Probe for oil gas gene exploration and detection

Abstract

The invention discloses a probe for oil gas gene exploration and detection, and belongs to the technical field of intelligent oil gas exploration. The method solves the problem that an existing probe still has limitation due to the fact that the existing probe mainly depends on manual result interpretation and lacks data intelligent processing capacity, by introducing edge calculation and large model technologies, after gene sequence data in a soil sample is detected by the probe, quantitative analysis is carried out by utilizing a fluorescent quantitative PCR technology, and the accuracy of the gene sequence data in the soil sample is improved. The target gene is amplified by using the primer sequence to obtain a gene copy number, and the oil and gas reservoir probability of the soil sample is obtained; and the real-time gene sequence data and the oil and gas reservoir probability data are processed through a lightweight convolutional neural network to generate an oil and gas reservoir probability distribution thermodynamic diagram, so that cloud dependence is reduced, a probe can process the acquired data in real time on site, and a drilling target in a target area is quickly obtained according to the oil and gas reservoir probability distribution thermodynamic diagram. Therefore, the invalid drilling period is effectively shortened, the exploration cost is reduced, and the exploration efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent oil and gas exploration, and in particular to a probe for oil and gas gene exploration and detection. Background Art

[0002] Oil and gas exploration is a high-risk business venture. Traditional oil and gas exploration techniques are primarily based on petroleum geology research, incorporating geophysics, geochemistry, and drilling techniques. However, these techniques suffer from technical limitations, such as high exploration costs and an inability to determine the oil and gas content of a target. In recent years, with the advancement of microbial detection technology, microbial oil and gas exploration has gained popularity. This method primarily exploits the correlation between microorganisms in near-surface soil layers and deep underground oil and gas reservoirs, predicting the presence of oil and gas resources by detecting the genes of these microorganisms.

[0003] However, although existing probes can detect biomarkers, they mainly rely on manual interpretation of results in data processing and result analysis, and lack the ability to intelligently process data, resulting in limitations in existing probes.

[0004] Therefore, the existing needs are not met, and we propose a probe for oil and gas gene exploration and detection. Summary of the Invention

[0005] The purpose of the present invention is to provide a probe for oil and gas gene exploration and detection. By introducing edge computing and large model technology, after detecting gene sequence data in soil samples, the probe uses fluorescent quantitative PCR technology for quantitative analysis, and uses primer sequences to amplify the target gene to obtain the gene copy number and derive the oil and gas reservoir probability of the soil sample; then, a lightweight convolutional neural network is used to process the real-time gene sequence data and oil and gas reservoir probability data to generate an oil and gas reservoir probability distribution heat map, thereby reducing cloud dependence and enabling the probe to process the collected data in real time on site, and quickly derive the drilling target in the target area based on the oil and gas reservoir probability distribution heat map, thereby effectively reducing invalid drilling cycles, reducing exploration costs, and improving exploration efficiency, thereby solving the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A probe for oil and gas gene exploration and detection, comprising: a probe body and an edge computing unit, wherein the probe body comprises a forward primer F: 5'-GACTACGTGCTACGT-3' and a reverse primer R: 5'-CATGCTAGCTACGTA-3', and the edge computing unit comprises: a data fusion module and an intelligent analysis module;

[0008] The probe body is configured to collect soil samples from the target area, extract microbial DNA using a magnetic bead method, and obtain gene sequence data; quantitatively analyze the microbial genes in the gene sequence data using fluorescent quantitative PCR technology, and amplify the target gene using primer sequences to obtain the Ct value of the oil and gas gene; then, convert the Ct value into the gene copy number based on the standard curve to obtain the probability of oil and gas accumulation in the soil sample;

[0009] The data fusion module is configured to extract biological features and oil and gas reservoir probabilities from gene sequence data based on a lightweight convolutional neural network model, obtain physical features and geological features in the target area, and introduce an attention mechanism to perform weighted fusion of the features to obtain fused features;

[0010] The intelligent analysis module is configured to predict the fusion features based on a lightweight convolutional neural network model and output a probability distribution heat map of the oil and gas reservoir; based on the probability distribution heat map of the oil and gas reservoir, drilling coordinates are recommended and areas with higher probability distribution are selected as drilling targets.

[0011] Furthermore, the data fusion module includes:

[0012] A data acquisition module is configured to acquire downhole pressure and temperature data based on downhole sensors to obtain physical data; and to acquire historical seismic wave data in the target area based on a seismic exploration record table;

[0013] a model building module configured to build a lightweight convolutional neural network model, taking the fluorescent signal of the gene marker in the historical detection data, the downhole pressure data and the temperature data, and the historical seismic wave data of the target area as an input set, and inputting the input into the lightweight convolutional neural network model;

[0014] The feature extraction module is configured to perform sliding window segmentation on the fluorescence signal based on a lightweight convolutional neural network model to extract biological features; construct a time series matrix to extract physical features; and convert the seismic wave signal into a spectrum graph, which is processed using a hexagonal convolution kernel to extract geological features.

[0015] Furthermore, the data fusion module further includes:

[0016] The weight calculation module is configured to build a channel attention mechanism and a spatial attention mechanism to calculate the importance weight of each feature channel and each feature spatial position; use global average pooling and global maximum pooling to extract global information, generate channel weights through fully connected layers and activation functions; then use convolutional layers to extract spatial features and generate spatial weights through activation functions;

[0017] The feature fusion module is configured to fuse the weighted biological features, physical features and geological features to obtain fused features for training a lightweight convolutional neural network model.

[0018] Furthermore, the intelligent analysis module includes:

[0019] A training simulation module is configured to train and test the lightweight convolutional neural network model based on the fusion features and historical detection data, so that the lightweight convolutional neural network model learns the patterns and features in the historical detection data and outputs a heat map of the probability distribution of oil and gas reservoirs in the target area and recommended drilling coordinates;

[0020] The model validation module is configured to compare the training results of the lightweight convolutional neural network model with the actual results in the historical exploration record log to determine the learning performance of the model.

[0021] Furthermore, the intelligent analysis module further includes:

[0022] The model implementation module is configured to integrate a trained lightweight convolutional neural network model into the chip and load it into the handle of the probe body. The probe body collects biological data, combines it with downhole physical data and geological data as detection data, and uses the lightweight convolutional neural network model in the chip for edge computing processing to derive the drilling target within the target area.

[0023] The model update module is configured to regularly update and iterate the lightweight convolutional neural network model based on real-time detection data to optimize the generalization ability of the lightweight convolutional neural network model.

[0024] Furthermore, when collecting soil samples from the target area based on the probe body, the method includes:

[0025] Sampling points were set up in the target exploration area, and soil samples were collected from 30-50 cm below the surface using a sterile drill bit. The samples were sealed and stored at 4°C as a sample set.

[0026] When sampling, GPS coordinates, soil moisture and pH value are recorded simultaneously;

[0027] Microbial DNA was extracted from the sample set using the magnetic bead method to obtain DNA data;

[0028] The purity and concentration of DNA were determined using an ultraviolet spectrophotometer;

[0029] The target genes were determined to be functional genes of hydrocarbon oxidizing bacteria, and DNA data were tested using fluorescent quantitative PCR technology;

[0030] Specificity verification: Blast comparison was performed to confirm that the primers matched only the target gene, melting curve analysis showed a single peak, and fluorescence quantitative PCR reaction system data was obtained;

[0031] The data of the fluorescent quantitative PCR reaction system were amplified and converted into gene copy number according to the standard curve;

[0032] The distribution probability of oil and gas zones is determined based on the copy number, and a geographic heat map is drawn.

[0033] Furthermore, the data acquisition module further includes:

[0034] The historical collection module is configured to collect historical biological data, historical physical data, and historical geological data in the target area based on historical exploration record logs; the historical biological data includes: gene marker fluorescence signals; the historical physical data includes: downhole pressure data and temperature data; the historical geological data includes: historical seismic wave data in the target area;

[0035] The data processing module is configured to clean, denoise and standardize the historical detection data, and divide the processed data into a training set and a test set.

[0036] Furthermore, the data acquisition module further includes:

[0037] A data storage module is configured to classify and save the collected biological data and its corresponding physical data and geological data according to time series and the region to which they belong, forming a database;

[0038] The data encryption module is configured to encrypt the database using a symmetric key and authenticate the visitor.

[0039] Furthermore, the probe further includes:

[0040] The human-computer interaction terminal is configured to visually display real-time detection data in the target area, heat maps of oil and gas reservoir probability distribution, drilling coordinates and drilling target areas. It is also used to input downhole pressure and temperature data collected by downhole sensors and historical seismic wave data in the target area.

[0041] Furthermore, the probe further includes:

[0042] The device integration unit is configured to electrically connect the probe body with a chip equipped with an edge computing unit and a human-computer interaction terminal based on GPRS short-range communication technology, so as to enable data interaction between the probe body, the edge computing unit and the human-computer interaction terminal.

[0043] Compared with the prior art, the present invention has the following beneficial effects:

[0044] In the present invention, by introducing edge computing and large model technology, after the probe detects the gene sequence data in the soil sample, it uses fluorescent quantitative PCR technology for quantitative analysis, and uses primer sequences to amplify the target gene to obtain the gene copy number and derive the oil and gas reservoir probability of the soil sample; then, the real-time gene sequence data and oil and gas reservoir probability data are processed by a lightweight convolutional neural network to generate an oil and gas reservoir probability distribution heat map, thereby reducing cloud dependence and enabling the probe to process the collected data in real time on site, and quickly derive the drilling target in the target area according to the oil and gas reservoir probability distribution heat map, thereby effectively reducing invalid drilling cycles, reducing exploration costs, and improving exploration efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 Schematic diagram of the probe for oil and gas gene exploration and detection according to the present invention;

[0046] Figure 2 This is a diagram showing the composition of the probe module for oil and gas genetic exploration and detection according to the present invention. DETAILED DESCRIPTION

[0047] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only 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 efforts are within the scope of protection of the present invention.

[0048] To address the technical issues that exist in existing probes, although they can detect biomarkers, they still rely on manual interpretation of results in data processing and analysis, and lack intelligent data processing capabilities, which leads to the limitations of existing probes. Figure 1 - Figure 2 , this embodiment provides the following technical solutions:

[0049] A probe for oil and gas gene exploration and detection, the probe comprising: a probe body, a device integration unit, an edge computing unit, and a human-computer interaction terminal;

[0050] The probe body includes: forward primer F: 5'-GACTACGTGCTACGT-3' (Tm = 58°C), reverse primer R: 5'-CATGCTAGCTACGTA-3' (Tm = 59°C);

[0051] The probe body is configured to collect soil samples from the target area, extract microbial DNA using the magnetic bead method, and obtain gene sequence data; use fluorescent quantitative PCR technology to quantitatively analyze the microbial genes in the gene sequence data, and use primer sequences to amplify the target genes to obtain the Ct value of the oil and gas gene; then, based on the standard curve, the Ct value is converted into the gene copy number to obtain the probability of oil and gas accumulation in the soil sample.

[0052] When collecting soil samples from the target area based on the probe body, it includes:

[0053] Sampling points were arranged in a 100m x 100m grid in the target exploration area. Soil samples were collected from 30-50cm below the surface using a sterile drill bit. Multiple soil samples (500g per point) were collected and sealed and stored at 4°C as a sample set.

[0054] GPS coordinates, soil moisture (needs to be controlled at 15-25%), and pH value (range 6.5-7.5) were recorded simultaneously during sampling;

[0055] Microbial DNA was extracted from the sample set using a magnetic bead method. Five grams of soil and lysis buffer (containing SDS and proteinase K) were shaken and lysed at 65°C for 1 hour. Magnetic beads (0.1 mm diameter) were added to adsorb the DNA. After magnetic separation, the supernatant was removed. The sample was then washed twice with 70% ethanol, and finally the DNA was released in the eluent (TE buffer) to obtain DNA data.

[0056] DNA purity and concentration were measured using a UV spectrophotometer to ensure high DNA purity, with a purity value between 1.8 and 2.0;

[0057] The target gene was determined to be a functional gene of hydrocarbon oxidizing bacteria, such as the alkB gene. DNA data was tested using fluorescent quantitative PCR technology. The primer sequences were: forward primer F: 5'-GACTACGTGCTACGT-3' (Tm = 58°C), reverse primer R: 5'-CATGCTAGCTACGTA-3' (Tm = 59°C).

[0058] Specificity verification was performed: Blast comparison was performed to confirm that the primers matched only the target gene (E value < 1e-10), melting curve analysis showed a single peak (Tm = 85 ± 1°C), and fluorescence quantitative PCR reaction system data was obtained;

[0059] The fluorescence quantitative PCR reaction system data was amplified, including: pre-denaturation: 95°C for 3 min; cycling phase (40 cycles): 95°C / 10 s → 60°C / 30 s (fluorescence signal acquisition); melting curve analysis: 65°C → 95°C, heating every 0.5°C for 5 s;

[0060] According to the standard curve, it is converted into gene copy number, such as: slope: -3.32 (efficiency ≈ 100%), R 2 Value: >0.99; for example, when Ct=25, 10^5 copies / g soil are calculated; when Ct=30, 10^4 copies / g soil are calculated;

[0061] Oil and gas reservoir probability assessment: Areas with a copy number ≥ 10^4 / g are identified as high-probability oil and gas zones. A geographic heat map is drawn using Python's matplotlib library, with a color gradient mapping the copy number (red > 10^5, orange 10^4-10^5, and yellow < 10^4), as shown in the following table:

[0062] serial number Ct value Copy number / g soil Oil and gas reservoir probability P-01 24.8 1.1×10^5 High (88%) P-02 28.3 3.2×10^4 Medium (62%) P-03 32.7 8.5×10^3 Low (35%)

[0063] Table 1. Oil and gas reservoir probability assessment table

[0064] Through the above operations, the genetic testing results of soil samples can be used as samples for training lightweight convolutional neural network models, thereby effectively improving the possibility of evaluating the distribution probability of oil and gas reservoirs.

[0065] The equipment integration unit is configured to electrically connect the probe body with a chip equipped with an edge computing unit and a human-computer interaction terminal based on GPRS short-range communication technology, so as to enable data interaction between the probe body, the edge computing unit and the human-computer interaction terminal. Specifically, after the probe body collects biological data and performs genetic testing to obtain the biological data, the operator customizes the input of downhole pressure data and temperature data, as well as seismic wave historical data in the target area through the human-computer interaction terminal, so as to determine the probability distribution heat map of oil and gas reservoirs in the target area and the recommended drilling coordinates through the edge computing unit.

[0066] A chip is embedded in the handle of the probe body, and an edge computing unit is installed based on the chip. The edge computing unit includes: a data fusion module and an intelligent analysis module;

[0067] The data fusion module is configured to extract biological features and oil and gas reservoir probabilities from gene sequence data based on a lightweight convolutional neural network model, obtain physical and geological features in the target area, and introduce an attention mechanism to perform weighted fusion of these features to obtain fused features. The features include: biological features (gene marker fluorescence signals), physical features (downhole pressure / temperature data), and geological features (regional seismic wave historical data). The attention mechanism can automatically learn the importance of different features and assign appropriate weights to each feature. The data fusion module includes:

[0068] The data acquisition module is configured to collect downhole pressure and temperature data based on downhole sensors, obtaining physical data that can directly reflect the physical state of oil and gas reservoirs and provide an important basis for oil and gas exploration. It also obtains historical seismic wave data within the target area based on seismic exploration records, including seismic wave propagation velocity and reflected wave intensity, as geological data, enabling it to reflect the geological structure and distribution of oil and gas reservoirs. Through the automated acquisition of the probe body and downhole sensors, combined with the chip's onboard edge computing unit, rapid data collection and preliminary processing are achieved, improving work efficiency. By combining biological, physical, and geological data, the possibility of the existence of oil and gas reservoirs is comprehensively analyzed from multiple dimensions, enabling more accurate prediction of their location compared to analysis methods based on a single data source.

[0069] The data acquisition module also includes:

[0070] The historical collection module is configured to collect historical biological data, historical physical data, and historical geological data in the target area based on historical exploration record logs; historical biological data includes: gene marker fluorescence signals; historical physical data includes: downhole pressure data and temperature data; historical geological data includes: historical seismic wave data in the target area; thereby providing more comprehensive information for subsequent analysis and model training to reduce duplication of work and errors.

[0071] The data processing module is configured to clean, denoise, and standardize historical detection data. Specifically, it removes outliers, duplicate values, and missing values ​​from the data, and removes noise in the data through filtering and other methods to improve data quality; normalizes data from different sources to the same scale to reduce analytical errors caused by data problems and facilitate subsequent analysis; and divides the processed data into training and test sets for model training and verification to improve model performance.

[0072] The data storage module is configured to classify and store the collected biological data and its corresponding physical and geological data by time series and region, facilitating rapid retrieval and analysis and improving work efficiency. This forms a database, providing a solid foundation for subsequent data analysis and research, and supporting long-term oil and gas exploration projects.

[0073] The data encryption module is configured to use symmetric keys to encrypt the database to prevent data leakage and unauthorized access, ensuring the security and confidentiality of exploration data; and authenticate visitors so that only authorized users can access the database, enhancing user trust in the system.

[0074] The model building module is configured to build a lightweight convolutional neural network model, which takes the fluorescence signals of gene markers in historical detection data, downhole pressure data and temperature data, and historical seismic wave data of the target area as input sets and inputs them into the lightweight convolutional neural network model; by deploying the lightweight convolutional neural network model in the edge computing unit, it can be processed near the source of data generation, reducing the delay in data transmission to the central server, so that it is suitable for oil and gas exploration scenarios, thereby quickly responding to and providing decision support at the exploration site.

[0075] The feature extraction module is configured to perform sliding window segmentation on the fluorescence signal based on the lightweight convolutional neural network model, with a window length of 1s and an overlap rate of 30%. This module can capture short-term dynamic changes in the fluorescence signal, enabling the lightweight convolutional neural network model to effectively extract biological features and improve the recognition ability of the fluorescence signals of genetic markers; construct a time series matrix with each 5 minutes as a sampling unit. The time series matrix can reflect the long-term trend and periodic changes of physical data, enabling the lightweight convolutional neural network model to extract physical features related to oil and gas reservoirs and improve the analysis ability of downhole pressure and temperature data; and convert the seismic wave signal into a spectrum diagram. The spectrum diagram can reflect the frequency distribution and energy distribution of the seismic wave. Using a hexagonal convolution kernel for processing can better capture the local features and directional information in the seismic wave signal, thereby extracting geological features and improving the analysis ability of geological structures.

[0076] The weight calculation module is configured to build a channel attention mechanism and a spatial attention mechanism to calculate the importance weight of each feature channel and each feature spatial position; use global average pooling and global maximum pooling to extract global information, and generate channel weights through a fully connected layer and an activation function; then use a convolutional layer to extract spatial features, and generate spatial weights through an activation function; by introducing the channel attention mechanism and the spatial attention mechanism, appropriate weights are assigned to each feature, which can better reflect the possibility of the existence of oil and gas reservoirs, enable the model to more comprehensively analyze the importance of features, and improve the effect of multi-source data fusion.

[0077] The feature fusion module is configured to fuse the weighted biological, physical, and geological features to obtain fused features for training the lightweight convolutional neural network model. By extracting and integrating biological, physical, and geological features, it can provide more comprehensive oil and gas reservoir feature information, ensuring that the lightweight convolutional neural network model can more accurately predict the location of oil and gas reservoirs, thereby improving the intelligence level and accuracy of oil and gas exploration.

[0078] The intelligent analysis module is configured to predict fused features based on a lightweight convolutional neural network model, learn patterns and regularities in historical data, and output a probability distribution heat map of oil and gas reservoirs. Based on the probability distribution heat map of oil and gas reservoirs, it recommends drilling coordinates and selects areas with higher probability distribution as drilling targets, thereby improving the success rate of oil and gas exploration. The intelligent analysis module includes:

[0079] The training simulation module is configured to train and test the lightweight convolutional neural network model based on the fusion features and historical detection data, so that the lightweight convolutional neural network model learns the rules and features in the historical detection data, and outputs the probability distribution heat map of the oil and gas reservoirs in the target area and the recommended drilling coordinates; specifically, by inputting the training set and test set processed by the data processing module into the model, the lightweight convolutional neural network model is trained using the training set to optimize the model parameters; the trained model is tested using the test set to evaluate the model performance; and the model outputs the probability distribution heat map of the oil and gas reservoirs in the target area and recommends drilling coordinates.

[0080] The model verification module is configured to compare the training results of the lightweight convolutional neural network model with the actual results in the historical exploration record log to determine the learning performance of the model; and adjust the model parameters and optimize the model structure based on the evaluation results.

[0081] The model update module is configured to regularly update and iterate the lightweight convolutional neural network model based on real-time detection data to optimize the generalization ability of the lightweight convolutional neural network model.

[0082] The model practice module is configured to convert the trained lightweight convolutional neural network model into a format suitable for edge computing devices, such as TensorFlow Lite, and integrate the converted model into a dedicated chip, such as NVIDIA Jetson Nano or Raspberry Pi; and load it into the handle of the probe body to ensure that the model can run efficiently on the edge device; biological data is collected through the probe body, including: gene marker fluorescence signals; downhole physical data and geological data are combined as detection data, including: downhole pressure and temperature data and seismic wave history data, and the lightweight convolutional neural network model in the chip is used to perform edge computing processing on the biological data, physical data and geological data, quickly output the probability distribution heat map of the oil and gas reservoir, and derive the drilling target in the target area; specifically, by integrating the lightweight convolutional neural network model in the probe body handle, real-time data processing and analysis are achieved, reducing data transmission time and computing delay; secondly, the lightweight model can run efficiently on resource-constrained edge devices to adapt to different exploration environments; thirdly, the lightweight model accurately predicts the location of oil and gas reservoirs and recommends drilling coordinates, reducing the number of invalid drillings and lowering exploration costs.

[0083] The human-computer interaction terminal is configured to visually display real-time detection data in the target area, oil and gas reservoir probability distribution heat map, drilling coordinates and drilling target areas, so as to intuitively display the potential location of the oil and gas reservoir, clearly mark the recommended drilling coordinates, and display areas with higher probability distribution in different colors as drilling target areas, providing a direct basis for exploration decision-making; it is also used to input downhole pressure and temperature data collected by downhole sensors and historical seismic wave data in the target area, so that the lightweight model can analyze its multimodal data and improve decision-making efficiency; at the same time, it enables users to easily view and analyze data, providing an intuitive and efficient operation interface for the oil and gas exploration system.

[0084] The beneficial effects achieved by the above content are as follows: by introducing edge computing and large model technology, and with the help of a chip loaded into the probe handle, after the probe detects the gene sequence data in the soil sample, it uses fluorescent quantitative PCR technology for quantitative analysis, and uses primer sequences to amplify the target gene to obtain the gene copy number and derive the oil and gas reservoir probability of the soil sample; then, a lightweight convolutional neural network is used to process the real-time gene sequence data and oil and gas reservoir probability data to generate an oil and gas reservoir probability distribution heat map, thereby reducing cloud dependence and enabling the probe to process the collected data in real time on site, and quickly derive the drilling target in the target area based on the oil and gas reservoir probability distribution heat map, thereby effectively reducing invalid drilling cycles, reducing exploration costs, and improving exploration efficiency.

[0085] Working principle: The edge computing unit is installed on the chip and loaded into the handle of the probe body. GPRS short-range communication technology is used to enable data sharing between the probe body, the edge computing unit and the human-computer interaction terminal. Biological data is obtained through the probe body, and physical and geological data are input into the human-computer interaction terminal. The lightweight convolutional neural network model in the edge computing unit is used to perform feature analysis on multimodal data to determine the probability distribution heat map of oil and gas reservoirs in the target area and the drilling coordinates. The operator optimizes oil and gas exploration efficiency by selecting areas with higher probability distribution as drilling targets.

[0086] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "including," "having," or any other variations thereof are intended to cover non-exclusive possessors, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or includes elements that are inherent to such process, method, article, or apparatus.

[0087] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions, and alterations may be made to the embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A probe for oil and gas gene exploration and detection, characterized in that: The probe includes: a probe body and an edge computing unit, the probe body includes a forward primer F: 5'-GACTACGTGCTACGT-3' and a reverse primer R: 5'-CATGCTAGCTACGTA-3', and the edge computing unit includes: a data fusion module and an intelligent analysis module; The probe body is configured to collect soil samples from the target area, extract microbial DNA using a magnetic bead method, and obtain gene sequence data; quantitatively analyze the microbial genes in the gene sequence data using fluorescent quantitative PCR technology, and amplify the target gene using primer sequences to obtain the Ct value of the oil and gas gene; then, convert the Ct value into the gene copy number based on the standard curve to obtain the probability of oil and gas accumulation in the soil sample; The data fusion module is configured to extract biological features and oil and gas reservoir probabilities from gene sequence data based on a lightweight convolutional neural network model, obtain physical features and geological features in the target area, and introduce an attention mechanism to perform weighted fusion of the features to obtain fused features; The intelligent analysis module is configured to learn the oil and gas reservoir probabilities of the fused features and soil samples based on a lightweight convolutional neural network model, and output a probability distribution heat map of the oil and gas reservoirs; based on the probability distribution heat map of the oil and gas reservoirs, it recommends drilling coordinates and selects areas with higher probability distributions as drilling targets.

2. The probe for oil and gas gene exploration and detection according to claim 1, characterized in that: When collecting soil samples from the target area based on the probe body, it includes: Sampling points were set up in the target exploration area, and soil samples were collected from 30-50 cm below the surface using a sterile drill bit. The samples were sealed and stored at 4°C as a sample set. When sampling, GPS coordinates, soil moisture and pH value are recorded simultaneously; Microbial DNA was extracted from the sample set using the magnetic bead method to obtain DNA data; The purity and concentration of DNA were determined using an ultraviolet spectrophotometer; The target genes were determined to be functional genes of hydrocarbon oxidizing bacteria, and DNA data were tested using fluorescent quantitative PCR technology; Specificity verification: Blast comparison was performed to confirm that the primers matched only the target gene, melting curve analysis showed a single peak, and fluorescence quantitative PCR reaction system data was obtained; The data of the fluorescent quantitative PCR reaction system were amplified and converted into gene copy number according to the standard curve; The distribution probability of oil and gas zones is determined based on the copy number, and a geographic heat map is drawn.

3. The probe for oil and gas gene exploration and detection according to claim 1, characterized in that: The data fusion module includes: A data acquisition module is configured to acquire downhole pressure and temperature data based on downhole sensors to obtain physical data; and to acquire historical seismic wave data in the target area based on a seismic exploration record table; a model building module configured to build a lightweight convolutional neural network model, taking the fluorescent signal of the gene marker in the historical detection data, the downhole pressure data and the temperature data, and the historical seismic wave data of the target area as an input set, and inputting the input into the lightweight convolutional neural network model; The feature extraction module is configured to perform sliding window segmentation on the fluorescence signal based on a lightweight convolutional neural network model to extract biological features; construct a time series matrix to extract physical features; and convert the seismic wave signal into a spectrum graph, which is processed using a hexagonal convolution kernel to extract geological features.

4. The probe for oil and gas gene exploration and detection according to claim 3, characterized in that: The data fusion module further includes: The weight calculation module is configured to build a channel attention mechanism and a spatial attention mechanism to calculate the importance weight of each feature channel and each feature spatial position; use global average pooling and global maximum pooling to extract global information, generate channel weights through fully connected layers and activation functions; then use convolutional layers to extract spatial features and generate spatial weights through activation functions; The feature fusion module is configured to fuse the weighted biological features, physical features and geological features to obtain fused features for training a lightweight convolutional neural network model.

5. The probe for oil and gas gene exploration and detection according to claim 1, characterized in that: The intelligent analysis module includes: A training simulation module is configured to train and test the lightweight convolutional neural network model based on the fusion features and historical detection data, so that the lightweight convolutional neural network model learns the patterns and features in the historical detection data and outputs a heat map of the probability distribution of oil and gas reservoirs in the target area and recommended drilling coordinates; The model validation module is configured to compare the training results of the lightweight convolutional neural network model with the actual results in the historical exploration record log to determine the learning performance of the model.

6. The probe for oil and gas gene exploration and detection according to claim 5, characterized in that: The intelligent analysis module further includes: The model implementation module is configured to integrate a trained lightweight convolutional neural network model into the chip and load it into the handle of the probe body. The probe body collects biological data, combines it with downhole physical data and geological data as detection data, and uses the lightweight convolutional neural network model in the chip for edge computing processing to derive the drilling target within the target area. The model update module is configured to regularly update and iterate the lightweight convolutional neural network model based on real-time detection data to optimize the generalization ability of the lightweight convolutional neural network model.

7. The probe for oil and gas gene exploration and detection according to claim 3, characterized in that: The data acquisition module further includes: The historical collection module is configured to collect historical biological data, historical physical data, and historical geological data in the target area based on historical exploration record logs; the historical biological data includes: gene marker fluorescence signals; the historical physical data includes: downhole pressure data and temperature data; the historical geological data includes: historical seismic wave data in the target area; The data processing module is configured to clean, denoise and standardize the historical detection data, and divide the processed data into a training set and a test set.

8. The probe for oil and gas gene exploration and detection according to claim 7, characterized in that: The data acquisition module also includes: A data storage module is configured to classify and save the collected biological data and its corresponding physical data and geological data according to time series and the region to which they belong, forming a database; The data encryption module is configured to encrypt the database using a symmetric key and authenticate the visitor.

9. The probe for oil and gas gene exploration and detection according to claim 1, characterized in that: The probe further comprises: The human-computer interaction terminal is configured to visually display real-time detection data in the target area, heat maps of oil and gas reservoir probability distribution, drilling coordinates and drilling target areas. It is also used to input downhole pressure and temperature data collected by downhole sensors and historical seismic wave data in the target area.

10. The probe for oil and gas gene exploration and detection according to claim 9, characterized in that: The probe further comprises: The device integration unit is configured to electrically connect the probe body with a chip equipped with an edge computing unit and a human-computer interaction terminal based on GPRS short-range communication technology, so as to enable data interaction between the probe body, the edge computing unit and the human-computer interaction terminal.

Images