43 results about "History matching" patented technology

The invention discloses an oil reservoir comprehensive production optimization method and device, and the method comprises the steps: carrying out the region level recognition, single well level recognition and production optimization model recognition according to the production historical data, and updating the oil reservoir dynamic recognition; based on production historical data, performing cyclic historical fitting on model parameters of the various production optimization models by adopting the following steps: performing historical fitting on the model parameters, and performing inter-well communication relation calibration on the various production optimization models; determining a constraint condition corresponding to each production optimization model according to the updated reservoir dynamic cognition, solving a target function corresponding to each production optimization model, obtaining an optimal decision variable corresponding to each production optimization model, and obtaining a development regulation scheme corresponding to each production optimization model; an oil reservoir comprehensive development regulation and control scheme is obtained; and according to the regulation and control effect obtained through real-time monitoring of the development operation, oil reservoir dynamic understanding is updated. According to the invention, the efficiency and precision of oil reservoir production optimization can be improved.

The invention discloses an end-to-end oil reservoir history fitting method based on flow matching, which belongs to the field of petroleum engineering, and comprises the following steps: step 1, collecting and sorting oil reservoir geological parameters and oil-water well production dynamic data, constructing a data set, and carrying out data preprocessing; step 2, constructing a flow matching infrastructure; step 3, constructing an improved U-shaped network for flow matching velocity field prediction; 4, constructing an ordinary differential equation solver to realize data generation; and 5, integrating and packaging the flow matching infrastructure, improving the U-shaped network and an ordinary differential equation solver to obtain a complete flow matching model, and realizing end-to-end dynamic history fitting of the oil reservoir. And directly inputting a production dynamic condition by utilizing the trained end-to-end flow matching model to generate an oil reservoir parameter field sample conforming to the condition. In the process, additional intermediate link conversion is not needed, and dynamic matching and accurate parameter adjustment of the oil reservoir model can be achieved.

Methods and systems are configured for obtain reservoir data representing one or more features of a reservoir. The reservoir data include a set of values that satisfy a probability distribution associated with that feature. The process includes performing a geological simulation of hydraulic fracturing in the reservoir, the geological simulation generating a production estimate for a well based on the reservoir data, the production estimate associated with the one or more features; generating training data using the production estimate associated with the one or more features. The process includes training, using the training data, a machine learning model to predict a fracture half-length value of the reservoir, the fracture half-length value corresponding to the hydraulic fracturing represented by the training data. The process includes determining, based on the training, a history-matched value for the fracture half-length of the reservoir.

The invention discloses an oil reservoir history fitting method based on graph lasso and set smooth multi-data assimilation, and relates to the technical field of oil and gas field development. The method comprises the following steps: firstly, acquiring to-be-optimized model parameters and observation data of a target oil reservoir, constructing an observation error covariance matrix and initializing a model parameter set, performing numerical simulation by utilizing an oil reservoir numerical simulator to obtain prediction data, constructing a joint state matrix, performing dimensionless standardization processing on the joint state matrix to obtain an empirical correlation coefficient matrix, and then, performing optimization on the joint state matrix. And executing a graph lasso algorithm combined with an extended Bayesian information criterion to obtain an optimal dimensionless sparse precision matrix, extracting correlation coefficient sub-blocks required by Kalman updating from the optimal dimensionless sparse precision matrix based on a Scherr's theorem, obtaining a robust data auto-covariance matrix through a reverse reduction physical quantity outline, calculating Kalman gain in combination with an observation error covariance matrix, and calculating a Kalman filter. And the model parameter set is updated until the preset condition is met, the reservoir history fitting model parameter set is output, and the stability and precision of reservoir automatic history fitting are improved.

The application provides a method and device for establishing a fine choke system in a middle-deep shale gas horizontal well flowback stage, and the method comprises the following steps: S100, collecting original data of a research area; S200, calculating a production pressure difference corresponding to different choke sizes; S300, obtaining a stress sensitivity curve of the production pressure difference corresponding to different choke sizes; S400, obtaining artificial fracture parameters through history matching; S500, establishing a flowback model; S600, establishing a flowback numerical model; and S700, analyzing based on a numerical simulation result of the flowback numerical model. The application solves the problem that a field horizontal well flowback system lacks theoretical support and technical guidance, can reduce the influence of stress sensitivity damage, support agent backflow and embedding, and ground process erosion or blockage in the flowback process, and lays a foundation for maximizing the productivity of a shale gas well.

The invention relates to a well location collaborative deployment method based on a dynamic oil reservoir model and related equipment, and the method comprises the steps: obtaining a plurality of digital-analog models corresponding to a target oil and gas field and oil field related data, and carrying out historical fitting based on the digital-analog models and the oil field related data to obtain corresponding historical fitting data; establishing a linkage relationship between the digital-analog models, and obtaining a target area based on the historical fitting data and the linkage relationship; obtaining an initial well location based on the target area and generating a well trajectory corresponding to the initial well location to obtain new well information corresponding to the initial well location; and S4, generating a corresponding new well production system based on the new well information, and obtaining corresponding new well predicted productivity according to the new well production system and the historical fitting data, so as to obtain a corresponding well location deployment result by taking the new well information corresponding to the initial well location as target well location information when the new well predicted productivity meets a preset target. By implementing the method, the complexity of a well location collaborative deployment process can be reduced, and rapid well location collaborative deployment is realized.

A computer implemented method for prediction of geomechanical performance including productivity index decline and completion integrity for a well or a hydrocarbon reservoir using a geomechanics informed machine intelligence (GIMI) algorithm. The method includes running a geomechanical reservoir simulator to generate training datasets for the hydrocarbon reservoir and incorporating physical models and identified variables into the GIMI algorithm. The method further includes training a neural network of the GIMI algorithm by using correlated training datasets that correlate to the physical models to produce a resulting prediction model and performing sensitivity analysis on the resulting prediction model. Additionally, the method includes identifying dominant variables for damage mechanisms through design of experiment statistics and performing history matching and blind test on the resulting prediction model. Lastly, the method includes updating the identified variables and models incorporated into the GIMI algorithm.

The invention discloses a gas reservoir three-dimensional geological modeling method and system and a medium. The method comprises the following steps: acquiring multi-source data of a target work area; in combination with vertical seismic section data and virtual well data, depth correction is performed on the horizon after time-depth conversion, and a three-dimensional tectonic framework model is established based on the corrected horizon and fault data; fusing and generating a sand body space development probability trend body based on the seismic attribute body, the seismic inversion body and the determined optimal weight coefficients of the seismic attribute body and the seismic inversion body; a three-dimensional tectonic framework model is used as a space framework, a sand body space development probability trend body is used as a transverse soft constraint, a plurality of isoprobable lithofacies model initial implementation is established, and a three-dimensional lithofacies model is obtained; a three-dimensional distribution model of reservoir physical property parameters is generated; and importing the three-dimensional lithofacies model and the three-dimensional distribution model into a numerical simulator, performing historical fitting by using gas reservoir dynamic production data, analyzing fitting differences, and performing iterative modeling to obtain a high-precision and low-uncertainty three-dimensional gas reservoir geologic model.

The invention discloses a gas injection development scheme optimization method based on carbon dioxide oil displacement, which comprises the following steps: 1) obtaining production characteristic related data of related wells using carbon dioxide oil displacement in a research area, and comparing and analyzing the production characteristics of each well; 2) selecting a general state equation set sub-simulator of numerical simulation software to carry out gas drive reservoir modeling according to reservoir characteristics of an oil area and the reality of an oil field and aiming at the characteristics of gas drive reservoir numerical simulation; 3) geologic modeling is carried out, wherein the geologic modeling comprises establishment of a grid data field, establishment of table data and establishment of a dynamic model; 4) performing oil reservoir development history fitting; and (5) according to the final oil reservoir model, injection-production parameter optimization and periodic water injection development index prediction are carried out, and an optimized test well group gas injection development scheme is obtained. According to the method, carbon dioxide oil displacement numerical simulation is carried out, corresponding reasonable mining method research and development scheme optimization research are carried out, and a scientific and reasonable water-gas alternate gas injection recovery efficiency improving scheme can be effectively guided and formulated.

The invention provides an automatic historical fitting data processing method and device for an oil reservoir. A grid structure of a target area where the target well is located, a prior oil reservoir parameter set and simulation observation data corresponding to the prior oil reservoir parameter set are obtained; according to the spatial position of the target well in the grid structure, the Euclidean distance from the center of each grid unit to the target well is calculated, and according to the Euclidean distance, a distance field corresponding to the grid structure is determined; determining a localization matrix corresponding to the grid structure according to the distance field and a preset localization function; according to the localization matrix, spatial weighted correction is carried out on a covariance matrix determined based on the prior oil reservoir parameter set and the simulation observation data corresponding to the prior oil reservoir parameter set, and a covariance matrix subjected to spatial localization correction is obtained; and according to the covariance matrix subjected to spatial localization correction, updating the priori oil reservoir parameter set to obtain a target oil reservoir parameter set. Therefore, the stability and accuracy of oil reservoir parameter fitting are improved.

A method for sequestering carbon dioxide (CO2) in a formation includes providing an injection well extending into the formation, injecting a tracer into the injection well to the formation, collecting the tracer at a surface of the formation, obtaining tracer data from the collected tracer, determining, by a computer system, a geological property with the obtained tracer data and a history matching algorithm, and optimizing a CO2 injection rate for CO2 sequestration in the formation based on the determined geological property. A system for CO2 sequestration and determining geological properties of a formation includes an injection well extending from a surface into a formation, an injection system in fluid communication with the injection well, a production well, a tracer detection system located at the surface, and a control system coupled to the tracer detection system and configured to sequester CO2 in the formation.

The present invention provides a set smoothing automatic history matching method based on adaptive hybrid gradients, which relates to the field of oil reservoir development technology. The method obtains an initialized priori oil reservoir model parameter set and production history dynamic data; dynamically adjusts the preprocessing gradient update direction with adaptive step size; performs covariance preprocessing on the reservoir automatic history matching objective function gradient to construct the preprocessing gradient update direction; calculates parameter sensitivity weights and updates parameters in the preprocessing gradient direction in combination with an adaptive step size dynamic adjustment strategy; constructs a hybrid update direction; introduces a momentum term to accelerate convergence and suppress oscillation; and determines the distribution of reservoir parameters and objective function changes based on the final oil reservoir parameter set obtained after iterative solution of the optimization algorithm, thereby completing the automatic history matching process. The technical solution of the present invention overcomes the problems of the prior art in that it cannot adapt to strongly nonlinear oil reservoir systems, has low parameter inversion accuracy, and has high computational costs.

The present application relates to the technical field of oil and gas field development, and particularly relates to a method and device for analyzing distribution pattern of remaining condensate gas in faulted carbonate rock, comprising the following steps: S1: different reservoir levels are divided according to reservoir development pattern and development structure level; S2: corresponding models are established according to different reservoir levels, and the models of different levels are fused to obtain a three-dimensional geological model; S3: after the three-dimensional geological model is coarsened, a numerical simulation is used to establish mathematical models of each single well and well group, and production history matching is completed; S4: distribution pattern of remaining condensate gas in faulted carbonate rock condensate gas reservoir is obtained according to the matching result. By establishing a geological model and using a numerical simulation method, distribution law of remaining condensate gas in faulted carbonate rock condensate gas reservoir is studied, and main control factors are determined, which provides effective theoretical guidance for deep understanding of characteristics of the reservoir and potential tapping and improvement of condensate gas recovery.

The present application relates to a kind of shale condensate gas well volume fracturing outer zone productivity contribution analysis method, method includes the following steps: based on the complex fracture network characteristics and oil-gas two-phase flow characteristics in the fracturing development process of shale condensate gas well, establish mathematical physics model based on three linear flow theory, the complex fracture network of shale condensate gas well is equivalent to be handled into fracturing reconstruction body, fracturing reconstruction body includes artificial main fracture, fracturing reconstruction inner zone and fracturing reconstruction outer zone;Based on the mathematical physics model of three linear flow theory, the history matching of shale condensate gas well dynamic production is carried out, and typical shale and fracturing crack parameter is obtained;Based on typical shale and fracturing crack parameter, the contribution degree of fracturing reconstruction outer zone to productivity under the condition of different fracturing reconstruction outer zone boundary length, different fracturing reconstruction outer zone permeability is analyzed, to give reasonable well spacing considering the contribution degree of fracturing reconstruction outer zone.

The application relates to a shale or coal rock gas evaluation method based on a gas-water two-phase flow isotope fractionation model, which comprises the following steps: 1, collecting data; 2, establishing a gas well model; 3, establishing a control equation set; 4, coupling the influence of real gas effect and effective stress; 5, establishing a gas transmission model and a relative permeability model; 6, establishing a capillary pressure and saturation relationship; 7, considering critical desorption pressure; 8, setting initial conditions and boundary conditions; 9, solving the control equation set; 10, calculating production dynamic parameters and isotope values; 11, performing history matching; 12, verifying the quality of history matching; 13, predicting productivity; 14, calculating gas dynamic output proportion; 15, identifying fractionation inflection points and gas production mechanism conversion; the application realizes the quantitative evaluation of deep reservoir production prediction and output gas dynamic proportion by coupling gas-water two-phase flow, multiple gas transmission mechanisms, stress sensitivity and methane isotope competitive adsorption theory.

The present application provides a kind of three-dimensional physical simulation experiment method for carbon dioxide geological storage reservoir-cap system, belong to physical simulation technical field, the present application is by setting hierarchical measuring point and interface control layer in three-dimensional reservoir-cap physical model, using pressure rise rate contrast determination interface breakthrough, utilize pressure field space gradient matrix to identify preferential percolation channel, by multi-index fusion algorithm comprehensively capillary breakthrough, seepage breakthrough and macro breakthrough three levels criterion, in combination with history matching inversion method dynamic adjustment model parameter, establish breakthrough characteristic parameter set and realize closed capacity quantitative grading, solved the technical problem that cannot accurately identify the multilevel process of carbon dioxide breakthrough cap in three-dimensional physical simulation experiment and quantitatively evaluate the sealing performance.

A kind of offshore oilfield water drive recovery prediction method, comprising the following steps: one: according to the geological reservoir characteristics of research object, determine analogy oilfield;Two: the production dynamic data of analogy oilfield is carried out history matching;Three: predict the recovery of analogy oilfield, and carry out recovery sensitivity analysis;Four: based on the recovery prediction result, machine learning model is constructed;Five: the trained machine learning model is used to predict the recovery of research object.The present application not only considers the particularity of offshore oilfield development;And, in the quantitative prediction of recovery, machine learning model is also constructed, subjective factors are avoided, and it has strong objectivity;At the same time, when machine learning, the problem of less sample points in conventional machine learning is avoided;Solve the problem of offshore oilfield water drive recovery prediction.

The application provides a reservoir production optimization method considering a complex fracture network, which comprises the following steps: step 1, dividing the reservoir into two regions according to the development scale of fractures in the reservoir; step 2, respectively establishing a single pipe model and a double medium model for the divided regions, and solving the models by dividing grids on the pipe; step 3, performing history matching on the observed values and the simulated values by means of a set multi-smooth data assimilation algorithm to obtain an optimized model; and step 4, using the optimized model to optimize the injection and production capacities of each well with the economic net present value of the reservoir as an objective function. The reservoir production optimization method considering the complex fracture network performs well in improving the oil recovery rate and achieves remarkable optimization effect. The method can effectively deal with the complex fracture network in the reservoir, improve the production efficiency and maximize the economic benefits of the oilfield.

The application relates to the technical field of oil exploration and development, and discloses a stratum blockage identification method and device based on oil and gas reservoir numerical simulation automatic history matching, which comprises the following steps: firstly, a three-dimensional numerical model of an oil and gas reservoir is established; then, dynamic and static characteristic parameters of a production stage of the oil and gas reservoir are used for automatic history matching; in the numerical simulation result, corresponding dynamic and static characteristic parameters of the production stage under the blockage condition are screened to establish a data set; further, a machine learning algorithm is used to establish an agent prediction model, and the data set is used to train the agent model; after the training is completed, the trained model is used to invert and identify main control factors causing the stratum blockage; finally, the main control factors are brought into the three-dimensional numerical model to perform automatic history matching, a non-dominated sorting genetic algorithm is used to optimize the numerical simulation result in a whole solution set space, and finally, an optimal solution set is obtained to identify the stratum blockage. The application can obtain a relatively accurate stratum blockage prediction result and greatly simplify the workload.

The invention provides a three-dimensional physical simulation experiment method for a carbon dioxide geological buried storage layer-cover layer system, and belongs to the technical field of physical simulation. Layering measuring points and an interface control layer are arranged in a three-dimensional reservoir-cover layer physical model, and interface breakthrough is judged through pressure rise rate comparison; a pressure field space gradient matrix is used for recognizing a preferential seepage channel, three level criteria of capillary breakthrough, seepage breakthrough and macroscopic breakthrough are integrated through a multi-index fusion algorithm, model parameters are dynamically adjusted in combination with a historical fitting inversion method, a breakthrough characteristic parameter set is established, and quantitative grading of the sealing capacity is achieved. The technical problem that the multi-level process that carbon dioxide breaks through the cover layer cannot be accurately identified and the sealing performance cannot be quantitatively evaluated in a three-dimensional physical simulation experiment is solved.

The application discloses a physical-data collaborative driving unconventional gas well production prediction method and system, and the method comprises the following steps: S1, obtaining production data of a target unconventional gas well, and preprocessing the production data; S2, based on a dynamic drainage area calculation model and a material balance equation coupled with the dynamic drainage area, a target function is established with the minimum cumulative production error as the target; S3, history matching is carried out to obtain model parameters; S4, the average pressure of the dynamic drainage area at different times is calculated; S5, the corresponding gas production index of the gas well at different times is calculated; S6, a gas production index-material balance pseudo-time curve is drawn, and a characteristic straight line is obtained; S7, the slope and the slope distance of the characteristic straight line are used to obtain fitting parameters; and S8, the gas production index corresponding to any time tj is calculated, and production prediction is carried out based on the recursion principle. The application realizes physical-data collaborative driving unconventional gas well production prediction, and can provide technical support for unconventional gas well development.

The application provides a method for quantitatively determining water inflow of a deep fault-dissolution body oil reservoir, which comprises the following steps: a water inflow direction determining step: according to the static connectivity analysis result of a well group to be analyzed, a water inflow direction result is obtained; a connectivity degree calculating step: the position of a bottom water and the position of a well perforation are determined, and then the initial value of a conductivity and a connected volume are set, and then the connectivity degree result of the well group to be analyzed at different production time points is calculated through production history matching; and a water inflow amount calculating step: the connectivity degree result obtained through normalization processing is combined with a water inflow amount calculation model to calculate the water inflow amount. The application uses the production dynamic data of an injection-production well group, performs fine history matching, quantitatively calculates the connectivity degree, calculates the water inflow amount distribution through normalized connectivity degree, proposes the treatment direction of the well group, forms the water inflow treatment and adjustment scheme of the injection-production well group, and realizes the differentiated and quantitative water inflow treatment of the fault-dissolution body oil reservoir injection-production well group.

The application discloses an automatic history matching method based on improved ensemble smoother multiple data assimilation, and relates to the technical field of petroleum engineering.The application firstly constructs multiple reservoir models, sets prior models, real models and initial inflation factors, obtains the production of each reservoir model by using numerical simulation of the reservoir model, calculates the residual error, improves the ensemble smoother based on the singular value decomposition of the modified data covariance matrix, the disturbance enhancement of the production observation data, the adaptive learning rate matrix scaling and the geological boundary constraint, updates the permeability of each reservoir model and performs numerical simulation again, judges whether the inflation factor meets the preset condition after updating the inflation factor, continues iteration if the preset condition is met, uses the updated inflation factor of the current iteration if the preset condition is not met, and ends the iteration, and finally outputs the updated permeability field of each reservoir model, so that the problems of parameter overshoot, covariance statistical deviation and low calculation efficiency in reservoir history matching are solved, and the history matching of complex reservoir production parameters is facilitated.

The application provides a kind of offshore high water cut oilfield injection-production structure adjustment method based on flow field characterization, belongs to the technical field of reservoir engineering, including the following steps: S1: fine history matching of reservoir numerical model and calculation of flow field basic attribute parameters;S2: taking the grid in the numerical model as a unit, calculate the dynamic efficiency coefficient in each grid and carry out normalization processing;S3: dynamic efficiency coefficient field grading evaluation, divide injection-production structure adjustment potential unit;S4: calculate the reserve abundance coefficient of each grid unit and carry out normalization processing;S5: combine the normalized dynamic efficiency coefficient and reserve abundance coefficient of grid unit, calculate and output dynamic splitting coefficient;S6: based on dynamic splitting coefficient, carry out flow field grading evaluation, and then guide injection-production structure adjustment.The application realizes real-time quantitative description of reservoir oil-water seepage law and remaining oil enrichment law, improves the accuracy and reliability of injection-production structure adjustment of offshore high water cut oilfield.

The present application belongs to the technical field of oil and gas field development, and particularly relates to a shale oil horizontal well support fracture width evaluation method. The evaluation method comprises the following steps: obtaining formation parameters, filling into Saphir to determine a diagnostic analysis model; importing pressure data and flow data into the diagnostic analysis model to obtain a pressure section history, a flow section history and a differential pressure-pressure derivative double logarithmic scatter plot; performing history matching on the differential pressure-pressure derivative double logarithmic scatter plot to obtain a differential pressure-pressure derivative double logarithmic curve, determine a variable wellbore storage pressure drop and a total pressure drop, and calculate a required fracture volume according to a main fracture half length, a fracture height and a variable wellbore storage time to calculate a fracture width. The evaluation method can effectively utilize field measured data, and the predicted fracture width has high accuracy.

The invention discloses a reservoir numerical simulation history fitting method based on artificial intelligence, and relates to the technical field of reservoir simulation, and the method comprises the following steps: constructing a permeability evolution prediction model; constructing a historical fitting fitness evaluation module; constructing a system collaboration deviation quantification module; and constructing a model parameter vector closed-loop correction module. The AI model and the oil reservoir physical model are deeply fused to construct an adaptive closed-loop correction system, so that the model can actively adapt to the oil reservoir state change, the prediction precision can be improved through real-time fine adjustment, strategic correction can be carried out when the system deviation is relatively large, intelligent optimization of model parameters and remarkable improvement of historical fitting precision are realized, and the prediction accuracy is improved. And the adaptability and reliability of complex oil reservoir development are enhanced.

The invention provides a saline water layer non-pure CO2 plume inversion method based on an artificial intelligence expert system, and the method comprises the following steps: (1) carrying out the initialization of the spatial heterogeneous distribution of the reservoir permeability of a saline water layer through taking the actual pressure data collected in a monitoring well as a reference, adopting a first artificial intelligence expert system to predict well bottom pressure data of the monitoring well; (2) inputting bottom hole pressure data of the monitoring well into a coupling system of a first artificial intelligence expert system and a global optimization algorithm, and solving an optimal reservoir description parameter through a minimum historical fitting error; and (3) taking the optimal reservoir description parameters as input data, and predicting the underground distribution condition of the non-pure CO2 plume by adopting a second artificial intelligence expert system to obtain an inversion result of the non-pure CO2 plume. The inversion method for the non-pure CO2 plume in the saline water layer is high in accuracy, low in calculation overhead, low in implementation cost and suitable for large-scale popularization and application.

The application discloses a hidden space delay acceptance deterministic layered assimilation reservoir history matching method and relates to the technical field of reservoir numerical simulation. The application selects an initial prior set matrix of a target reservoir and historical observation data, projects a high-dimensional physical parameter space to a low-dimensional hidden space through singular value decomposition, constructs a deterministic reconstruction mapping operator and initializes a hidden space state set, evaluates the physical mismatch potential energy of particles by using a reservoir numerical simulator, calculates the temperature increment of adaptive annealing and updates the annealing target temperature, obtains a predicted response normalized anomaly matrix in the hidden space and executes low-dimensional singular value decomposition, generates a preconditioned hidden set and obtains the physical response thereof, constructs a local linear approximate response operator with a dynamic extreme value soft truncation boundary constraint, updates the hidden space state of each proposed particle and the accurate predicted observation response set at each annealing temperature, obtains a reservoir posteriori history matching result, and accurately realizes automatic history matching of the reservoir.

The invention discloses a tight sandstone gas reservoir dynamic reserve calculation method and device, and the method comprises the steps: carrying out the initialization of a gas reservoir numerical simulation model based on a three-dimensional texture model and gas reservoir data, and completing the building of the gas reservoir numerical simulation model; carrying out production history fitting on the gas reservoir numerical simulation model to realize verification of the gas reservoir numerical simulation model; performing production dynamic prediction and single-well dynamic reserve time-sharing calculation by using the gas reservoir numerical simulation model to obtain single-well dynamic reserves at a certain moment; and performing data processing and final value determination on the single-well dynamic reserves at different moments to obtain the single-well dynamic reserves at any moment. The method can solve the problems that the calculation of the tight sandstone gas reservoir single well dynamic reserves is greatly influenced by production time and is greatly influenced by the flow state and the like.

The present invention relates to an automatic history matching system for an oil reservoir based on transfer learning, comprising a data reading module, a population reinitializing module, an optimization module, a simulated calculation module, a comparative judgment module and an output module, wherein the data reading module reads an optimized result of an existing oil reservoir, outputs the optimized result to the population reinitializing module, obtains an initial population of a new oil reservoir by calculation and outputs the initial population to the optimization module.