179 results about "Logging while drilling" patented technology

The invention discloses a logging-while-drilling reservoir parameter dynamic prediction method based on double attention convolution LSTM, and the method comprises the steps: obtaining logging-while-drilling data and drilling data in real time, and carrying out the preprocessing of the data; establishing a DTW dynamic time window model based on a gradual transition strategy, and adaptively adjusting the length of an input sequence according to the stratum change speed; a double attention convolution LSTM model is combined with the DTW dynamic time window model to establish a feature-time double attention mechanism, and two dimensions of features and time are adaptively fused based on the convolution LSTM; setting a buffer area, storing data and controlling data quality; setting an online increment fine tuning strategy; and performing instance training, and performing multi-angle evaluation on the performance of the double attention convolution LSTM model by adopting multiple indexes. According to the scheme, the influence of features and time on the prediction accuracy is fully considered, and the prediction accuracy is improved.

The present invention relates to a computer implemented method, the main objective of which is to predict real-time LWD profiles in the intervals from the drill tip to the limit of the profile offsets. This becomes a methodology that serves as yet another input to improve assertiveness in decision-making during well drilling operations.

The invention discloses a while-drilling electromagnetic wave advanced detection sensitivity analysis method, and belongs to the field of oil-gas exploration and development. The method comprises the following steps: constructing a while-drilling electromagnetic wave advanced detection model, quantifying the influence of different measurement parameters on advanced detection signals, then analyzing the response characteristics of the advanced detection signals, extracting key sensitivity factors, and defining a single-parameter sensitivity function and a multi-parameter coupling sensitivity function; the sensitivity of advanced detection signals is qualitatively and quantitatively analyzed from the four angles of the distance to the interface, the formation resistivity, the interlayer and the stratigraphic dip angle, so that the while-drilling electromagnetic wave advanced detection range is optimized, and the adaptability in the well logging process is enhanced. And meanwhile, the advanced detection second-order sensitivity is approximately solved by adopting a central difference formula, influence factors suffered by advanced detection signals are analyzed from a numerical angle, the interaction of multi-parameter coupling is analyzed, and a scientific basis is provided for electromagnetic wave logging while drilling accurate inversion and instrument design.

The invention relates to the technical field of logging while drilling, and particularly provides a logging while drilling electromagnetic wave resistivity logging inversion method. The method comprises the following steps: constructing an XGBoost feature screening module, and optimizing a data set through the XGBoost feature screening module to obtain an optimized data set; according to the method, a BiLSTM-Transformer model is constructed, an optimized data set is input into the model, an inversion result is obtained, the inversion precision and the calculation efficiency are improved, and inversion of the target stratum resistivity, the adjacent layer surrounding rock resistivity and the boundary distance under the anisotropic stratum model is achieved.

The invention provides a petroleum drilling geomechanical characteristic estimation method based on deep learning, and aims to solve the problems of high measurement cost and poor real-time performance of shear wave velocity Vs in traditional geomechanical analysis by fusing logging while drilling LWD data and real-time drilling engineering parameters. According to the method, a Transformer model is adopted to carry out depth correction on logging-while-drilling data, depth offset of a drill bit and a logging sensor is eliminated, and real-time drilling parameters such as torque T, bit pressure WOB and drilling speed ROP are combined and input into a multi-layer perceptron MLP network to predict the shear wave speed Vs. And further calculating key geomechanical parameters based on the predicted shear wave velocity Vs. In practical application, the method improves the shear wave velocity prediction accuracy to 97.2%, the mean absolute error MAE of the shear modulus is reduced from 0.186 to 0.059, and the bulk modulus is reduced from 0.189 to 0.040. The method can be used for outputting stratum elastic parameters, optimizing bit pressure, pre-warning well wall instability and adjusting a well track, well drilling safety and efficiency are improved, and meanwhile dependence on an expensive well logging technology is reduced.

Systems and methods include obtaining well log data and core sample data of a subsurface formation; generating, based on the well log data and the core sample data, an unconfined compressive strength log for the subsurface formation; using an unsupervised machine learning model to form rock type clusters based on the unconfined compressive strength log and the well log data; forming a training dataset including the well log data, the training dataset labeled based on the rock type clusters; training a supervised machine learning model using the training dataset. While drilling a well in the subsurface formation, logging-while-drilling data is obtained from drilling equipment used to drill the well; and rock types in the subsurface formation are determined using the supervised machine learning model and the logging-while-drilling data.

The invention discloses a formation pressure prediction method and device, electronic equipment and a storage medium. The method comprises the steps that logging-while-drilling data of a target area are acquired; performing variable sequence characteristic quantification on the continuous time sequence logging-while-drilling data by using a recurrent neural network to obtain sequence characteristics; on the basis of the sequential features, weighted values of all logging parameters in the logging-while-drilling data are constructed through an attention mechanism, and then stratum deposition features are constructed through a preset weight parameter matrix; inputting the stratum deposition characteristics into a prediction model to predict and obtain a stratum pressure prediction value of the target area; wherein the prediction model is obtained based on a full-connection network through training of formation feature representation marked with actual formation pressure. The method can efficiently and accurately realize formation pressure prediction, and can be widely applied to the technical field of data prediction.

The invention discloses a rotary steerable drilling trajectory prediction method based on deep learning and knowledge distillation, and belongs to the technical field of drilling trajectory prediction, and the method comprises the following steps: collecting rotary steerable drilling site ground data and while-drilling well logging data, and carrying out data preprocessing; constructing a teacher model and training; the teacher model comprises a well drilling feature enhancement module and a sequence change prediction module; a teaching assistant model is constructed and trained, and the teacher model is assisted to carry out multi-interlayer data adaptive compression; a composite knowledge distillation framework and a student model are built, the student model is trained, and the trained student model is a lightweight borehole trajectory prediction model; during field application, ground data and logging-while-drilling data are preprocessed and then sent into the lightweight well track prediction model, a well track prediction value in the rotary steering drilling process is obtained, and therefore drilling operation parameters are adjusted in time according to a prediction result. According to the method, high-precision and light-weight well track prediction is realized.

The invention relates to the technical field of logging-while-drilling instruments, in particular to a signal construction method based on an azimuth logging-while-drilling instrument. The method comprises the following steps: forming a plurality of groups of measurement modes under different source distances on the basis of a multi-transmitting multi-receiving antenna system of a while-drilling azimuth logging instrument, and compensating a calculation definition formula of a voltage component; symmetrical compensation combinations are added on multiple groups of measurement modes under different source distances, symmetrical positive compensation and symmetrical inverse compensation are carried out on signals, and a combined strategy of multi-source-distance focusing and symmetrical compensation combinations is constructed; according to the joint strategy, the geological signal is defined, the error is reduced, and the measurement precision is improved.

The invention discloses a reservoir fracture distribution prediction method and system based on big data, and relates to the technical field of oil and gas field development, and the method comprises the steps of multi-source data collection and fusion, multi-scale feature extraction and optimization, hybrid model construction and three-dimensional fracture probability body generation, and fracture distribution prediction and visual display. According to the method, prediction blind areas are reduced by collecting multi-source heterogeneous data, physical constraints are achieved by introducing a Griffith criterion so as to enhance geological rationality, dynamic data and static geological data of fracturing construction are subjected to combined modeling, dynamic fusion of multi-source data is achieved, timeliness of crack prediction is improved, in addition, the physical constraint hybrid model is adopted, and the prediction efficiency is improved. And spatial characteristics and dynamic response are considered through a CNN-GBDT hybrid architecture, a fracture mechanics criterion is embedded to enhance interpretability, real-time access of logging-while-drilling data in the drilling process is supported, and key technical support can be provided for unconventional oil and gas development.

The embodiment of the invention discloses a quality control method and device for logging-while-drilling data and computing equipment. The method comprises the following steps: acquiring actually measured signal values of different sampling orientations acquired by the electromagnetic wave logging-while-drilling instrument at a target depth; performing waveform fitting on the actually measured signal values of different sampling orientations to generate a waveform coefficient corresponding to the target depth; fitting signal values of different sampling directions are generated based on the waveform coefficients; extracting a waveform baseline from the waveform coefficient, processing the measured signal value by using the waveform baseline to obtain a corrected measured signal value, and processing the fitted signal value by using the waveform baseline to obtain a corrected fitted signal value; calculating a data error of the sampling orientation according to the corrected actual measurement signal value and the corrected fitting signal value of the sampling orientation; and generating a quality index value of the target depth according to the data errors of different sampling orientations at the target depth. According to the scheme, accurate quality detection and accurate control of the logging-while-drilling data can be achieved, and the subsequent logging data analysis accuracy is improved.

The invention discloses a guiding method, a guiding system, a guiding device and a guiding medium based on a dynamic three-dimensional geologic model for logging while drilling. According to the guiding method, the guiding system and the guiding device, a three-dimensional geologic model around a horizontal well is constructed by utilizing conventional logging data of a vertical well around the horizontal well; latest horizontal well drilling and logging data and horizontal well conventional logging-while-drilling data information are supplemented into the model, and a three-dimensional geologic model body is continuously iterated and updated, so that the unification of the three-dimensional geologic model, drilling, logging and logging-while-drilling understanding is realized; the spatial relationship between the current drilling direction and a high-quality reservoir is judged in a mode of slicing along the track of the forward drilling horizontal well, decision data is provided for horizontal well geosteering, and the purpose of improving the drilling rate of an oil layer and a gas layer of the horizontal well is achieved.

According to the logging-while-drilling data lag compensation method based on deep learning provided by the invention, the prediction precision of the lag logging-while-drilling data is improved by combining the drilling engineering data and the historical logging-while-drilling data. According to the method, an encoder-decoder neural network model based on a self-attention mechanism is utilized, lag logging-while-drilling data is predicted in combination with drilling engineering features, and prediction precision is improved through a convolutional layer. Results show that when the method is used for predicting the gamma curve in the logging-while-drilling data, the root mean square error within the prediction range of 30 meters is lower than 0.589, and the average absolute percentage error is lower than 1.587. The problem of lag of the logging while drilling data is effectively solved, the operation risk under the complex stratum condition is reduced, the drilling strategy can be optimized in real time, and the safety and efficiency of the drilling process are improved.

The present application relates to the field of logging while drilling and intelligent steering control technology in oil and gas drilling engineering, in particular to a rotary steering drilling adaptive trajectory control method for image processing. It comprises: receiving original logging while drilling image data frames of downhole transmission channels; generating a high-resolution predicted geological texture map; synchronously generating a feature uncertainty distribution map representing the pixel probability distribution characteristics in the high-resolution predicted geological texture map; calculating the matching degree of the high-resolution predicted geological texture map and a preset stress-induced artifact feature library; generating a sensor contamination marker signal; calling a safety baseline geological model constructed before drilling; calculating the geological texture structure difference value relative to the safety baseline geological model; generating a geological illusion entropy; receiving the geological illusion entropy and the sensor contamination marker signal; and driving a downhole steering tool according to the activated mode. The present application ensures that the steering instruction is always based on the real existing geological information, and avoids the deviation of the drilling trajectory caused by the fictitious texture.

The invention relates to the technical field of centralizers, in particular to a logging-while-drilling tool anti-disengaging damping centralizing device and a using method, the logging-while-drilling tool anti-disengaging damping centralizing device comprises at least two centralizing damping short sections, and each centralizing damping short section comprises an outer sleeve, a centralizing damping part and a fastening adjusting part. The fishing device is reasonable and compact in structure and convenient to use, the logging-while-drilling instrument string can be fixedly installed in the middle of the inner side of the centralizing and damping short section through the centralizing and damping part, the fastening and adjusting part and the stop part, and when underground complexity and drilling tool jamming occur, the fishing device and the upper end of the logging-while-drilling instrument string are clamped and fixed together, and the centralizing and damping short section is fixed to the centralizing and damping short section. When a well logging while drilling instrument string is fished out of a well, the well logging while drilling instrument string can be separated from the centralizing and damping part by dragging the fisher upwards, penetrates through the inner side of a small water hole of the accident handling tool upwards and is fished out of the well quickly, and then underground complex and drill jamming accidents are handled through the accident handling tool, so that the fishing operation efficiency of the well logging while drilling instrument string is effectively improved; and the logging while drilling instrument string is prevented from being buried underground to cause heavy property loss.

The present invention discloses a horizontal rotary test bench for logging while drilling instruments, which improves the technical problem of electromagnetic interference to logging instruments. The device comprises a front support seat, a middle support seat and a rear support seat which are arranged in sequence, and a limiting part is respectively arranged on the middle support seat and the rear support seat, and the two limiting parts are used to jointly support and limit the logging instrument and keep the logging instrument horizontal; a shielding box is arranged on the front support seat; a power shaft, one end of which is located inside the shielding box and the other end extends to the outside of the shielding box; a motor power source is arranged in the shielding box and is used to control the rotation of the power shaft; a transmission part, one end of which is connected to the power shaft and the other end is used to connect to the logging instrument, and the transmission part is used to transmit the torque of the power shaft to the logging instrument, so that the logging instrument rotates along its own axis. The present invention can reduce electromagnetic interference, ensure the debugging accuracy of the logging instrument, and thus ensure the smooth progress of the drilling operation.

The invention discloses a section cluster optimization design method based on geological guidance mechanical specific energy intelligent correction, which comprises the following steps: determining a target stable point set according to a multi-scale local stability index in each lithologic layer section, and obtaining the lithologic layer section by performing lithologic layered clustering on logging while drilling data along a horizontal well section and corresponding well depth data; in each lithologic layer section, according to the mechanical specific energy median and / or the mechanical specific energy prediction result of the target stable point, performing layered correction on the original mechanical specific energy, the mechanical specific energy prediction result being obtained based on prediction of the target stable point set; according to the geological dessert labels and the corresponding well depth data, fracturing sections are divided; and in the fracturing section, according to the mechanical specific energy after layering correction and the geological dessert label, a cluster arrangement strategy is formulated, and perforation clusters are optimally designed based on the cluster arrangement strategy. According to the method, the mechanical specific energy can be corrected accurately and efficiently at low cost, so that the accuracy and practicability of fracturing segment cluster optimization design can be improved based on the corrected mechanical specific energy.

The invention discloses a three-dimensional crustal stress intelligent prediction method fusing physical information, and relates to the technical field of reservoir numerical simulation. The method comprises the following steps: acquiring three-dimensional crustal stress, logging-while-drilling parameters related to the three-dimensional crustal stress and lithologic parameters of developed reservoirs in the same block, matching the three-dimensional crustal stress, the logging-while-drilling parameters and the lithologic parameters, and then performing data preprocessing on the three-dimensional crustal stress, the logging-while-drilling parameters and the lithologic parameters; on the basis of a CAE self-encoder, feature extraction is conducted on the preprocessed logging-while-drilling parameters and the preprocessed core parameters, and parameters obtained after dimension reduction are obtained; based on an AVE auto-encoder, performing data enhancement on the parameters after dimension reduction and the preprocessed three-dimensional crustal stress to generate more samples; based on a PINNs neural network, training is carried out in combination with an Adam optimizer, and a trained model is used as a prediction model. According to the method provided by the invention, the prediction precision and robustness are improved when the method faces heterogeneous and anisotropic reservoirs, so that accurate crustal stress prediction is provided for efficient development of compact conglomerate reservoirs.

The invention belongs to the technical field of logging instruments, and particularly relates to a logging-while-drilling device. The logging-while-drilling device comprises a shell, the shell is detachably arranged on a downhole tool, and at least one antenna is arranged in the shell; a plurality of magnetic bars are uniformly arranged in the shell along the circumferential direction, and the central axis of each magnetic bar is parallel to the central axis of the shell; and the measuring circuit is arranged in the underground tool in a sealing manner and is connected with the antenna. The operation and maintenance difficulty of the antenna can be reduced, and installation is convenient.

The invention provides a logging-while-drilling data transmission system and method based on an underwater acoustic communication technology, and the method comprises the steps: enabling logging-while-drilling data to be intelligently compressed through an annular underwater acoustic communication transmitting transducer disposed at the upper end of a logging-while-drilling short section and an annular underwater acoustic communication receiving transducer at a well mouth, and converting the data into a sound wave signal at a well bottom; the drilling fluid is sent from the well bottom to the well mouth in an underwater acoustic communication or acoustic wave mode through a drilling fluid channel in the drill hole. The real-time transmission efficiency of the logging-while-drilling data can be greatly improved through the rate of underwater acoustic communication, and the logging-while-drilling data can be rapidly provided for well drilling engineers in real time. Real-time processing and comprehensive explanation are carried out on logging while drilling data transmitted in real time on a drilling site, and real-time optimization and adjustment of a well track can be carried out according to a comprehensive explanation result, so that intelligent geosteering drilling is realized, the drilling rate of a high-quality reservoir is finally improved, and the oil and gas yield of a single well is increased.

The invention discloses a while-drilling pressure formation force monitoring method and a while-drilling pressure formation force monitoring device. The method comprises the steps that logging-while-drilling parameters of at least one normal-pressure well section in drilled sections of a well to be monitored are obtained, and the first drilling fluid outlet temperature slope of the normal-pressure well section is determined according to the logging-while-drilling parameters of the normal-pressure well section; acquiring logging-while-drilling parameters of the current drilling well section of the to-be-monitored well, and determining a second drilling fluid outlet temperature slope of the current drilling well section according to the logging-while-drilling parameters of the current drilling well section; according to the first drilling fluid outlet temperature slope and the second drilling fluid outlet temperature slope, the formation pressure coefficient change condition is determined; and determining whether the formation pressure is normal or not according to the formation pressure coefficient change condition and a preset formation pressure coefficient change condition threshold value. According to the pressure transition zone logging parameter response characteristics, the formation pressure change can be found in the actual drilling process, and therefore the abnormal pressure early warning effect is achieved.

The invention discloses a while-drilling azimuth acoustic logging data acquisition method, and belongs to the technical field of while-drilling logging. The invention discloses a while-drilling azimuth acoustic logging data acquisition method. The method comprises the following steps: S1, acquiring natural gamma intensity, formation resistivity, three-axis vibration signals and a tool face angle of a current depth formation; s2, geological events of the current stratum are recognized according to the natural gamma intensity, the stratum resistivity and the three-axis vibration signals, and the geological events comprise a first geological event and a second geological event; s3, a first collection strategy is executed on the first geological event to collect waveform data, a second collection strategy is executed on the second geological event to collect waveform data, the first collection strategy and the second collection strategy both adopt an equal azimuth interval collection scheme, and the azimuth interval of the first collection strategy is larger than that of the second collection strategy; and S4, transmitting the waveform data acquired by the first acquisition strategy to a time difference statistical value, compressing the waveform data acquired by the second acquisition strategy, and preferentially transmitting the waveform data.

The present invention provides a method for quantitatively evaluating the sweet spot quality of conglomerate horizontal wells based on logging-while-drilling (LWD) data. Because some wells lack completion electrical logging data and dipole acoustic wave data, density, P-wave transit time, and S-wave transit time data cannot be directly obtained. LWD and gas logging data must be used to model and effectively model density, acoustic wave, and S-wave transit time. Based on the calculated density, acoustic wave, and S-wave transit time, geological parameters such as porosity, saturation, and permeability can be calculated. Based on S-wave transit time, density, and P-wave transit time, reservoir engineering parameters such as Young's modulus, Poisson's ratio, brittleness index, and maximum and minimum principal stresses can be calculated. The quality of the horizontal well sweet spot can be quantitatively evaluated by comprehensively utilizing the geological sweet spot and engineering sweet spot parameters.

The present invention proposes a three-dimensional vibration fiber grating sensor for a while-drilling logging engineering parameter pup joint, which belongs to the technical field of fiber grating sensors; it comprises a hollow shell, and the two ends of the shell extending along a first preset direction are respectively provided with a fiber input port and a fiber output port, the first preset direction is the length extension direction of the shell, and the shell is fixedly connected to the engineering parameter pup joint; three support components are sequentially arranged in the shell along the first preset direction, and the three support components are all provided with a through embedding groove, and the normal directions of the end faces of the three support components with the embedding grooves are different from each other; a bend-insensitive optical fiber is inserted into the shell and passes through the fiber input port, the embedding grooves on the three support components and the fiber output port in sequence, and the parts of the bend-insensitive optical fiber passing through the three support components are all provided with a grid area; wherein the grid area and the support component in which it is located together constitute a sensing body, and the sensing body is used to obtain vibration signals of different axes during the process of while-drilling logging.

A continuous wear measuring device for measuring a wear in a well tool includes a probe configured to wear simultaneously with the well tool, and a control unit electrically connected to the probe and configured to continuously measure a parameter of the probe. The control unit is configured to map a measured value of the parameter to a wear amount of the well tool.

The application discloses a downhole multi-parameter logging device and belongs to the technical field of logging while drilling, comprising a drill collar, a battery module and a circuit module are respectively arranged at the two ends of the inner cavity of the drill collar and are axially fixed through locking modules, a metal centralizer is fixedly arranged in the inner cavity of the drill collar, a plurality of independent centralizing blocks are arranged at the center of the metal centralizer, the drill collar comprises a drill collar body, an antenna housing, a strip-shaped magnetic core and a coil, and stable signal transmission is guaranteed, the battery module comprises annular batteries and butterfly springs, can be replaced on site and has a damping function, the circuit module comprises resistivity, azimuth gamma and inclination measurement modules, is integrally designed and has strong anti-interference performance, the metal centralizer realizes self-adaptive centralizing through centrifugal force and hydraulic damping, and the independent centralizing blocks can compensate temperature errors, the device is stable in structure, is suitable for complex well conditions, is convenient to maintain and can realize accurate and synchronous measurement of multiple parameters, thereby providing reliable data support for logging while drilling.

The invention provides a while-drilling radioactive logging system, a logging method and a drilling tool assembly, and belongs to the technical field of oil-gas exploration, and the system comprises at least one perovskite semiconductor detector which is used for directly absorbing gamma rays from a stratum and generating corresponding electric pulse signals; the electronics module is electrically connected with the perovskite semiconductor detector and used for processing the electric pulse signal to obtain while-drilling radioactive logging data, and the while-drilling radioactive logging data comprises at least one of natural gamma counting data, natural gamma energy spectrum data and gamma-gamma density data; the protection packaging body comprises a metal shock-resistant and high-pressure-resistant shell, a heat insulation and heat preservation layer and a heat channel which are arranged in the metal shock-resistant and high-pressure-resistant shell, and a high shock-resistant and impact-resistant connecting piece for fixing internal devices, and the protection packaging body is used for accommodating the perovskite semiconductor detector and the electronics module; and the safety and the accuracy of radioactive logging while drilling are improved.

The present invention discloses an adaptive scale coarsening method for logging-while-drilling curves, belonging to the field of logging-while-drilling data processing in oil and gas exploration and development, including: intelligently analyzing logging curves through deep learning technology and dynamically adjusting the coarsening strategy to achieve the collaborative optimization of data dimensionality reduction, noise filtering, and retention of key geological features. The present invention is applicable to multiple scenarios such as intelligent directional drilling, reservoir parameter calculation, 3D geological modeling, and well-seismic joint inversion, and involves technical fields such as signal processing, deep learning, and geological data analysis. The cross-integration of these fields makes the adaptive coarsening of logging curves possible, laying a technical foundation for improving the efficiency of logging data processing and the accuracy of geological interpretation.

The invention discloses a drilling risk prediction method based on geophysical data and related equipment. The method comprises the steps that geophysical parameters in the seabed drilling process are collected in real time through logging while drilling equipment; preprocessing the geophysical parameters to generate target logging data; the target logging data are constructed into a space sequence with the depth as the axis, and a dynamic extrapolation model is trained and generated by using a preset time sequence prediction algorithm based on the space sequence of the drilled stratum; inputting real-time target logging data into the dynamic extrapolation model, and predicting to obtain stratum parameters of an undrilled stratum within a preset distance in front of the drill bit; and coupling the stratum parameters with a preset geomechanical model to generate a dynamic safe drilling window containing a drilling parameter safe range. The method can effectively predict front geological anomalies, remarkably improves the safety and operation efficiency of seabed oil and gas drilling, and can be widely applied to the technical field of seabed oil and gas resource exploration and development.

The invention provides a logging-while-drilling high-speed transmission system and method based on multi-modal combination, and belongs to the technical field of oil exploration and logging-while-drilling, the logging-while-drilling high-speed transmission system comprises an underground acquisition unit, a transmission unit and a ground decoding unit, the underground acquisition unit continuously acquires multiple types of stratum parameters in the drilling process of a drilling tool and generates standardized data frames, and the transmission unit transmits the standardized data frames to the ground decoding unit; the transmission unit selects different transmission modes to transmit standardized data frames in real time, the ground decoding unit comprises an independent decoding layer, a redundancy check layer and a fusion layer, and the independent decoding layer performs wavelet transformation on the standardized data frames to separate modal signals corresponding to the transmission modes; the redundancy check layer marks conflict data with low consistency by calculating correlation among different modal signals and repairs the conflict data, the fusion layer constructs a probability fusion model based on a Bayesian network, and output data of the redundancy check layer is input into the probability fusion model to obtain decoded stratum parameters. According to the invention, efficient and reliable transmission of underground data can be effectively improved.