Geothermal resource exploration intelligent decision-making method and system based on big data

Abstract

The invention relates to the technical field of geothermal resource exploration, in particular to an intelligent decision-making method and system for geothermal resource exploration based on big data. The method comprises the following steps: acquiring target data of a target area, preprocessing the target data to obtain preprocessed data, inputting the preprocessed data as a training sample into an exploration model to train the exploration model to obtain a trained exploration model, and finally acquiring the target data of a detected area. And inputting the target data of the detected area to the trained exploration model, obtaining the geothermal resource evaluation of the detected area, carrying out geological logging according to the geothermal resource evaluation of the detected area, obtaining geological logging data, and screening the geothermal area based on the geological logging data. According to the method, multi-source data fusion is realized, and the multi-source data is used as a training sample of the exploration model, so that the efficiency of geothermal resource evaluation of the detected area by the trained exploration model is improved, and the accuracy of the exploration model is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of geothermal resource exploration, and in particular to a geothermal resource exploration intelligent decision-making method and system based on big data. Background Art

[0002] Geothermal energy is natural heat energy extracted from the earth's crust. This energy comes from the molten rock inside the earth and exists in the form of heat. It is the energy that causes volcanic eruptions and earthquakes. As natural heat energy hidden inside the earth and a highly competitive clean and renewable energy source, geothermal energy has the advantages of large reserves, wide distribution, green and low-carbon, strong applicability, and good stability. High-temperature geothermal resources will become one of the main forces for future energy structure adjustment and development of clean energy.

[0003] Publication number CN118938351A discloses a deep geothermal resource exploration method and system based on big data. The method comprises the following steps: determining the location of an exploration hole in a determined upper plate area of ​​a fault, drilling and logging at the exploration hole location according to a set drilling depth, and obtaining logging results. The logging results are analyzed for coincidence with geophysical interpretation to determine geothermal energy reserves. However, in the prior art, the judgment of geothermal resources is highly subjective, which makes it difficult to determine the area where the geothermal resources are located. Summary of the invention

[0004] The purpose of the present invention is to address the problems existing in the background technology and to propose an intelligent decision-making method and system for geothermal resource exploration based on big data.

[0005] The technical solution of the present invention:

[0006] On the one hand, the present application provides a geothermal resource exploration intelligent decision-making method based on big data, comprising:

[0007] Set acquisition parameters, acquire target data of the target area, pre-process the target data, and obtain pre-processed data;

[0008] Creating an exploration model, inputting the preprocessed data as training samples into the exploration model to train the exploration model, and obtaining a trained exploration model;

[0009] Collect target data of the measured area, input the target data of the measured area into the trained exploration model, obtain geothermal resource evaluation of the measured area, perform geological logging based on the geothermal resource evaluation of the measured area, obtain geological logging data, and screen geothermal areas based on the geological logging data.

[0010] Preferably, setting acquisition parameters, acquiring target data of the target area, and preprocessing the target data to obtain preprocessed data include:

[0011] Create geothermal data table;

[0012] Setting acquisition parameters, and acquiring a plurality of target data of the target area based on the acquisition parameters; the target data includes fracture layer data and target images;

[0013] A target data is randomly selected from the geothermal data table, and the selected target data is preprocessed to obtain preprocessed data, until all the target data in the geothermal data table are selected.

[0014] Preferably, a target data is randomly selected from the geothermal data table, and the selected target data is preprocessed to obtain preprocessed data, until all target data in the geothermal data table are selected, including:

[0015] Create a preprocessing model; the preprocessing model is based on a CNN convolutional neural network and includes a convolutional layer, a pooling layer, and a fully connected layer;

[0016] A target image of target data is randomly selected from the geothermal data table;

[0017] The target image is input into the preprocessing model, the image features of the target image are extracted through the convolution layer of the preprocessing model, and the image features are recognized through the fully connected layer to obtain preprocessed data, and the preprocessed data is put into the geothermal data table;

[0018] Return a target image of a target data randomly selected from the geothermal data table until all the target data in the geothermal data table are selected, thereby obtaining a plurality of preprocessed data.

[0019] Preferably, creating an exploration model, inputting the preprocessed data as training samples into the exploration model to train the exploration model, and obtaining the trained exploration model, comprises:

[0020] Create exploration models;

[0021] All preprocessed data are divided into test set and training set according to random proportions;

[0022] Inputting the training set into the exploration model to train the exploration model, so that the exploration model continuously learns the coupling relationship between the target data and the geothermal resources, and obtains the trained exploration model;

[0023] The test set is input into the trained exploration model to determine whether the trained exploration model has the ability to automatically output geothermal resource evaluation based on the input target data.

[0024] Preferably, the training set is input into the exploration model to train the exploration model, so that the exploration model continuously learns the coupling relationship between the target data and the geothermal resources, and obtains the trained exploration model, including:

[0025] Randomly select a preprocessed data from the geothermal data table and the geothermal resource evaluation corresponding to the preprocessed data;

[0026] Establishing a coupling relationship between preprocessing data and geothermal resource evaluation, inputting the preprocessing data, geothermal resource evaluation, and the coupling relationship between preprocessing data and geothermal resource evaluation into the exploration model to train the exploration model, so that the exploration model continuously learns the corresponding relationship between the preprocessing data and the geothermal resource evaluation;

[0027] Return to randomly select a preprocessed data from the geothermal data table and the geothermal resource evaluation corresponding to the preprocessed data, until all preprocessed data in the geothermal data table are selected, and the trained exploration model is obtained.

[0028] Preferably, collecting target data of the measured area, inputting the target data of the measured area into the trained exploration model, obtaining geothermal resource evaluation of the measured area, performing geological logging according to the geothermal resource evaluation of the measured area, obtaining geological logging data, and screening geothermal areas based on the geological logging data, including:

[0029] Collect target data of the measured area;

[0030] Inputting target data of the measured area into the trained exploration model, and obtaining geothermal resource evaluation of the measured area output by the trained exploration model;

[0031] Determine whether there are geothermal resources in the measured area based on the geothermal resource evaluation of the measured area;

[0032] If geothermal resources exist in the measured area, drilling is carried out in the measured area and drilling data is collected, and the geothermal resources are determined based on the drilling data.

[0033] Preferably, if there are geothermal resources in the measured area, drilling is performed in the measured area and drilling data is collected, and the geothermal resources are determined based on the drilling data, including:

[0034] Set drilling data;

[0035] Drilling is performed based on the drilling data, and geological logging is performed during the drilling process to obtain geological logging data; the geological logging includes drilling logging, cuttings logging, and drilling fluid logging;

[0036] Collect historical drilling data in the surrounding area to determine whether the drilling distance between the new well and the historical well is qualified;

[0037] If the drilling distance between the new well and the historical well is qualified, the geothermal resources are judged and analyzed based on the geological logging data.

[0038] Preferably, if the drilling distance between the new well and the historical well is qualified, the geothermal resources are judged and analyzed based on the geological logging data, including:

[0039] Obtain geological logging data;

[0040] Obtaining gamma-ray anomaly areas based on gamma-ray cross-plot;

[0041] Acquiring the abnormal area of ​​acoustic waves based on the acoustic wave cross-graph;

[0042] The geothermal area is generated by superimposing the gamma-ray anomaly area and the sound wave anomaly area.

[0043] Preferably, the drilling data includes drilling depth, production depth, maximum well inclination angle, and drilling spacing.

[0044] On the other hand, the present application also provides an intelligent decision-making system for geothermal resource exploration based on big data, including a multi-source data acquisition component and a processing component. Multiple target data are collected through the multi-source data acquisition component. The processing component is communicated with the multi-source data acquisition component. The processing component is used to preprocess the target data, and the geothermal resources in the measured area are judged in combination with the preprocessing.

[0045] Compared with the prior art, the above technical solution of the present invention has the following beneficial technical effects:

[0046] By setting acquisition parameters, target data of the target area is collected, the target data is preprocessed to obtain preprocessed data, and then an exploration model is created. The preprocessed data is input as a training sample into the exploration model to train the exploration model to obtain a trained exploration model. Finally, target data of the measured area is collected, and the target data of the measured area is input into the trained exploration model to obtain a geothermal resource evaluation of the measured area. Geological logging is performed based on the geothermal resource evaluation of the measured area to obtain geological logging data, and geothermal areas are screened based on the geological logging data. The present application realizes multi-source data fusion by fusing target images and geological data, and uses multi-source data as training samples for the exploration model, thereby improving the efficiency of the trained exploration model for geothermal resource evaluation of the measured area and improving the accuracy of the exploration model. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 A schematic diagram of a process of an intelligent decision-making method for geothermal resource exploration based on big data proposed by the present invention;

[0048] Figure 2 This is a schematic diagram of the structure of a geothermal resource exploration intelligent decision-making system based on big data proposed by the present invention;

[0049] Figure 3 A drilling schematic diagram of an intelligent decision-making method for geothermal resource exploration based on big data proposed in the present invention.

[0050] Reference numerals: 100, multi-source data acquisition component; 200, processing component. DETAILED DESCRIPTION

[0051] Embodiment 1, as Figure 1 As shown, the present invention proposes an intelligent decision-making method for geothermal resource exploration based on big data, including:

[0052] S100, setting acquisition parameters, acquiring target data of a target area, and preprocessing the target data to obtain preprocessed data;

[0053] S200, creating an exploration model, inputting the preprocessed data as training samples into the exploration model to train the exploration model, and obtaining a trained exploration model;

[0054] S300, collecting target data of the measured area, inputting the target data of the measured area into the trained exploration model, obtaining geothermal resource evaluation of the measured area, performing geological logging according to the geothermal resource evaluation of the measured area, obtaining geological logging data, and screening geothermal areas based on the geological logging data.

[0055] In the present invention, by setting acquisition parameters, target data of the target area is acquired, the target data is preprocessed to obtain preprocessed data, and then an exploration model is created. The preprocessed data is input as a training sample into the exploration model to train the exploration model to obtain a trained exploration model. Finally, target data of the measured area is acquired, the target data of the measured area is input into the trained exploration model, a geothermal resource evaluation of the measured area is obtained, and geological logging is performed according to the geothermal resource evaluation of the measured area to obtain geological logging data, and geothermal areas are screened based on the geological logging data. The present application realizes multi-source data fusion by fusing target images and geological data, and uses multi-source data as training samples of the exploration model, thereby improving the efficiency of the trained exploration model for geothermal resource evaluation of the measured area and improving the accuracy of the exploration model.

[0056] In an optional embodiment, the S100 includes:

[0057] S110, creating a geothermal data table;

[0058] S120, setting acquisition parameters, and acquiring a plurality of target data of the target area based on the acquisition parameters; the target data includes geological data and target images;

[0059] Specifically, the fault layer is composed of the main fault plane and the broken rock blocks, secondary faults or fracture surfaces on both sides. Tectonic rocks will develop in the area close to the main fault plane, spreading to both sides along the axis of the main fault plane to form a series of rock types, including fault mud, mylonite, fault breccia, cataclase, etc.

[0060] S130, randomly selecting a target data from the geothermal data table, preprocessing the selected target data to obtain preprocessed data, until all the target data in the geothermal data table are selected.

[0061] It should be noted that after collecting multiple target data according to the acquisition parameters, the target image is preprocessed to extract the target area of ​​the target image, and the training set and the test set are created by combining the target data and the target area to facilitate the subsequent training of the exploration model.

[0062] In an optional embodiment, the S130 includes:

[0063] S131, creating a preprocessing model; the preprocessing model is based on a CNN convolutional neural network and includes a convolutional layer, a pooling layer, and a fully connected layer;

[0064] S132, randomly selecting a target image of target data from the geothermal data table;

[0065] S133, input the target image into the preprocessing model, extract the image features of the target image through the convolution layer of the preprocessing model, and recognize the image features through the fully connected layer to obtain preprocessed data, and put the preprocessed data into the geothermal data table;

[0066] S134, returning to randomly select a target image of target data from the geothermal data table until all target data in the geothermal data table are selected, thereby obtaining a plurality of pre-processed data.

[0067] It should be noted that CNN convolutional neural network is a deep learning algorithm suitable for processing image data. Its structure is mainly composed of convolutional layer, pooling layer and fully connected layer. After the target image is preprocessed by CNN convolutional neural network, the measured area in the target image can be detected. The preprocessing model created based on CNN convolutional neural network has better accuracy and generalization ability.

[0068] In an optional embodiment, the S200 includes:

[0069] S210, creating an exploration model;

[0070] S220, dividing all preprocessed data into a test set and a training set according to a random ratio;

[0071] S230, inputting the training set into the exploration model to train the exploration model, so that the exploration model continuously learns the coupling relationship between the target data and the geothermal resources, and obtains a trained exploration model;

[0072] S240, inputting the test set into the trained exploration model to determine whether the trained exploration model has the ability to automatically output geothermal resource evaluation according to the input target data;

[0073] Specifically, geothermal resource evaluation is an evaluation of the richness of geothermal resources contained in the measured area or target area. Geothermal resource evaluation includes rich geothermal resources, average geothermal resources and poor geothermal resources. The richer the geothermal resources, the more potential it has for development.

[0074] It should be noted that by creating and training the exploration model, the exploration model continuously learns the coupling relationship between the target data and geothermal resources, so that the trained exploration model has the ability to automatically output geothermal resource evaluation according to the input target data, so that when evaluating the geothermal resources in the measured area in the future, it is only necessary to input the target data of the measured area into the trained exploration model, thereby improving the efficiency of geothermal resource evaluation in the measured area.

[0075] The exploration model in this application combines the target image of the target area with the target data to perform multi-source data fusion, so that the exploration model can more accurately evaluate the geothermal resources in the measured area, thereby improving the subsequent exploration efficiency of geothermal resources.

[0076] In an optional embodiment, the S230 includes:

[0077] S231, randomly selecting a preprocessed data and a geothermal resource evaluation corresponding to the preprocessed data from the geothermal data table;

[0078] S232, establishing a coupling relationship between preprocessing data and geothermal resource evaluation, inputting the preprocessing data, geothermal resource evaluation, and the coupling relationship between preprocessing data and geothermal resource evaluation into the exploration model to train the exploration model so that the exploration model continuously learns the corresponding relationship between the preprocessing data and the geothermal resource evaluation;

[0079] Specifically, the coupling relationship between preprocessed data and geothermal resource evaluation is the correspondence between preprocessed data and geothermal resource evaluation;

[0080] S233, returning to randomly select a preprocessed data from the geothermal data table and the geothermal resource evaluation corresponding to the preprocessed data, until all preprocessed data in the geothermal data table are selected, and the trained exploration model is obtained.

[0081] It should be noted that by inputting multiple preprocessed data, multiple geothermal resource evaluations, and the coupling relationship between preprocessed data and geothermal resource evaluations into the exploration model, the exploration model continuously learns the correspondence between the preprocessed data and the geothermal resource evaluations until the exploration model has the ability to automatically output geothermal resource evaluations based on the input target data.

[0082] In an optional embodiment, the S300 includes:

[0083] S310, collecting target data of the measured area;

[0084] S320, inputting the target data of the measured area into the trained exploration model, and obtaining the geothermal resource evaluation of the measured area output by the trained exploration model;

[0085] S330, judging whether there are geothermal resources in the measured area according to the geothermal resource evaluation of the measured area;

[0086] S340: If geothermal resources exist in the measured area, drilling is performed in the measured area and drilling data is collected, and the geothermal resources are determined based on the drilling data.

[0087] It should be noted that after the trained exploration model outputs the geothermal resource evaluation of the measured area, it is necessary to judge whether it is necessary to drill in the area based on the geothermal resource evaluation results of the measured area. Drilling is only required in areas where the geothermal resources are evaluated as rich or average.

[0088] In an optional embodiment, the S340 includes:

[0089] S341, setting drilling data;

[0090] S342, drilling based on the drilling data, and performing geological logging during the drilling process to obtain geological logging data; the geological logging includes drilling logging, cuttings logging, and drilling fluid logging, and the geological logging data includes gamma ray data, acoustic wave data, and temperature data;

[0091] S343, collecting historical drilling data of the surrounding area to determine whether the drilling distance between the new well and the historical well is qualified;

[0092] S344, if the drilling distance between the new well and the historical well is qualified, the geothermal resources are judged and analyzed based on the geological logging data.

[0093] It should be noted that when drilling in the measured area, geological logging needs to be performed and geological logging data needs to be collected at the same time, so as to determine the geothermal resources in the measured area based on the geological logging data.

[0094] The reason why it is necessary to first determine the drilling distance between the new well and the historical well is to avoid the risk of wellbore collision between the new well and the historical well, thereby ensuring construction safety.

[0095] In an optional embodiment, the S344 includes:

[0096] S345, obtaining geological logging data;

[0097] S346, obtaining a gamma-ray abnormal area based on the gamma-ray cross map;

[0098] S347, obtaining an abnormal acoustic wave region based on an acoustic wave cross-graph;

[0099] S348, the geothermal area is generated by superimposing the gamma-ray anomaly area and the sound wave anomaly area.

[0100] It should be noted that gamma rays, also known as gamma particle streams, are rays released when the energy level of an atomic nucleus transitions and de-excites. They are electromagnetic waves with a wavelength shorter than 0.1 angstroms (1 angstrom = 10-10m), an energy higher than 124keV, and a frequency exceeding 30EHz (3×1019Hz). Gamma rays have strong penetrating power and can be used in industry for flaw detection or automatic control of assembly lines.

[0101] The present application selects geothermal areas from the measured area by combining gamma ray abnormal areas and acoustic wave abnormal areas, thereby detecting the actual quality of geothermal resources in the measured area.

[0102] In an optional embodiment, the drilling data includes drilling depth, production depth, maximum well inclination angle, and drilling spacing.

[0103] It should be noted that if Figure 3 The figure shows a three-dimensional schematic diagram of a geothermal well with a drilling depth set to 3000 meters (vertical depth) / 3090.45 meters (inclined depth) and a maximum well inclination angle of 24.29°. Since multiple wells need to be drilled in one area, the drilling spacing needs to be set to avoid impact between adjacent wells, thereby ensuring the safety of drilling.

[0104] like Figure 2 As shown, the present application also provides an intelligent decision-making system for geothermal resource exploration based on big data, including a multi-source data acquisition component 100 and a processing component 200. Multiple target data are collected by the multi-source data acquisition component 100, and the processing component 200 is communicated with the multi-source data acquisition component 100. The processing component 200 is used to preprocess the target data, and the geothermal resources in the measured area are judged in combination with the preprocessing.

[0105] It should be noted that the target image and the target data are collected respectively by the multi-source data collection component 100, and all the collected data are input into the processing component 200 for preprocessing to obtain preprocessed data, and then the exploration model is trained based on the preprocessed data to obtain a trained exploration model.

[0106] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited thereto, and various changes can be made within the knowledge scope of technicians in the relevant technical field without departing from the purpose of the present invention.

Claims

1. An intelligent decision-making method for geothermal resource exploration based on big data, characterized in that: include: Set acquisition parameters, acquire target data of the target area, pre-process the target data, and obtain pre-processed data; Creating an exploration model, inputting the preprocessed data as training samples into the exploration model to train the exploration model, and obtaining a trained exploration model; Collect target data of the measured area, input the target data of the measured area into the trained exploration model, obtain geothermal resource evaluation of the measured area, perform geological logging based on the geothermal resource evaluation of the measured area, obtain geological logging data, and screen geothermal areas based on the geological logging data.

2. According to claim 1, a geothermal resource exploration intelligent decision-making method based on big data is characterized in that: Set acquisition parameters, acquire target data of the target area, pre-process the target data, and obtain pre-processed data, including: Create geothermal data table; Setting acquisition parameters, and acquiring a plurality of target data of the target area based on the acquisition parameters; the target data includes fracture layer data and target images; A target data is randomly selected from the geothermal data table, and the selected target data is preprocessed to obtain preprocessed data, until all the target data in the geothermal data table are selected.

3. The intelligent decision-making method for geothermal resource exploration based on big data according to claim 2 is characterized in that: A target data is randomly selected from the geothermal data table, and the selected target data is preprocessed to obtain preprocessed data, until all the target data in the geothermal data table are selected, including: Create a preprocessing model; the preprocessing model is based on a CNN convolutional neural network and includes a convolutional layer, a pooling layer, and a fully connected layer; A target image of target data is randomly selected from the geothermal data table; The target image is input into the preprocessing model, the image features of the target image are extracted through the convolution layer of the preprocessing model, and the image features are recognized through the fully connected layer to obtain preprocessed data, and the preprocessed data is put into the geothermal data table; Return a target image of a target data randomly selected from the geothermal data table until all the target data in the geothermal data table are selected, thereby obtaining a plurality of preprocessed data.

4. The intelligent decision-making method for geothermal resource exploration based on big data according to claim 3 is characterized in that: Create an exploration model, input the preprocessed data as training samples into the exploration model to train the exploration model, and obtain the trained exploration model, including: Create exploration models; All preprocessed data are divided into test set and training set according to random proportions; Inputting the training set into the exploration model to train the exploration model, so that the exploration model continuously learns the coupling relationship between the target data and the geothermal resources, and obtains the trained exploration model; The test set is input into the trained exploration model to determine whether the trained exploration model has the ability to automatically output geothermal resource evaluation based on the input target data.

5. The intelligent decision-making method for geothermal resource exploration based on big data according to claim 4 is characterized in that: The training set is input into the exploration model to train the exploration model, so that the exploration model continuously learns the coupling relationship between the target data and the geothermal resources, and obtains the trained exploration model, including: Randomly select a preprocessed data from the geothermal data table and the geothermal resource evaluation corresponding to the preprocessed data; Establishing a coupling relationship between preprocessing data and geothermal resource evaluation, inputting the preprocessing data, geothermal resource evaluation, and the coupling relationship between preprocessing data and geothermal resource evaluation into the exploration model to train the exploration model, so that the exploration model continuously learns the corresponding relationship between the preprocessing data and the geothermal resource evaluation; Return to randomly select a preprocessed data from the geothermal data table and the geothermal resource evaluation corresponding to the preprocessed data, until all preprocessed data in the geothermal data table are selected, and the trained exploration model is obtained.

6. The intelligent decision-making method for geothermal resource exploration based on big data according to claim 5 is characterized in that: Collect target data of the measured area, input the target data of the measured area into the trained exploration model, obtain geothermal resource evaluation of the measured area, perform geological logging according to the geothermal resource evaluation of the measured area, obtain geological logging data, and screen geothermal areas based on the geological logging data, including: Collect target data of the measured area; Inputting target data of the measured area into the trained exploration model, and obtaining geothermal resource evaluation of the measured area output by the trained exploration model; Determine whether there are geothermal resources in the measured area based on the geothermal resource evaluation of the measured area; If geothermal resources exist in the measured area, drilling is carried out in the measured area and drilling data is collected, and the geothermal resources are determined based on the drilling data.

7. The intelligent decision-making method for geothermal resource exploration based on big data according to claim 6 is characterized in that: If there are geothermal resources in the measured area, drilling is carried out in the measured area and drilling data is collected. The geothermal resources are determined based on the drilling data, including: Set drilling data; Drilling is performed based on the drilling data, and geological logging is performed during the drilling process to obtain geological logging data; the geological logging includes drilling logging, cuttings logging, and drilling fluid logging; Collect historical drilling data in the surrounding area to determine whether the drilling distance between the new well and the historical well is qualified; If the drilling distance between the new well and the historical well is qualified, the geothermal resources are judged and analyzed based on the geological logging data.

8. The intelligent decision-making method for geothermal resource exploration based on big data according to claim 7 is characterized in that: If the drilling distance between the new well and the historical well is qualified, the geothermal resources are judged and analyzed based on the geological logging data, including: Obtain geological logging data; Obtaining gamma-ray anomaly areas based on gamma-ray cross-plot; Acquiring the abnormal area of ​​acoustic waves based on the acoustic wave cross-graph; The geothermal area is generated by superimposing the gamma-ray anomaly area and the sound wave anomaly area.

9. The intelligent decision-making method for geothermal resource exploration based on big data according to claim 8 is characterized in that: The drilling data includes drilling depth, production depth, maximum well inclination angle, and drilling spacing.

10. An intelligent decision-making system for geothermal resource exploration based on big data, applied to the intelligent decision-making method for geothermal resource exploration based on big data described in any one of claims 1 to 9, characterized in that: include: Multi-source data acquisition component; Collect multiple target data through the multi-source data collection component; A processing component is communicatively connected with the multi-source data acquisition component, and the target data is preprocessed by the processing component, and the geothermal resources of the measured area are determined in combination with the preprocessing.

Images