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.

Hybrid data-driven framework for shale gas production performance analysis via game theory, machine learning, and optimization approaches

A comprehensive and precise analysis of shale gas production performance is crucial for evaluating resource potential,designing a field development plan,and making investment decisions.However,quantitative analysis can be challenging because production performance is dominated by the complex interaction among a series of geological and engineering factors.In fact,each factor can be viewed as a player who makes cooperative contributions to the production payoff within the constraints of physical laws and models.Inspired by the idea,we propose a hybrid data-driven analysis framework in this study,where the contributions of dominant factors are quantitatively evaluated,the productions are precisely forecasted,and the development optimization suggestions are comprehensively generated.More specifically,game theory and machine learning models are coupled to determine the dominating geological and engineering factors.The Shapley value with definite physical meaning is employed to quantitatively measure the effects of individual factors.A multi-model-fused stacked model is trained for production forecast,which provides the basis for derivative-free optimization algorithms to optimize the development plan.The complete workflow is validated with actual production data collected from the Fuling shale gas field,Sichuan Basin,China.The validation results show that the proposed procedure can draw rigorous conclusions with quantified evidence and thereby provide specific and reliable suggestions for development plan optimization.Comparing with traditional and experience-based approaches,the hybrid data-driven procedure is advanced in terms of both efficiency and accuracy.

Evolutionary-assisted reinforcement learning for reservoir real-time production optimization under uncertainty

Production optimization has gained increasing attention from the smart oilfield community because it can increase economic benefits and oil recovery substantially.While existing methods could produce high-optimality results,they cannot be applied to real-time optimization for large-scale reservoirs due to high computational demands.In addition,most methods generally assume that the reservoir model is deterministic and ignore the uncertainty of the subsurface environment,making the obtained scheme unreliable for practical deployment.In this work,an efficient and robust method,namely evolutionaryassisted reinforcement learning(EARL),is proposed to achieve real-time production optimization under uncertainty.Specifically,the production optimization problem is modeled as a Markov decision process in which a reinforcement learning agent interacts with the reservoir simulator to train a control policy that maximizes the specified goals.To deal with the problems of brittle convergence properties and lack of efficient exploration strategies of reinforcement learning approaches,a population-based evolutionary algorithm is introduced to assist the training of agents,which provides diverse exploration experiences and promotes stability and robustness due to its inherent redundancy.Compared with prior methods that only optimize a solution for a particular scenario,the proposed approach trains a policy that can adapt to uncertain environments and make real-time decisions to cope with unknown changes.The trained policy,represented by a deep convolutional neural network,can adaptively adjust the well controls based on different reservoir states.Simulation results on two reservoir models show that the proposed approach not only outperforms the RL and EA methods in terms of optimization efficiency but also has strong robustness and real-time decision capacity.

Data-driven production optimization using particle swarm algorithm based on the ensemble-learning proxy model

Production optimization is of significance for carbonate reservoirs,directly affecting the sustainability and profitability of reservoir development.Traditional physics-based numerical simulations suffer from insufficient calculation accuracy and excessive time consumption when performing production optimization.We establish an ensemble proxy-model-assisted optimization framework combining the Bayesian random forest(BRF)with the particle swarm optimization algorithm(PSO).The BRF method is implemented to construct a proxy model of the injectioneproduction system that can accurately predict the dynamic parameters of producers based on injection data and production measures.With the help of proxy model,PSO is applied to search the optimal injection pattern integrating Pareto front analysis.After experimental testing,the proxy model not only boasts higher prediction accuracy compared to deep learning,but it also requires 8 times less time for training.In addition,the injection mode adjusted by the PSO algorithm can effectively reduce the gaseoil ratio and increase the oil production by more than 10% for carbonate reservoirs.The proposed proxy-model-assisted optimization protocol brings new perspectives on the multi-objective optimization problems in the petroleum industry,which can provide more options for the project decision-makers to balance the oil production and the gaseoil ratio considering physical and operational constraints.

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.

Better use of experience from other reservoirs for accurate production forecasting by learn-to-learn method

To assess whether a development strategy will be profitable enough,production forecasting is a crucial and difficult step in the process.The development history of other reservoirs in the same class tends to be studied to make predictions accurate.However,the permeability field,well patterns,and development regime must all be similar for two reservoirs to be considered in the same class.This results in very few available experiences from other reservoirs even though there is a lot of historical information on numerous reservoirs because it is difficult to find such similar reservoirs.This paper proposes a learn-to-learn method,which can better utilize a vast amount of historical data from various reservoirs.Intuitively,the proposed method first learns how to learn samples before directly learning rules in samples.Technically,by utilizing gradients from networks with independent parameters and copied structure in each class of reservoirs,the proposed network obtains the optimal shared initial parameters which are regarded as transferable information across different classes.Based on that,the network is able to predict future production indices for the target reservoir by only training with very limited samples collected from reservoirs in the same class.Two cases further demonstrate its superiority in accuracy to other widely-used network methods.

Multi-surrogate framework with an adaptive selection mechanism for production optimization

Data-driven surrogate models that assist with efficient evolutionary algorithms to find the optimal development scheme have been widely used to solve reservoir production optimization problems.However,existing research suggests that the effectiveness of a surrogate model can vary depending on the complexity of the design problem.A surrogate model that has demonstrated success in one scenario may not perform as well in others.In the absence of prior knowledge,finding a promising surrogate model that performs well for an unknown reservoir is challenging.Moreover,the optimization process often relies on a single evolutionary algorithm,which can yield varying results across different cases.To address these limitations,this paper introduces a novel approach called the multi-surrogate framework with an adaptive selection mechanism(MSFASM)to tackle production optimization problems.MSFASM consists of two stages.In the first stage,a reduced-dimensional broad learning system(BLS)is used to adaptively select the evolutionary algorithm with the best performance during the current optimization period.In the second stage,the multi-objective algorithm,non-dominated sorting genetic algorithm II(NSGA-II),is used as an optimizer to find a set of Pareto solutions with good performance on multiple surrogate models.A novel optimal point criterion is utilized in this stage to select the Pareto solutions,thereby obtaining the desired development schemes without increasing the computational load of the numerical simulator.The two stages are combined using sequential transfer learning.From the two most important perspectives of an evolutionary algorithm and a surrogate model,the proposed method improves adaptability to optimization problems of various reservoir types.To verify the effectiveness of the proposed method,four 100-dimensional benchmark functions and two reservoir models are tested,and the results are compared with those obtained by six other surrogate-model-based methods.The results demonstrate that our approach can obtain the maximum net present value(NPV)of the target production optimization problems.

Using deep neural networks coupled with principal component analysis for ore production forecasting at open-pit mines

Ore production is usually affected by multiple influencing inputs at open-pit mines.Nevertheless,the complex nonlinear relationships between these inputs and ore production remain unclear.This becomes even more challenging when training data(e.g.truck haulage information and weather conditions)are massive.In machine learning(ML)algorithms,deep neural network(DNN)is a superior method for processing nonlinear and massive data by adjusting the amount of neurons and hidden layers.This study adopted DNN to forecast ore production using truck haulage information and weather conditions at open-pit mines as training data.Before the prediction models were built,principal component analysis(PCA)was employed to reduce the data dimensionality and eliminate the multicollinearity among highly correlated input variables.To verify the superiority of DNN,three ANNs containing only one hidden layer and six traditional ML models were established as benchmark models.The DNN model with multiple hidden layers performed better than the ANN models with a single hidden layer.The DNN model outperformed the extensively applied benchmark models in predicting ore production.This can provide engineers and researchers with an accurate method to forecast ore production,which helps make sound budgetary decisions and mine planning at open-pit mines.

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.

Optimal production lot sizing model in a supply chain with periodically fixed demand considering learning effect

This paper presents an optimal production model for manufacturer in a supply chain with a fixed demand at a fixed interval with respect to the learning effect on production capacity. An algorithm is employed to find the optimal delay time for production and production time sequentially. It is found that the optimal delay time for production and the production time are not static, but dynamic and variant with time. It is important for a manufacturer to schedule the production so as to prevent facilities and workers from idling.

Deep-Learning-Based Production Decline Curve Analysis in the Gas Reservoir through Sequence Learning Models

Production performance prediction of tight gas reservoirs is crucial to the estimation of ultimate recovery,which has an important impact on gas field development planning and economic evaluation.Owing to the model’s simplicity,the decline curve analysis method has been widely used to predict production performance.The advancement of deep-learning methods provides an intelligent way of analyzing production performance in tight gas reservoirs.In this paper,a sequence learning method to improve the accuracy and efficiency of tight gas production forecasting is proposed.The sequence learning methods used in production performance analysis herein include the recurrent neural network(RNN),long short-term memory(LSTM)neural network,and gated recurrent unit(GRU)neural network,and their performance in the tight gas reservoir production prediction is investigated and compared.To further improve the performance of the sequence learning method,the hyperparameters in the sequence learning methods are optimized through a particle swarm optimization algorithm,which can greatly simplify the optimization process of the neural network model in an automated manner.Results show that the optimized GRU and RNN models have more compact neural network structures than the LSTM model and that the GRU is more efficiently trained.The predictive performance of LSTM and GRU is similar,and both are better than the RNN and the decline curve analysis model and thus can be used to predict tight gas production.

Teaching Reform Path of Architecture Specialty under the Collaborative Education of “Production, Teaching and Research” and Its Application Effect

To break through the development bottleneck of architecture specialty,it should think about how to realize the high-quality cultivation of talents through the adjustment of teaching system.Reform path of architecture under the collaborative education of "production,teaching and research" and its application effect are systemically studied.By analyzing the relationship among production,teaching and scientific research,the essence of integration and mutual assistance among the three is analyzed.A series of teaching reforms are carried out on the reform principles of teaching content,setting principles of practice curriculum and implementation paths of practice curriculum in the theoretical and practical courses of architecture specialty.The purpose is to make the achievements of scientific research and production contribute to the development of teaching,so that the three sections of "production,teaching and research" form a benign interaction,mutual assistance and common promotion.This teaching system under the collaborative education of "production,teaching and research" is of positive significance to the cultivation of architectural talents’ innovative ability,practical ability and comprehensive ability.

Cultivation Mode of Innovative Talents in Architecture under the Cooperation of Production, Teaching and Research

In current training of architectural talents, there are a series of problems, such as separation of teaching, scientific research and production, rigid teaching mode and lack of practice and innovative ability, which make it hard to achieve the goal of training the idealized innovative talents. In order to solve these problems and the confusion of architecture in the cultivation of innovative talents, a production-teachingresearch integrated iterative platform was built at the levels of teacher and student. In the platform, the 32 modules involving teaching, researching and producing can be linked according to the logical relationship and teaching needs, thus helping the development of teaching by scientific research and production achievements. The teaching system under this kind of mutual assistance has positive significance to the training of innovative talents in architecture.

A Comparison of Chinese, Japanese, and Korean shipyard production technology

This paper compares Chinese, Korean, and Japanese shipyard production technology. Development in the world shipbuilding over recent years has influenced focus areas related to shipyard manufacturing technologies and product performance. Software systems, information technology, production technology, and local challenges of shipyards are compared with shipbuilding outputs among these three countries. Various technologies developments, shipyard production and the problems in Chinese, Japanese, and Korean shipyards are discussed respectively. Finally, future areas of research are pointed out.

Theoretical Study on Quantitative Characterization of Interlayer Interference in Multi-Layer Commingled Production

X oilfield is a typical layered reservoir with a large vertical span and many oil-bearing formations. There are significant differences in reservoir types and fluid properties among various formations. The interlayer interference is severe in the development process. At present, the interlayer interference research based on dynamic monitoring data cannot meet development adjustment needs. Combined with the field test results, through the indoor physical simulation experiment method, dynamic inversion method, and reservoir engineering method, this paper analyzes the main control factors and interference mechanism of interlayer interference, studies the variation law of interference coefficient, improves and forms the quantitative characteristic Theory of interlayer interference in multi-layer commingled production, and provides theoretical guidance for the total adjustment of the middle strata division in the oilfield.

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.

Changeover Cost in Garment Production

With the increase of style variation, decrease of lot size and shortening of lead-time, production planning becomes a big problem in clothing industry. It was found that changeover cost is one of key factors in garment production. In this paper, based on time measurement and the data collected, considering the relevent elements such as previous experience, lot size, number of lines,the effect of changeover cost on garment production is calculated and analyzed. Then some suggestions are put forward for manufacturers to balance their production planning.

Research Progress of Application of Microbial Inoculants in Agricultural Production

Microbial inoculants have received increasing attention in strengthening plant biological barriers,antagonizing and inhibiting harmful microorganisms,and ensuring the safe production of agricultural products.This paper summarized the research status of agricultural microbial inoculants,the application of microbial inoculants in agriculture,and the trends and prospects of agricultural microbial research.

Artificial neural network based production forecasting for a hydrocarbon reservoir under water injection

As the conventional prediction methods for production of waterflooding reservoirs have some drawbacks, a production forecasting model based on artificial neural network was proposed, the simulation process by this method was presented, and some examples were illustrated. A workflow that involves a physics-based extraction of features was proposed for fluid production forecasting to improve the prediction effect. The Bayesian regularization algorithm was selected as the training algorithm of the model. This algorithm, although taking longer time, can better generalize oil, gas and water production data sets. The model was evaluated by calculating mean square error and determination coefficient, drawing error distribution histogram and the cross-plot between simulation data and verification data etc. The model structure was trained, validated and tested with 90% of the historical data, and blindly evaluated using the remaining. The predictive model consumes minimal information and computational cost and is capable of predicting fluid production rate with a coefficient of determination of more than 0.9, which has the simulation results consistent with the practical data.

Studies on Production Planning of Dispersion Type U3Si2-Al Fuel in Plate-Type Fuel Elements for Nuclear Research Reactors

Several fuel plants that supply nuclear research reactors need to increase their production capacity in order to meet the growing demand for this kind of nuclear fuel. After the enlargement of the production capacity of such plants, there will be the need of managing the new production level. That level is usually the industrial one, which poses challenges to the managerial staff. Such challenges come from the fact that several of those plants operate today on a laboratorial basis and do not carry inventory. The change to the industrial production pace asks for new actions regarding planning and control. The production process based on the hydrolysis of UF6 is not a frequent production route for nuclear fuel. Production planning and control of the industrial level of fuel production on that production route is a new field of studies. The approach of the paper consists in the creation of a mathematical linear model for minimization of costs. We also carried out a sensitivity analysis of the model. The results help in minimizing costs in different production schemes and show the need of inventory. The mathematical model is dynamic, so that it issues better results if performed monthly. The management team will therefore have a clearer view of the costs and of the new, necessary production and inventory levels.