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.

Optimization of Gas Production from Hydrate-Bearing Sediments with Fluctuation Characteristics

As an important source of low-carbon,clean fossil energy,natural gas hydrate plays an important role in improving the global energy consumption structure.Developing the hydrate industry in the South China Sea is important to achieving‘carbon peak and carbon neutrality’goals as soon as possible.Deep-water areas subjected to the action of long-term stress and tectonic movement have developed complex and volatile terrains,and as such,the morphologies of hydrate-bearing sediments(HBSs)fluctuate correspondingly.The key to numerically simulating HBS morphologies is the establishment of the conceptual model,which represents the objective and real description of the actual geological body.However,current numerical simulation models have characterized HBSs into horizontal strata without considering the fluctuation characteristics.Simply representing the HBS as a horizontal element reduces simulation accuracy.Therefore,the commonly used horizontal HBS model and a model considering the HBS’s fluctuation characteristics with the data of the SH2 site in the Shenhu Sea area were first constructed in this paper.Then,their production behaviors were compared,and the huge impact of the fluctuation characteristics on HBS production was determined.On this basis,the key parameters affecting the depressurization production of the fluctuating HBSs were studied and optimized.The research results show that the fluctuation characteristics have an obvious influence on the hydrate production of HBSs by affecting their temperatures and pressure distributions,as well as the transmission of the pressure drop and methane gas discharge.Furthermore,the results show that the gas productivity of fluctuating HBSs was about 5%less than that of horizontal HBSs.By optimizing the depressurization amplitude,well length,and layout location of vertical wells,the productivity of fluctuating HBSs increased by about 56.6%.

Production optimization in a fractured carbonate reservoir with high producing GOR

The Gas-Oil Ratio(GOR)is a crucial production parameter in oil reservoirs.An increase in GOR results in higher gas production and lower oil production,potentially leading to well shut-ins due to economic infeasibility.This study focuses on a real fractured oil field that requires urgent production operations to reduce the producing GOR.In this study,the static model for the field was developed using commercial software,involving steps such as data collection,fault modeling,meshing,and statistical analysis to prepare for dynamic simulation.The dynamic model incorporates geometry,gridding,and rock properties from the static model,utilizing a dual-porosity approach for the naturally fractured reservoir and the Peng-Robinson equation for fluid phase behavior.Initial reservoir conditions,production history,and rock-fluid interactions were defined,with relative permeability curves indicating a water-wet reservoir and low critical gas saturation affecting the GOR.To better understand the relationship between reservoir and production parameters,a detailed sensitivity analysis was performed using the Response Surface Methodology(RSM).Following the sensitivity analysis,a history matching process was conducted using the Designed Exploration and Controlled Evolution(DECE)optimizer to validate the model for future forecasts.Six operational scenarios were defined to decrease the production GOR and enhance final recovery from the field.The results indicate that the water injection scenario is effective in preventing the GOR increase by maintaining reservoir pressure,thereby sustaining production over a longer period.This scenario also improves oil recovery by approximately 6%compared to the base case.Finally,optimization was carried out using the DECE optimizer for each scenario to fine-tune the operational parameters.The goal was to maximize oil revenue for each scenario during the optimization process.This study stands out as one of the few that provides a comprehensive analysis of production behavior and development planning for a real fractured reservoir with high producing GOR.

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.

Multi-Objective Image Optimization of Product Appearance Based on Improved NSGA-II

A second-generation fast Non-dominated Sorting Genetic Algorithm product shape multi-objective imagery optimization model based on degradation(DNSGA-II)strategy is proposed to make the product appearance optimization scheme meet the complex emotional needs of users for the product.First,the semantic differential method and K-Means cluster analysis are applied to extract the multi-objective imagery of users;then,the product multidimensional scale analysis is applied to classify the research objects,and again the reference samples are screened by the semantic differentialmethod,and the samples are parametrized in two dimensions by using elliptic Fourier analysis;finally,the fuzzy dynamic evaluation function is used as the objective function of the algorithm,and the coordinates of key points of product contours Finally,with the fuzzy dynamic evaluation function as the objective function of the algorithm and the coordinates of key points of the product profile as the decision variables,the optimal product profile solution set is solved by DNSGA-II.The validity of the model is verified by taking the optimization of the shape scheme of the hospital connection site as an example.For comparison with DNSGA-II,other multi-objective optimization algorithms are also presented.To evaluate the performance of each algorithm,the performance evaluation index values of the five multi-objective optimization algorithms are calculated in this paper.The results show that DNSGA-II is superior in improving individual diversity and has better overall performance.

Intelligent production optimization method for a low pressure and low productivity shale gas well

Shale gas wells frequently suffer from liquid loading and insufficient formation pressure in the late stage of production.To address this issue,an intelligent production optimization method for low pressure and low productivity shale gas well is proposed.Based on the artificial intelligence algorithms,this method realizes automatic production and monitoring of gas well.The method can forecast the production performance of a single well by using the long short-term memory neural network and then guide gas well production accordingly,to fulfill liquid loading warning and automatic intermittent production.Combined with adjustable nozzle,the method can keep production and pressure of gas wells stable automatically,extend normal production time of shale gas wells,enhance automatic level of well sites,and reach the goal of refined production management by making production regime for each well.Field tests show that wells with production regime optimized by this method increased 15%in estimated ultimate reserve(EUR).Compared with the development mode of drainage after depletion recovery,this method is more economical and can increase and stabilize production effectively,so it has a bright application prospect.

Efficient production optimization for naturally fractured reservoir using EDFM

Naturally fractured reservoirs make important contributions to global oil and gas reserves and production.The modeling and simulation of naturally fractured reservoirs are different from conventional reservoirs as the existence of natural fractures.To address the development optimization problem of naturally fractured reservoirs,we propose an optimization workflow by coupling the optimization methods with the embedded discrete fracture model(EDFM).Firstly,the effective and superior performance of the workflow is verified based on the conceptual model.The stochastic simplex approximate gradient(StoSAG)algorithm,the ensemble optimization(EnOpt)algorithm,and the particle swarm optimization(PSO)algorithm are implemented for the production optimization of naturally fractured reservoirs based on the improved versions of the Egg model and the PUNQ-S3 model.The results of the two cases demonstrate the effectiveness of this optimization workflow by finding the optimal well controls which yield the maximum net present value(NPV).Compared to the initial well control guess,the final NPV obtained from the production optimization of fractured reservoirs based on all three optimization algorithms is significantly enhanced.Compared with the optimization results of the PSO algorithm,StoSAG and EnOpt have significant advantages in terms of final NPV and computational efficiency.The results also show that fractures have a significant impact on reservoir production.The economic efficiency of fractured reservoir development can be significantly improved by the optimization workflow.

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.

Simulation and optimization of scrap wagon dismantling system based on Plant Simulation

Based on the existing plant layout and process flow,a simulation analysis was conducted using the Plant Simulation platform with the utilization efficiency of each station and production capacity of the dismantling system as indicators.A problem with long-term suspension in the disassembly process was determined.Based on the two optimization directions of increasing material transportation equipment and expanding the buffer capacity,a cost-oriented optimization model is established.A genetic algorithm and model simulation were used to solve the model.An optimization scheme that satisfies the production needs and has the lowest cost is proposed.The results show that the optimized dismantling system solves the suspended work problem at the dismantling station and a significant improvement in productivity and station utilization efficiency compared with the previous system.

Flow Dynamics during the Hydrocarbon Exploitation for Prevention and Management of Water Venues in Oil Field: A Study Case of Crystal Field in Badila/Chad

The southern part of the Lake Chad basin is under the gas and oil petroleum industry due to its hydrocarbon potential for about twenty years. This project stands out as the main challenges of the hydrocarbon production and the management of fluxes particularly the groundwater venues. A comprehensive study is thus conducted to develop a dynamic and analytic model for diagnosing the production performances with a particular view on the management of groundwater venues. The three main concerned reservoirs subdivided on subunits evidence their proper characteristics. The porous media, their densities, the internal flows and the water injection techniques such as water flooding were thus adopted. The oil viscosity variability within the reservoirs creates different levels of mobility between water and oil, highlighting the challenges of water management. The material balance model and the behavior of the well analysis were taken in consideration within the identified aquifer, emphasizing the importance of keeping the pressure through injection. The control of water productions, the management of the reservoir, the well strategical position and the specific completions lead to the model functioning. In addition, the CO log and the Pulsed Neutron indicate their limitations as a result of the water salinity and the porosity of the aquifer. The management of groundwater venues at Badila requires various approaches throughout the lifetime of the Crystal field such as the data acquisition and remediation actions and prevention, under a permanent monitoring of the dynamic fluxes in the reservoirs.

Agricultural Production Structure Optimization: A Case Study of Major Grain Producing Areas, China

A large number of mathematical models were developed for supporting agricultural production structure optimization decisions; however, few of them can address various uncertainties existing in many factors （e.g., eco-social benefit maximization, food security, employment stability and ecosystem balance）. In this study, an interval-probabilistic agricultural production structure optimization model （IPAPSOM） is formulated for tackling uncertainty presented as discrete intervals and/or probability distribution. The developed model improves upon the existing probabilistic programming and inexact optimization approaches. The IPAPSOM considers not only food security policy constraints, but also involves rural households’income increase and eco-environmental conversation, which can effectively reflect various interrelations among different aspects in an agricultural production structure optimization system. Moreover, it can also help examine the reliability of satisfying （or risk of violating） system constraints under uncertainty. The model is applied to a real case of long-term agricultural production structure optimization in Dancheng County, which is located in Henan Province of Central China as one of the major grain producing areas. Interval solutions associated with different risk levels of constraint violation are obtained. The results are useful for generating a range of decision alternatives under various system benefit conditions, and thus helping decision makers to identify the desired agricultural production structure optimization strategy under uncertainty.

MODIFIED GENETIC ALGORITHM APPLIED TO SOLVE PRODUCT FAMILY OPTIMIZATION PROBLEM

The product family design problem solved by evolutionary algorithms is discussed. A successful product family design method should achieve an optimal tradeoff among a set of competing objectives, which involves maximizing commonality across the family of products and optimizing the performances of each product in the family. A 2-level chromosome structured genetic algorithm （2LCGA） is proposed to solve this class of problems and its performance is analyzed in comparing its results with those obtained with other methods. By interpreting the chromosome as a 2-level linear structure, the variable commonality genetic algorithm （GA） is constructed to vary the amount of platform commonality and automatically searches across varying levels of commonality for the platform while trying to resolve the tradeoff between commonality and individual product performance within the product family during optimization process. By incorporating a commonality assessing index to the problem formulation, the 2LCGA optimize the product platform and its corresponding family of products in a single stage, which can yield improvements in the overall performance of the product family compared with two-stage approaches （the first stage involves determining the best settings for the platform variables and values of unique variables are found for each product in the second stage）. The scope of the algorithm is also expanded by introducing a classification mechanism to allow mul- tiple platforms to be considered during product family optimization, offering opportunities for superior overall design by more efficacious tradeoffs between commonality and performance. The effectiveness of 2LCGA is demonstrated through the design of a family of universal electric motors and comparison against previous results.

Availability-based simulation and optimization modeling framework for open-pit mine truck allocation under dynamic constraints

We present a novel system productivity simulation and optimization modeling framework in which equipment availability is a variable in the expected productivity function of the system. The framework is used for allocating trucks by route according to their operating performances in a truck-shovel system of an open-pit mine, so as to maximize the overall productivity of the fleet. We implement the framework in an originally designed and specifically developed simulator-optimizer software tool. We make an application on a real open-pit mine case study taking into account the stochasticity of the equipment behavior and environment. The total system production values obtained with and without considering the equipment reliability, availability and maintainability (RAM) characteristics are compared. We show that by taking into account the truck and shovel RAM aspects, we can maximize the total production of the system and obtain specific information on the production availability and productivity of its components.

Fuzzy linear model for production optimization of mining systems with multiple entities

Planning and production optimization within multiple mines or several work sites （entities） mining systems by using fuzzy linear programming （LP） was studied. LP is the most commonly used operations research methods in mining engineering. After the introductory review of properties and limitations of applying LP, short reviews of the general settings of deterministic and fuzzy LP models are presented. With the purpose of comparative analysis, the application of both LP models is presented using the example of the Bauxite Basin Niksic with five mines. After the assessment, LP is an efficient mathematical modeling tool in production planning and solving many other single-criteria optimization problems of mining engineering. After the comparison of advantages and deficiencies of both deterministic and fuzzy LP models, the conclusion presents benefits of the fuzzy LP model but is also stating that seeking the optimal plan of production means to accomplish the overall analysis that will encompass the LP model approaches.

A novel 3-layer mixed cultural evolutionary optimization framework for optimal operation of syngas production in a Texaco coal-water slurry gasifier

Optimizing operational parameters for syngas production of Texaco coal-water slurry gasifier studied in this paper is a complicated nonlinear constrained problem concerning 3 BP(Error Back Propagation) neural networks. To solve this model, a new 3-layer cultural evolving algorithm framework which has a population space, a medium space and a belief space is firstly conceived. Standard differential evolution algorithm(DE), genetic algorithm(GA), and particle swarm optimization algorithm(PSO) are embedded in this framework to build 3-layer mixed cultural DE/GA/PSO(3LM-CDE, 3LM-CGA, and 3LM-CPSO) algorithms. The accuracy and efficiency of the proposed hybrid algorithms are firstly tested in 20 benchmark nonlinear constrained functions. Then, the operational optimization model for syngas production in a Texaco coal-water slurry gasifier of a real-world chemical plant is solved effectively. The simulation results are encouraging that the 3-layer cultural algorithm evolving framework suggests ways in which the performance of DE, GA, PSO and other population-based evolutionary algorithms(EAs) can be improved,and the optimal operational parameters based on 3LM-CDE algorithm of the syngas production in the Texaco coalwater slurry gasifier shows outstanding computing results than actual industry use and other algorithms.

Rapid diagnostic method for transplutonium isotope production in high flux reactors

Transplutonium isotopes are scarce and need to be produced by irradiation in high flux reactors.However,their production is inefficient,and optimization studies are necessary.This study analyzes the physical nature of transplutonium isotope produc-tion using ^(252)Cf,^(244)Cm,^(242)Cm,and ^(238)Pu as examples.Traditional methods based on the Monte Carlo burnup calculation have the limitations of many calculations and cannot analyze the individual energy intervals in detail;thus,they cannot sup-port the refined evaluation,screening,and optimization of the irradiation schemes.After understanding the physical nature and simplifying the complexity of the production process,we propose a rapid diagnostic method for evaluating radiation schemes based on the concepts“single energy interval value(SEIV)”and“energy spectrum total value(ESTV)”.The rapid diagnostic method not only avoids tedious burnup calculations,but also provides a direction for optimization.The optimal irradiation schemes for producing ^(252)Cf,^(244)Cm,^(242)Cm,and ^(238)Pu are determined based on a rapid diagnostic method.Optimal irradiation schemes can significantly improve production efficiency.Compared with the initial scheme,the optimal scheme improved the production efficiency of ^(238)Pu by 7.41 times;^(242)Cm,11.98 times;^(244)Cm,65.20 times;and ^(252)Cf,15.08 times.Thus,a refined analysis of transplutonium isotope production is conducted and provides a theoretical basis for improving production efficiency.

A new method to evaluate integrated production system capacity in oil fields

To improve the design and management of an integrated production system(IPS),a set of mathematical models and workflows are developed for evaluating the capacity of an IPS at steady-state conditions.Combining the conservation laws with applicable multiphase fluid and choke models,these mathematical models are solved to characterize the hydraulics of an integrated system of reservoir,wells,chokes,flowlines,and separator at steady state.The controllable variables such as well count,choke size and separator pressure are adjusted to optimize the performance of the IPs at a specific time.It is found that increasing the well count can increase the bulk flow rate of the production network,but too many wells may increase the manifold pressure,leading to decline of single-well production.Increasing the choke size can improve the capacity of the IPs.The production of the IPs is negatively correlated with the separator pressure.With increasing separator pressure and decreasing choke size,the increment of total fluid production(the capacity of IPS)induced by increasing well count decreases.Validation tests with field examples show a maximum absolute deviation is 1.5%,demonstrating the robustness and validity of the proposed mathematical models and workflows.

Assessment of Climate for Agricultural Suitability and Optimal Allocation of Agricultural Production in the Guanzhong Region, Shaanxi Province

Based on spatial climatic data of agriculture and the experiment data, the models of agro-ecological assessment of climate for agricultural suitability in this study were developed using the fuzzy mathematical method. Three coefficients, in- cluding the resource coefficient （Cr）, the efficiency coefficient （Ce）, and the utility co- efficient （K）, were used in the models, which were calculated based on temperature, moisture, and sunshine duration data of Guanzhong region, Shaanxi Province. The results indicated that resource coefficient was higher in west of the region than that in east, and higher in south （especially in the Central Shaanxi Plain） than that in the Weibei plateau. The value of Cr changed from 6.5 to 9.2 from north to plain area. Spatial change of efficiency coefficient was obvious, lower in the northeast than in the central plain, and the value of Ce changed from 2.3 to 6.5 from the northeast to the central plain. As for utility coefficient, it was lower in northeastern part of the Weibei plateau and in southern mountain areas than that in the central plain, showing significant latitudinal zonality. Furthermore, the value of K increased from 0.35 to 0.78 from northeast to the central plain, and decreased from 0.78 to 0.53 from the central plain to southern mountain areas. These indicated that climate resource in the central plain region was more abundant and potential, compared with other regions. GuanZhong region was classified into three larger agricultural zones and three small independent zones, according to agro-ecological assessment. Light, heat and water resources should be made use of in an efficient way in spatial allo- cation of agricultural production. For example, water facilities should also be im- proved in Weibei plateau region where highly-qualified fruit should be enhanced and fruit processing industrial chain should be shaped. Large-scale production area of wheat should be increased in central irrigation region and more vegetable bases should be developed around large and medium-scale cities. Thanks for outstanding water conservation function, the three-dimensional agriculture including medicine and other sideline production should be developed in Qinling Mountains and the special- ized commercial agriculture should be accelerated in independent small zones, ac- cording to local conditions. In the research, different crop varieties were developed in corresponding regions as per current eco-climatic conditions.

Optimized Strategy for Layout of Crop Production Areas in Hunan Province

The optimized strategy made a comprehensive consideration of resources, technology, market orientation, production scale, industry basis and layout based on the principle of crop security and farmers’ income increasing, and determined the general planning on layout and structure optimization of future crop production ar-eas, with present crop production, market outlook, future industry development, con-cluding crop production characteristics of the 4 crop regions, and proposing function orientation and highlights.

A review of closed-loop reservoir management

The closed-loop reservoir management technique enables a dynamic and real-time optimal production schedule under the existing reservoir conditions to be achieved by adjusting the injection and production strategies. This is one of the most effective ways to exploit limited oil reserves more economically and efficiently. There are two steps in closed-loop reservoir management： automatic history matching and reservoir production opti- mization. Both of the steps are large-scale complicated optimization problems. This paper gives a general review of the two basic techniques in closed-loop reservoir man- agement; summarizes the applications of gradient-based algorithms, gradient-free algorithms, and artificial intelligence algorithms; analyzes the characteristics and application conditions of these optimization methods; and finally discusses the emphases and directions of future research on both automatic history matching and reservoir production optimization.