Deposit microfacies identification method based on deep belief network

Abstract

The invention discloses a sedimentary microfacies identification method based on a deep belief network, and relates to the technical field of intelligent information processing, and the method comprises the steps: firstly, carrying out the combination and optimization of a logging curve, then employing the original data of the selected logging curve to construct a binary image reflecting the morphological characteristics of the curve, taking the binary image as the sample feature data, and carrying out the recognition of the sedimentary microfacies; a deep belief network constructed based on a restricted Boltzmann machine is utilized to capture a mapping relation between sample features and categories of the sample features. And the trained network can be used for single well profile sedimentary microfacies identification. By means of the strong learning ability and generalization ability of the deep belief network, the identification precision of the sedimentary microfacies can be effectively improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent information processing, and in particular relates to a sedimentary microfacies identification method based on a deep belief network. Background Art

[0002] Reservoir description is a qualitative or quantitative description of the reservoir, which allows for a comprehensive evaluation of the reservoir. Determining the types and spatial distribution of sedimentary facies and sedimentary microfacies is one of its important research contents. Traditional sedimentary facies research mainly identifies the sedimentary environment by observing information such as the rock composition, structure, and sedimentary rhythm of the core. The premise for conducting sedimentary facies research is that sufficient coring data has been obtained from the study well section, but the high cost of coring work cannot guarantee the provision of sufficient coring data. The ever-developing logging technology provides each well with rich logging data. These logging data contain rich sedimentary environment information. Therefore, combining coring data and logging data to conduct sedimentary facies research has become the current mainstream.

[0003] Many geologists have used the morphological characteristics of curves in well logging data to identify sedimentary environments through manual processing. However, the identification results are highly subjective, labor-intensive, and vary from person to person. To more fully utilize well logging data to objectively, efficiently, and accurately identify sedimentary microfacies, many scholars have begun to introduce various advanced technologies into sedimentary microfacies identification, with neural network technology being the most popular. Xu Shaohua et al. proposed an automatic sedimentary microfacies identification method based on a combination of neural networks and image processing techniques, effectively improving the learning efficiency and generalization ability of neural networks. They also applied regularized fuzzy neural networks to the identification of reservoir sedimentary microfacies, demonstrating excellent adaptability and practicality in solving the sedimentary microfacies identification problem. Luo Li, Lu Song, Wu Cancan, Cheng Liangbing, Pang Guoyin, Ma Kui, Liu Lingyun, and others used BP (Back Propagation) neural networks, genetic neural networks, and probabilistic neural networks to automatically identify sedimentary microfacies from actual well logging data, achieving significant results. Zhang Fuming et al. extracted characteristic parameters from well logging curves and, based on core observations and regional experience, selected curve combinations with high sensitivity in indicating sedimentary facies. They then used neural network pattern recognition technology to automatically interpret sedimentary microfacies in single-well profiles. Li Wenbo et al. applied a self-organizing neural network approach to sedimentary microfacies identification. This method extracted geometric and image characteristic parameters from well logging curves, used these characteristic parameters to establish a relationship between sedimentary microfacies type and logging curve shape, developed a microfacies pattern recognition system, and applied it to achieve good identification results. Xia Zhen et al. studied an automatic sedimentary microfacies identification algorithm based on active layering. This algorithm derived the changing characteristics of the logging curves from a set of well logging response data, used these characteristics to divide the stratigraphic profile into several electrical facies, established a facies relationship model, and thus automatically identified sedimentary microfacies. Duan Dongpeng et al. used an improved BP neural network method to automatically identify sedimentary microfacies in areas with high well spacing density and a large number of wells, reducing work time and improving efficiency. Mei Junwei et al. constructed a system for sedimentary microfacies identification. This system uses principal component analysis and back-propagation neural networks based on genetic algorithms, which can reduce computational complexity and avoid falling into local optimal solutions, effectively improving identification efficiency and maintaining high recognition accuracy. Li Guorong et al., addressing the multi-solution and fuzzy nature of the correspondence between sedimentary environments and sedimentary characteristics, applied a neural network fuzzy clustering analysis method, combining the adaptability and fault tolerance of neural networks with the fuzzy comprehensive discriminant characteristics of fuzzy logic to achieve multi-factor fuzzy comprehensive identification of sedimentary microfacies. Liu Hui et al. generated a deep process neural network by superimposing a deep belief network and a BP classifier, and based on this, achieved automatic identification of sedimentary microfacies.Luo Renze et al. proposed an intelligent sedimentary microfacies identification method based on characteristic structures (DMC) and bidirectional long short-term memory (BiLSTM) for well logging sedimentary microfacies, making the algorithm more robust. He Xu et al. established a well logging facies identification model by training a convolutional neural network. They then used wavelet basis functions of different scales and extreme value segmentation to process and segment the well logging data, ultimately achieving the classification of sedimentary units of different scales. Valentin et al. established a deep residual network to automatically identify reservoir lithofacies using single-point ultrasonic and microresistivity borehole imaging log data as input. Xiao He et al. used a hierarchical clustering method and Fisher discriminant to divide strata in uncored wells, thereby achieving automatic sedimentary microfacies classification. Camila Martins Saporetti et al. optimized an artificial neural network using an adaptive differential evolution (DE) algorithm. The improved ANN network architecture demonstrated promising results in solving the lithologic identification problem in the South Provence Basin, providing improved classification accuracy. Hong et al. designed a target detection strategy that uses integrated classifiers to detect the type and location of sedimentary microfacies. Their application to clastic strata in the Ordos Basin demonstrated the effectiveness and consistency of this method for automatic classification of sedimentary microfacies. Luo Xin et al. deepened their understanding of the internal structure of sandstone-conglomerate sedimentary microfacies through FMI image extraction and learning using a CNN network, providing evidence for the detailed characterization and effective prediction of sedimentary microfacies. While these techniques have achieved some success in sedimentary microfacies identification, they require the pre-process extraction of feature parameters from well logging curves, which results in a certain amount of information loss. Furthermore, the network learning and testing process uses relatively few sample features and input nodes, which directly affects the network's generalization ability and leads to low recognition accuracy.

[0004] Deep belief networks (DBNs) are a relatively mature model in the field of deep learning. Based on restricted Boltzmann machines (RBMs), they employ a layered training approach to effectively overcome the vanishing gradient problem inherent in deep networks. They have been successfully applied in fields such as classification, detection, monitoring, and identification. However, the application of DNBs in sedimentary microfacies identification has yet to be reported. Summary of the Invention

[0005] To solve the above technical problems, the present invention proposes a sedimentary microfacies identification method based on deep belief network. The constructed samples can make full use of the morphological characteristics of small-layer microfacies to improve the identification accuracy of sedimentary microfacies. By utilizing the multi-layer mapping mechanism of deep learning, the identification accuracy and generalization ability of sedimentary microfacies can be improved.

[0006] To achieve the above objectives, the present invention provides a sedimentary microfacies identification method based on a deep belief network, comprising: Converting raw data of the target well's logging curve combination into a binary image, wherein the binary image reflects the morphological characteristics of the logging curve; The binary image is trained using a deep belief network, where the deep belief network is composed of a multi-layer restricted Boltzmann machine and learns the mapping relationship between the well logging curve morphology and the sedimentary microfacies category through unsupervised pre-training and supervised fine-tuning; The trained deep belief network is used to classify the sedimentary microfacies of the well logging data of the identified wells.

[0007] Technical effect of the present invention: The present invention discloses a sedimentary microfacies identification method based on a deep belief network, which uses the original curve morphological features in the logging data as input samples and combines the deep belief network to perform sedimentary microfacies identification. It can free geologists from manual processing and hand over the logging data processing and sedimentary microfacies identification work to the computer for automatic completion. While reducing the manual workload, it can also ensure the information integrity of the sampled logging data and ensure the objectivity, efficiency and accuracy of the identification results. The results show that compared with the core analysis results, the sedimentary microfacies identification rate of this method can reach more than 87%, so this method is suitable for single-well sedimentary microfacies identification in uncored areas. Compared with other artificial intelligence methods, the deep belief network method adopted by the present invention has good adaptability, good network robustness, strong generalization ability, high recognition efficiency, and is very suitable for handling sedimentary microfacies identification problems. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] The accompanying drawings, which constitute part of this application, are intended to provide a further understanding of this application. The exemplary embodiments and descriptions of this application are intended to explain this application and do not constitute an improper limitation on this application. In the accompanying drawings: Figure 1 is a restricted Boltzmann machine model according to an embodiment of the present invention; Figure 2 The structure of the deep belief network of an embodiment of the present invention; Figure 3 Schematic diagram of the basic scheme of using DBN to identify sedimentary microfacies in a single well profile according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the number of small layer data points in an embodiment of the present invention; Figure 5 Schematic diagram of small layer logging curve data according to an embodiment of the present invention, wherein (a) is GR, (b) is HAC, (c) is RLLD, and (d) is RLLS.

[0009] Figure 6 Schematic diagram of 20 randomly selected image samples in an embodiment of the present invention; Figure 7This is a schematic diagram comparing the microfacies identification results of a well based on the DBN method and geological data in an embodiment of the present invention. DETAILED DESCRIPTION

[0010] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in this application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0011] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0012] This embodiment provides a sedimentary microfacies identification method based on a deep belief network, including: Converting raw data of the target well's logging curve combination into a binary image, wherein the binary image reflects the morphological characteristics of the logging curve; The binary image is trained using a deep belief network, where the deep belief network is composed of a multi-layer restricted Boltzmann machine and learns the mapping relationship between the well logging curve morphology and the sedimentary microfacies category through unsupervised pre-training and supervised fine-tuning; The trained deep belief network is used to classify the sedimentary microfacies of the well logging data of the identified wells.

[0013] Furthermore, the logging curve combination includes natural gamma, sonic transit time, deep lateral resistivity and shallow lateral resistivity.

[0014] Furthermore, the binary image generation process includes: unifying the data length of the logging curves of each small layer, truncating the excess part, and filling the insufficient part with zero; according to the value range of the logging curve, mapping the data points to the corresponding pixel positions of the binary image, where 1 represents the data point and 0 represents the background.

[0015] Furthermore, the hidden layer nodes of the deep belief network are configured as 50, 50, and 250.

[0016] Furthermore, the unsupervised pre-training includes training each layer of restricted Boltzmann machine in sequence, where the first layer RBM is iterated 10 times, the second layer RBM is iterated 50 times, and the top layer RBM is iterated 100 times.

[0017] Furthermore, supervised fine-tuning uses the back-propagation algorithm to optimize network parameters.

[0018] Furthermore, the sedimentary microfacies categories include distributary channels, non-main channel, sheet sand, sheet sand margin and distributary inter-channel mud.

[0019] Furthermore, the logging curve data also includes small layer thickness, which is mapped to a binary image after being magnified.

[0020] The Restricted Boltzmann Machine (RBM) is a probabilistic graphical model with only two layers. Strictly speaking, it is not a true deep learning model. It has attracted attention because it can be used to construct deep learning models such as autoencoders and deep belief networks. From the perspective of probabilistic graphical models, this article introduces the standard model and learning algorithm of RBM. Figure 1 As shown in the figure Is the visual layer node, is the visual layer bias; is a hidden layer node, is the hidden layer bias.

[0021] First, the RBM energy function is defined as follows: (1); Based on this function, the joint probability of the RMB visual layer and hidden layer can be further defined: (2); According to formula (2), we can get the following two edge laws: (3); (4); make , , the training algorithm of RBM is given below.

[0022]

[0023] Mathematically, the Deep Belief Net (DBN) is a hybrid graph model consisting of two parts: undirected and directed. Figure 2 As shown in Figure 2. The top two layers are undirected graphs, forming an associative memory network, and the remaining layers are directed graphs. These layers are essentially restricted Boltzmann machines. Figure 2 middle, is the visible layer vector, is the hidden layer vector, . is the weight between the visible layer and the first hidden layer, For the hidden layers and the The weights between the hidden layers, For the label layer and The weights between the hidden layers, is the visual layer bias, For the The bias of the hidden layer, is the label layer bias.

[0024] make , , the joint probability distribution of the deep belief network is: (5); in, ,and You can press is the visible vector, is the restricted Boltzmann machine calculation of the implicit vector. By eliminating the label vector through the probability summation operation, Equation (5) can be simplified as follows: (6); At the label level In , only one component is 1, and the rest are 0. The component subscript with a value of 1 represents the category of the sample. Based on the above results, the input and output relationship of the deep belief network is as follows.

[0025] (7); in, For the A vector with dimension 1 and the rest all 0s, ; .

[0026] The training of the deep belief network adopts a combination of unsupervised and supervised methods. First, the RBM algorithm is used to perform unsupervised pre-training on the restricted Boltzmann machines of each layer, and then the back-propagation algorithm is used for supervised tuning.

[0027] The sedimentary microfacies types determined by the expert analysis results of the core data and the original morphological data of the optimized well logging curve combination are used as learning samples and input into the deep belief network model for learning, thereby identifying the sedimentary microfacies types of a single well. The basic scheme for using the deep belief network to identify the sedimentary microfacies of a single well profile is as follows: Figure 3 shown.

[0028] DBN, a key deep learning model, is still essentially a neural network. Before it can be used for microfacies identification, it must first be trained using a standard small layer. To construct the training sample, all microfacies types contained in the target reservoir must be determined, i.e., a microfacies type set must be constructed. Each element in this set will serve as the expected output of the training sample.

[0029] Taking the experimental oil layer of the experimental oil field as an example, according to the expert analysis results of coring and electrical logging data, the study area belongs to the large-scale river-controlled delta front subfacies, mainly including five sedimentary microfacies: distributary channel, non-main channel, sheet sand, sheet sand edge and distributary inter-mud.

[0030] In actual facies identification, each well has multiple well logs, each contributing to varying degrees of success in identifying sedimentary microfacies. Furthermore, different oilfield regions have distinct sedimentary environments. Therefore, when conducting sedimentary microfacies research, it is necessary to optimize well log combinations based on regional characteristics. This optimization facilitates the construction of high-quality training samples, thereby improving network training efficiency and generalization capabilities.

[0031] The basic method for optimizing well logging curve combinations is to evaluate the correspondence between the characteristics of the well logging curve combinations and known sedimentary microfacies types based on existing coring data. This allows the selection of well logging curve combinations with high sensitivity for sedimentary microfacies identification to construct sub-layer characteristic samples. Taking the experimental reservoirs in the experimental oil field as an example, analysis and comparison revealed that the following four well logging curve combinations are highly correlated with sedimentary microfacies types: natural gamma ray, acoustic transit time, deep lateral resistivity, and shallow lateral resistivity. Furthermore, formation thickness is also a significant factor influencing microfacies types.

[0032] The key to using artificial intelligence to identify sedimentary microfacies lies in enabling computers to automatically identify individual well logging curves, analyze their morphological differences, and then determine the sedimentary microfacies type corresponding to each combination of well logging curves based on their combined characteristics. The advantage of deep belief networks is that they can directly utilize the raw features of sample data for training and identification. Therefore, once a well logging curve combination is selected, only simple processing of the raw curve data is required.

[0033] Unlike traditional neural network sample construction methods, the DBN-based recognition method proposed in this paper directly uses all the data from the selected well log curves to construct training samples. This fully utilizes the feature data and improves recognition results, which is also a prominent advantage of this method. Specifically, the well log curve corresponding to each sublayer is mapped into a binary image. This image becomes the DBN input sample, and the microfacies category of the sublayer becomes the corresponding DBN output sample.

[0034] First, for all wells involved in the training, the number of data points in each small layer on the logging curve is counted. According to the statistical results, all small layer data points are adjusted to the same length. If the length exceeds this value, the data will be truncated; if the length is less than this value, the data will be padded with zeros.

[0035] Then count the maximum and minimum values of each logging curve of all small layers. Assume that the logging curve determined by the combination optimization is The maximum and minimum values of each bar are , The maximum and minimum thickness of all small layers are .

[0036] For each small layer, first generate a The binary empty image, at this time all pixel gray values are 0. Then The logging curve is mapped to the image from left to right. The specific method is as follows: The first curve data points , first Round to the nearest integer , and then the pixels in the binary image For small layer thickness, the corresponding pixel is set to 1 using this method. The only difference is that the small layer data points, all with the same thickness. This completes the construction of the input sample. For the desired output sample, it can be constructed as dimensional vector , the vector has only The dimension is 1, and the rest are all 0, where is the number of microfacies categories.

[0037] First, the DBN is trained using training samples, enabling it to learn the highly complex nonlinear mapping relationship between sub-layer characteristics and microfacies types. Once trained, it can be used to identify sedimentary microfacies types in a single well. Specifically, the target sub-layer in the well to be identified is constructed using the same method used to construct training samples, creating identification samples in the form of binary images. These identification samples are then submitted to the DBN, and the sedimentary facies type of the target sub-layer in the well is obtained at the DBN's output endpoint.

[0038] Using Matlab to program a deep belief network, we simulated sedimentary microfacies identification for 10,400 substrata in the experimental oil formations of 800 wells in an experimental oil field. As mentioned previously, these substrata have five types of sedimentary microfacies. After optimizing well log combinations, we ultimately selected gamma ray (GR), acoustic transit time (HAC), deep lateral resistivity (RLLD), shallow lateral resistivity (RLLS), and substratum thickness as substratum characteristic data.

[0039] According to the statistical results, the number of data points contained in all 10400 small layers is as follows Figure 4 As shown, the maximum value is 240 and the minimum value is 12. The four logging curve data are as follows Figure 5As shown, (a) is GR, (b) is HAC, (c) is RLLD, and (d) is RLLS. The maximum values of the four curves are 1075, 471, 361, and 306 after rounding, and the minimum values are 0, 0, 0, and -70 after rounding. To simplify the calculation, according to Figure 4 , the present invention uniformly sets the number of points of all small layers to ;according to Figure 5 The four curves are restricted to the ranges of GR: [0, 300], HAC: [150, 400], RLLD: [0, 100], and RLLS: [0, 100]. Data outside this range are truncated. All sub-layer thicknesses are between 0 and 8 cm. To improve discrimination, all sub-layer thicknesses are multiplied by 10 and rounded to the nearest integer. The sub-layer thickness range is now [0, 80].

[0040] Based on the sample construction method, each small layer feature data can be converted into a magnitude 20 images randomly selected from 10400 image samples are shown in the figure below. Figure 6 These binary images can be directly submitted to the deep belief network for network training.

[0041] In order to fully verify the excellent performance of the deep belief network, the following experimental scheme is adopted: the small layer data of all 800 wells are divided into training samples and test samples according to different proportions. and Represent the number of training samples and the number of test samples respectively. This scheme will The specific settings are: 7:1, 6:2, 5:3, 4:4, 3:5, 2:6, and 1:7. The DBN parameters are set as follows: 150300 visible layer nodes, 200 first hidden layer nodes, 200 second hidden layer nodes, 1000 top layer nodes, and 6 label layer nodes. The number of iterations for each layer is 10 for the first RBM, 50 for the second RBM, and 100 for the top RBM.

[0042] To ensure the objectivity of the experimental results, the DBN network was trained 10 times for each ratio, and the recognition effect after each training was statistically analyzed. The specific results are shown in Table 1.

[0043] Table 1

[0044] According to the experimental results, when The experimental results are best when , with both recognition rates being the highest, as shown in bold in Table 1. This demonstrates that DBN can achieve high recognition results for sedimentary microfacies types even with a small number of training samples. Furthermore, the DBN's recognition performance varies little across various ratios, indicating that DBN is minimally affected by sample ratio, demonstrating the robustness of the DBN recognition method.

[0045] In order to comprehensively examine the recognition performance of the DBN method for sedimentary microfacies types, based on the above experimental results, the recognition effect of DBN under different hidden layer node configurations is further examined. The specific scheme is: The hidden layer nodes were set to the following seven configurations: (10, 10, 50), (30, 30, 150), (50, 50, 250), (100, 100, 500), (150, 150, 750), (200, 200, 1000), and (250, 250, 1250). The number of iterations for each RBM was set the same as before. For each node configuration, the DBN was trained independently 10 times, and the recognition performance after each training session was analyzed. The specific results are shown in Table 2.

[0046] Table 2

[0047] According to the experimental results, the best performance was achieved when the hidden layer node configuration was (50, 50, 250), achieving the highest recognition rates for both methods. This is shown in bold in Table 2. This demonstrates that a relatively simple node configuration can achieve ideal results in identifying sedimentary microfacies using DBNs. Furthermore, the DBN's recognition performance remained similar across various hidden layer node combinations, indicating that DBNs are minimally affected by the node configuration, further demonstrating the robustness of the DBN recognition method.

[0048] The comparison of the microfacies identification results based on the DBN method and the geological data shows that the microfacies identified by the DBN method are basically consistent with the microfacies divided by the geological data. Only the sheet sand microfacies of the PI11 layer was identified as a non-main channel. The reason for this may be that the bottom of the curve has the scour surface characteristics of the channel, and the sandstone thickness is between the sheet sand and the non-main channel microfacies, resulting in an identification error. Finally, 600 test wells were used to identify sedimentary microfacies and compared with the sedimentary microfacies identification results of these logging experts. Figure 7 As shown in the figure, the results show that the sedimentary microfacies identification method based on the DBN method is feasible and effective.

[0049] The network structure of the present invention is simple and has strong generalization capabilities. This is because DBN, unlike conventional neural networks, employs a novel learning mechanism and multi-layer structure, overcoming the vanishing gradient phenomenon often associated with the BP algorithm when a large number of layers are used, thereby effectively improving the DBN's recognition capabilities. Furthermore, converting raw sample data into binary images and fully utilizing the morphological features of well logging curves for recognition is also a key factor in improving DBN's recognition performance and a unique feature of the present invention.

[0050] By using the original curve morphological features in the logging data as input samples and combining them with a deep belief network for sedimentary microfacies identification, geologists can be freed from manual processing and the logging data processing and sedimentary microfacies identification work can be automatically completed by computers. While reducing the manual workload, it can also ensure the information integrity of the sampled logging data and the objectivity, efficiency and accuracy of the identification results. The results show that compared with the core analysis results, the sedimentary microfacies identification rate of this method can reach more than 87%. Therefore, this method is suitable for single-well sedimentary microfacies identification in uncored areas. Compared with other artificial intelligence methods, the deep belief network method adopted in the present invention has good adaptability, good network robustness, strong generalization ability, high recognition efficiency, and is very suitable for handling sedimentary microfacies identification problems.

[0051] The above are merely preferred embodiments of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A sedimentary microfacies identification method based on deep belief network, characterized by: include: Converting raw data of the target well's logging curve combination into a binary image, wherein the binary image reflects the morphological characteristics of the logging curve; The well logging curve combination includes natural gamma, sonic transit time, deep lateral resistivity and shallow lateral resistivity; The binary image is trained using a deep belief network, where the deep belief network is composed of a multi-layer restricted Boltzmann machine and learns the mapping relationship between the well logging curve morphology and the sedimentary microfacies category through unsupervised pre-training and supervised fine-tuning; The trained deep belief network is used to classify the sedimentary microfacies of the well logging data of the identified wells.

2. The sedimentary microfacies identification method based on deep belief network according to claim 1, characterized in that: The binary image generation process includes: unifying the data length of the logging curves of each small layer, truncating the excess part, and filling the insufficient part with zero; according to the value range of the logging curve, mapping the data points to the corresponding pixel positions of the binary image, where 1 represents the data point and 0 represents the background.

3. The sedimentary microfacies identification method based on deep belief network according to claim 1, characterized in that: The hidden layer node configuration of the deep belief network is 50, 50, 250.

4. The sedimentary microfacies identification method based on deep belief network according to claim 1, characterized in that: The unsupervised pre-training involves training each layer of restricted Boltzmann machines in sequence, where the first layer RBM is iterated 10 times, the second layer RBM is iterated 50 times, and the top layer RBM is iterated 100 times.

5. The sedimentary microfacies identification method based on deep belief network according to claim 1, characterized in that: Supervised fine-tuning uses the back-propagation algorithm to optimize network parameters.

6. The sedimentary microfacies identification method based on deep belief network according to claim 1, characterized in that: Sedimentary microfacies categories include distributary channels, non-main channel, sheet sand, sheet sand margin and distributary inter-channel mud.

7. The sedimentary microfacies identification method based on deep belief network according to claim 1, characterized in that: The well logging curve data also includes the thickness of small layers, which is mapped to a binary image after being magnified.

Images