Three-dimensional geologic model construction program of groundwater in the Black River Basin

(A) The process of constructing the three-dimensional geological model of groundwater in the Black River Basin

The model construction needs to formulate the process of its construction so that the model can be completed in an orderly manner. After the practice of software development, we have formulated the process of modeling:

(1) Data collection, organization and checking;

(2) Data importing into the modeling system software, and modeling of individual basins;

(3) According to the imported data, clarifying the corresponding relationship, and establishing the geological bodies such as faults, stratigraphic layers and lenticular bodies;

(4) Checking and calibrating the model, so that the model can be completed methodically. p>(4) Check and verify the established sub-basin 3D visualization model, and repeat the previous steps to modify the model if there is any error;

(5) Combine the sub-basin models to form an overall 3D visualization model.

(2) Scale design of three-dimensional geological model of groundwater in Heihe River Basin

The modeling area of the three-dimensional model of groundwater is characterized by a wide transverse area and shallow vertical depth, which makes it necessary to set different transverse and vertical scales, and the comparison of transverse and vertical scales should be appropriate, so as to build up the model, and the model appearance is more beautiful.

The working area of Black River Basin is about seven to eight hundred kilometers long and 400 kilometers wide, while the depth of underground exploration is only a few hundred meters or even a few dozen meters, so if the model construction is carried out with a uniform scale, then the model built will be reduced to a surface. The geological structure of the underground is not expressed and cannot be navigated. If the vertical scale enlargement, then in the watershed territory of the Qilian Mountains at an altitude of more than five thousand meters high, compared with the depth of the exploration data underground, there are also dozens of times the gap between the model built in the height of the higher, the underground part will be a very thin piece.

In this case, it is necessary to determine the appropriate aspect ratio, so that the model is more beautiful, and in line with the actual requirements. Modeling system data input uniformly in meters as a unit, after many tests, we determined that the Black River Basin modeling horizontal ratio and vertical ratio of 2000 is more appropriate. That is, the XY direction is reduced by 2000 times in the input, and the vertical data, such as the depth of the profile, remain unchanged in the input. After such a change, the model constructed with proper aspect ratio can better reflect the geological structure of the Black River Basin.

(III) Design basis of block modeling and determination of the boundaries of each module

The Heihe River has a complex geological structure, and several basin units are developed in different tectonic backgrounds, which leads to the independence of the basin in this basin, and at the same time determines the spatial changes in the structure of the groundwater system and the complexity of the groundwater movement.

Constructing a geological model of such a large area as the Heihe River Basin, the system can read in hundreds of megabytes of raw data, plus the number of triangular grids generated by the stratigraphy of the memory occupied by the system, the resource requirements on the machine will be quite high, and our software is positioned to use a microcomputer rather than a workstation for modeling, which must be considered to resolve this conflict.

The basin can be divided into a number of basin units according to its geology, and each basin unit is modeled independently. This helps us to solve the problem by building sub-models of each basin under a unified framework, and then assembling and combining the triangular grid surfaces of the sub-models to form a total model of the basin. This not only overcomes the limitation of machine speed so that each model can be completed separately, but also ensures the integrity of the overall model. It can be said to be multi-purpose.

The construction process of the overall model is the process of combining - dividing - combining. First, the overall framework is established, i.e., the fault and surface model of the whole work area is built, then the sub-models are constructed according to the geological conditions strictly according to the coordinates of the subdivisions, and finally the models are assembled under the same surface according to the coordinates.

The zoning of each basin unit in the basin is according to the geological structural features. Faults and uplifts play a great role in the transportation and change of groundwater, and are the natural dividing line of each basin. Based on the geologic data, the dividing line can be marked on the basin map. Then, the geologic data are partitioned according to the dividing line, and then modeled based on the geologic data of each partition. The seven blocks modeled separately are: the Damaying Basin, the Shandan Basin and the Daqingyang Basin, the Zhangye Basin, the Jiuquan East Basin, the Jiuquan West Basin, the Jinta Basin, and the Ejinagi Basin. Specifically, the location of the dividing lines of the basins is described as follows:

(1) the General Zhai-Yonggu Fault is the dividing line between the Zhangye Basin and the Damaying Basin;

(2) the Yonggu Uplift is the dividing line between the Zhangye Basin and the Shandan Basin;

(3) the arc between the Bitterwater Mountain Uplift at the western end of the Yumu Mountains and the Gautai Concealed Uplift constitutes the dividing line between the Jiuquan East Basin and the Zhangye Basin;

(4) the faults north and south of the central mountains north of Jiayuguan are the respective dividing lines of the Jiuquan Basin and the Jinta Basin;

(5) the Jiayuguan Fracture is the dividing line between the Jiuquan East Basin and the Jiuquan West Basin;

(6) the Dewan Dongliang Uplift is the dividing line between the Jinta Basin and the Ejinaqi Basin.

(4) Design of sub-model construction

For the sub-model, that is, a specific basin, the geological structure is relatively complex, and there are more geologic bodies to be represented, and these geologic bodies have the relationship of cutting, intersecting, etc., which requires the construction of these geologic bodies according to a certain order, class by class, one by one, and in the order of the model. The order of model construction is briefly explained here, i.e., read in the original data, generate the surface, establish the fault model according to the fault line on the surface and the fault line on the selected section, after that, construct each stratum, designate the effective area of the stratum, construct the lenticular body, adjust the spatial position between the stratum and the lenticular body, and finally, generate the geologic body. After generating the geoid, it is possible to perform visualization and graphical output.

In order to illustrate the modeling in this order, which involves modeling the relationship between the geological body, in general, all the data need to have a definite geological body in the modeling process to play a role in reference, the ground surface can be very good to take on this role, when the ground surface is input, the fault line, the profile line can be drawn on the ground surface, and in accordance with the changes in the undulation of the surface of the ground. and undulation according to the undulation of the ground surface. Therefore, it is necessary to model the ground surface first. Once the surface has been modeled, basin boundaries and surface geographic information map elements can be input to the surface. When constructing the stratum and fault geoid, since the ground surface is cut by the fault surface to form the upper and lower disks of the fault, and the fault surface is not subject to change, it is necessary to prioritize the construction of the fault surface, and after the construction of the fault surface is completed, the ground surface is constructed and the ground surface is processed for automatic cutting of the ground surface. After the ground plane is constructed, the lens body follows, because the lens body belongs to different strata, so it needs to be constructed on the basis of the ground plane. After the generation of the lenticular bodies, all the preparatory work has been carried out up to this point, and the generation of the geologic blocks can be carried out. This process is an optimization of the construction of the model, which has been proved to be correct in practice.

(E) the surface data points of the thinning and elimination

Surface data from the ARCINFO format data conversion, composed of contour lines of the plane of the line segment and its elevation value corresponds to one by one, so that there is a three-dimensional coordinate set of the surface. When the system is inputted, the points that form the contour lines are read in, i.e., the points of each line segment are read in, and combined with the elevation values corresponding to the line segments, the discrete points required for generating the model are formed, and the system uses these discrete points for interpolation to generate the ground surface. Since the data volume of all the 1:250,000 data in the Heihe River Basin is quite large, and the density of discrete points is also large, and the data capacity stored in MAPGIS plain code files reaches more than 150 megabytes, if all these data are inputted into the system and interpolated to generate the ground surface, the number of small triangles on the generated ground surface will be quite large, which will take up too many computer resources and have a great impact on the construction of the subsequent model. At the same time, generating the ground surface is slow and unsatisfactory. Therefore, it is necessary to thin these discrete points. The effect of thinning is to reduce the number of points on the contour, due to the existence of a large number of points on the contour, thinning by ratio will not affect its accuracy. After thinning, the density of discrete points required to generate the model is reduced, the number of discrete points is reduced, the system generates the surface faster, and the smoothness of the surface is improved.

For the surface of some bad points, such as the elevation value is too high or too low points, that is, the elevation is higher than the highest point of the surface, or the elevation is lower than the elevation of the lowest point of the surface, these points are due to errors or data conversion, the use of such a point of interpolation, will result in the surface of the undulation of the surface of the change in the dramatic, the surface of the surface of the rough, affecting the surface of the smoothness of the surface, for this kind of point needs to be input into the system when the For such points, they need to be eliminated when inputting into the system. That is, in the system input module, the threshold value is used to limit the points that are too high or too low to be rejected and not allowed to participate in the modeling.

(F) Design of the grid size of the three-dimensional geological model of groundwater in the Black River Basin

By the principle of modeling, it can be known that the geologic bodies built by the model, such as strata, faults, etc., are connected by the grid to form a surface, which is wrapped around the surface to form a body. The smallest unit of the model is a small triangle, and the number of triangles has a direct impact on the accuracy of the model and the speed of the system.

Generally speaking, generating the model of the triangle mesh is too large, the model face is relatively rough, the model is not fine, and even can not show the morphological characteristics of the face, the mesh is too small, the mesh density is large, the triangles need to take up a large number of system resources, so that the amount of data processed by the computer dramatically increase the machine's operating speed is slow, if the amount of data is too much or too little discrete points, then the model face is more complex, and can not show the overall model face. complex and cannot express the overall characteristics of the model surface. Therefore, when constructing the model, it is necessary to choose an appropriate grid size.

In the process of constructing the model of Heihe Basin, after practice, the grid size of the surface is better with a grid spacing of 50-200m, and the speed of the machine and the smoothness of the surface can be coordinated and harmonized. Generally, in the initial construction of the model, the grid spacing of 200m is selected, the machine speed is fast, and when the construction of the model of each small basin is completed, the grid spacing of 50m is selected for the generation of the ground surface, so that the surface of the ground is relatively fine.

Stratigraphy and faults are modeled using the line segments on the profile, and the system automatically encrypts the discrete points on these line segments. 200m grid spacing is chosen, which does not have much influence on the generation of stratigraphy and faults, and the fineness meets the requirements. Because of the small area of the lens body, the accuracy requirements are moderate, so the grid size of 100m grid spacing can meet the needs.

For raster data, such as remote sensing image pictures, its spatial resolution can be adjusted according to the accuracy of the model display pixel size. Generally, 100-300 dpi is sufficient. The fine remote sensing image pictures with more than 800 megabytes can be generated into BMP, JPG and other graphic formats according to the needs to ensure the pixel accuracy and reduce the need for memory usage.

(VII) Automatic Generation and Local Adjustment of the Model

The fault surface, ground surface, and top and bottom surfaces of the lens body of the model are all discretized by the system according to the line segments selected by the technicians in the section, and then these discrete points are used for interpolation to form a series of small triangles connected to form the surfaces of the different geologic bodies, which generates the surface process and generates the surface stretching according to the tendency of discrete points, causing the geologic body to be discretized and to be connected to a series of small triangles. The process of generating the surface in this way will cause the surface to stretch according to the trend of discrete points, which will lead to the occurrence of phenomena such as deformation and distortion of the geological surface, which is not in line with the actual geological situation. Such as faults cut each other, faults cut strata, strata intersection cut and a series of geologic surface relationship can not be established.

In this case, it is necessary to use the control points to carry out local fine-tuning of the face to change the shape of the face. So that the generated surface as close as possible to the interpretation of geological data results, expressing the real geological conditions. At the same time, it can also make faults cut each other, faults cut strata, strata intersection cuts and other geological body relationships are established. The system then defines the relationship between these surfaces according to the technician, and automatically generates the geologic body that meets the conditions.

Therefore, in the process of model construction, it is necessary to combine the actual situation and use the function of automatic generation and local adjustment of control points to complete the model construction.

(VIII) Error analysis and treatment of the three-dimensional geological model of groundwater in the Black River Basin

In the process of model construction, it will inevitably lead to the generation, transmission and accumulation of errors, and the reasons for errors should be analyzed, and the error control mechanism should be established in order to improve the quality and accuracy of the model constructed. From the source analysis of error generation, there are reasons in terms of basic information, as well as errors generated by the system in the modeling process, which are discussed separately below.

1. Errors in basic information

The model is mainly constructed by using profile data and drilling data, and the drilling data will make the generated stratigraphy undulate too much due to the overly messy stratigraphic changes of the drilling holes. On the profile side, errors can also occur due to human factors in the drawing process.

In summary, different technicians have different understanding and treatment of the geological structure, which will inevitably be reflected in the stratigraphic changes and values of the profile. In addition, different geological data are used, and there are different arguments or even contradictory views between some of them, so as to cause errors. In the process of drawing, the technicians use different information for the process of synthesizing and processing, such as when translucent diagrams, so that the error arises. When drawing by hand on the square grid calculation paper, especially when drawing long profiles, due to the folds of the paper, the process of receiving the map is bound to have errors. Even some calculation grid paper itself is not standardized, and the condition of tilted grid lines occurs. At the same time, the process of scanning and vectorization of drawings will also transmit errors. The prominent performance is that in the cross position of the profile, the position of the stratigraphic line can not be intersected, or the phenomenon of stratigraphic trend inconsistency with other profiles.

2. Errors in system operation

Since the basic data for model construction will produce certain errors in the process of data production, when the data with errors are inputted into the modeling system, it will result in the transmission and amplification of errors, which will reduce the accuracy of the model. In the case of trying to avoid errors in the basic information data, the system should also be analyzed for errors in the operation process.

Analysis of errors in the modeling process can be concluded that in all the steps of modeling, there is a possibility of errors due to operational errors or system accuracy settings. The following analysis can be made specifically. In the data processing stage, errors will occur when various data and graphic information are data fused, such as projection transformation, errors in error correction. In addition, the system modeling in the interpolation process, the local point is more dense, such as the selected profile line, the rest of the region of the discrete point sparse, almost no, in accordance with the mathematical method of interpolation, the interpolated surface and the actual surface does not match the phenomenon, in order to make the surface conform to the reality of some of the method of the use of the control point of the pull or pressure, which will inevitably cause a certain amount of error.

In addition, when the distance between model layers is small, it is easy to cross and overlap the ground plane, and when the control points are used for correction, the lens body or the ground plane is changed, and the error is very easy to occur. Also, when the position between the ground plane level and the lens body is adjusted, a chain reaction is very likely to occur, resulting in errors.

3. Error control

Some of the errors are inevitable, and some of them are accidental. For these errors, our main strategy is to verify and calibrate the data for the errors in the basic data, such as in the process of drawing the profile, strengthen the quality control, enhance the sense of responsibility of the technicians, in the same area, let the geotechnical staff to coordinate the calibration of the map, and hire geologists to check and audit, and these measures play a very good role in quality control. As for the errors in the system operation, the project team members analyzed the reasons, strengthened the debugging, and took a lot of measures to try to minimize the errors. For example, they precisely adjusted the position of the profiles and tried to make the position of the strata consistent. For obvious errors, such as mismatch of stratigraphic lines, they analyzed the changes of the stratigraphic layers based on other profiles and used the editing function of the system to edit the stratigraphic lines with the errors, so as to reduce the errors to the minimum. In the process of generating layers, try to add more points on the profile line, so that the formed stratigraphic mesh passes through the line of the profile to which it belongs, and so on. These measures play a good role in minimizing the error.

Model checking in modeling is also a way to reduce the error. The model checking includes the accuracy checking of the original data, i.e. whether the formed surface is consistent with the original data points and whether the data of the original points are retained; the checking of the geological rationality can be checked at the same time by using the geological data such as boreholes and profiles. In addition, the geological boundaries of each layer (including the fault plane) can be checked by making a profile near the original profile by modeling and comparing the two profiles. For some important sections, it can be used to check the intersection of faults, the intersection of faults and strata, and the consistency of stratigraphic patterns between different layers and on both sides of faults. The model checking process has the role of real-time verification in the error control of the model.