Deep prospecting detection system and method based on in-well and inter-well high-power induced polarization method

Abstract

The invention discloses a deep prospecting detection system and method based on an in-well and inter-well high-power induced polarization method, and relates to the field of mining area detection, and the system comprises an uploading module which is used for uploading induced polarization data of a deep geologic body of an old mining area and storing the data; the transmission module is used for receiving the data stored in the uploading module, cleaning the data and forwarding the cleaned data to the summarizing module after cleaning is completed; according to the invention, deep detection work can be carried out in an old mine area with complex electromagnetic interference, the mine environment is adapted through high-power induced polarization transmitting and receiving, the data acquisition repeatability is good, the precision is high, and deep geologic structures and ore body distribution conditions can be clearly presented through in-well and inter-well induced polarization observation in combination with a three-dimensional inversion technology.

Description

Technical Field

[0001] The present invention relates to the technical field of mine area exploration, and specifically to a deep prospecting detection system and method based on high-power induced polarization method in boreholes and between boreholes. Background Art

[0002] Deep prospecting in old mining areas is a work of great strategic significance. As shallow resources are gradually depleted, it becomes inevitable to turn to deep exploration for new treasures. Through advanced geological exploration technologies, such as geophysical and geochemical methods, potential deep ore bodies can be accurately located. This can not only extend the lifespan of old mining areas, but also alleviate the resource shortage situation and provide strong support for sustainable economic development.

[0003] The invention patent application with the application number 202410605678.6 discloses a method for separating motion noise of a ground-air electromagnetic detection system, including: constructing a data set, the data set including; a motion parameter sequence of the ground-air electromagnetic detection system, a magnetic field signal containing motion noise, and a motion noise time series; using the data set to train a preset MNRformer network model to obtain a motion noise separation model; where the preset MNRformer network model is obtained by improving the iTransformer network model; improving the iTransformer network model includes; introducing an improved multi-attention mechanism layer combining self-attention mechanism and cross-attention mechanism to replace the original multi-attention mechanism layer of the iTransformer network model; using the data set to train the MNRformer network model includes; using the motion parameter sequence and the magnetic field signal containing motion noise as inputs, the motion noise time series as outputs, and defining the magnetic field signal containing motion noise as an internal variable and the motion parameter sequence as an external variable to train the MNRformer network model. This application solves the problem that "LSTM is difficult to achieve parallel computing and is less efficient in processing large-scale ground-air / aerial electromagnetic data. In addition, LSTM only interprets the short-term correlation and change trend of a single time series itself, cannot establish long-term dependencies and cannot reveal the correlation between multiple variables, and is difficult to meet the requirements of separating low-frequency motion noise of the ground-air electromagnetic detection system".

[0004] However, for the deep prospecting detection scenario in old mining areas, there is strong electromagnetic interference in old mining areas, and it is difficult for conventional surface geophysical methods to obtain accurate deep geological data, so it is difficult to carry out prospecting work.

[0005] Therefore, a deep prospecting detection system and method based on high-power induced polarization method in boreholes and between boreholes are proposed. Summary of the Invention

[0006] In view of the above-mentioned disadvantages of the prior art, the present invention provides a deep prospecting detection system and method based on high-power induced polarization method in boreholes and between boreholes, which can effectively solve the problems of the prior art.

[0007] To achieve the above objectives, the present invention is realized through the following technical solutions; The present invention discloses a deep prospecting detection system based on high-power induced polarization method in boreholes and between boreholes, including: An upload module, which is used to upload the induced polarization data of deep geological bodies in old mining areas and store the data; a transmission module, which is used to receive the data stored in the upload module, clean the data, and after the cleaning is completed, forward the cleaned data to the summary module; a summary module, which is used to continuously obtain the cleaned data in the transmission module, summarize the continuously obtained cleaned data, and import it into the inversion module; an inversion module, which is used to generate a geological model by applying the data and identify potential ore body scale information in the geological model; a storage module, which is used to receive the identification result of the potential ore body scale information in the inversion module and store the identification result; a visualization module, which is used to read the ore body scale information stored in the storage module and generate a three-dimensional model of the ore body based on the ore body scale information; Furthermore, when the upload module stores the data, several intervals are set, and the data is stored separately based on the several intervals; Each interval for storing data corresponds to four spatial coordinates and a height value. The four spatial coordinates corresponding to each interval define a plane, and combined with the height value, a spatial area in the deep geological body of the old mining area is defined. The data stored in the interval all comes from the corresponding spatial area; Among them, the specification parameters and sizes of the spatial areas corresponding to each interval for storing data are equal.

[0008] Furthermore, the induced polarization data of the deep geological body of the old mining area uploaded during the operation stage of the upload module includes: potential data, resistivity data, polarization rate data, in-well electromagnetic background noise, and complex resistivity, density, and magnetic susceptibility of rock samples; When the transmission module cleans the data, the content of the cleaning operation includes: deleting duplicate data, discarding incomplete data, denoising, and normalization processing; Among them, when the transmission module performs the cleaning operation, the found incomplete data is fed back to the system-side user, and the system-side user customizes the decision whether to re-upload the complete data corresponding to the incomplete data. After the re-upload operation is performed, the cleaning operation of the data in the transmission module is synchronously refreshed and executed until there is no data re-uploaded and then it ends.

[0009] Further, a data retrieval unit and a control unit are provided at a lower level of the transmission module. The data retrieval unit is configured to retrieve the data stored in the upload module and send the retrieved data to the transmission module. The control unit is configured to monitor the operating states of the transmission module and the summary module. After the transmission module finishes the forwarding operation of the cleaned data and the summary module finishes running based on the received data, the control unit performs a jump and returns to the operating stage of the transmission module; During the operation stage of the data retrieval unit, the range of the data stored in the upload module is used as the target for retrieving data in a single run, and all the data in the retrieved range is sent to the transmission module.

[0010] Further, the summary module performs an abnormal enhancement summarization operation guided by the geoelectric structure tensor on the continuously acquired induced polarization data. The operation includes the following steps: a) Based on the induced polarization data points and their three-dimensional spatial coordinates corresponding to each interval collected by the upload module, an initial three-dimensional geoelectric parameter field covering the entire detection area is constructed as a spatial region matrix; b) For each cell c in the initial three-dimensional geoelectric parameter field , calculate the local geoelectric structure tensor within its neighborhood : , where is the local geoelectric structure tensor at cell c; is the local calculation window centered on cell c; is the weighting function within the window; is the gradient vector of the initial geoelectric parameter field at point r; denotes the transpose of the vector; the local geoelectric structure tensor is used to characterize the local change direction and anisotropy degree of the parameter field; c) Based on the eigenvalues of the local geoelectric structure tensor and the corresponding principal eigenvectors , and in combination with the preset preferred geological structure direction vector , define and apply a directional anomaly response function to calculate the enhanced anomaly response value at cell c; where is expressed as: , where is the enhanced anomaly response value at cell c; is a positive modulation coefficient; is a small positive constant to prevent the denominator from being zero; The deviation of the original induced polarization parameter value at cell c relative to the background value ; d) Generate a three-dimensional geoelectric parameter field after anomaly enhancement summary , where the value of each cell c is given by and transmit it to the inversion module.

[0011] Furthermore, the inversion module adopts a guided hybrid regularization inversion framework and is integrated by guiding the regularization process of the RES3DINV inversion software through the model structure generated by the Occam inversion algorithm; the guided hybrid regularization inversion framework includes the following two stages executed in sequence: Stage 1: Receive the three-dimensional geoelectric parameter field after anomaly enhancement summary from the summary module as observation data and perform preliminary inversion using the Occam inversion algorithm; among them, the smoothing constraint weight in the model constraint term of the preliminary inversion is adjusted spatially adaptively according to the geometric distribution of the boreholes and the sensitivity of each model unit to the observation data, so as to allow the model to have a higher parameter change gradient in the region with strong data constraints and apply a strong smoothing constraint in the region with weak data constraints, generating a preliminary three-dimensional geoelectric parameter model containing the structural characteristics determined by the spatially adaptive adjusted smoothing constraint; Stage 2: Use the RES3DINV inversion software to perform fine inversion on the preliminary three-dimensional geoelectric parameter model and its contained structural characteristics; among them, extract the structural information from the preliminary three-dimensional geoelectric parameter model and introduce a structural guiding constraint in the regularization process of the RES3DINV inversion software using this structural information. This structural guiding constraint makes a strong smoothing constraint be applied in the region where the preliminary three-dimensional geoelectric parameter model shows smoothness, while in the region where the preliminary three-dimensional geoelectric parameter model shows significant structural changes, the model is allowed to generate sharper boundaries or more significant anomaly changes, and finally obtain a fine three-dimensional geoelectric parameter distribution for identifying the position, shape, and extension span of the ore body.

[0012] Furthermore: The Occam inversion algorithm in Stage 1 solves the preliminary three-dimensional geoelectric parameter model by minimizing the objective function , and the objective function is expressed as: , where is the observation data; is the three-dimensional geoelectric parameter model to be inverted; is the data fitting term; is the forward response operator calculated according to the model ; is the data weighting matrix; is the model constraint term; is a prior reference model; is a spatial weighting operator for model parameters; is a regularization parameter that controls the balance between the data fitting term and the model constraint term; The structure-guided constraint introduced in the second stage is achieved by adding a structure-guided hybrid regularization term to the objective function of the regularization process in the RES3DINV inversion software The regularization term is designed as: , where is the three-dimensional geoelectric parameter model being optimized by RES3DINV; is the preliminary three-dimensional geoelectric parameter model; the first term is the structure-preserving smoothing constraint term, and its spatial weight function is anti-correlated with the gradient magnitude at position r; the second term is the reference model constraint term, and its spatial weight is set according to the degree of trust in different regions of at position r; , are regularization parameters for balancing each term.

[0013] Furthermore, the visualization module is integrated by a computer device with a display function, and the three-dimensional ore body model generated by the operation of the visualization module is displayed in real time through the computer device for visual reading by the system-end user.

[0014] Furthermore, the upload module is interconnected with a transmission module and a summary module through a local area network. The lower level of the transmission module is interconnected with a retrieval unit and a control unit through a local area network. The retrieval unit is interconnected with the upload module through a local area network. The control unit is interconnected with the summary module through a local area network. The summary module is interconnected with the inversion module, the storage module, and the visualization module through a local area network.

[0015] A deep prospecting detection method based on the high-power induced polarization method in boreholes and between boreholes includes: Uploading the induced polarization data of the deep geological body in the old mining area and storing the data; retrieving and cleaning the data from the stored data, and after all the data is cleaned, summarizing the data; performing inversion on the summarized data based on the Occam inversion algorithm and RES3DINV to obtain the scale information of the potential ore bodies in the deep geological body of the old mining area; storing the potential ore body scale information, and constructing a three-dimensional model of the potential ore body according to the potential ore body scale information; sending the three-dimensional ore body model to a computer device with a display function, and displaying the three-dimensional ore body model through the computer device for visual reading by the user.

[0016] Compared with the known prior art, the technical solution provided by the present invention has the following beneficial effects: 1. The present invention can carry out deep exploration in old mining areas with complex electromagnetic interference. It can adapt to the mine environment through high-power induced polarization transmission and reception. The data collection has good repeatability and high accuracy. Through in-well and inter-well induced polarization observations, combined with three-dimensional inversion technology, it can clearly present the deep geological structure and ore body distribution.

[0017] 2. The present invention can effectively explore the distribution characteristics of ore bodies around known drill holes, discover deep mineral anomalies, accurately locate potential ore bodies, improve prospecting efficiency, reduce blind exploration, provide key technical support for the secondary development of resources in old mining areas, and promote the sustainable development of the mining industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0019] Figure 1 It is a structural diagram of a deep prospecting and detection system based on the high-power induced polarization method in wells and between wells; Figure 2 It is a flow chart of the deep prospecting and detection method based on the high-power induced polarization method in wells and between wells. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0021] The present invention will be further described below in conjunction with the embodiments.

[0022] Embodiment 1: The deep prospecting detection system based on the high-power IP method in the well and between wells of this embodiment is as follows: Figure 1 As shown, including: Upload module 1, used to upload the IP data of deep geological bodies in old mining areas and store the data; When the upload module 1 stores data, several intervals are set, and the data is stored separately based on the several intervals; Each interval for storing data corresponds to four spatial coordinates and a height value. The four spatial coordinates corresponding to each interval define a plane, and together with the height value, they define a spatial region in the deep geological body of the old mining area. The data stored in the interval all come from the corresponding spatial region; Among them, the specification parameters and sizes of the spatial regions corresponding to each interval for storing data are equal; The induced polarization data of the deep geological body of the old mining area uploaded during the operation stage of the upload module 1 include: potential data, resistivity data, polarization rate data, in-well electromagnetic background noise, and the complex resistivity, density, and magnetic susceptibility of rock samples; When the data is cleaned in the transmission module 2, the content of the cleaning operation includes: deletion of duplicate data, discarding of incomplete data, denoising, and normalization processing; Among them, when the transmission module 2 performs the cleaning operation, the incomplete data found is fed back to the system-side user. The system-side user customizes the decision whether to re-upload the complete data corresponding to the incomplete data. After the re-upload operation is executed, the cleaning operation of the data in the transmission module 2 is synchronously refreshed and executed until there is no data re-uploaded and then it ends; The transmission module 2 is used to receive the data stored in the upload module 1, clean the data, and after the cleaning is completed, forward the cleaned data to the summary module 3; There is a retrieval unit 21 and a control unit 22 under the transmission module 2. The retrieval unit 21 is used to retrieve the data stored in the upload module 1 and send the retrieved data to the transmission module 2. The control unit 22 is used to monitor the operating states of the transmission module 2 and the summary module 3. After the transmission module 2 completes the forwarding operation of the cleaned data and the summary module 3 finishes running based on the received data, it controls the jump and returns to the operation stage of the transmission module 2 again; Among them, during the operation stage of the retrieval unit 21, the intervals of the data stored in the upload module 1 are used as the targets for retrieving data in a single run, and all the data in the intervals are retrieved and sent to the transmission module 2; The summary module 3 is used to continuously obtain the cleaned data in the transmission module 2, summarize the continuously obtained cleaned data, and import it into the inversion module 4; The summary operation logic of continuously obtaining data in the summary module 3 is: Based on the spatial region matrix constructed from the spatial regions corresponding to each interval in the upload module 1, place the data belonging to each spatial region, and perform anomaly enhancement and summarization on the data in the spatial region matrix using geoelectric structure tensor analysis and directional response function; After the summarization is completed, it jumps to the operation of the inversion module 4.

[0023] Specifically, after receiving the data cleaned by the transmission module 2, the summary module 3 first constructs a spatial region matrix and executes an anomaly enhancement summary method guided by the tensor of geoelectric structure, and the specific steps are as follows: 1. Construction and preliminary fusion of the spatial region matrix: The spatial region matrix is a structured expression of the corresponding spatial regions in each interval of the upload module (1). It is not just a simple division of the physical space, but a data structure with data organization and analysis functions. The construction method of the spatial region matrix is as follows: First, based on the spatial region defined by the four spatial coordinates and height values corresponding to each interval, a three-dimensional grid structure is constructed. The resolution of this grid can be dynamically determined based on exploration requirements and data density to ensure that it can fully express the geological change characteristics without causing redundant calculations. Each grid cell corresponds to a three-dimensional spatial voxel for storing the geoelectric parameters at that position.

[0024] Formally expressed as: For the spatial region corresponding to the k-th interval, construct the spatial region matrix , where the matrix element represents the geoelectric parameter value at the spatial position . If there are no data points directly falling into some cells, they are temporarily marked as empty or preliminary interpolation is performed using neighboring data. Through this process, an initial three-dimensional geoelectric parameter field covering the entire detection area is formed. This parameter field is a spatialized data set, where each cell c has a preliminary induced polarization parameter value .

[0025] 2. Calculation of the local geoelectric structure tensor: To capture the local structural characteristics of the electrical property parameters of the underground medium, for each cell c in the initial geoelectric parameter field , calculate the local geoelectric structure tensor within its neighborhood. This tensor can describe the change direction and anisotropy degree of the parameter field in the local area. For three-dimensional data, the structure tensor can be obtained by weighted averaging the outer product of the gradients within a local window : , In the formula, is the local geoelectric structure tensor at cell c (with the center coordinate ). is the local calculation window centered on cell c. is the weighting function within the window. is the gradient vector of the initial geoelectric parameter field at point r. represents the transpose of the vector. The eigenvalues and the corresponding eigenvectors Reveals the variation characteristics of the local geoelectric parameter field: Indicates the direction with the largest parameter change, Characterizes the change intensity in this direction.

[0026] 3. Definition of the directional anomaly response function: Based on geological prior information, set one or more preferred geological structure direction vectors . Then, define a directional anomaly response function , which is used to enhance the geoelectric anomalies that are consistent with the direction and have significant structures: , where is the enhanced anomaly response value at cell c. is a positive modulation coefficient used to control the intensity of structure information enhancement. is an index characterizing the linearity or planarity of the structure, is a small positive constant to prevent the denominator from being zero. represents the square of the alignment degree between the main change direction of the parameter field and the preferred geological structure direction, with a value ranging from [0,1], and the higher the alignment degree, the larger the value. is the deviation of the original induced polarization parameter value at cell c from the background value , representing the amplitude of the basic anomaly signal.

[0027] 4. Generate enhanced summary data: Finally, the summary data field output by the summary module 3 for use by the inversion module 4 , and the value of each cell c is given by . That is: , such summary data not only synthesizes the measurement information of all intervals, but also enhances the geoelectric anomaly signals related to the target geological structure and mineralization through the application of structural tensor analysis and directional response functions, while possibly suppressing some random noise and non-target responses.

[0028] The inversion module 4 is used to apply the data to generate a geological model and identify potential ore body scale information in the geological model; During the operation stage of the inversion module 4, an initial geoelectric model containing structure information is established through preliminary inversion considering borehole geometry and data sensitivity and combined with the geological structure, ore deposit physical property characteristics and existing geological data in the study area. The structure information of the initial geoelectric model is used to guide the regularization process of the multi-source induced polarization three-dimensional inversion software. The processed induced polarization data is input, and based on iterative calculations, the model parameters are repeatedly adjusted to make the model response reach the best fit with the measurement data, so as to obtain the three-dimensional spatial distribution of underground resistivity and polarization rate, and identify the position, shape and extension span of the ore body in the model; Among them, in the geoelectric model construction stage, it is based on the geological model, and the potential ore body scale information is the position, shape, and extension span of the ore body; The inversion module 4 is integrated by guiding the regularization process of the RES3DINV inversion software through the model structure generated by the Occam inversion algorithm; Among them, RES3DINV is the inversion software.

[0029] Specifically, in order to more effectively utilize the high-power induced polarization data in boreholes and between boreholes and adapt to the complex geological conditions of old mining areas, the present invention adopts a guided hybrid regularization inversion framework in the inversion module 4. The content of this framework is as follows: Stage 1: Occam-like preliminary inversion based on borehole geometry and data sensitivity constraints This stage aims to use the Occam-like algorithm to perform preliminary inversion on the enhanced summary data output by the summary module 3, generating a relatively smooth background / reference model in terms of structure but capable of well fitting the main data features and reflecting the borehole geometry constraints and data sensitivity distribution .

[0030] Its objective function can be expressed as: , , , Among them, is the three-dimensional geoelectric parameter model to be inverted; is the data fitting term; is the summarized observed data the actual data vector used for inversion after being sorted out; is the forward response operator calculated according to the model ; is the data weighting matrix, and its weight can be determined according to the signal-to-noise ratio of the original data and the contribution degree of each data point to the structure enhancement calculated in the GST-AEA process; is the regularization parameter, controlling the balance between the data fitting term and the model constraint term; is the model constraint term (regularization term); is a prior reference model, which can be a uniform half-space model or an initial model based on the geological understanding of a large area; is the spatial weighting sum (or difference) operator of the model parameters, and the present invention has improved it: The weight coefficients will be adaptively adjusted spatially according to the geometric distribution of boreholes and the sensitivity of each model unit to the in-well and between-well observation data (which can be obtained from the Jacobian matrix or its approximation). Specifically, in the areas near boreholes and with high data sensitivity between boreholes, relatively weak smoothing constraints are imposed (allowing the model to have a higher parameter change gradient); while in the deep or edge areas far from boreholes and with weak data constraints, strong smoothing constraints are imposed to ensure the stability and geological rationality of the model. In addition, directional smoothing can also be introduced. For example, greater parameter continuity is allowed in the direction parallel to the main geological structures (such as bedding planes, faults), while more significant parameter jumps are allowed in the vertical direction.

[0031] Phase II: Structure-guided RES3DINV fine inversion In this phase, iterative inversion is carried out using three-dimensional inversion software. The core lies in using the model generated in Phase I to guide the regularization process to better recover the ore bodies in the deep and complex shapes while maintaining the stability of the model.

[0032] The inversion of RES3DINV-like software usually also optimizes an objective function containing a data fitting term and a model regularization term. The present invention improves its regularization term to form a structure-guided hybrid regularization as follows: , wherein, is the three-dimensional geoelectric parameter model being optimized by RES3DINV.

[0033] The first term is the structure-preserving smoothing constraint term: is the L2 norm of the model parameter gradient, promoting model smoothing; is the spatial weight function, and this weight is anti-correlated with the structural information extracted from , that is, anti-correlated with the gradient magnitude at position r . That is, in the areas where appears smooth ( is small), is large, encouraging to also remain smooth; while in the areas where shows significant structural changes ( is large, which may indicate geological boundaries or the edges of abnormal bodies), is small, allowing to generate sharper boundaries or more significant abnormal changes in these areas, thus better recovering local details.

[0034] The second term is the reference model constraint term: Penalize the current model and the reference model obtained in Phase I for the difference between them. Its spatial weight can be set according to the degree of trust in different regions at position r.

[0035] , is the hyperparameter that balances these two regularization terms and the data fitting term.

[0036] In addition, during the RES3DINV inversion process, the mesh discretization can also be locally refined according to the key regions identified in . Data weighting should also consider the characteristics of the high-power IP method in boreholes and between boreholes, such as assigning higher weights to data with high sensitivity and high signal-to-noise ratio in the deep part.

[0037] Through this two-stage, structure-guided hybrid regularization inversion framework, the inversion module 4 can make more full use of the high-power IP data in boreholes and between boreholes, overcome the influence of complex geological conditions and electromagnetic interference in old mining areas, and the generated geological model can not only better fit the observed data, but also clearly and accurately reflect the spatial position, occurrence and scale information of potential ore bodies in the deep part.

[0038] The storage module 5 is used to receive the identification results of the scale information of potential ore bodies in the inversion module 4 and store the identification results; The visualization module 6 is used to read the ore body scale information stored in the storage module and generate a three-dimensional model of the ore body based on the ore body scale information; The visualization module 6 is integrated by a computer device with a display function, and the three-dimensional model of the ore body generated by the operation of the visualization module 6 is displayed in real time through the computer device for visual reading by the system-end users; The upload module 1 is interconnected with the transmission module 2 and the summary module 3 through local area network interaction. The lower level of the transmission module 2 is interconnected with the retrieval unit 21 and the control unit 22 through local area network interaction. The retrieval unit 21 is interconnected with the upload module 1 through local area network, and the control unit 22 is interconnected with the summary module 3 through local area network. The summary module 3 is interconnected with the inversion module 4, the storage module 5 and the visualization module 6 through local area network.

[0039] ​In this embodiment, the uploading module 1 operates to upload the induced polarization data of the deep geological bodies in the old mining area, store the data, and the transmission module 2 operates later to receive the data stored in the uploading module 1, clean the data, and after the cleaning is completed, forward the cleaned data to the summarizing module 3. The retrieving unit 21 synchronously retrieves the data stored in the uploading module 1 and sends the retrieved data to the transmission module 2. The control unit 22 monitors the operating states of the transmission module 2 and the summarizing module 3 in real time. After the transmission module 2 completes the forwarding operation of the cleaned data and the summarizing module 3 finishes running based on the received data, it controls the jump and returns to the running stage of the transmission module 2 again. Then, the summarizing module 3 continuously obtains the cleaned data in the transmission module 2, summarizes the continuously obtained cleaned data, and imports it into the inversion module 4. The inversion module 4 simultaneously applies the data to generate a geological model, identifies the potential ore body scale information in the geological model, and finally, the storage module 5 receives the identification result of the potential ore body scale information in the inversion module 4 and stores the identification result. The visualization module 6 reads the ore body scale information stored in the storage module and generates a three-dimensional ore body model based on the ore body scale information.

[0040] Through the operation of the system in the above embodiment, it provides technical support for deep prospecting in the old mining area, enabling further development of the deep minerals in the old mining area.

[0041] Embodiment 2: At the specific implementation level, on the basis of Embodiment 1, this embodiment refers to Figure 2 to further specifically describe the deep prospecting detection system based on the in-well and cross-well high-power induced polarization method in Embodiment 1: The deep prospecting detection method based on the in-well and cross-well high-power induced polarization method includes: Step 1: Upload the induced polarization data of the deep geological bodies in the old mining area and store the data; Step 2: Retrieve and clean the data from the stored data. After all the data is cleaned, summarize the data; Step 3: Invert the summarized data based on the Occam inversion algorithm and RES3DINV to obtain the scale information of the potential ore bodies in the deep geological bodies of the old mining area; Step 4: Store the potential ore body scale information and construct a three-dimensional model of the potential ore body according to the scale information of the potential ore body; Step 5: Send the three-dimensional ore body model to a computer device with a display function, and display the three-dimensional ore body model through the computer device for visual reading by the user.

[0042] In summary, the system and method in the above embodiments can carry out deep exploration in old mining areas with complex electromagnetic interference, adapt to the mine environment through high-power induced polarization transmission and reception, and have good data acquisition repeatability and high accuracy. Through in-well and inter-well induced polarization observations, combined with three-dimensional inversion technology, the deep geological structure and ore body distribution can be clearly presented. At the same time, for the expansion and deepening of old mining areas, it can effectively explore the distribution characteristics of ore bodies around known drill holes, discover deep mineral anomalies, accurately locate potential ore bodies, improve prospecting efficiency, reduce exploration blindness, provide key technical support for the secondary development of resources in old mining areas, and promote the sustainable development of the mining industry.

[0043] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A deep prospecting detection system based on high-power IP method in borehole and cross-borehole, characterized in that, Including: An upload module (1) for uploading IP data of deep geological bodies in old mining areas and storing the data; A transmission module (2) for receiving the data stored in the upload module (1), cleaning the data, and after the cleaning is completed, forwarding the cleaned data to a summary module (3); A summary module (3) for continuously obtaining the cleaned data in the transmission module (2), summarizing the continuously obtained cleaned data, and importing it into an inversion module (4); An inversion module (4) for applying the data to generate a geological model and identifying potential ore body scale information in the geological model; A storage module (5) for receiving the identification results of the potential ore body scale information in the inversion module (4) and storing the identification results; A visualization module (6) for reading the ore body scale information stored in the storage module and generating a three-dimensional ore body model based on the ore body scale information; The summary module (3) performs an anomaly enhancement summary operation guided by the geoelectric structure tensor on the continuously obtained IP data, and this operation includes the following steps: a) Based on the induced polarization data points corresponding to each interval collected by the upload module (1) and their three-dimensional space coordinates, construct an initial three-dimensional geoelectric parameter field covering the entire detection area as a spatial region matrix; b) Calculate the initial three-dimensional geoelectric parameter field For each cell c in , calculate the local geoelectric structure tensor within its neighborhood : ， In the formula, is the local electrogeometric structure tensor at cell c; is the local calculation window centered on cell c; is the weighting function within the window; is the gradient vector of the initial electrogeometric parameter field at point r; represents the transpose of the vector; the local electrogeometric structure tensor is used to characterize the local change direction and anisotropy degree of the parameter field; c) Based on the local electro-structural tensor eigenvalues and the corresponding principal eigenvectors , and combined with a preset preferred geological structure direction vector , define and apply a directional anomaly response function to calculate the enhanced anomaly response value at cell c; where is expressed as: ， Among them, is the enhanced abnormal response value at cell c; is a positive modulation coefficient; is a small normal constant to prevent the denominator from being zero; is the deviation of the original induced polarization parameter value at cell c relative to the background value ; d) Generate a three-dimensional geoelectric parameter field after abnormal enhancement summary , where the value of each cell c is given by and is passed to the inversion module (4).

2. The deep prospecting detection system based on the high-power induced polarization method in and between wells according to claim 1, characterized in that, When the upload module (1) stores the data, a number of intervals are set, and the data is stored separately based on the number of intervals; Each interval for storing data corresponds to four spatial coordinates and a height value. The four spatial coordinates corresponding to each interval define a plane, and in combination with the height value, a spatial area in the deep geological body of the old mining area is defined. The data stored in the interval all comes from the corresponding spatial area; Among them, the specification parameters and sizes of the spatial areas corresponding to each interval for storing data are equal.

3. The deep prospecting detection system based on the high-power induced polarization method in and between wells according to claim 1, wherein, The IP data of the deep geological body of the old mining area uploaded during the operation stage of the upload module (1) includes: potential data, resistivity data, polarizability data, in-well electromagnetic background noise, and complex resistivity, density, and magnetic susceptibility of rock samples; When the transmission module (2) cleans the data, the content of the cleaning operation includes: deleting duplicate data, discarding incomplete data, denoising, and normalization processing; Among them, when the transmission module (2) performs the cleaning operation, the found incomplete data is fed back to the system-end user. The system-end user customizes the decision whether to re-upload the complete data corresponding to the incomplete data. After the re-upload operation is performed, the cleaning operation of the data in the transmission module (2) is synchronously refreshed and executed until there is no data re-uploaded and then it ends.

4. The deep prospecting detection system based on high-power IP method in borehole and cross-hole according to claim 1, characterized in that, A retrieval unit (21) and a control unit (22) are provided at the lower level of the transmission module (2). The retrieval unit (21) is used to retrieve the data stored in the upload module (1) and send the retrieved data to the transmission module (2). The control unit (22) is used to monitor the operating states of the transmission module (2) and the summary module (3). After the transmission module (2) completes the forwarding operation of the cleaned data and the summary module (3) finishes running based on the received data, it controls the jump and returns to the operation stage of the transmission module (2) again; Among them, during the operation stage of the retrieval unit (21), the intervals of the data stored in the upload module (1) are used as the targets for retrieving data in a single run, and all the data in the retrieved intervals is sent to the transmission module (2).

5. The deep prospecting detection system based on the high-power induced polarization method in boreholes and between boreholes according to claim 1, characterized in that, The inversion module (4) adopts a guided hybrid regularization inversion framework, which is integrated by guiding the regularization process of the RES3DINV inversion software through the model structure generated by the Occam inversion algorithm; the guided hybrid regularization inversion framework includes the following two stages executed in sequence: Stage 1: Receive the three-dimensional geoelectric parameter field after anomaly-enhanced summarization from the summarization module (3) as the observed data, and perform preliminary inversion using the Occam inversion algorithm; among them, the smoothing constraint weight in the model constraint term of the preliminary inversion is adjusted spatially adaptively according to the geometric distribution of the boreholes and the sensitivity of each model unit to the observed data, so as to allow the model to have a higher parameter change gradient in the region with strong data constraints, and apply a strong smoothing constraint in the region with weak data constraints, generating a preliminary three-dimensional geoelectric parameter model containing the structural features determined by the spatially adaptive adjustment of the smoothing constraint; Stage 2: Use the RES3DINV inversion software to perform fine inversion on the preliminary three-dimensional geoelectric parameter model and its contained structural features; among them, extract the structural information from the preliminary three-dimensional geoelectric parameter model, and introduce a structural guiding constraint in the regularization process of the RES3DINV inversion software using this structural information. This structural guiding constraint makes a strong smoothing constraint applied in the region where the preliminary three-dimensional geoelectric parameter model shows smoothness, while in the region where the preliminary three-dimensional geoelectric parameter model shows significant structural changes, the model is allowed to generate a sharper boundary or more significant anomaly changes, and finally obtain the fine three-dimensional geoelectric parameter distribution for identifying the position, shape, and extension span of the ore body.

6. The deep prospecting detection system based on the in-well and cross-well high-power induced polarization method according to claim 5, characterized in that: The Occam inversion algorithm in the first stage solves the preliminary three-dimensional geoelectric parameter model by minimizing the objective function as follows, where the objective function is expressed as: ， Among them, is the observed data; is the three-dimensional geoelectric parameter model to be inverted; is the data fitting term; is the forward response operator calculated according to the model ; is the data weighting matrix; is the model constraint term; is the prior reference model; is the spatial weighting operator of the model parameters; is the regularization parameter, which controls the balance between the data fitting term and the model constraint term; The structural guidance constraint introduced in the second stage is achieved by adding a structurally guided hybrid regularization term to the objective function in the regularization process of the RES3DINV inversion software. This regularization term is designed as follows: ， Among them, is the three-dimensional geoelectric parameter model being optimized by RES3DINV; is the preliminary three-dimensional geoelectric parameter model; the first term is the smooth constraint term for structure preservation, and its spatial weight function is inversely correlated with the gradient magnitude at position r; the second term is the reference model constraint term, and its spatial weight is set according to the degree of trust in different regions at position r; , are the regularization parameters for balancing each term.

7. The deep prospecting detection system based on the high-power induced polarization method in borehole and cross-hole according to claim 1, characterized in that, The visualization module (6) is integrated by a computer device with a display function, and the three-dimensional model of the ore body generated by the operation of the visualization module (6) is displayed in real time through the computer device for visual reading by the system-side user.

8. The deep prospecting detection system based on the high-power induced polarization method in and between wells according to claim 1, characterized in that, The uploading module (1) is interconnected with a transmission module (2) and a summarization module (3) through local area network interaction. The lower level of the transmission module (2) is interconnected with a retrieval unit (21) and a control unit (22) through local area network interaction. The retrieval unit (21) is interconnected with the uploading module (1) through local area network. The control unit (22) is interconnected with the summarization module (3) through local area network. The summarization module (3) is interconnected with the inversion module (4), the storage module (5), and the visualization module (6) through local area network.

9. A deep prospecting detection method based on in-well and cross-well high-power induced polarization method, which is an implementation method of the deep prospecting detection system based on in-well and cross-well high-power induced polarization method according to any one of claims 1-8, characterized in that, Including: Step 1: Upload the induced polarization data of the deep geological body in the old mining area and store the data; Step 2: Retrieve and clean the data from the stored data, and summarize the data after all the data is cleaned; Step 3: Invert the summarized data based on the Occam inversion algorithm and RES3DINV to obtain the scale information of the potential ore body in the deep geological body of the old mining area; Step 4: Store the scale information of the potential ore body, and construct a three-dimensional model of the potential ore body according to the scale information of the potential ore body; Step 5: Send the 3D model of the ore body to a computer device with a display function, and display the 3D model of the ore body through the computer device for visual reading by the user.

Images