Microbial oil-gas exploration method and device, terminal and storage medium

Abstract

The invention provides a microbial oil-gas exploration method and device, a terminal and a storage medium, and relates to the technical field of microbial oil-gas exploration. The method comprises the following steps: acquiring the number of light hydrocarbon oxidizing bacteria of a to-be-detected soil sample of each sampling point in a target area; fusing the number of the light hydrocarbon oxidizing bacteria of each to-be-detected soil sample with the environmental features of the corresponding sampling point and the geological features of the corresponding sampling point to obtain a fused feature matrix; and inputting the fusion feature matrix into an oil and gas reservoir probability distribution model, outputting the oil and gas reservoir distribution probability of each sampling point in the target region, and determining an oil and gas reservoir existence region of the target region based on the oil and gas reservoir distribution probability of each sampling point. According to the invention, the accuracy and reliability of oil and gas reservoir distribution judgment can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of microbial oil and gas exploration, and particularly relates to a microbial oil and gas exploration method, device, terminal and storage medium. BACKGROUND

[0002] The oil and gas microbial exploration technology is a branch of the near-surface geochemical exploration technology, and has the characteristics of directness, effectiveness and low cost. It can be used as an independent technology to determine the prospective area, evaluate the nature and reserves of the oil reservoir in the area without geological or seismic data, and be combined with three-dimensional seismic and chemical exploration to more accurately determine the well site, expand the reserves, find out the undiscovered oil and gas traps and lost oil and gas reservoirs, increase the success rate, and reduce the exploration cost and risk.

[0003] The technical principle of the oil and gas microbial exploration is that the light hydrocarbon gas (mainly methane, ethane, propyl butane, etc.) of the oil and gas reservoir buried in the deep underground continuously diffuses and migrates to the surface through micro-leakage under the action of various powers, and the specific microorganism in the soil takes the light hydrocarbon gas as the only energy source, and is very developed in the surface soil directly above the oil and gas reservoir and forms a microbial anomaly, and the existence of the underlying oil and gas reservoir can be predicted by detecting the microbial anomaly. The light hydrocarbon gas micro-leakage has three characteristics: one is verticality, the migration direction of the light hydrocarbon micro-leakage is generally vertical, so the range of the microbial anomaly approximately corresponds to the boundary of the underground oil and gas reservoir, and the intensity of the microbial anomaly corresponds to the original content of the oil and gas reservoir; the second is universality, that is, most of the oil and gas reservoirs have light hydrocarbon gas micro-leakage, so the microbial detection method is suitable for the exploration of most oil and gas reservoirs; the third is dynamic, that is, with the continuous development of the oil and gas reservoir, the intensity of the light hydrocarbon micro-leakage will be reduced accordingly, and this dynamic change can represent the distribution change of the remaining oil and gas in the oil and gas field development.

[0004] It is reported that more than 90% of the detected light hydrocarbon gas in the soil above the natural gas field is methane, and less than 30% of the light hydrocarbon gas in the soil above the oil reservoir with low gas content and high oil content is methane, and most of it is propyl butane. It is reported in the related research on microbial oil and gas exploration that the number of light hydrocarbon oxidizing bacteria is positively correlated with the concentration of methane, so the light hydrocarbon oxidizing bacteria are important indicator bacteria for microbial oil and gas exploration. Combined with the analysis of the oil and gas reservoir formation mode, the rapid detection of the light hydrocarbon oxidizing bacteria can provide support for accurately delineating the natural gas reservoir. At the same time, the comparison of the number of light hydrocarbon oxidizing bacteria and propyl butane oxidizing bacteria can provide a basis for judging the nature of the oil and gas reservoir, and can provide a method for evaluating the yield change of the oil and gas field.

[0005] Currently, the detection of light hydrocarbon-oxidizing bacteria mainly adopts traditional culture method or molecular biology method, which cannot complete on-site rapid detection. Among them, the traditional culture method mainly obtains the number of light hydrocarbon-oxidizing bacteria in the sample through culture medium culture and MPN (most probable number) counting. Since the data interval obtained by the counting method is limited in magnitude, the obtained oil and gas distribution prediction result is qualitative-semiquantitative, which can only be used to qualitatively answer whether there is oil and gas reservoir at a certain point, and cannot be used as quantitative data to interpolate and predict the oil and gas distribution characteristics. Some researchers introduce molecular biology technology into the detection of light hydrocarbon-oxidizing bacteria, mainly using real-time fluorescent quantitative PCR technology (qPCR), gene chip technology, digital PCR technology and the like to detect the number of light hydrocarbon-oxidizing bacteria.

[0006] However, in the aspect of dividing the favorable area of light hydrocarbon-oxidizing bacteria in soil, there is a common problem that the abnormal grade division is not clear:

[0007] Researchers often use drilling data to assist in quantification, but the number of drilling data is often limited, and it is almost impossible to obtain continuous and complete data, so the threshold of the microbial anomaly system determined by this method has great uncertainty, and it is difficult to accurately delineate the range of favorable area for oil and gas exploration.

[0008] Deep learning has been applied to geological data analysis and prediction, and its typical model architecture includes convolutional neural network (CNN), recurrent neural network (RNN), long short-term memory network (LSTM), and generative adversarial network (GAN). CNN extracts local features of images through convolutional layers, pooling layers and fully connected layers, and is suitable for image recognition and geological image processing tasks in space. RNN and LSTM are suitable for processing sequence data and can capture long-term dependencies in time series, and are often used for geological time series prediction. GAN consists of a generator and a discriminator, and can generate realistic geological data, and is suitable for geological data generation and enhancement. In addition, graph neural network (GNN) and Transformer are also used for geological prediction to model complex geological structures and relationships. However, there is no deep learning model designed for light hydrocarbon-oxidizing bacteria and their characteristics related to natural gas reported so far. SUMMARY

[0009] The present application provides a microbial oil and gas exploration method, device, terminal and storage medium, to solve the problem that the number of light hydrocarbon-oxidizing bacteria is difficult to be quantified in the prior art, resulting in inaccurate identification of oil and gas reservoir distribution characteristics according to the number of light hydrocarbon-oxidizing bacteria.

[0010] In a first aspect, the present application provides a microbial oil and gas exploration method, comprising:

[0011] obtaining the number of light hydrocarbon-oxidizing bacteria in the soil sample to be tested at each sampling point in the target area;

[0012] The number of light hydrocarbon oxidizing bacteria of each to-be-tested soil sample is fused with the environmental characteristics of the corresponding sampling point and the geological characteristics of the corresponding sampling point to obtain a fused feature matrix;

[0013] The fused feature matrix is input into an oil and gas reservoir probability distribution model, and the oil and gas reservoir distribution probability of each sampling point in the target region is output, and the oil and gas reservoir existing region of the target region is determined based on the oil and gas reservoir distribution probability of each sampling point.

[0014] In a second aspect, the present application provides a microbial oil and gas exploration device for implementing the microbial oil and gas exploration method as described in the first aspect, and the microbial oil and gas exploration device comprises:

[0015] The acquisition module is configured to acquire the number of light hydrocarbon oxidizing bacteria of each to-be-tested soil sample in the target region.

[0016] The fusion module is configured to fuse the number of light hydrocarbon oxidizing bacteria of each to-be-tested soil sample with the environmental characteristics of the corresponding sampling point and the geological characteristics of the corresponding sampling point to obtain a fused feature matrix.

[0017] The determination module is configured to input the fused feature matrix into an oil and gas reservoir probability distribution model, output the oil and gas reservoir distribution probability of each sampling point in the target region, and determine the oil and gas reservoir existing region of the target region based on the oil and gas reservoir distribution probability of each sampling point.

[0018] In a third aspect, the present application provides a terminal comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method as described in the first aspect or any possible implementation manner of the first aspect when executing the computer program.

[0019] In a fourth aspect, the present application provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program is executable by a processor to implement the steps of the method as described in the first aspect or any possible implementation manner of the first aspect.

[0020] The application provides a microbial oil and gas exploration method, device, terminal and storage medium. The number of light hydrocarbon oxidizing bacteria of a to-be-tested soil sample of each sampling point in a target area is obtained; the number of light hydrocarbon oxidizing bacteria of each to-be-tested soil sample is fused with the environmental characteristics of the corresponding sampling point and the geological characteristics of the corresponding sampling point to obtain a fusion feature matrix; the fusion feature matrix is input into an oil and gas reservoir probability distribution model to output the oil and gas reservoir distribution probability of each sampling point in the target area, and based on the oil and gas reservoir distribution probability of each sampling point, the oil and gas reservoir existing area of the target area is determined. The integration of the multi-dimensional information of the number of light hydrocarbon oxidizing bacteria of the to-be-tested soil sample, the environmental characteristics and the geological characteristics can more comprehensively and accurately reflect the geological and microbial activity conditions of the target area, compared with a single information source, the accuracy and reliability of the oil and gas reservoir distribution judgment can be improved; and by using the constructed oil and gas reservoir probability distribution model, the oil and gas reservoir distribution probability of each sampling point in the target area can be quickly output, which greatly shortens the time required for a large number of field tests and long-time analysis in the traditional exploration method, significantly improves the exploration efficiency, helps to quickly lock the potential oil and gas reservoir area, reduces the exploration cost and time input; at the same time, based on the oil and gas reservoir distribution probability of each sampling point, the oil and gas reservoir existing area of the target area can be more accurately determined, which avoids the blindness and uncertainty that may exist in the traditional method, provides a clear target area for subsequent oil and gas exploitation work, and improves the success rate and economic benefit of exploitation. BRIEF DESCRIPTION OF DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the application, the drawings needed to be used in the embodiments or prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor.

[0022] Figure 1 is the implementation flowchart of the microbial oil and gas exploration method provided by the embodiments of the application;

[0023] Figure 2 is the structural schematic diagram of the microbial oil and gas exploration device provided by the embodiments of the application;

[0024] Figure 3 is the schematic diagram of the terminal provided by the embodiments of the application. DETAILED DESCRIPTION

[0025] In the following description, for purposes of explanation and not limitation, specific details are set forth such as particular architectures, technologies, techniques, etc. in order to provide a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application can be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present application with unnecessary detail.

[0026] In order to make the objects, technical solutions and advantages of the present application clearer, the following will be described by specific embodiments in conjunction with the accompanying drawings.

[0027] Figure 1 The implementation flowchart of the microbial oil and gas exploration method provided by the embodiments of the present application is described in detail as follows.

[0028] In step 101, the number of light hydrocarbon oxidizing bacteria of the to-be-tested soil sample of each sampling point in the target area is obtained.

[0029] In the embodiments of the present application, first, sample point arrangement is performed, according to the national oil and gas exploration method, sampling points are arranged by using different grid densities mainly in the grid method according to different purposes and requirements such as reconnaissance, general survey, detailed survey and intensive survey. Then, sample collection is performed, the to-be-tested soil sample should be avoided from being polluted, and necessary sampling depth can be set, such as 50cm-70cm. At the sampling depth, the to-be-tested soil sample of each sampling point in the target area is obtained by using a sampling tool, and the sampling amount of the to-be-tested soil sample can be 20g-50g. Finally, according to the to-be-tested soil sample of each sampling point, the number of light hydrocarbon oxidizing bacteria of the corresponding to-be-tested soil sample is detected by using real-time fluorescent quantitative PCR, gene chip or culture method and the like.

[0030] For example, taking microbial oil and gas exploration of an oilfield in North China as an example (the examples in the following are the same), the investigation accuracy is general survey, and the process of sample point arrangement and sample collection can be as follows:

[0031] Step 1.1, according to the accuracy of general survey in the oil and gas exploration specification, the sampling points are arranged according to the grid with a line distance of 1km and a point distance of 1km, and the investigation scale is 1:100000.

[0032] Step 1.2, in order to prevent the influence of surface microorganisms, after disinfecting the sampling tool, the soil sample below 50cm is collected, put into a sterile sampling bag and placed into a thermos box for cold storage.

[0033] In the embodiments of the present application, the quantity of light hydrocarbon oxidizing bacteria can be obtained by real-time fluorescent quantitative PCR, gene chip or culture method to obtain the quantity of light hydrocarbon oxidizing bacteria in each soil sample to be tested. The process of fluorescent quantitative PCR includes DNA extraction, PCR amplification of sample DNA, preparation of plasmid standard, and fluorescent quantitative PCR determination. The process of culture method includes: first, the light hydrocarbon oxidizing bacteria in each soil sample to be tested is rapidly separated by using specific rapid culture medium of light hydrocarbon oxidizing bacteria, and the quantity of light hydrocarbon oxidizing bacteria is obtained by counting method or luciferase method.

[0034] Among them, the light hydrocarbon oxidizing bacteria include methane oxidizing bacteria, ethane oxidizing bacteria, propane oxidizing bacteria and the like.

[0035] For example, taking methane oxidizing bacteria as an example, the process of obtaining the quantity of light hydrocarbon oxidizing bacteria in each soil sample by using fluorescent quantitative PCR is as follows:

[0036] Step 2.1, sample DNA extraction, mainly using DNA extraction kit to extract DNA, and the specific execution refers to the instruction manual of DNA extraction kit.

[0037] Step 2.2, PCR amplification of sample DNA: select methane oxidizing bacteria functional gene pmoA primers A189f (5'-GGNGAC TGG GAC TTC TGG-3') and mb661R (5'-CCG GMG CAA CGT CYT TAC C-3'). The reaction system is 10x buffer 5L, MgCl2(25mM) 4L, dNTP(10mM) 2.5L, BSA(20g / L) 0.2L, Taq DNA polymerase(5u / L) 0.5L, 20pmol of primers and 2L of DNA template are added, and ddH2O is added to 50L. Select touchdown PCR for amplification: 94℃ pre-denaturation for 5min; 94℃ denaturation for 1min, 60℃ annealing for 1min, 72℃ extension for 3min, 20 cycles of amplification, and the annealing temperature decreases by 0.5℃ for each cycle; then the annealing temperature is kept at 50℃ for 10 cycles of amplification; finally, 72℃ extension for 10min.

[0038] Step 2.3, preparation of plasmid standard, select DNA that can successfully amplify the target fragment of pmoA functional gene. The obtained target band is purified by using a commercially available DNA purification and recovery kit (such as Omega) to purify the gel and determine the recovery concentration. The correct size of the fragment is obtained by 1.5% agarose gel electrophoresis, and the gel is cut and purified by PCR product gel recovery.

[0039] Next, the connection work is carried out, and the connection system is as follows: PCR recovery product 7 μl, R2.1 vector 1, T4 ligase 1 μl, 10x buffer 1 μl.

[0040] Subsequently, the transformation work is carried out, and the steps are as follows:

[0041] (1) The connection product is placed in a water bath or PCR instrument, and connected at 14℃ overnight.

[0042] (2) Take out the frozen DH5α competent cells 1 tube, and dissolve in ice.

[0043] (3) Take 5 μl of the connection product and add it to 100 μl of the competent cells, mix gently, and ice bath for 30 minutes.

[0044] (4) Heat shock at 42℃ for 60-90 seconds, and ice bath for 2 minutes.

[0045] (5) Add 900 μl of non-resistant LB medium to the shaker at 37℃, 200 rpm, and culture for 60-90 minutes.

[0046] (6) Take 200 μl of the pre-transformed bacterial solution, add 40 μl of 20 mg / ml Xgal and 4 μl of 50 mg / ml IPTG in the clean bench, mix well, and then spread on the plate containing 100 ug / ml ampicillin.

[0047] (7) Place the plate in a constant temperature incubator at 37℃ and invert culture overnight until the clones grow. Pick the clones for verification PCR, and if passed, use the plasmid extraction to extract the plasmid of the clones, and perform electrophoresis. Use the micro ultraviolet nucleic acid quantifier to measure the nucleic acid concentration, and calculate the copy number of the methanotrophic bacteria gene fragment.

[0048] Step 2.4, determination of fluorescent quantitative PCR: test with a fluorescent quantitative PCR instrument, total reaction system of 20 μL: SYBR Premix Ex TaqTM(2X) 10 μL, Forward Primer(10 μM) 0.2 μL, Reverse Primer(10 μM) 0.2 μL, ROX Reference Dye(50x) 0.4 μL, ddH2O 7.2 μL, DNA template 2 μL, adopt two-step amplification program: first stage 95℃ pre-denaturation 30s; second stage 95℃ 5s, 60℃ 30s, 84℃ 15s (collect fluorescence signal) amplification 40 cycles; third stage draw melting curve 95℃ 15s, 60℃ 60s, 60℃-95℃ every 0.3℃ reading, reading time 15s for melting curve analysis.

[0049] In step 102, the light hydrocarbon oxidizing bacteria number of each to-be-tested soil sample is fused with the environmental characteristics and geological characteristics of the corresponding sampling point to obtain a fused feature matrix.

[0050] In the embodiments of the present application, the environmental characteristics and geological characteristics of each sampling point in the target region are acquired, and then the light hydrocarbon oxidizing bacteria number of each to-be-tested soil sample of the sampling point acquired in step 101 is fused with the corresponding environmental characteristics and geological characteristics to obtain a fused feature matrix.

[0051] The embodiments of the present application can more comprehensively and accurately reflect the geological and microbial activity conditions of the target region through integration of multi-dimensional information, and compared with a single information source, can improve the accuracy and reliability of the judgment of the oil and gas reservoir distribution.

[0052] In a possible implementation, fusing the light hydrocarbon oxidizing bacteria number of each to-be-tested soil sample with the environmental characteristics of the corresponding sampling point and the geological characteristics of the corresponding sampling point to obtain a fused feature matrix can include:

[0053] Based on the light hydrocarbon oxidizing bacteria number of the to-be-tested soil sample of each sampling point in the target region, a Kriging interpolation algorithm is used to acquire a gene space distribution discrete point matrix of the target region;

[0054] An environmental feature vector is constructed using the environmental characteristics of each sampling point in the target region;

[0055] A geological feature vector is constructed using the geological characteristics of each sampling point in the target region;

[0056] The gene space distribution discrete point matrix, the environmental feature vector, and the geological feature vector are transversely spliced to construct a preliminary fused feature matrix;

[0057] The preliminary fused feature matrix is used to construct derived features, and a principal component analysis method is used to perform dimension reduction processing on the derived features to obtain a fused feature matrix.

[0058] Optionally, based on the light hydrocarbon oxidizing bacteria number of the to-be-tested soil sample of each sampling point in the target region, a gridding sampling (interval according to investigation requirements) is used to acquire gene space distribution discrete point data, which is measured by a high-precision fluorescence detector (detection range 0-10 5 RLU). Then, the Kriging interpolation algorithm is used to generate a regular grid matrix, i.e., a gene space distribution discrete point matrix (dimension N×H×W), from the gene space distribution discrete point data. For example, the difference error can be set to ≤5%.

[0059] Then the multi-parameter water quality detector is used to collect the environmental characteristics of each sampling point in the target area in real time, including but not limited to salinity, water content, etc. The geological characteristics of each sampling point in the target area are obtained by drilling core / seismic inversion, including but not limited to porosity, permeability (unit: mD), reservoir thickness (unit: m), oil and gas saturation (unit: %) and the like. The environmental characteristic vector (dimension N x 6) is constructed by using the environmental characteristics, and the geological characteristic matrix (dimension N x 8) is constructed by using the geological characteristics.

[0060] The gene space distribution discrete point matrix, the environmental characteristic matrix and the geological characteristic matrix are horizontally spliced according to the sampling points, and a preliminary fusion feature matrix is formed. Then, the derived features are constructed, and the dimension reduction processing is performed on the derived features by using the principal component analysis method, and the fusion feature matrix is obtained.

[0061] In a possible implementation, the environmental characteristic vector of each sampling point in the target area can include:

[0062] The environmental characteristics of each sampling point in the target area are subjected to z-score standardization, and the environmental characteristic vector is constructed by using the z-score standardized environmental characteristics.

[0063] Optionally, the z-score standardization is performed on the environmental characteristics, and the calculation formula of the z-score standardization is:

[0064]

[0065] Wherein, x ′ is the standardized environmental characteristic, x is the environmental characteristic, μ is the mean, and σ is the standard deviation.

[0066] Then, the abnormal values are removed through the box plot (IQR coefficient 1.5).

[0067] Correspondingly, the geological characteristics are also preprocessed by using the z-score standardization, and the preprocessing process is consistent with the preprocessing process of the environmental characteristics.

[0068] In a possible implementation, the gene space distribution discrete point matrix, the environmental characteristic vector and the geological characteristic vector are horizontally spliced to construct the preliminary fusion feature matrix, which can include:

[0069] The gene space distribution discrete point matrix, the environmental characteristic vector and the geological characteristic vector are spatially aligned under the WGS84 geographic coordinate system;

[0070] The gene space distribution discrete point matrix, the environmental characteristic vector and the geological characteristic vector after spatial alignment are horizontally spliced according to the sampling point sequence to construct the preliminary fusion feature matrix.

[0071] Optionally, the obtained gene space distribution discrete point matrix, environmental feature vector and geological feature vector are spatially aligned according to the WGS84 geographic coordinate system to construct a structured database. Then, the aligned gene space distribution discrete point matrix, environmental feature vector and geological feature vector are horizontally spliced according to the sampling point sequence to obtain a preliminary fusion feature matrix.

[0072] In one possible implementation, using the preliminary fusion feature matrix, derivative features are constructed, and the principal component analysis method is used to reduce the dimension of the derivative features to obtain a fusion feature matrix, which can include:

[0073] The product of the number of light hydrocarbon oxidizing bacteria of each sampling point and the permeability in the geological features of the sampling point is taken as the first derivative feature of the sampling point.

[0074] The ratio of the salinity in the environmental features of each sampling point to the water content in the environmental features of the sampling point is taken as the second derivative feature of the sampling point.

[0075] The principal component analysis method is used to reduce the dimension of the first derivative features and the second derivative features of all sampling points to obtain a fusion feature matrix.

[0076] Optionally, the derivative features include the first derivative features and the second derivative features. Specifically:

[0077] The "light hydrocarbon oxidizing bacteria number x permeability" (unit: RLU·mD) of each sampling point is taken as the first derivative feature of the corresponding sampling point, which is used to represent the methane diffusion flux. The "salinity / water content" (dimensionless) of each sampling point is taken as the second derivative feature of the corresponding sampling point, which is used to reflect the stability of the microbial living environment. Then, the principal component analysis is used to reduce the dimension of all first derivative features and second derivative features (cumulative variance contribution rate ≥ 85%), to obtain a fusion feature matrix.

[0078] For example, the determination process of the fusion feature matrix is as follows:

[0079] Gene space distribution discrete point matrix: The number of light hydrocarbon oxidizing bacteria detected at 12 sampling points is used to generate a network matrix with a resolution of 1 km x 1 km using the Kriging interpolation algorithm, with an interpolation error of ≤ 4.2%, forming standardized data that can reflect the gene space distribution, i.e., the gene space distribution discrete point matrix.

[0080] Environmental feature vector: The salinity (range: 0.1-3.2%) and water content (range: 8-25%) data of each sampling point are collected in real time using a multi-parameter soil detector, and are preprocessed using z-score standardization to eliminate dimensional effects, and the preprocessed environmental matrix is used to construct the environmental feature vector.

[0081] Geological feature vector: The geological feature data of each sampling point, such as porosity (range: 8-22%), permeability (range: 5-120 mD), reservoir thickness (range: 5-25 m), and oil and gas saturation (range: 20-60%), are obtained through drilling core analysis, well logging interpretation, and seismic inversion.

[0082] Then, all the gene space distribution discrete point matrices, environmental feature vectors, and geological feature vectors are spatially aligned according to the WGS84 geographic coordinate system to construct a structured database, ensuring the consistency and accuracy of the data characteristics fusion and derivation.

[0083] Feature fusion: The preprocessed gene space distribution discrete point matrix (dimension: 120x1x1), environmental feature vector (dimension: 120x2), and geological feature vector (dimension: 120x4) are horizontally spliced according to the sampling points to form a preliminary fusion feature matrix.

[0084] Derivative feature construction: Based on the principles of geological fluid dynamics, "light hydrocarbon oxidizing bacteria number x permeability" (unit: RLU.mD) is constructed to represent the methane diffusion flux, and "salinity / water content" (dimensionless) is constructed to reflect the stability of the microbial survival environment. Principal component analysis is performed on these derivative features to reduce the dimension, with the cumulative variance contribution rate reaching 88.3%, removing redundant information, and finally forming a fusion feature matrix with a dimension of 120x12.

[0085] In step 103, the fusion feature matrix is input into the oil and gas reservoir probability distribution model, and the oil and gas reservoir distribution probability of each sampling point in the target area is output. Based on the oil and gas reservoir distribution probability of each sampling point, the oil and gas reservoir existence area of the target area is determined.

[0086] In the embodiments of the present application, the fusion feature matrix is input into the oil and gas reservoir probability distribution model constructed based on the 3D-CNN model and the LSTM model, and the oil and gas reservoir distribution probability of each sampling point in the target area is output. Then, based on the oil and gas reservoir distribution probability of each sampling point in the target area, the oil and gas reservoir existence area of the target area is determined.

[0087] The construction process of the oil and gas reservoir probability distribution model is as follows:

[0088] Data input module: build a multi-source data parallel input interface, adapt the gene space distribution discrete point matrix to a tensor form (dimension N x H x W x D), and map the environmental feature vector and the geological feature vector to a 128-dimensional high-dimensional space through an embedding layer, and stack them in the channel dimension to form a unified input. For example, convert the gene space distribution discrete point matrix to a tensor form, with a dimension of 120 x 1 x 1 (N x H x W x D); map the environmental feature vector and the geological feature vector to a 128-dimensional high-dimensional space through an embedding layer, and then stack them in the channel dimension to form a unified input tensor with an input dimension of 120 x 10 x 10 x 129.

[0089] Multi-module feature extraction: a mixed architecture of 3D-CNN model and LSTM model is adopted: a 3D-CNN layer (convolution kernel is 3 x 3 x 3, step is 1) is used to capture the three-dimensional space features of the number of light hydrocarbon oxidizing bacteria, and extract the local structure and global distribution rule; an LSTM layer with 128 hidden units is used to process the environmental features and geological features, and learn the spatial dependency through a gating mechanism; an SE attention module is introduced to automatically allocate weights (weight range is 0.1-0.9) according to the feature contribution, and to strengthen the association between microorganisms and geological features. For example, the automatically allocated weight of the microbial feature is 0.65, and the weight of the geological feature is 0.35, highlighting the contribution of key features to the prediction of oil and gas reservoir distribution.

[0090] Feature fusion layer: the tensor output by the 3D-CNN (dimension N x 16 x 16 x 16) is spliced with the feature vector output by the LSTM through a fully connected layer, and then connected to a 2-layer multilayer perceptron (MLP) with 256 and 128 neurons respectively, and the neurons are transformed multiple times through a ReLU activation function to enhance the feature expression capability. Add Dropout (0.3) between network layers to randomly discard part of the neurons to prevent model overfitting.

[0091] Output layer design: since the implementation is a binary classification prediction of the existence of oil and gas reservoirs, the output layer uses a single neuron with a Sigmoid function as the activation function, and the output value ranges from 0 to 1, representing the probability of the existence of oil and gas reservoirs. For regression tasks, set the neurons according to the target dimension + Linear activation (directly output the reserves or range value).

[0092] In the embodiments of the present application, after the oil and gas reservoir probability distribution model is constructed, model training and optimization are still needed, and the specific process is as follows:

[0093] Data division: divide the fused multi-source data into a training set, a validation set and a test set according to a 7:1:2 ratio. For example, there are 120 samples, the training set includes 84 samples, the validation set includes 12 samples, and the test set includes 24 samples.

[0094] In the division, stratified sampling strategy is adopted to ensure the balanced distribution of data in different geological conditions and different reservoir states in each set, so that the model can learn more comprehensive data features.

[0095] Data augmentation: The gene space distribution discrete point matrix is enhanced, including ±15° rotation, ±5 grid unit translation, 0.8-1.2 times scaling, etc. Geometric transformation generates 168 virtual samples, expands the data scale, and improves the model's adaptability to different spatial distribution patterns of light hydrocarbon oxidation bacteria.

[0096] Add Gaussian noise with mean 0 and standard deviation 0.1 to the environmental feature vector and geological feature vector to simulate the noise interference in the actual data and enhance the robustness of the model.

[0097] Parameter training includes loss function, optimizer and training period:

[0098] 1) Loss function:

[0099] Since it is a binary classification task, the cross-entropy loss function is used, and the formula is:

[0100]

[0101] Where L1 is the cross-entropy loss function, N is the number of samples, is the true label of sample i (0 or 1), p i is the probability of the model predicting that sample i is a positive class.

[0102] The regression task uses the mean square error loss function, that is:

[0103]

[0104] Where L2 is the mean square error loss function, is the true value of sample i, is the predicted value of the model for sample i.

[0105] 2) Optimizer:

[0106] The Adam optimization algorithm is used, and the initial learning rate is set to 0.001. During training, the learning rate is dynamically adjusted according to the performance of the validation set. When the validation set loss no longer decreases in continuous 10 rounds of training, the learning rate is multiplied by 0.1 to decay, to avoid the model falling into local optimum and speed up the convergence speed.

[0107] 3) Training period:

[0108] The training period is set to 50 rounds, and the model is evaluated using the validation set after completing every 5 training periods. That is, the accuracy, precision, recall, F1 value of binary classification, or the MSE, MAE of regression is calculated every 5 training periods, and the hyperparameters such as the 3D-CNN convolution kernel size (such as 3x3x3), the LSTM hidden layer unit number (such as 128), and the Dropout ratio (0.3) are optimized through grid search. For example, after the training is completed, the final model is comprehensively evaluated using the test set, and the test set accuracy reaches 92.3%, the F1 value is 0.88, and the AUC-ROC is 0.94. Compared with the traditional random forest model, the prediction and actual oil and gas reservoir overlap rate (IoU) is improved by 27.5%, indicating that the model has better prediction performance.

[0109] The embodiments of the present application use multi-source data fusion combined with 3D-CNN and LSTM models for oil and gas reservoir prediction, which improves the prediction accuracy by more than 20% compared with traditional methods, and the coincidence rate of the delineated favorable area through drilling verification reaches 85%, fully proving the application of the embodiments of the present application in actual oil and gas exploration.

[0110] In a possible implementation, determining the oil and gas reservoir distribution area of the target area based on the oil and gas reservoir distribution probability of each sampling point can include:

[0111] determining whether the oil and gas reservoir distribution probability of the target sampling point is greater than a preset probability, the target sampling point being any sampling point in the target area;

[0112] if the oil and gas reservoir distribution probability of the target sampling point is greater than the preset probability, determining that the region corresponding to the target sampling point is an oil and gas reservoir existing region;

[0113] if the oil and gas reservoir distribution probability of the target sampling point is not greater than the preset probability, determining that the region corresponding to the target sampling point is not an oil and gas reservoir existing region.

[0114] Optionally, after obtaining the oil and gas reservoir distribution probability of the target sampling point, it is determined whether the oil and gas reservoir distribution probability is greater than a preset probability. If yes, it is determined that the target sampling point is an oil and gas reservoir existing region; if not, it is determined that the target sampling point is not an oil and gas reservoir existing region.

[0115] In addition, the embodiments of the present application can also use the output of the Sigmoid activation function in the oil and gas reservoir probability distribution model as the basis for judging the oil and gas reservoir existing region.

[0116] Exemplary, the multi-source data of the target sampling points is converted into tensor data required by the model, and is input into the trained oil and gas reservoir probability distribution model. In the binary classification prediction, the region with the output probability of the Sigmoid activation function greater than 0.5 is determined as the oil and gas reservoir existing region, and finally 3 favorable areas are delineated. Comparing the seismic exploration results of these favorable areas, it is found that the coincidence rate of the favorable area and the amplitude anomaly area reaches 81%, and according to the drilling experience, the oil saturation of these areas is greater than 38%, verifying the accuracy of the prediction result.

[0117] In a possible implementation, after determining the oil and gas reservoir existing region of the target area based on the oil and gas reservoir distribution probability of each sampling point, the method can further include:

[0118] Visualizing and displaying the oil and gas reservoir existing region of each sampling point in the target area, wherein the visualization includes two-dimensional visualization, three-dimensional visualization and dynamic visualization.

[0119] The two-dimensional visualization: the prediction result is displayed in the form of a two-dimensional map by using GIS software. The hierarchical color setting method is adopted, the region with a probability between 0-0.3 is set to light blue and marked as a low favorable area; the region with a probability greater than 0.7 is set to dark red and marked as a high favorable area; the region with a probability between 0.3-0.7 is set to yellow and marked as a medium favorable area. The contour line with a precision of 10m, the measured fault line and the sampling point coordinates are superimposed on the map to intuitively display the relationship between the oil and gas reservoir distribution and the geological structure, and to enhance the professionalism and readability of the visualization result.

[0120] The three-dimensional visualization: based on the three-dimensional geological modeling software, a geological body model is generated in combination with the seismic data. By integrating the predicted oil and gas reservoir distribution, the three-dimensional distribution of the oil and gas reservoir is displayed by adjusting the transparency (0-100%), supporting ±360° rotation, 1-10 times scaling and cross-section cutting interaction. For example, by adjusting the model transparency to 60%, the three-dimensional shape of the oil and gas reservoir in the three-dimensional space is clearly displayed, and at the same time, the user is supported to perform 360° rotation, 2 times scaling and cross-section cutting interaction, which facilitates the observation of the spatial relationship between the oil and gas reservoir and the surrounding strata and structure from different perspectives, and provides strong support for geological analysis and exploration decision-making.

[0121] The dynamic visualization: a prediction result change animation is made with time series (for example, time interval 1-3 months) as the axis, data labels (such as light hydrocarbon oxidizing bacteria number and oil and gas saturation) and geological annotations are added in the animation, and the frame rate is set to 10-15 frames / second.

[0122] The application provides a microbial oil and gas exploration method, which comprises the following steps: acquiring the number of light hydrocarbon oxidizing bacteria of a to-be-tested soil sample of each sampling point in a target region; fusing the number of light hydrocarbon oxidizing bacteria of each to-be-tested soil sample with the environmental characteristics of the corresponding sampling point and the geological characteristics of the corresponding sampling point to obtain a fusion feature matrix; inputting the fusion feature matrix into an oil and gas reservoir probability distribution model to output the oil and gas reservoir distribution probability of each sampling point in the target region, and determining the oil and gas reservoir existing region of the target region based on the oil and gas reservoir distribution probability of each sampling point. The integration of the number of light hydrocarbon oxidizing bacteria of the to-be-tested soil sample, the environmental characteristics and the geological characteristics can more comprehensively and accurately reflect the geological and microbial activity conditions of the target region, and compared with a single information source, the accuracy and reliability of the oil and gas reservoir distribution judgment can be improved. The oil and gas reservoir probability distribution model can be used to quickly output the oil and gas reservoir distribution probability of each sampling point in the target region, which greatly shortens the time required for a large number of field tests and long-time analysis in the traditional exploration method, significantly improves the exploration efficiency, helps to quickly lock the potential oil and gas reservoir region, reduces the exploration cost and time input, and based on the oil and gas reservoir distribution probability of each sampling point, the oil and gas reservoir existing region of the target region can be more accurately determined, which avoids the blindness and uncertainty in the traditional method, provides a clear target region for subsequent oil and gas exploitation work, and improves the success rate and economic benefits of exploitation.

[0123] It should be understood that the size of the serial number of each step in the above embodiment does not mean the order of execution, and the execution order of each process should be determined according to its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the application.

[0124] The following is a device embodiment of the application. For details not described in detail, reference can be made to the corresponding method embodiments described above.

[0125] Figure 2 A structure schematic diagram of a microbial oil and gas exploration device provided by an embodiment of the application is shown. For ease of illustration, only parts related to the embodiments of the application are shown, and details are described as follows:

[0126] As shown in Figure 2 The microbial oil and gas exploration device 2 comprises:

[0127] The acquisition module 21 is configured to acquire the number of light hydrocarbon oxidizing bacteria of a to-be-tested soil sample of each sampling point in a target region.

[0128] The fusion module 22 is configured to fuse the number of light hydrocarbon oxidizing bacteria of each to-be-tested soil sample with the environmental characteristics of the corresponding sampling point and the geological characteristics of the corresponding sampling point to obtain a fusion feature matrix.

[0129] The determining module 23 is configured to input the fusion feature matrix into an oil and gas reservoir probability distribution model, output the oil and gas reservoir distribution probability of each sampling point in the target region, and determine the oil and gas reservoir existing region of the target region based on the oil and gas reservoir distribution probability of each sampling point.

[0130] The present application provides a kind of unmanned aerial vehicle life prediction device, by obtaining the number of light hydrocarbon oxidizing bacteria of the soil sample to be measured of each sampling point in target region;The number of light hydrocarbon oxidizing bacteria of each soil sample to be measured is fused with the environmental characteristics of corresponding sampling point and the geological characteristics of corresponding sampling point, and fusion feature matrix is obtained;The fusion feature matrix is input into oil and gas reservoir probability distribution model, and the oil and gas reservoir distribution probability of each sampling point in the target region is output, and the oil and gas reservoir existing region of the target region is determined based on the oil and gas reservoir distribution probability of each sampling point.The present application can more comprehensively and accurately reflect the geological and microbial activity situation of target region by the integration of the multi-dimensional information of the number of light hydrocarbon oxidizing bacteria of soil sample to be measured, environmental characteristics and geological characteristics, compared with single information source, the accuracy and reliability of oil and gas reservoir distribution judgment can be improved;And using the oil and gas reservoir probability distribution model that has been constructed, the oil and gas reservoir distribution probability of each sampling point in the target region can be quickly output, which greatly shortens the time required for a large number of field tests and long time analysis in traditional exploration method, significantly improves the exploration efficiency, helps to quickly lock the potential oil and gas reservoir region, reduces the exploration cost and time input;At the same time, based on the oil and gas reservoir distribution probability of each sampling point, the oil and gas reservoir existing region of the target region can be determined more accurately, which avoids the blindness and uncertainty that may exist in traditional method, provides a clear target region for subsequent oil and gas exploitation work, and improves the success rate and economic benefit of exploitation.

[0131] In a possible implementation manner, the fusion module can be configured to:

[0132] Based on the number of light hydrocarbon oxidizing bacteria of the soil sample to be measured of each sampling point in the target region, the gene space distribution discrete point matrix of the target region is obtained by using Kriging interpolation algorithm;

[0133] The environmental feature vector is constructed by using the environmental characteristics of each sampling point in the target region;

[0134] The geological feature vector is constructed by using the geological characteristics of each sampling point in the target region;

[0135] The gene space distribution discrete point matrix, the environmental feature vector and the geological feature vector are transversely spliced to construct the preliminary fusion feature matrix;

[0136] The derived feature is constructed by using the preliminary fusion feature matrix, and the fusion feature matrix is obtained by using principal component analysis method to reduce the dimension of the derived feature.

[0137] In a possible implementation, the fusion module can be further configured to:

[0138] The environmental features of each sampling point in the target area are subjected to z-score standardization, and the environmental feature vectors are constructed by using the z-score standardized environmental features.

[0139] In a possible implementation, the fusion module can be further configured to:

[0140] The gene space distribution discrete point matrix, the environmental feature vectors and the geological feature vectors are subjected to spatial alignment in the WGS84 geographic coordinate system;

[0141] The gene space distribution discrete point matrix, the environmental feature vectors and the geological feature vectors subjected to spatial alignment are horizontally spliced according to the sampling point sequence to construct a preliminary fusion feature matrix.

[0142] In a possible implementation, the environmental features include salinity and water content, and the geological features include permeability; the fusion module can be further configured to:

[0143] The product of the number of light hydrocarbon oxidizing bacteria of each sampling point and the permeability in the geological features of the sampling point is taken as a first derived feature of the sampling point;

[0144] The ratio of the salinity in the environmental features of each sampling point to the water content in the environmental features of the sampling point is taken as a second derived feature of the sampling point;

[0145] The first derived features and the second derived features of all the sampling points are subjected to dimension reduction processing by using the principal component analysis method to obtain a fusion feature matrix.

[0146] In a possible implementation, the determination module can be configured to:

[0147] It is determined whether the oil and gas reservoir distribution probability of the target sampling point is greater than a preset probability, the target sampling point being any sampling point in the target area;

[0148] If the oil and gas reservoir distribution probability of the target sampling point is greater than the preset probability, it is determined that the region corresponding to the target sampling point is an oil and gas reservoir existing region;

[0149] If the oil and gas reservoir distribution probability of the target sampling point is not greater than the preset probability, it is determined that the region corresponding to the target sampling point is not an oil and gas reservoir existing region.

[0150] Figure 3 is a schematic diagram of a terminal provided by an embodiment of the present application. As shown in Figure 3As shown, the terminal 3 of this embodiment includes a processor 30, a memory 31, and a computer program 32 stored in the memory 31 and executable on the processor 30. The processor 30 implements the steps in each of the above-described microbial oil and gas exploration method embodiments when executing the computer program 32, such as Figure 1 As shown, steps 101 to 103. Alternatively, the processor 30 implements the functions of each module / unit in each of the above-described apparatus embodiments when executing the computer program 32, such as Figure 2 As shown, the functions of each module.

[0151] For example, the computer program 32 can be segmented into one or more modules / units stored in the memory 31 and executed by the processor 30 to complete the present application. The one or more modules / units can be a series of computer program instruction segments capable of completing a specific function, which are used to describe the execution process of the computer program 32 in the terminal 3. For example, the computer program 32 can be segmented into Figure 2 As shown, each module.

[0152] The terminal 3 can be a desktop computer, a notebook, a palm computer, a cloud server, and other computing devices. The terminal 3 can include, but is not limited to, the processor 30, the memory 31. Those skilled in the art can understand that Figure 3 The terminal 3 is only an example and does not constitute a limitation on the terminal 3, which can include more or fewer components than those shown, or combine some components, or different components, for example, the terminal can also include an input / output device, a network access device, a bus, etc.

[0153] The processor 30 can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor.

[0154] The memory 31 can be an internal storage unit of the terminal 3, such as a hard disk or a memory of the terminal 3. The memory 31 can also be an external storage device of the terminal 3, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the terminal 3. Further, the memory 31 can also include both the internal storage unit and the external storage device of the terminal 3. The memory 31 is used to store the computer program and other programs and data required by the terminal. The memory 31 can also be used to temporarily store data that has been output or will be output.

[0155] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above functional units and modules is exemplified, and in actual application, the above functions can be completed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiment can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit, and the integrated unit can be realized in the form of hardware or software. In addition, the specific names of each functional unit and module are only for easy distinction, and do not limit the protection scope of the application. The specific working process of the units and modules in the system can refer to the corresponding process in the foregoing method embodiments, which will not be described here.

[0156] In the above embodiments, the description of each embodiment has its own emphasis, and the parts not described or recorded in detail in a certain embodiment can refer to the relevant description of other embodiments.

[0157] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized in electronic hardware or a combination of computer software and electronic hardware. Whether the functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. A person skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the application.

[0158] In the embodiments of the present application, it should be understood that the disclosed apparatus / terminal and method can be implemented by other manners. For example, the apparatus / terminal embodiments described above are only illustrative, for example, the division of the modules or units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed units can be indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0159] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, i.e., can be located in one place or can be distributed on a plurality of network units. Part or all of the units can be selected to achieve the purpose of the embodiment according to actual needs.

[0160] In addition, the functional units in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0161] The integrated module / unit, if realized in the form of a software functional unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on this understanding, all or part of the processes in the above-mentioned embodiment methods can also be completed by a computer program instructing related hardware, and the computer program can be stored in a computer readable storage medium. When the processor executes the computer program, the steps of each microbial oil and gas exploration method embodiment can be implemented. The computer program includes computer program code, which can be in the form of source code, object code, executable file or some intermediate form, etc. The computer readable medium can include any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0162] The above-described embodiments are only used to illustrate the technical solutions of the present application, but not limit them; although the present application is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A microbial oil and gas exploration method, characterized in that: include: Obtain the number of light hydrocarbon oxidizing bacteria in the soil samples to be tested at each sampling point in the target area; The number of light hydrocarbon oxidizing bacteria in each soil sample to be tested is fused with the environmental characteristics and geological characteristics of the corresponding sampling point to obtain a fusion feature matrix; The fused feature matrix is ​​input into the oil and gas reservoir probability distribution model, the oil and gas reservoir distribution probability of each sampling point in the target area is output, and the oil and gas reservoir existence area in the target area is determined based on the oil and gas reservoir distribution probability of each sampling point.

2. The microbial oil and gas exploration method according to claim 1, characterized in that: The number of light hydrocarbon oxidizing bacteria in each soil sample to be tested is fused with the environmental characteristics of the corresponding sampling point and the geological characteristics of the corresponding sampling point to obtain a fused feature matrix, including: Based on the number of light hydrocarbon oxidizing bacteria in the soil samples to be tested at each sampling point in the target area, a Kriging interpolation algorithm is used to obtain a discrete point matrix of gene spatial distribution in the target area; Constructing an environmental feature vector using the environmental features of each sampling point in the target area; constructing a geological feature vector using the geological features of each sampling point in the target area; horizontally splicing the gene spatial distribution discrete point matrix, the environmental feature vector, and the geological feature vector to construct a preliminary fusion feature matrix; The preliminary fusion feature matrix is ​​used to construct derived features, and the principal component analysis method is used to perform dimensionality reduction processing on the derived features to obtain the fusion feature matrix.

3. The microbial oil and gas exploration method according to claim 2, characterized in that: The step of constructing an environmental feature vector using the environmental features of each sampling point in the target area includes: The environmental characteristics of each sampling point in the target area are z-score standardized, and the environmental characteristic vector is constructed using the z-score standardized environmental characteristics.

4. The microbial oil and gas exploration method according to claim 2, characterized in that: The step of horizontally splicing the gene spatial distribution discrete point matrix, the environmental feature vector, and the geological feature vector to construct a preliminary fusion feature matrix includes: Spatially aligning the gene spatial distribution discrete point matrix, the environmental characteristic vector, and the geological characteristic vector in the WGS84 geographic coordinate system; The spatially aligned gene spatial distribution discrete point matrix, the environmental feature vector and the geological feature vector are horizontally spliced ​​in the order of sampling points to construct the preliminary fusion feature matrix.

5. The microbial oil and gas exploration method according to claim 2, characterized in that: The environmental characteristics include salinity and water content, and the geological characteristics include permeability. The preliminary fusion feature matrix is ​​used to construct derived features, and the principal component analysis method is used to perform dimensionality reduction processing on the derived features to obtain the fusion feature matrix, including: The product of the number of light hydrocarbon oxidizing bacteria at each sampling point and the permeability in the geological characteristics of the sampling point is used as the first derived characteristic of the sampling point; The ratio of the salinity in the environmental characteristics of each sampling point to the water content in the environmental characteristics of the sampling point is used as the second derived characteristic of the sampling point; The principal component analysis method is used to perform dimensionality reduction processing on the first derived features and the second derived features of all sampling points to obtain the fusion feature matrix.

6. The microbial oil and gas exploration method according to claim 1, characterized in that: The determining of the oil and gas reservoir distribution area of ​​the target area based on the oil and gas reservoir distribution probability of each sampling point includes: Determining whether the oil and gas reservoir distribution probability of a target sampling point is greater than a preset probability, the target sampling point being any sampling point in the target area; If the oil and gas reservoir distribution probability of the target sampling point is greater than the preset probability, then determining that the area corresponding to the target sampling point is an oil and gas reservoir existence area; If the oil and gas reservoir distribution probability of the target sampling point is not greater than the preset probability, it is determined that the area corresponding to the target sampling point is not an area where oil and gas reservoirs exist.

7. A microbial oil and gas exploration device, characterized in that: For implementing the microbial oil and gas exploration method according to any one of claims 1 to 6, the microbial oil and gas exploration device comprises: An acquisition module is used to obtain the number of light hydrocarbon oxidizing bacteria in the soil samples to be tested at each sampling point in the target area; A fusion module is used to fuse the number of light hydrocarbon oxidizing bacteria in each soil sample to be tested with the environmental characteristics and geological characteristics of the corresponding sampling point to obtain a fusion feature matrix; The determination module is used to input the fused feature matrix into the oil and gas reservoir probability distribution model, output the oil and gas reservoir distribution probability of each sampling point in the target area, and determine the oil and gas reservoir existence area in the target area based on the oil and gas reservoir distribution probability of each sampling point.

8. The microbial oil and gas exploration device according to claim 7, characterized in that: The fusion module is used to: Based on the number of light hydrocarbon oxidizing bacteria in the soil samples to be tested at each sampling point in the target area, a Kriging interpolation algorithm is used to obtain a discrete point matrix of gene spatial distribution in the target area; Constructing an environmental feature vector using the environmental features of each sampling point in the target area; constructing a geological feature vector using the geological features of each sampling point in the target area; horizontally splicing the gene spatial distribution discrete point matrix, the environmental feature vector, and the geological feature vector to construct a preliminary fusion feature matrix; The preliminary fusion feature matrix is ​​used to construct derived features, and the principal component analysis method is used to perform dimensionality reduction processing on the derived features to obtain the fusion feature matrix.

9. A terminal comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the microbial oil and gas exploration method according to any one of claims 1 to 6 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the microbial oil and gas exploration method according to any one of claims 1 to 6 are implemented.

Images