Engineering survey geological data error checking method

Abstract

This invention provides a method for error detection in geological data for engineering exploration based on a neural network algorithm. The specific steps of the method include: determining the complexity of the site; identifying the stratigraphic layers of the target borehole; determining the calculation boreholes surrounding the target borehole; establishing a distance influence function; establishing a weighted formula for calculating the stratigraphic depth of the target borehole using the depths of n comparison boreholes surrounding each target borehole; and using this formula to calculate the theoretical stratigraphic depth value of the target borehole. The actual stratigraphic depth of the target borehole is compared with the calculated stratigraphic depth to obtain the difference value η between the actual and calculated stratigraphic depths. This difference value is then used to determine whether the stratigraphic depth of the target borehole is abnormal. This invention only requires geological data to complete data verification and identify abnormal stratigraphic data, has high computational efficiency, effectively solves the problem of falsification at the exploration site, and can also provide an auxiliary means for reviewing exploration results.

Description

Technical Field

[0001] This invention relates to the field of engineering surveying, specifically a method for checking errors in geological data used in engineering surveying. Background Technology

[0002] Geotechnical engineering investigation is an essential and crucial step in construction projects, and the accuracy of its data directly affects project costs and even quality and safety. Currently, the review of investigation results is based on independent investigation projects, relying on manual identification to determine the authenticity of stratigraphic data. This method has several drawbacks: First, for large-scale investigation projects with thousands of exploration points, manual identification is time-consuming, labor-intensive, and inaccurate. Second, manual identification is highly dependent on the reviewer's professional level; different people may yield different results. Third, even if a single investigation project passes review, inconsistencies may arise when compared with nearby existing investigation results, a problem that the current review mechanism cannot resolve.

[0003] With the application of information technology in the field of exploration, exploration results can be stored in the form of spatial data, providing a foundation for automated data identification. Currently, the supervision of the authenticity of exploration data is still focused on the original data collection, mainly judging the compliance of the behavior, and there is no research on the identification of data accuracy. For the judgment of exploration results, three-dimensional geological models can play a certain auxiliary role, but because the learning and use costs of this technology are relatively high, it is sometimes unable to complete the modeling of complex strata, which greatly limits its application. Summary of the Invention

[0004] To address the problems existing in the prior art, this application provides a method for checking errors in geological data for engineering exploration. This method only requires geological data to complete the verification of data and the identification of abnormal strata data. It is convenient to use and has high computational efficiency. It can effectively solve the problem of falsification at the exploration site and can also provide auxiliary means for the review of exploration results.

[0005] To achieve the above-mentioned technical objectives, the present invention provides a method for error detection in engineering survey geological data, characterized by the following specific steps:

[0006] (1) Determine the complexity of the site according to the "Code for Geotechnical Investigation";

[0007] (2) Determine the stratigraphic structure of the target borehole: Obtain the stratigraphic information of all target boreholes in the field, establish a stratigraphic table, and then determine the stratigraphic information of the target borehole, treating each stratigraphic layer as a calculation unit.

[0008] (3) Determine the calculation boreholes around the target borehole: Determine the appropriate radiation range of the target borehole based on the complexity of the site, and use all boreholes within this range as calculation boreholes.

[0009] (4) Establish the distance influence function and calculate the formation influence coefficient at the location of the target borehole using the distance influence function;

[0010] a. Since the influence of stratigraphic relationships is between 0 and 1, and the influence trend slows down with distance, the Sigmoid function is used to map the output to between 0 and 1, serving as the transfer function. The function expression is: x represents the input value, and f(x) is the output value between 0 and 1;

[0011] b. Considering geological variation patterns, the function is modified to accommodate foundation influence patterns of varying complexity. The modified functional relationship is as follows: ;

[0012] The target borehole distance influence coefficient;

[0013] x is the distance between the comparison borehole and the target borehole, x≥0;

[0014] 'a' is the factor affecting the complexity of the stratigraphy, and 'a' ≥ 0;

[0015] From x=0, =1, we can get b=2, thus obtaining the distance influence function: ;

[0016] Among them, parameter 'a' is related to the geological complexity. The more complex the geology, the steeper the curvature. Adjusting the value of 'a' can make the fitting more accurate.

[0017] (6) Establish a weighted formula for calculating the stratum depth of the target borehole based on the depth of n comparative boreholes around the target borehole, and use the formula to calculate the theoretical stratum depth of the target borehole.

[0018] ,

[0019] The theoretical formation depth for drilling the target borehole;

[0020] n This represents the depth value of the stratum where the nth borehole is located.

[0021] n The distance influence coefficient of the nth target borehole;

[0022] (6) By comparing the actual formation depth of the target borehole with the calculated formation depth, the difference between the actual formation depth and the calculated formation depth of the target borehole is obtained. The difference between the actual and calculated formation depths of the target borehole is used to determine whether the formation depth is abnormal. The calculation formula is as follows:

[0023] .

[0024] A further technical solution of the present invention: The complexity of the foundation in step (1) is divided into three levels according to the "Code for Geotechnical Engineering Investigation", specifically including:

[0025] A. Primary foundation, i.e. complex strata: diverse types of soil and rock, highly heterogeneous, and with significant variations in properties;

[0026] B. Secondary foundation, i.e., medium-complex strata: there are many types of soil and rock, which are uneven and vary greatly in properties;

[0027] C. Level III foundation is simply a single layer: the soil and rock types are uniform and their properties do not vary much;

[0028] A further technical solution of the present invention: In step (2), when there is a missing stratum, the thickness of the stratum is set to 0, and it is also included in the calculation.

[0029] A further technical solution of the present invention: In step (3), the radiation range of the target borehole in complex strata is 40-60m, the radiation range of medium complex foundation is 60-80m, and the radiation range of simple strata is 80-100m.

[0030] A further technical solution of the present invention: the function correction of step b in step (4) is based on the distance influence relationship. When the distance between the two boreholes is 0, the stratum depth should be consistent, that is, the correlation is 1. As the distance gradually increases, the burial depth correlation tends to 0.

[0031] A further technical solution of the present invention: In the case of faulting or geological discontinuity caused by artificial alteration of strata in step (4), it is necessary to establish a second layer of influencing factors. , The value should be set from 0 to 1 based on the strength of the actual influencing factor's impact on geological continuity. This factor can be arranged as a linear function on the borehole layout plan, with the starting and ending points forming a line. When the line connecting the target borehole and the comparison borehole crosses this line, the distance influence function... The automatic reduction, and its final distance influence function relationship is:

[0032] .

[0033] A further technical solution of the present invention: In step (4), when a stratum lens appears in the borehole, the target borehole does not have the same layer in the adjacent comparison borehole, and a warning message is given. When it is set as a lens, it participates in the calculation of the stratum of the adjacent target borehole. When a geological interlayer appears in the borehole, first count whether the adjacent comparison borehole has the same sequence. If it does, compare them one by one. Otherwise, calculate according to the depth of the lowest interlayer.

[0034] A further technical solution of the present invention: the difference value in step (6) is also closely related to the type of soil and rock and the heterogeneity of the site; that is... Value, of which primary foundation Take 0.1, secondary foundation Take 0.06, Level III foundation Take 0.04.

[0035] A further technical solution of the present invention: In step (6), the threshold value of the formation depth difference is set to 30%, and the calculated formation depth difference value is compared with the formation depth difference value. Threshold comparison, when the difference in stratum depth When the threshold is exceeded, it can be determined that the stratum depth where the target borehole is located is abnormal.

[0036] Geological stratification is generally based on factors such as the age, mechanical properties, color, structure, and inclusions of the rock and soil. Under normal circumstances, under historical sedimentary processes, geology follows a stratification pattern where older strata lie below and younger strata rise above. However, tectonic movements can cause discontinuities or even reversals in the stratigraphic sequence. Geology often forms traceable patterns over a large area. For a small area, such as an industrial park or a residential complex, the area is usually small, and the stratigraphic sequence also exhibits relatively simple variation patterns. From an engineering survey perspective, a particular stratum in a borehole often shows similarity to adjacent strata, or exhibits a certain continuity in a certain direction. The strata are inferable in terms of burial depth. Within a plane, the closer two boreholes are, the stronger their similarity; the farther apart they are, the weaker the similarity. The similarity relationship between them is between 0 and 1. This calculation method is consistent with neural network calculation models. Therefore, the relationship between the strata of each borehole and surrounding boreholes can be established based on their distance, allowing for the identification of abrupt changes in strata with unreasonable relationships to surrounding strata.

[0037] The method in this application only requires geological data to complete the verification of data and the identification of abnormal strata data. It is convenient to use and has high computational efficiency. It effectively solves the problem of falsification at the exploration site and can also provide auxiliary means for the review of exploration results. Attached Figure Description

[0038] Figure 1 This is a flowchart of the present invention;

[0039] Figure 2 This is a diagram showing the distribution of the neural network in this invention;

[0040] Figure 3 This is a relationship diagram established between the target borehole and the surrounding comparative boreholes in this invention;

[0041] Figure 4 Is the stratigraphic complexity influencing factor a on Value influence curve;

[0042] Figure 5 This is a schematic diagram of the distribution of boreholes in abnormal formations in the embodiment;

[0043] Figure 6 In this embodiment, the formation distribution pattern is viewed through a formation depth cloud map after the borehole data is repaired. Detailed Implementation

[0044] The present invention will be further described below with reference to the accompanying drawings and embodiments. The drawings are drawn in a simplified manner and are only used to clearly and concisely illustrate the embodiments of the present invention. The technical solutions shown in the drawings below are specific solutions of embodiments of the present invention and are not intended to limit the scope of the claimed invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0045] The present invention establishes a method for error detection in engineering exploration geological data based on a neural network algorithm. The workflow is as follows: Figure 1 As shown, the specific steps include:

[0046] (1) Determine the complexity of the site; since the complexity of the foundation affects the continuity of the strata, the simpler the strata, the better the continuity of the strata, and the stronger the correlation between the strata of the surrounding boreholes and the strata of the target borehole, and vice versa. Determining the complexity of the site is used on the one hand to determine the comparison boreholes within the target borehole range, and on the other hand to determine the changes in the transfer function; according to the "Code for Geotechnical Investigation" (GB 50021-2001), the complexity of the foundation can be divided into three levels:

[0047] A. Primary foundation (complex foundation): Diverse soil and rock types, highly heterogeneous, with significant variations in properties;

[0048] B. Secondary foundation (moderately complex foundation): The soil and rock types are diverse, uneven, and vary considerably in properties.

[0049] C. Level III foundation (simple foundation): The soil and rock types are uniform and their properties do not vary much.

[0050] (2) Determine the target borehole and decompose the strata: Obtain the strata information of all boreholes in the site and establish a strata table. Then determine the strata information of the target borehole, and treat each stratum as a calculation unit. When there is a missing stratum, to avoid unreasonable omissions, the thickness of the missing stratum can be set to 0, and it can still participate in the calculation.

[0051] (3) Determine the surrounding boreholes of the target borehole: such as Figure 3 As shown, the appropriate radiation range of the target borehole is determined according to the complexity of the site to avoid the calculation area being too large and affecting the analysis efficiency. All boreholes within this range are used as calculation boreholes. According to the influence of actual strata, 40-60m can be selected for complex strata, 60-80m for moderately complex foundations, and 80-100m for simple strata.

[0052] (4) Establishing the distance influence function: The influence of stratigraphic relationships is between 0 and 1, and the influence trend slows down with distance. Therefore, the Sigmoid function can be used to map the output to between 0 and 1, and it can be used as a transfer function; the function expression is: ;

[0053] Considering geological variations and the influence of distance, when the distance between two boreholes is 0, the stratum depth should be consistent, i.e., the correlation is 1. As the distance gradually increases, the correlation of burial depth tends to 0. This function needs to be modified to accommodate foundation influence patterns of varying complexity. The modified function is: b. Considering geological variations, this function is modified to accommodate foundation influence patterns of varying complexity. The modified function is: ;

[0054] The target borehole distance influence coefficient;

[0055] x is the distance between the comparison borehole and the target borehole, x≥0;

[0056] 'a' is the factor affecting the complexity of the stratigraphy, and 'a' ≥ 0;

[0057] From x=0, =1, we can get b=2, thus obtaining the distance influence function: ;

[0058] Wherein: parameter 'a' is related to the geological complexity, and the specific values ​​of parameter 'a' are as follows: Level 1 foundation Take 0.1, secondary foundation Take 0.06, Level III foundation The value is set to 0.04; the relationship between parameter a and geological influence is shown in [reference needed]. Figure 4 As shown, the more complex the geology and the steeper the curvature, the more accurate the fitting can be by adjusting the value of 'a'.

[0059] (5) Establish a weighted formula: the depth value of the formation where a single comparison borehole is located Calculate the formation depth of the target borehole. The formula is To compare a specific borehole (node) with its surrounding n boreholes and establish a weighted value for the surrounding n boreholes, the network in this process can use a linear transfer function, and the theoretical formation depth of the target borehole can be calculated using this formula.

[0060] ,

[0061] The theoretical formation depth for drilling the target borehole;

[0062] n This represents the depth value of the stratum where the nth borehole is located.

[0063] n The distance influence coefficient of the nth target borehole;

[0064] (7) By comparing the actual formation depth of the target borehole with the calculated formation depth, the difference between the actual formation depth and the calculated formation depth of the target borehole is obtained. The difference between the actual formation depth and the calculated formation depth is used to determine whether the formation depth of the target borehole is abnormal. The calculation formula is as follows:

[0065] ;

[0066] The difference between the actual formation depth and the calculated formation depth of the target borehole.

[0067] The depth of abnormal formations is determined based on the difference value. This difference value is highly dependent on the type of soil and rock and the heterogeneity of the site; different sites may have different values. However, for specific long-term exploration sites, such as steel plant sites, a threshold of 30% for the difference value of formation depth can be set based on expert experience. If the difference value exceeds this threshold, the borehole and its formation are considered abnormal, and the system will identify and remove the abnormal borehole.

[0068] The invention will be further described below with reference to an embodiment. This embodiment pertains to a steel company that has been operating for over 50 years and has accumulated a large amount of exploration data, all stored in paper form in an archive. Currently, the site is undergoing a series of renovations and upgrades. To effectively utilize existing exploration data, accelerate the construction period, and reduce project costs, the exploration results were digitized, ultimately inputting approximately 60,000 boreholes. Due to the large time span of the data, problems arose such as inconsistent stratigraphic names, inconsistent stratigraphic codes, and inconsistent stratification standards, causing significant difficulties for regional geological analysis and application. It is necessary to standardize the stratigraphy of the site. During the standardization process, a large number of stratigraphic variations with excessively large differences were found. To solve this problem, the engineering exploration geological data error-checking method described in this application is adopted, the specific steps of which are as follows:

[0069] (1) Establishing standard land for the site: Based on the experience of the site, establish standard land for the site, totaling 16 layers, including 9 major layers such as fill, silty clay, silty clay, fine sand, and sandstone, and 7 minor layers such as plain fill, slag layer, and cobblestone.

[0070] (2) Original strata matching standard: Based on the spatial distribution of strata, lithology and lithofacies of the same geological age in the region, a unified standard strata is reasonably established and a corresponding relationship is established with the original strata. Batch changes are made so that the strata of all boreholes in the site are within the standard strata range.

[0071] (3) Set the complexity of the site: Based on the regional geology and expert judgment, the complexity of the foundation of the site is set as level two; the borehole influence distance limit is set to 80m, that is, boreholes within 80m around the target borehole are searched for calculation, and the distance influence factor a is set to 0.6.

[0072] (4) Locating the depth of formation anomalies: Following the method provided in this patent, a starting borehole is selected as the target borehole. The calculation is performed using the method described above, with a difference threshold set at 30%. Depths exceeding this threshold are considered formation anomalies. The system marks the anomalous formation and freezes it from the calculation database, preventing it from participating in the formation calculations of other boreholes. After the calculation of this borehole is completed, the next borehole is cycled through, and the anomalous formation is identified and marked again. This process is repeated until all formation calculations are completed. All anomalous formations are marked on the plan view, such as... Figure 5As shown in the figure; all points in the figure represent the borehole planar positions, and the points marked with pentagrams represent boreholes with abnormal formations calculated using the calculation method of this patent. Boreholes A14, 37, ZK47, ZK77, 18, and 44 are selected from the figure for summarization. The actual recorded elevation of the silty clay layer in borehole A14# (5-1-0) is 16.52m. The theoretical depth calculated for this stratum in the surrounding 8 boreholes is 12.61m, a deviation of 31.01%, indicating a problem with the layer. After checking the original records and core photographs, some residual soil in layer 6-0-0 should be classified as layer 5-1-0, with an adjusted depth of 13.25m. The limestone layer in borehole 18# (8-1-0) is a rock stratum. Because rock strata have downward extension, the borehole depth does not represent the bottom depth of the rock stratum; therefore, the bottom elevation is not judged and is not considered abnormal.

[0073] A 4-2-0 silty fine sand layer was found in borehole #44. Calculations from surrounding boreholes did not reveal any alterations. Based on geological stratigraphy, this layer is considered geologically possible and can therefore be identified as a lens. The bottom elevation of the 5-1-0 silty clay layer was recorded as 6.33m, while the theoretical elevation calculated from 18 surrounding boreholes was 9.42m, resulting in a deviation of 32.80%, thus classifying it as an abnormal borehole.

[0074]

[0075] (5) Handling Abnormal Strata: Analyze abnormal strata, find the causes, and make corrections. If the strata are too coarse, the borehole strata can be subdivided based on the surrounding strata; if the depth offset is too large, the depth can be reasonably corrected according to the geological variation law. After correction, to verify the rationality and prevent the abnormal data from reappearing, this technique can be used again for one cycle, and the strata distribution law can be viewed in conjunction with the contour cloud map, specifically as follows: Figure 6 As mentioned above, by examining the stratigraphic distribution pattern through the stratigraphic depth cloud map, it can be seen that the corrected verification data is reasonable.

[0076] This application solves the problem of how to quickly identify abnormal boreholes (data falsification or misjudgment) among tens of thousands of boreholes.

[0077] The above description is merely one embodiment of the present invention, and while it is detailed and specific, it should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the appended claims.

Claims

1. A method for checking errors in geological data during engineering surveys, characterized in that... The specific steps are as follows: (1) Determine the complexity of the site according to the "Code for Geotechnical Investigation"; (2) Determine the stratigraphic structure of the target borehole: Obtain the stratigraphic information of all target boreholes in the field, establish a stratigraphic table, and then determine the stratigraphic information of the target borehole, treating each stratigraphic layer as a calculation unit. (3) Determine the calculation boreholes around the target borehole: Determine the appropriate radiation range of the target borehole based on the complexity of the site, and use all boreholes within this range as calculation boreholes. (4) Establish the distance influence function and calculate the formation influence coefficient at the location of the target borehole using the distance influence function; a. Since the influence of stratigraphic relationships is between 0 and 1, the Sigmoid function is used to map the output to the range of 0 to 1, serving as the transfer function. The function expression is: x represents the input value, and f(x) is the output value between 0 and 1; b. Considering geological variation patterns, the function is modified to accommodate foundation influence patterns of varying complexity. The modified functional relationship is as follows: ; The target borehole distance influence coefficient; x is the distance between the comparison borehole and the target borehole, x≥0; 'a' is the factor affecting the complexity of the stratigraphy, and 'a' ≥ 0; From x=0, =1, we can get b=2, thus obtaining the distance influence function: ; Among them, parameter 'a' is related to the geological complexity. The more complex the geology, the steeper the curvature. Adjusting the value of 'a' can make the fitting more accurate. (5) Establish a weighted formula for calculating the stratum depth of the target borehole based on the depth of n comparative boreholes around the target borehole, and use the formula to calculate the theoretical stratum depth of the target borehole. ， The theoretical formation depth for drilling the target borehole; n This represents the depth value of the stratum where the nth borehole is located. n The distance influence coefficient of the nth target borehole; (6) By comparing the actual formation depth of the target borehole with the calculated theoretical formation depth, the difference between the actual formation depth and the calculated theoretical formation depth is obtained. The difference between the actual and calculated theoretical formation depths of the target borehole is used to determine whether the depth is abnormal. The calculation formula is as follows: 。 2. The method for checking errors in engineering survey geological data according to claim 1, characterized in that... The site complexity in step (1) is divided into three levels according to the "Code for Geotechnical Investigation", specifically including: A. Primary foundation, i.e. complex strata: diverse types of soil and rock, highly heterogeneous, and with significant variations in properties; B. Secondary foundation, i.e., medium-complex strata: there are many types of soil and rock, which are uneven and vary greatly in properties; C. Level III foundation is simply a single layer: the soil and rock types are uniform and their properties do not vary much.

3. The method for checking errors in engineering survey geological data according to claim 1 or 2, characterized in that: In step (2), when there is a missing stratum, the thickness of the missing stratum is set to 0, and it is still included in the calculation.

4. A method for checking errors in engineering geological survey data according to claim 1 or 2, characterized in that: In step (3), the radiation range of the target borehole in complex strata is 40-60m, the radiation range of the medium complex foundation is 60-80m, and the radiation range of the simple strata is 80-100m.

5. A method for checking errors in engineering geological survey data according to claim 1 or 2, characterized in that: The function correction in step (4) is determined based on the distance influence relationship. When the distance between the two boreholes is 0, the formation depth should be consistent, that is, the formation relationship influence is 1. As the distance gradually increases, the formation relationship influence tends to 0.

6. A method for checking errors in engineering survey geological data according to claim 1 or 2, characterized in that: In step (4), if a fault occurs or geological discontinuity is caused by artificial alteration of the strata, a second layer of influencing factors needs to be established. , The value should be set from 0 to 1 based on the strength of the actual influencing factor's impact on geological continuity. This factor can be arranged as a linear function on the borehole layout plan, with the starting and ending points forming a line. When the line connecting the target borehole and the comparison borehole crosses this line, the distance influence function... The automatic reduction, and its final distance influence function relationship is: 。 7. A method for checking errors in engineering survey geological data according to claim 1 or 2, characterized in that: In step (4), when a stratum lens appears in the borehole, and the target borehole does not have the same layer in the adjacent comparison borehole, a warning message is given. When it is set as a lens, it participates in the calculation of the stratum of the adjacent target borehole. When a geological interlayer appears in the borehole, first count whether the adjacent comparison borehole has the same sequence. If it does, compare them one by one. Otherwise, calculate according to the depth of the lowest interlayer.

8. A method for checking errors in engineering survey geological data according to claim 1 or 2, characterized in that: In step (6), the difference value is also closely related to the type of soil and rock and the heterogeneity of the site. The influencing factor of stratigraphic complexity Value, of which primary foundation Take 0.1, secondary foundation Take 0.06, Level III foundation Take 0.

04.

9. A method for checking errors in engineering geological survey data according to claim 1 or 2, characterized in that: Step (6) sets the threshold for the formation depth difference value to 30%, and compares the calculated formation depth difference value with the formation depth difference value. Threshold comparison, when the difference in stratum depth When the threshold is exceeded, it can be determined that the stratum depth where the target borehole is located is abnormal.

Images