibility Assessment with Machine Learning Algorithms

Marjanovic

Bajat

Kovacevic

2009

Case study addresses NW slopes of Fruška GoraMountain, Serbia. Landslide activity is quite notorious in this region, especially along the Danube's right river bank, and recently intensified seismicity coupled with atmospheric precipitation might be critical for triggering new landslide occurrences. Hence, it is not a moment too soon for serious landslide susceptibility assessment in this region. State-of-theart approaches had been taken into consideration, cutting down to the Support Vector Machine (SVM) and k-Nearest Neighbor (k-NN) algorithms, trained upon expert based model of landslide susceptibility (a multi-criteria analysis). The latter involved Analytical Hierarchy Process (AHP) for weighting influences of different input parameters. These included elevation, slope angle, aspect, distance from flows, vegetation cover, lithology, and rainfall, to represent the natural factors of the slope stability. Processed in a GIS environment (as discrete or float raster layers) trough AHP, those parameters yielded susceptibility pattern, classified by the entropy model into four classes. Subsequently the susceptibility pattern has been featured as training set in SVM and k-NN algorithms. Detailed fitting involved several cases, among which SVM with Gaussian kernel over geo-dataset (coordinates and input parameters) reached the highest accuracy (88%) outperforming other considered cases by far.