A Productivity Prediction Method Based on Artificial Neural Networks and Particle Swarm Optimization for Shale-Gas Horizontal Wells

In order to overcome the deficiencies of current methods for the prediction of the productivity of shale gas hor-izontal wells after fracturing,a new sophisticated approach is proposed in this study.This new model stems from the combination several techniques,namely,artificial neural network(ANN),particle swarm optimization(PSO),Imperialist Competitive Algorithms(ICA),and Ant Clony Optimization(ACO).These are properly implemented by using the geological and engineering parameters collected from 317 wells.The results show that the optimum PSO-ANN model has a high accuracy,obtaining a R2 of 0.847 on the testing.The partial dependence plots(PDP)indicate that liquid consumption intensity and the proportion of quartz sand are the two most sensitive factors affecting the model’s performance.

A transient production prediction method for tight condensate gas wells with multiphase flow

Considering the phase behaviors in condensate gas reservoirs and the oil-gas two-phase linear flow and boundary-dominated flow in the reservoir,a method for predicting the relationship between oil saturation and pressure in the full-path of tight condensate gas well is proposed,and a model for predicting the transient production from tight condensate gas wells with multiphase flow is established.The research indicates that the relationship curve between condensate oil saturation and pressure is crucial for calculating the pseudo-pressure.In the early stage of production or in areas far from the wellbore with high reservoir pressure,the condensate oil saturation can be calculated using early-stage production dynamic data through material balance models.In the late stage of production or in areas close to the wellbore with low reservoir pressure,the condensate oil saturation can be calculated using the data of constant composition expansion test.In the middle stages of production or when reservoir pressure is at an intermediate level,the data obtained from the previous two stages can be interpolated to form a complete full-path relationship curve between oil saturation and pressure.Through simulation and field application,the new method is verified to be reliable and practical.It can be applied for prediction of middle-stage and late-stage production of tight condensate gas wells and assessment of single-well recoverable reserves.

Product quality prediction based on RBF optimized by firefly algorithm

With the development of information technology,a large number of product quality data in the entire manufacturing process is accumulated,but it is not explored and used effectively.The traditional product quality prediction models have many disadvantages,such as high complexity and low accuracy.To overcome the above problems,we propose an optimized data equalization method to pre-process dataset and design a simple but effective product quality prediction model:radial basis function model optimized by the firefly algorithm with Levy flight mechanism(RBFFALM).First,the new data equalization method is introduced to pre-process the dataset,which reduces the dimension of the data,removes redundant features,and improves the data distribution.Then the RBFFALFM is used to predict product quality.Comprehensive expe riments conducted on real-world product quality datasets validate that the new model RBFFALFM combining with the new data pre-processing method outperforms other previous me thods on predicting product quality.

A systematic machine learning method for reservoir identification and production prediction

Reservoir identification and production prediction are two of the most important tasks in petroleum exploration and development.Machine learning(ML)methods are used for petroleum-related studies,but have not been applied to reservoir identification and production prediction based on reservoir identification.Production forecasting studies are typically based on overall reservoir thickness and lack accuracy when reservoirs contain a water or dry layer without oil production.In this paper,a systematic ML method was developed using classification models for reservoir identification,and regression models for production prediction.The production models are based on the reservoir identification results.To realize the reservoir identification,seven optimized ML methods were used:four typical single ML methods and three ensemble ML methods.These methods classify the reservoir into five types of layers:water,dry and three levels of oil(I oil layer,II oil layer,III oil layer).The validation and test results of these seven optimized ML methods suggest the three ensemble methods perform better than the four single ML methods in reservoir identification.The XGBoost produced the model with the highest accuracy;up to 99%.The effective thickness of I and II oil layers determined during the reservoir identification was fed into the models for predicting production.Effective thickness considers the distribution of the water and the oil resulting in a more reasonable production prediction compared to predictions based on the overall reservoir thickness.To validate the superiority of the ML methods,reference models using overall reservoir thickness were built for comparison.The models based on effective thickness outperformed the reference models in every evaluation metric.The prediction accuracy of the ML models using effective thickness were 10%higher than that of reference model.Without the personal error or data distortion existing in traditional methods,this novel system realizes rapid analysis of data while reducing the time required to resolve reservoir classification and production prediction challenges.The ML models using the effective thickness obtained from reservoir identification were more accurate when predicting oil production compared to previous studies which use overall reservoir thickness.

A Data-Driven Oil Production Prediction Method Based on the Gradient Boosting Decision Tree Regression

Accurate prediction ofmonthly oil and gas production is essential for oil enterprises tomake reasonable production plans,avoid blind investment and realize sustainable development.Traditional oil well production trend prediction methods are based on years of oil field production experience and expertise,and the application conditions are very demanding.With the rapid development of artificial intelligence technology,big data analysis methods are gradually applied in various sub-fields of the oil and gas reservoir development.Based on the data-driven artificial intelligence algorithmGradient BoostingDecision Tree(GBDT),this paper predicts the initial single-layer production by considering geological data,fluid PVT data and well data.The results show that the GBDT algorithm prediction model has great accuracy,significantly improving efficiency and strong universal applicability.The GBDTmethod trained in this paper can predict production,which is helpful for well site optimization,perforation layer optimization and engineering parameter optimization and has guiding significance for oilfield development.

Fully coupled fluid-solid productivity numerical simulation of multistage fractured horizontal well in tight oil reservoirs

A mathematical model, fully coupling multiple porous media deformation and fluid flow, was established based on the elastic theory of porous media and fluid-solid coupling mechanism in tight oil reservoirs. The finite element method was used to determine the numerical solution and the accuracy of the model was verified. On this basis, the model was used to simulate productivity of multistage fractured horizontal wells in tight oil reservoirs. The results show that during the production of tight oil wells, the reservoir region close to artificial fractures deteriorated in physical properties significantly, e.g. the aperture and conductivity of artificial fractures dropped by 52.12% and 89.02% respectively. The simulations of 3000-day production of a horizontal well in tight oil reservoir showed that the predicted productivity by the uncoupled model had an error of 38.30% from that by the fully-coupled model. Apparently, ignoring the influence of fluid-solid interaction effect led to serious deviations of the productivity prediction results. The productivity of horizontal well in tight oil reservoir was most sensitive to the start-up pressure gradient, and second most sensitive to the opening of artificial fractures. Enhancing the initial conductivity of artificial fractures was helpful to improve the productivity of tight oil wells. The influence of conductivity, spacing, number and length of artificial fractures should be considered comprehensively in fracturing design. Increasing the number of artificial fractures unilaterally could not achieve the expected increase in production.

Productivity simulation of hydraulically fractured wells based on hybrid local grid refinement and embedded discrete fracture model

Using current Embedded Discrete Fracture Models(EDFM) to predict the productivity of fractured wells has some drawbacks, such as not supporting corner grid, low precision in the near wellbore zone, and disregarding the heterogeneity of conductivity brought by non-uniform sand concentration. An EDFM is developed based on the corner grid, which enables high efficient calculation of the transmissibility between the embedded fractures and matrix grids, and calculation of the permeability of each polygon in the embedded fractures by the lattice data of the artificial fracture aperture. On this basis, a coupling method of local grid refinement(LGR) and embedded discrete fracture model is designed, which is verified by comparing the calculation results with the Discrete Fracture Network(DFN) method and fitting the actual production data of the first hydraulically fractured well in Iraq. By using this method and orthogonal experimental design, the optimization of the parameters of the first multi-stage fractured horizontal well in the same block is completed. The results show the proposed method has theoretical and practical significance for improving the adaptability of EDFM and the accuracy of productivity prediction of fractured wells, and enables the coupling of fracture modeling and numerical productivity simulation at reservoir scale.

Integrated Modelling of Microstructure Evolution and Mechanical Properties Prediction for Q&P Hot Stamping Process of Ultra‑High Strength Steel

High strength steel products with good ductility can be produced via Q&P hot stamping process,while the phase transformation of the process is more complicated than common hot stamping since two-step quenching and one-step carbon partitioning processes are involved.In this study,an integrated model of microstructure evolution relating to Q&P hot stamping was presented with a persuasively predicted results of mechanical properties.The transformation of diffusional phase and non-diffusional phase,including original austenite grain size individually,were considered,as well as the carbon partitioning process which affects the secondary martensite transformation temperature and the subsequent phase transformations.Afterwards,the mechanical properties including hardness,strength,and elongation were calculated through a series of theoretical and empirical models in accordance with phase contents.Especially,a modified elongation prediction model was generated ultimately with higher accuracy than the existed Mileiko’s model.In the end,the unified model was applied to simulate the Q&P hot stamping process of a U-cup part based on the finite element software LS-DYNA,where the calculated outputs were coincident with the measured consequences.

3-D FRACTURE PROPAGATION SIMULATION AND PRODUCTION PREDICTION IN COALBED

In accordance with the fracturing and producing mechanism in coalbed methane well, and combining the knowledge of fluid mechanics, linear elastic fracture mechanics, thermal transfer, computing mathematics and software engineering, the three-dimensional hydraulic fracture propagating and dynamical production predicting models for coalbed methane well is put forward. The fracture propagation model takes the variation of rock mechanical properties and in-situ stress distribution into consideration. The dynamic performance prediction model takes the gas production mechanism into consideration. With these models, a three-dimensional hydraulic fracturing optimum design software for coalbed methane well is developed, and its practicality and reliability have been proved by ex-ample computation.

Production prediction at ultra-high water cut stage via Recurrent Neural Network

A deep learning method for predicting oil field production at ultra-high water cut stage from the existing oil field production data was presented,and the experimental verification and application effect analysis were carried out.Since the traditional Fully Connected Neural Network(FCNN)is incapable of preserving the correlation of time series data,the Long Short-Term Memory(LSTM)network,which is a kind of Recurrent Neural Network(RNN),was utilized to establish a model for oil field production prediction.By this model,oil field production can be predicted from the relationship between oil production index and its influencing factors and the trend and correlation of oil production over time.Production data of a medium and high permeability sandstone oilfield in China developed by water flooding was used to predict its production at ultra-high water cut stage,and the results were compared with the results from the traditional FCNN and water drive characteristic curves.The LSTM based on deep learning has higher precision,and gives more accurate production prediction for complex time series in oil field production.The LSTM model was used to predict the monthly oil production of another two oil fields.The prediction results are good,which verifies the versatility of the method.

A production prediction method of single well in water flooding oilfield based on integrated temporal convolutional network model

Since the oil production of single well in water flooding reservoir varies greatly and is hard to predict, an oil production prediction method of single well based on temporal convolutional network(TCN) is proposed and verified. This method is started from data processing, the correspondence between water injectors and oil producers is determined according to the influence radius of the water injectors, the influence degree of a water injector on an oil producer in the month concerned is added as a model feature, and a Random Forest(RF) model is built to fill the dynamic data of water flooding. The single well history is divided into 4 stages according to its water cut, that is, low water cut, middle water cut, high water cut and extra-high water cut stages. In each stage, a TCN based prediction model is established, hyperparameters of the model are optimized by the Sparrow Search Algorithm(SSA). Finally, the models of the 4 stages are integrated into one whole-life model of the well for production prediction. The application of this method in Daqing Oilfield, NE China shows that:(1) Compared with conventional data processing methods, the data obtained by this processing method are more close to the actual production, and the data set obtained is more authentic and complete.(2) The TCN model has higher prediction accuracy than other 11 models such as Long Short Term Memory(LSTM).(3) Compared with the conventional full-life-cycle models, the model of integrated stages can significantly reduce the error of production prediction.

Modelling Manure Production in Beef Calves: Development, Evaluation, and Application of a Complete vs. Simplified Prediction Model

There has been increased interest in quantifying the manure production of livestock, primarily driven by public authorities, who aim to evaluate the environmental impact of livestock production, but also at the farm level, to manage manure storage and availability of fertilizer for crop production. Moreover, current manure production estimates from intensively reared beef calves are higher than actual production due to changes in farming systems, advances in animal genetics and feed efficiency. This study aims to redefine and update manure production estimates in intensively reared beef calves to predict manure production as a policy and planning tool, as there are no current models available. A trial was conducted to collect data on manure production during the growing-finishing period (243 d) of 54 Limousine calves (from 346.7 to 674.0 kg live weight, LW). Such data were used to develop two models to predict manure excretion: (1) a complex mechanistic model (CompM), and (2) a simplified empirical model (SimpM). Both models were evaluated against an independent dataset including a total of 4,692 animals on 31 farms and 5 breeds. Results from CompM require interpretation because the model does not output a single value but a range of manure production (minimum, medium and maximum), and would therefore be more suitable for professional use. The SimpM could be considered simple, reliable, and versatile for predicting manure excretion at farm level. SimpM could be refined and improved by including data from other studies on beef cattle with distinct characteristics and management.

The Meteorological Prediction Model of Lemon Production in Anyue County Based on Correlation

Using the meteorological data during 1971- 2013 and lemon growth and yield data during 2003- 2013 in Anyue,the suitability problem of lemon growth and correlation problem between meteorological factors and lemon growth in Anyue area were studied. According to relevance between the selected meteorological factors and yield of lemon,meteorological prediction model of lemon yield was established in Anyue,and the prediction accuracy was higher. The research had certain guiding significance for management work of lemon production in Anyue area.

Subsection and superposition method for reservoir formation damage evaluation of complex-structure wells

Kinds of complex-structure wells can effectively improve production,which are widely used.However,in the process of drilling and completion,complex-structure wells with long drilling cycle and large exposed area of reservoir can lead to the fact that reservoir near wellbore is more vulnerable to the working fluid invasion,resulting in more serious formation damage.In order to quantitatively describe the reservoir formation damage in the construction of complex-structure well,taking the inclined well section as the research object,the coordinate transformation method and conformal transformation method are given according to the flow characteristics of reservoir near wellbore in anisotropic reservoir.Then the local skin factor in orthogonal plane of wellbore is deduced.Considering the un-even distribution of local skin factor along the wellbore,the oscillation decreasing model and empirical equation model of damage zone radius distribution along the wellbore direction are established and then the total skin factor model of the whole well is superimposed to realize the reservoir damage evaluation of complex-structure wells.Combining the skin factor model with the production model,the production of complex-structure wells can be predicted more accurately.The two field application cases show that the accuracy of the model can be more than 90%,which can also fully reflect the invasion characteristics of drilling and completion fluid in any well section of complex-structure wells in anisotropic reservoir,so as to further provide guidance for the scientific establish-ment of reservoir production system.

Shale gas production evaluation framework based on data-driven models

Increasing the production and utilization of shale gas is of great significance for building a clean and low-carbon energy system.Sharp decline of gas production has been widely observed in shale gas reservoirs.How to forecast shale gas production is still challenging due to complex fracture networks,dynamic fracture properties,frac hits,complicated multiphase flow,and multi-scale flow as well as data quality and uncertainty.This work develops an integrated framework for evaluating shale gas well production based on data-driven models.Firstly,a comprehensive dominated-factor system has been established,including geological,drilling,fracturing,and production factors.Data processing and visualization are required to ensure data quality and determine final data set.A shale gas production evaluation model is developed to evaluate shale gas production levels.Finally,the random forest algorithm is used to forecast shale gas production.The prediction accuracy of shale gas production level is higher than 95%based on the shale gas reservoirs in China.Forty-one wells are randomly selected to predict cumulative gas production using the optimal regression model.The proposed shale gas production evaluation frame-work overcomes too many assumptions of analytical or semi-analytical models and avoids huge computation cost and poor generalization for numerical modelling.

A New Elastoplastic 3D Sand Production Model for Fractured Gas Fields

Reservoirs characterized by high temperature,high-pressure,medium high cementation strength,low porosity,and low permeability,in general,are not affected by sand production issues.Since 2009,however,it is known that cases exists where sand is present and may represent a significant technical problem(e.g.,the the Dina II condensate gas field).In the present study,the main factors affecting sand production in this type of reservoir are considered(mechanical properties,stress fields,production system,completion method and gas flow pattern changes during the production process).On this basis,a new liquid-solid coupled porous elasto-plastic 3D sand production model is introduced and validated through comparison with effective sand production data.The related prediction errors are found to be within 15%,which represents the necessary prerequisite for the utilization of such a model for the elaboration of sand prevention measures.

Application of Deep Learning to Production Forecasting in Intelligent Agricultural Product Supply Chain

Production prediction is an important factor influencing the realization of an intelligent agricultural supply chain.In an Internet of Things(IoT)environment,accurate yield prediction is one of the prerequisites for achieving an efficient response in an intelligent agricultural supply chain.As an example,this study applied a conventional prediction method and deep learning prediction model to predict the yield of a characteristic regional fruit(the Shatian pomelo)in a comparative study.The root means square error(RMSE)values of regression analysis,exponential smoothing,grey prediction,grey neural network,support vector regression(SVR),and long short-term memory(LSTM)neural network methods were 53.715,6.707,18.440,1.580,and 1.436,respectively.Among these,the mean square error(MSE)values of the grey neural network,SVR,and LSTM neural network methods were 2.4979,31.652,and 2.0618,respectively;and theirRvalues were 0.99905,0.94,and 0.94501,respectively.The results demonstrated that the RMSE of the deep learning model is generally lower than that of a traditional prediction model,and the prediction results are more accurate.The prediction performance of the grey neural network was shown to be superior to that of SVR,and LSTM neural network,based on the comparison of parameters.

Revolutionizing Tight Reservoir Production: A Novel Dual-Medium Unsteady Seepage Model for Optimizing Volumetrically Fractured Horizontal Wells

This study presents an avant-garde approach for predicting and optimizing production in tight reservoirs,employing a dual-medium unsteady seepage model specifically fashioned for volumetrically fractured horizontal wells.Traditional models often fail to fully capture the complex dynamics associated with these unconventional reservoirs.In a significant departure from these models,our approach incorporates an initiation pressure gradient and a discrete fracture seepage network,providing a more realistic representation of the seepage process.The model also integrates an enhanced fluid-solid interaction,which allows for a more comprehensive understanding of the fluid-structure interactions in the reservoir.This is achieved through the incorporation of improved permeability and stress coupling,leading to more precise predictions of reservoir behavior.The numerical solutions derived from the model are obtained through the sophisticated finite element method,ensuring high accuracy and computational efficiency.To ensure the model’s reliability and accuracy,the outcomes were tested against a real-world case,with results demonstrating strong alignment.A key revelation from the study is the significant difference between uncoupled and fully coupled volumetrically fractured horizontal wells,challenging conventional wisdom in the field.Additionally,the study delves into the effects of stress,fracture length,and fracture number on reservoir production,contributing valuable insights for the design and optimization of tight reservoirs.The findings from this study have the potential to revolutionize the field of tight reservoir prediction and management,offering significant advancements in petroleum engineering.The proposed approach brings forth a more nuanced understanding of tight reservoir systems and opens up new avenues for optimizing reservoir management and production.

Application of New Water Flooding Characteristic Curve in the High Water-Cut Stage of an Oilfield

The oil production predicted by means of the conventional water-drive characteristic curve is typically affected by large deviations with respect to the actual value when the so-called high water-cut stage is entered.In order to solve this problem,a new characteristic relationship between the relative permeability ratio and the average water saturation is proposed.By comparing the outcomes of different matching methods,it is verified that it can well reflect the variation characteristics of the relative permeability ratio curve.Combining the new formula with a reservoir engineering method,two new formulas are derived for the water flooding characteristic curve in the high water-cut stage.Their practicability is verified by using the production data of Mawangmiao and Xijiakou blocks.The results show that the error between the predicted cumulative oil production and production data of the two new water drive characteristic curves is less than the error between the B-type water drive characteristic curve and the other two water drive characteristic curves.It is concluded that the two new characteristic curves can be used to estimate more accurately the recoverable reserves,the final recovery and to estimate the effects of water flooding.

Production performance forecasting method based on multivariate time series and vector autoregressive machine learning model for waterflooding reservoirs

A forecasting method of oil well production based on multivariate time series(MTS)and vector autoregressive(VAR)machine learning model for waterflooding reservoir is proposed,and an example application is carried out.This method first uses MTS analysis to optimize injection and production data on the basis of well pattern analysis.The oil production of different production wells and water injection of injection wells in the well group are regarded as mutually related time series.Then a VAR model is established to mine the linear relationship from MTS data and forecast the oil well production by model fitting.The analysis of history production data of waterflooding reservoirs shows that,compared with history matching results of numerical reservoir simulation,the production forecasting results from the machine learning model are more accurate,and uncertainty analysis can improve the safety of forecasting results.Furthermore,impulse response analysis can evaluate the oil production contribution of the injection well,which can provide theoretical guidance for adjustment of waterflooding development plan.