Multi-source geological modeling process

Multi-source geological modeling includes the following steps, and the whole process is shown in Figure 5. 1.

Figure 5. 1 Multi-source geological modeling data organization

Topographic and geological data processing

Three-dimensional terrain representation is a research hotspot in GIS field in recent years, and it is usually represented by digital elevation model (DEM). Traditionally, contour line is the most commonly used terrain representation method, and contour data obtained by various means is still the most commonly used terrain data source in large areas (Zhong et al., 2006). With the development of GIS, people hope to observe the terrain features in a more intuitive way, and regular grid and TIN become the first choice (Wu Lixin et al., 2005). DEM model represented by contour lines can be transformed into grid or TIN model by some algorithm (Li et al., 2006), or TIN model can be established by editing topological information and adopting strip algorithm (et al., 2008). TIN mode is widely used at present.

The topographic map shown in fig. 5.2a is a TIN grid generated by triangulation after the contour map is digitized and its elevation information is restored. It can be found that this area belongs to valley landform.

The geological map shows the projection information of exposed strata on the plane, which is expressed in the form of two-dimensional multi-line segments. Figure 5.2b shows the distribution map of geological unit boundaries in the study area. By adding geological boundaries, the plane distribution map of exposed strata is formed.

The polygon that constitutes the outcrop features of rock strata is vertically projected onto the terrain TIN model, as shown in Figure 5.2c, and the terrain TIN is cut by this geological boundary to obtain the distribution of different rock strata on the ground in three-dimensional state, as shown in Figure 5.2d, and different regions are given corresponding rock attributes as the first step of multi-source geological modeling.

Figure 5.2 3D Modeling of Terrain and Geological Data

A. terrain DEM model; B. Geological regional distribution map; C, projecting the geological area distribution map onto the topographic map; D, according to the geographical distribution of topographic map.

Geological modeling of 5.2.2.2.

The main data source of geological level modeling is still the stratigraphic line on the sequence geological profile. According to the stratigraphic line on the profile, the three-dimensional surface of the stratum is generated by using the three-dimensional reconstruction algorithm of contour lines. In addition to the stratum line on the geological profile, in order to make the stratum surface as close as possible to the actual situation, we should fully consider the stratum point information revealed by the existing borehole data and the geological sketch data in the construction process, and use these constraint information to locally correct the stratum surface to make it as close as possible to the real spatial distribution of the stratum.

Figure 5.3 Grid Classification

A, directly constructing a network from a section line; B. grid subdivision

Considering the reliability of information sources, although the profile stratigraphic line has certain characteristics of human experience, it is still considered as reliable data without obvious errors, and the layer grid accurately passes through the profile stratigraphic boundary. What needs to be smoothed is the part between sections, and the defect of contour reconstruction algorithm is to directly establish triangulation between section lines, that is, the feature points between sections are absolutely linear (Figure 5.3a). New nodes can be added to these initial triangular elements by linear interpolation (Figure 5.3b), which is also called mesh classification, so as to achieve the purpose of thinning the initial mesh. The points inserted in the subdivision process are uncertain, and these uncertain nodes can be readjusted by using data points such as drilling to achieve the purpose of subdivision.

The elevation values of newly inserted grid nodes after subdivision are linear interpolation of corresponding feature points on the section line, which needs to be re-interpolated or fitted by combining known and newly added borehole and geological sketch data. The newly inserted points between sections are local. Therefore, in interpolation, the moving interpolation method is used to set the upper and lower thresholds of the search radius respectively, and then the elevation value of the insertion point is determined according to the following steps:

(1) Search within the lower limit. If there is a data point, directly replace it with the nearest point and exit;

(2) Otherwise, search within the upper threshold, and if there are data points, interpolate according to certain interpolation methods (such as Kriging method, anti-weighted distance method, etc.). ) get the elevation value of the insertion point and exit;

(3) Otherwise, nothing will be done.

Ground simulation is the core of multi-source geological modeling. Figure 5.4a shows the initial triangular surface mesh composed of two section lines. The grid is divided into four levels, and the data inserted in the middle is the linear interpolation of two sections of section lines. From the stratum information revealed by borehole data, it can be seen that some areas have large deviation, so borehole data should be taken into account as constraint information. After adopting the above processing method, the constraint point C3 replaces the adjacent corner points on the stratum surface, and the positions C0, C 1 and C2 achieve the approximation effect. Except the section line as the reliability data, the stratum surface tends to be smooth on the whole, which is more in line with the actual situation (Figure 5.4b).

Figure 5.4 Interpolation Update of Formation Surface

A. original formation surface and drilling constraint information; B, adding constraint information to locally update the stratum surface. It can be seen that among the point source constraint information in the four boreholes, the constraint information C3 replaces the interpolated adjacent corner points, and the positions C0, C 1 and C2 achieve approximate results.

5.2.2.3 ·SGM model

For a complete stratum, it is usually surrounded by upper layer, lower layer and boundary surface, which is the essence of SGM model. For the stratum affected by fault cutting, it is necessary to adjust the ground plane and fault plane through surface clipping and fault dislocation, and finally form a closed solid model.

The geological model shown in Figure 5.5 is a monoclinic structure composed of four strata, which belongs to valley landform. For example, S3 stratum is composed of topographic plane T2, inclined ground planes F2 and F3, lower bottom surface D2 and front and rear boundary surfaces not shown. That is to say, each stratum is composed of several strata, but it is not a simple combination, which requires the strata to maintain geometric and topological consistency between triangular elements at the boundary. If triangular patches are generated by contour lines through contour algorithm, then they are consistent at the boundary; At the fault cutting place, when cutting the surface, it is necessary to rebuild the ground plane and structural plane to keep the boundary consistent.

Figure 5.5 Formation principle of the model

A. Geomorphological geological model of river valley composed of four monoclinic strata; B. strata in geological model 3; C, decompose stratum 3.