Method and system for improving tunnel electrical prospecting anomaly interpretation precision

Abstract

The invention discloses a method and a system for improving tunnel electrical prospecting anomaly interpretation precision. The method comprises the following steps: acquiring a resistivity profile or a three-dimensional resistivity slice to form a resistivity data volume; for the resistivity data body, calculating the gradient of each point by using the position of each node and the resistivity value to obtain a gradient data body; normalizing the resistivity data volume and the gradient data volume; calculating the abnormal weight of each node to obtain an abnormal weight data body; carrying out secondary normalization on the abnormal weighted data body, obtaining abnormal indexes of each node and forming an abnormal index data body; and abnormal indexes are extracted, an abnormal index result map is drawn, and tunnel abnormal grade division is carried out according to numerical values. According to the method, the resistivity value and the resistivity gradient are weighted and summed after the abnormal points are removed, the abnormal index is formed after secondary normalization, and the abnormal degree is visually reflected through the abnormal index value, so that the abnormal interpretation precision of tunnel electrical prospecting is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of tunnel surveying, and in particular to a method and a system for improving the accuracy of anomaly interpretation in tunnel electrical prospecting. Background Art

[0002] When conducting tunnel surveys using electrical methods, the resistivity contour map is still mainly used to intuitively reflect the changes in profile or regional resistivity. However, many years of construction experience have shown that in tunnel electrical surveys, the places where the resistivity contour changes dramatically or the resistivity value is extremely low are the first to approach the location where groundwater or saturated soft rock and soil may exist, which is prone to geological disasters such as landslides, roof falls, and sudden water and mud. Therefore, low resistivity areas and areas where the resistivity contour changes dramatically have become the focus of technical personnel. However, there are two problems that cannot be ignored when relying solely on resistivity results maps for anomaly delineation. First, due to changes in lithology or geological structures, there may be regional low resistance in the resistivity results map, but it is difficult to judge whether the low resistance area is a geological anomaly. Second, the contour line spacing in the resistivity results map is usually set by the technicians themselves to ensure the aesthetics of the map, and there is no unified standard. The degree of drastic change of the contour line is judged only by the resistivity results map, which is highly subjective and does not start from the data itself, which is easy to cause misjudgment or omission. The root cause of these two problems is that the current geophysical tunnel exploration anomaly classification is mainly based on subjective judgment by technicians through experience, and no quantitative indicators have been established. Therefore, how to comprehensively consider the resistivity value and gradient change and convert them into objective indicators for delineating geological anomalies has become a topic that needs to be studied and verified. The existing patent (application number 201510870801.8) uses wavelet transform to detect singular points and superimposes them with resistivity contour maps. Its fundamental purpose is to use resistivity values ​​and resistivity gradient values ​​for anomaly delineation. The existing patent (application number 202211518847.X) uses interpolated resistivity and normalized resistivity gradient to multiply to obtain normalized resistivity values. Both of the above anomaly delineation methods comprehensively consider resistivity and resistivity gradient, but they also have certain defects. The former only shows two factors separately and is not converted into a unified indicator; the latter uses product for synthesis, which is greatly affected by absolute values. Therefore, it is necessary to propose a more complete tunnel anomaly interpretation method. Summary of the invention

[0003] Therefore, the purpose of the present invention is to provide a method and system for improving the accuracy of anomaly interpretation in tunnel electrical exploration, by weighted summing the resistivity value and the resistivity gradient after removing the abnormal points, and forming an anomaly index after secondary normalization. The anomaly index value can intuitively reflect the degree of anomaly, so as to improve the accuracy of anomaly interpretation in tunnel electrical exploration.

[0004] In order to achieve the above object, the present invention provides a method for improving the accuracy of anomaly interpretation in tunnel electrical exploration, comprising the following steps: S1. Obtain resistivity profile or three-dimensional resistivity slice to form resistivity data volume; specifically, the resistivity profile or three-dimensional resistivity slice of the work area can be obtained through data acquisition, inversion and other steps, and the resistivity of each measuring line or measuring network can be extracted to form a resistivity data volume. ; S2. For the resistivity data volume, using the position and resistivity value of each node, calculate the gradient of each point , get the gradient data volume .

[0005] S3. Normalize the resistivity data volume and the gradient data volume to obtain an inverse normalized resistivity data volume. and the normalized gradient data volume ; S4. According to the inverse normalized resistivity data With normalized gradient data volume , calculate the abnormal weight of each node and obtain the abnormal weighted data body ; S5. Perform secondary normalization on the abnormal weighted data body to obtain the abnormal index of each node and form an abnormal index data body ; S6. Extract the anomaly index, draw an anomaly index result map, and classify the tunnel anomaly level according to the numerical value.

[0006] Further, preferably, in S1, the resistivity data volume It is a two-dimensional array containing the plane coordinates, elevation and resistivity values ​​of each node.

[0007] Furthermore, preferably, if three-dimensional resistivity slices are used to form a resistivity data volume, the gradient of each point is calculated using the position and resistivity value of each node. When including: S201, respectively obtain the values ​​along the survey line, vertical survey line and along the depth direction ( , and direction); S202: For data nodes at non-border locations , whose three-dimensional coordinates are , calculate its absolute value of gradient: in, Represents data nodes respectively In three-dimensional coordinates The resistivity data value at , , Represents data nodes respectively In three-dimensional coordinates The resistivity gradient value at ; , , Represents data nodes respectively The coordinates of the next position. Note that , , For the convenience of representation, At , it indicates the same resistivity value.

[0008] Furthermore, preferably, if a two-dimensional resistivity profile is used to form a resistivity data volume, the gradient of each point is calculated using the position and resistivity value of each node. When including: then it is necessary to find the direction along the survey line and along the depth direction ( and direction), for non-boundary data nodes , whose two-dimensional coordinates are , calculate its absolute value of gradient: in, , Represents data nodes respectively At coordinates The resistivity data value at , Represents data nodes respectively At coordinates The resistivity gradient value at ; , Represents data nodes respectively The coordinates of the next position. Note that , For the convenience of representation, At , it indicates the same resistivity value.

[0009] Further, preferably, in S3, the normalizing the resistivity data volume and the gradient data volume comprises: Calculate the mean and standard deviation for the resistivity and gradient data volumes; According to the actual distribution of data, the values ​​exceeding n times the standard deviation of the mean are eliminated as extreme values; The Min-Max normalization method is used for normalization so that the minimum value of each data volume is 0 and the maximum value is 1.

[0010] Furthermore, preferably, it also includes, for the resistivity data volume after normalization, using 1 to subtract each normalized value to obtain an inverse normalized resistivity data volume with opposite weights. .

[0011] Further, preferably, in S4, for the data node , according to the inverse normalized resistivity data volume With normalized gradient data volume The calculated outlier weighted value at this point is in, For data nodes The inverted normalized resistivity data value at , and ; , , They are , and Normalized resistivity gradient values ​​in three directions; , and They are , and The weights of the three directions and their proportions are selected independently according to the exploration objectives.

[0012] Further, preferably, in S5, performing secondary normalization on the abnormal weighted data volume includes: Weighted data set for anomalies , calculate the mean and variance, and remove the data outside the 99% confidence interval. The retained data are normalized using the Min-Max normalization method so that the minimum value of each data volume is 0 and the maximum value is 1; form an abnormal index data volume .

[0013] Further, preferably, in S6, for the abnormal index data body Medium data node , according to the data node The abnormal index value of Classify abnormal levels: when When , the rock mass belongs to the V-type anomaly, which corresponds to the complete rock mass, undeveloped or closed structural surface, and good stability; when When , the rock mass belongs to the type IV anomaly, which corresponds to a relatively complete rock mass, a well-developed structural surface, and good stability; when When , the rock mass belongs to the type III anomaly, which corresponds to the relatively broken rock mass, relatively developed structural surface, and general stability; when When , the rock mass belongs to type II anomaly, corresponding to rock mass fragmentation, structural surface development, and poor stability; when When , the rock mass belongs to Class I anomaly, which corresponds to extremely fragmented rock mass, extremely developed structural surfaces and poor stability.

[0014] The present invention also provides a system for improving the accuracy of interpretation of anomalies in tunnel electrical exploration, which is used to implement the steps of the method for improving the accuracy of interpretation of anomalies in tunnel electrical exploration, including: A data acquisition module acquires resistivity profiles or three-dimensional resistivity slices to form a resistivity data volume; The gradient calculation module calculates the gradient of each point of the resistivity data volume using the position and resistivity value of each node. , get the gradient data volume ; Normalization module, which performs normalization processing on resistivity data volume and gradient data volume; The anomaly weighting module calculates the anomaly weight of each node based on the inverted normalized resistivity data volume and the normalized gradient data volume to obtain the anomaly weighted data volume. ; The secondary normalization module performs secondary normalization on the abnormal weighted data body, obtains the abnormal index of each node and forms the abnormal index data body ; The graphic interpretation module extracts the anomaly index, draws the anomaly index result map, and divides the tunnel anomaly level according to the numerical value.

[0015] The method and system disclosed in the present application for improving the accuracy of interpretation of anomalies in tunnel electrical exploration have at least the following advantages over the prior art: The present application normalizes the resistivity data volume and the gradient data volume after removing abnormal points to obtain a normalized data volume. The normalized resistivity data volume and the normalized gradient data volume are weighted and summed, and secondary normalization is performed to obtain an abnormal index data volume. The abnormal index is extracted, a normalized abnormal result map is drawn, and the tunnel anomaly is delineated according to the abnormal division principle. The present invention can digitize the abnormal index by weighted summation of resistivity data anomalies and gradient anomalies, and effectively improves the ability to identify tunnel geological anomalies. The multiple effects of resistivity and resistivity gradient are comprehensively considered to avoid the absolute influence of a single factor, and the tunnel anomaly interpretation result obtained is more objective and accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A flow chart of a method for improving the accuracy of anomaly interpretation in tunnel electrical exploration provided by the present invention.

[0017] Figure 2 A schematic diagram of the structure of a system for improving the accuracy of anomaly interpretation in tunnel electrical exploration provided by the present invention.

[0018] Figure 3It is a resistivity contour map of an embodiment of the present invention.

[0019] Figure 4 It is a contour map of anomaly index according to an embodiment of the present invention.

[0020] Figure 5 4 is a tunnel anomaly level diagram of the embodiment. DETAILED DESCRIPTION

[0021] The present invention is further described below by taking the two-dimensional resistivity profile anomaly index extraction and geological anomaly identification as examples.

[0022] like Figure 1 As shown, an embodiment of the present invention provides a method for improving the accuracy of anomaly interpretation in tunnel electrical exploration, which adopts a geological anomaly delineation method based on resistivity and resistivity gradient to construct anomaly indicators to improve the accuracy of anomaly interpretation in tunnel electrical exploration, and specifically includes the following steps: S1. Obtain a resistivity profile or a three-dimensional resistivity slice to form a resistivity data volume; Figure 2 As shown in the figure, the resistivity profile or three-dimensional resistivity slice of the work area can be obtained through data acquisition and inversion, and the resistivity of each measuring line or measuring network can be extracted to form a resistivity data volume. In S1, the resistivity data body is a two-dimensional array containing the plane coordinates, elevation and resistivity values ​​of each node. It should be noted that since the depth division corresponding to each measuring point is the same, it can be regarded that the resistivity of each measuring point conforms to the grid division. Therefore, the node of each grid is recorded as the node of the resistivity data body.

[0023] S2. For the resistivity data volume, using the position and resistivity value of each node, calculate the gradient of each point , get the gradient data volume .

[0024] Furthermore, in S2, if three-dimensional resistivity slices are used to form a resistivity data volume, the gradient of each point is calculated using the position and resistivity value of each node. When including: S201, respectively obtain the values ​​along the survey line, vertical survey line and along the depth direction ( , and direction); S202: For data nodes not at the border , whose three-dimensional coordinates are , calculate its absolute value of gradient: in, , , Represents data nodes respectively In three-dimensional coordinates The resistivity data value at , , Represents data nodes respectively In three-dimensional coordinates The resistivity gradient value at ; , , Represents data nodes respectively The coordinates of the next position. Note that , , For the convenience of representation, at the node At , it indicates the same resistivity value.

[0025] As a preferred embodiment, if a two-dimensional resistivity profile is used to form a resistivity data volume, the gradient of each point is calculated using the position and resistivity value of each node. When including: then it is necessary to find the direction along the survey line and along the depth direction ( and direction), for non-boundary data nodes , whose two-dimensional coordinates are , calculate its absolute value of gradient: in, , Represents data nodes respectively At coordinates The resistivity data value at , Represents data nodes respectively At coordinates The resistivity gradient value at ; , Represents data nodes respectively The 2D coordinates of the next position. Note that , For the convenience of representation, at the node At , it indicates the same resistivity value.

[0026] Further preferably, in S2, for the data points at the boundary On the basis of obtaining the gradient of non-boundary points, the gradient value is obtained by extrapolating the spline interpolation, Hermite interpolation and other interpolation methods in each direction.

[0027] S3. Normalize the resistivity data volume and the gradient data volume to obtain an inverse normalized resistivity data volume. and the normalized gradient data volume ; Since the resistivity value range is often large, as small as a few , up to tens of thousands If the resistivity value and the gradient value are directly superimposed, it is often difficult to grasp the order of magnitude between the resistivity value and the gradient value, resulting in a large proportion of a certain item with a large value when performing abnormal weighting. and gradient data volume First, normalization is performed to obtain the normalized resistivity data volume and gradient data volume .

[0028] Furthermore, in S3, considering that the resistivity data and gradient data as a whole are subject to or substantially subject to normal distribution, the present invention first calculates the mean and standard deviation for the resistivity data volume and the gradient data volume, and then removes the values ​​exceeding the mean n times the standard deviation as extreme values ​​according to the actual distribution of the data. After removing the outliers and extreme values, the Min-Max normalization method is used for normalization processing, so that the minimum value of each data volume is 0 and the maximum value is 1. In S3, for the resistivity data volume after normalization processing, 1 is subtracted from each standardized value to obtain a normalized resistivity data volume with opposite weights. .

[0029] S4. According to the inverse normalized resistivity data With normalized gradient data volume , calculate the abnormal weight of each node and obtain the abnormal weighted data body ; In S4, for data nodes According to the inverse normalized resistivity data volume and the normalized gradient data volume, the anomaly weighted value at this point is calculated as in, For data nodes The inverted normalized resistivity data value at , and ; , , They are , and Normalized resistivity gradient values ​​in three directions; , and They are , and The weights of the three directions are selected according to the exploration target. For example, when detecting target geological bodies such as boulders, the same weight can be used in each direction; when detecting strata, the weight can be increased appropriately. Direction weight ; If you are looking for faults, then the horizontal weight can be larger, for example and β is taken as 0.5, and α is taken as 2.

[0030] S5. Perform secondary normalization on the abnormal weighted data body to obtain the abnormal index of each node and form an abnormal index data body .

[0031] Further preferably, in S5, the abnormal weighted data body , calculate the mean and variance, and remove the data outside the 99% confidence interval. Then, the newly formed data body is normalized using the Min-Max normalization method so that the minimum value of each data body is 0 and the maximum value is 1, and the abnormal index data body is constructed. .

[0032] S6. Extract the anomaly index, draw an anomaly index result map, and classify the tunnel anomaly level according to the numerical value.

[0033] In S6, the anomaly index data body It is a data matrix containing the plane coordinates, elevation and anomaly index values ​​of each data point. For this data body, the anomaly result map is drawn using conventional drawing software. Figure 4 shown.

[0034] like Figure 5 As shown, in S6, it also includes data for abnormal index Medium data node , according to the abnormal index value of the point Classify abnormal levels: ① , the rock mass belongs to the V-type anomaly, corresponding to the complete rock mass, undeveloped or closed structural planes, and good stability; ② , the rock mass belongs to the type IV anomaly, corresponding to a relatively complete rock mass, a relatively developed structural surface, and good stability; ③ , the rock mass belongs to the type III anomaly, corresponding to the relatively broken rock mass, relatively developed structural planes, and general stability; ④ , the rock mass belongs to type II anomaly, corresponding to rock mass fragmentation, structural surface development, and poor stability; ⑤ The rock mass belongs to Class I anomaly, which corresponds to extremely fragmented rock mass, extremely developed structural surfaces and poor stability.

[0035] like Figure 2 As shown, the present invention also provides a system for improving the accuracy of anomaly interpretation in tunnel electrical exploration, which is used to implement the above anomaly index calculation process, including: The data acquisition module is used to extract all resistivity values ​​from the resistivity inversion results and form a resistivity data volume.

[0036] The gradient calculation module is used to calculate the gradients of each node in the resistivity data volume along different directions and form a gradient data volume.

[0037] The normalization module is used to normalize the resistivity data volume and resistivity gradient data volume after removing abnormal points and extreme points using the Min-Max normalization method.

[0038] The anomaly weighting module is used to add the inverted normalized resistivity data volume and the normalized gradient data volume through different weights to form an anomaly weighted data volume.

[0039] The secondary normalization module is used to normalize the abnormal weighted data body after removing the abnormal values ​​and extreme values ​​using the Min-Max normalization method, so as to obtain the abnormal index data body.

[0040] The mapping interpretation module is used to draw anomaly index result map based on the anomaly index data volume. The anomaly level is divided according to the anomaly index result map and the tunnel anomaly is circled.

[0041] Compared with the prior art, the present invention has the following beneficial effects: (1) The resistivity value and gradient value are normalized and then weighted summed. This can take into account both resistivity value anomalies and gradient anomalies when dividing anomalies, and can more accurately identify anomalies and delineate abnormal locations.

[0042] (2) Eliminating outliers and extreme values ​​from the abnormal weighted data can effectively avoid the impact of abnormal data on the overall rating; performing secondary normalization so that the anomaly index is in the range of [0, 1] can present the anomaly level as an objective indicator, avoiding the subjectivity of artificial classification of anomalies.

[0043] Obviously, the above embodiments are only examples for clear explanation, and are not intended to limit the implementation methods. For ordinary technicians in the relevant field, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived from them are still within the protection scope of the invention.

Claims

1. A method for improving the accuracy of anomaly interpretation in tunnel electrical exploration, characterized in that: The following steps are involved: S1. Obtain resistivity profile or three-dimensional resistivity slice to form resistivity data volume ; S2. For the resistivity data volume, using the position and resistivity value of each node, calculate the gradient of each point , get the gradient data volume ; S3. Normalize the resistivity data volume and the gradient data volume to obtain an inverse normalized resistivity data volume. and the normalized gradient data volume ; S4. According to the inverse normalized resistivity data With normalized gradient data volume , calculate the abnormal weight of each node and obtain the abnormal weighted data body ; S5. Perform secondary normalization on the abnormal weighted data body to obtain the abnormal index of each node and form an abnormal index data body ; S6. Extract the anomaly index, draw an anomaly index result map, and classify the tunnel anomaly level according to the numerical value.

2. The method for improving the accuracy of anomaly interpretation in tunnel electrical exploration according to claim 1, characterized in that: In S1, the resistivity data volume It is a two-dimensional array containing the plane coordinates, elevation and resistivity values ​​of each node.

3. The method for improving the accuracy of anomaly interpretation in tunnel electrical exploration according to claim 1, characterized in that: If three-dimensional resistivity slicing is used to form a resistivity data volume, the gradient of each point is calculated using the position and resistivity value of each node. When including: S201, respectively obtain the values ​​along the survey line , vertical line and along the depth direction The gradient of S202: For data nodes not at the border , whose three-dimensional coordinates are , calculate the absolute value of the gradient: in, , , Represents data nodes respectively In three-dimensional coordinates The resistivity data value at , , Represents data nodes respectively In three-dimensional coordinates The resistivity gradient value at ; , , Represents data nodes respectively The coordinates of the next position, , , For the convenience of representation, At , it indicates the same resistivity value.

4. The method for improving the accuracy of anomaly interpretation in tunnel electrical exploration according to claim 1, characterized in that: If a two-dimensional resistivity profile is used to form a resistivity data volume, the gradient of each point is calculated using the position and resistivity value of each node. When including: obtaining along the survey line as needed and along the depth Directional gradient, for data nodes not at the boundary , whose two-dimensional coordinates are , calculate the absolute value of the gradient: in, , Represents data nodes respectively At coordinates The resistivity data value at , Represents data nodes respectively At coordinates The resistivity gradient value at ; , Represents data nodes respectively The coordinates of the next position, , For the convenience of representation, At , it indicates the same resistivity value.

5. The method for improving the accuracy of interpretation of anomalies in tunnel electrical exploration according to claim 1, characterized in that: In S3, the normalizing the resistivity data volume and the gradient data volume includes: Calculate the mean and standard deviation for the resistivity and gradient data volumes; According to the actual distribution of data, the values ​​exceeding n times the standard deviation of the mean are eliminated as extreme values; The Min-Max normalization method is used for normalization so that the minimum value of each data volume is 0 and the maximum value is 1.

6. The method for improving the accuracy of interpretation of anomalies in tunnel electrical exploration according to claim 5 is characterized in that: The method also includes subtracting each normalized value from 1 for the normalized resistivity data volume to obtain an inverse normalized resistivity data volume with opposite weights. .

7. The method for improving the accuracy of interpretation of anomalies in tunnel electrical exploration according to claim 6, characterized in that: In S4, for data nodes , according to the resistivity data The abnormal weighted value at this point calculated from the resistivity gradient data volume is: in, For data nodes The inverted normalized resistivity data value at , and ; , , They are , and Normalized resistivity gradient values ​​in three directions; , and They are , and The weights of the three directions and their proportions are selected independently according to the exploration objectives.

8. The method for improving the accuracy of anomaly interpretation in tunnel electrical exploration according to claim 1, characterized in that: In S5, performing secondary normalization on the abnormal weighted data body includes: Weighted data set for anomalies , calculate the mean and variance; and remove the data outside the 99% confidence interval, and use the Min-Max normalization method to normalize the retained data so that the minimum value of the data body is 0 and the maximum value is 1; form an abnormal index data body .

9. The method for improving the accuracy of interpretation of anomalies in tunnel electrical exploration according to claim 1, characterized in that: In S6, for the abnormal index data body Medium data node , according to the data node The abnormal index value of Classify abnormal levels: when When , the rock mass belongs to the V-type anomaly, which corresponds to the complete rock mass, undeveloped or closed structural surface, and good stability; when When , the rock mass belongs to the type IV anomaly, which corresponds to a relatively complete rock mass, a well-developed structural surface, and good stability; when When , the rock mass belongs to the type III anomaly, which corresponds to the relatively broken rock mass, relatively developed structural surface, and general stability; when When , the rock mass belongs to type II anomaly, corresponding to rock mass fragmentation, structural surface development, and poor stability; when When , the rock mass belongs to Class I anomaly, which corresponds to extremely fragmented rock mass, extremely developed structural surfaces and poor stability.

10. A system for improving the accuracy of interpretation of anomalies in tunnel electrical exploration, characterized in that: The method for improving the accuracy of interpretation of anomalies in tunnel electrical exploration according to any one of claims 1 to 9 comprises: Data acquisition module, obtains resistivity profile or three-dimensional resistivity slice to form resistivity data volume ; The gradient calculation module calculates the gradient of each point of the resistivity data volume using the position and resistivity value of each node. , get the gradient data volume ; A normalization module performs normalization processing on the resistivity data volume and the gradient data volume to obtain a normalized gradient data volume and an inverse normalized resistivity data volume; The anomaly weighting module calculates the anomaly weight of each node based on the inverted normalized resistivity data volume and the normalized gradient data volume to obtain the anomaly weighted data volume. ; The secondary normalization module performs secondary normalization on the abnormal weighted data body, obtains the abnormal index of each node and forms the abnormal index data body ; The graphic interpretation module extracts the anomaly index, draws the anomaly index result map, and divides the tunnel anomaly level according to the numerical value.

Images