Global and Local Surrogate-Model-Assisted Differential Evolution for Waterflooding Production Optimization

SPE Journal ◽  
2019 ◽  
Vol 25 (01) ◽  
pp. 105-118 ◽  
Author(s):  
Guodong Chen ◽  
Kai Zhang ◽  
Liming Zhang ◽  
Xiaoming Xue ◽  
Dezhuang Ji ◽  
...  
Keyword(s):  
Differential Evolution ◽  
Surrogate Model ◽  
Net Present Value ◽  
Channel Model ◽  
Optimization Problems ◽  
Optimization Methods ◽  
Production Optimization ◽  
Optimization Process ◽  
Promising Area ◽  
Global And Local

Summary Surrogate models, which have become a popular approach to oil-reservoir production-optimization problems, use a computationally inexpensive approximation function to replace the computationally expensive objective function computed by a numerical simulator. In this paper, a new optimization algorithm called global and local surrogate-model-assisted differential evolution (GLSADE) is introduced for waterflooding production-optimization problems. The proposed method consists of two parts: (1) a global surrogate-model-assisted differential-evolution (DE) part, in which DE is used to generate multiple offspring, and (2) a local surrogate-model-assisted DE part, in which DE is used to search for the optimum of the surrogate. The cooperation between global optimization and local search helps the production-optimization process become more efficient and more effective. Compared with the conventional one-shot surrogate-based approach, the developed method iteratively selects data points to enhance the accuracy of the promising area of the surrogate model, which can substantially improve the optimization process. To the best of our knowledge, the proposed method uses a state-of-the-art surrogate framework for production-optimization problems. The approach is tested on two 100-dimensional benchmark functions, a three-channel model, and the egg model. The results show that the proposed method can achieve higher net present value (NPV) and better convergence speed in comparison with the traditional evolutionary algorithm and other surrogate-assisted optimization methods for production-optimization problems.

A Classification-Based Surrogate-Assisted Multiobjective Evolutionary Algorithm for Production Optimization under Geological Uncertainty

SPE Journal ◽  
2020 ◽  
Vol 25 (05) ◽  
pp. 2450-2469
Author(s):  
Mengjie Zhao ◽  
Kai Zhang ◽  
Guodong Chen ◽  
Xinggang Zhao ◽  
Jun Yao ◽  
...  
Keyword(s):  
Evolutionary Algorithm ◽  
Net Present Value ◽  
Optimization Problems ◽  
Optimization Methods ◽  
Production Optimization ◽  
Benchmark Problems ◽  
Geological Uncertainty ◽  
Rbf Network ◽  
Multiobjective Evolutionary Algorithm ◽  
Computationally Expensive

Summary Multiobjective optimization (MOO) is a popular procedure for waterflooding optimization under geological uncertainty that maximizes the expectation of net present value (NPV) over all possible uncertainty models and minimizes the variance simultaneously. However, the optimization process involves a large number of decision variables, and the problem is computationally expensive. Surrogate-assisted evolutionary algorithms (SAEAs), which have proved to be an effective way to solve expensive problems, design computationally inexpensive functions to approximate each objective function. On the basis of characterization, we have designed an efficient multiobjective evolutionary algorithm (MOEA) to effectively deal with computationally expensive simulation-based optimization problems. The uniqueness of this algorithm is that it incorporates a Pareto-rank-learning scheme with surrogate-assisted infill criterion. The first is to introduce a multiclass error-correcting output codes (ECOC) model that directly predicts the dominance relationship between candidate solutions and prescreens, and the second is to train a radial-basis function (RBF) network that predicts the objective functions of prescreened solutions to calculate the hypervolume (HV) improvement that maintains convergence and diversity. Compared with typical surrogate-based methods, the developed method provides a classifier first that can enhance the accuracy in high dimensions and reduce computational complexity. To the best of our knowledge, the proposed method compares with state-of-the-art surrogate frameworks for multiobjective production-optimization problems. In this paper, the proposed approach is applied to two 200D benchmark problems and two synthetic reservoir models. The results show that the proposed method can achieve more comprehensive and efficient reservoir management (RM) with a higher convergence speed compared with traditional MOEAs and surrogate-assisted optimization methods.

scholarly journals APPLICATION OF GLOBAL OPTIMIZATION TO PREDICT STRAINS IN RC COMPRESSED COLUMNS

2021 ◽  
Vol 61 (1) ◽  
pp. 242-252
Author(s):  
Marek Lechman ◽  
Andrzej Stachurski
Keyword(s):  
Least Squares Method ◽  
Optimization Problems ◽  
Particle Swarm ◽  
Trust Region ◽  
Global Solutions ◽  
Optimization Methods ◽  
Box Constraints ◽  
Particle Swarm Method ◽  
Global And Local Optimization ◽  
Global And Local

In this paper, the results of an application of global and local optimization methods to solve a problem of determination of strains in RC compressed structure members are presented. Solutions of appropriate sets of nonlinear equations in the presence of box constraints have to be found. The use of the least squares method leads to finding global solutions of optimization problems with box constraints. Numerical examples illustrate the effects of the loading value and the loading eccentricity on the strains in concrete and reinforcing steel in the a cross-section.Three different minimization methods were applied to compute them: trust region reflective, genetic algorithm tailored to problems with real double variables and particle swarm method. Numerical results on practical data are presented. In some cases, several solutions were found. Their existence has been detected by the local search with multistart, while the genetic and particle swarm methods failed to recognize their presence.

Drive optimization of a pulsatile total artificial heart

2014 ◽  
Vol 33 (3) ◽  
pp. 941-952
Author(s):  
André Pohlmann ◽  
Kay Hameyer
Keyword(s):  
Cardiovascular Diseases ◽  
Differential Evolution ◽  
Cost Function ◽  
Human Body ◽  
Differential Evolution Algorithm ◽  
Optimization Methods ◽  
Parameter Variation ◽  
Optimization Process ◽  
Total Artificial Heart ◽  
Content Type

Purpose – Total artificial hearts (TAHs) are required for the treatment of cardiovascular diseases. In order to replace the native heart a TAH must provide a sufficient perfusion of the human body, prevent blood damage and meet the implantation constraints. Until today there is no TAH on the market which meets all constraints. So the purpose of this paper is to design a drive in such a way that the operated TAH meets all predefined constraints. Design/methodology/approach – The drive is designed in terms of weight and electric losses. In setting up a cost function containing those constraints, the drive design can be included in a optimization process. When reaching the global minimum of the cost function the optimum drive design is found. In this paper the optimization methods manual parameter variation and differential evolution are applied. Findings – At the end of the optimization process the drive's weight amounts to 460 g and its mean losses sum up to 10 W. This design meets all predefined constraints. Further it is proposed to start the optimization process with a parameter variation to reduce the amount of optimization parameters for the time consuming differential evolution algorithm. Practical implications – This TAH has the potential to provide a therapy for all patients suffering from cardiovascular diseases as it is independent of donor organs. Originality/value – The optimization-based design process yields an optimum drive for a TAH in terms of weight and electrical losses. In this way a TAH is developed which meets all implantation constraints and provides sufficient perfusion of the human body at the same time.

scholarly journals Development of methodology for calculation of optimal distribution of electric power between power units of condensing power plant

2021 ◽  
Vol 1 (2) ◽  
pp. 18-25
Author(s):  
V.Y. Ilichev ◽  
◽  
E.A. Yurik ◽  
Keyword(s):  
Power Plant ◽  
Differential Evolution ◽  
Fuel Consumption ◽  
Power Plants ◽  
Optimization Problems ◽  
Economic Effect ◽  
Optimization Methods ◽  
Local Extremum ◽  
Fuel Savings ◽  
Total Capacity

Optimization methods are used to solve many problems in the field of energy. One of such tasks is the problem of optimal redistribution of power between power units in order to achieve minimum fuel consumption. This is especially important for powerful condensation power plants, where even relatively small fuel savings have significant economic effect. The article is devoted to description of developed method of such optimization, based on the ap-plication of differential evolution, which has many advantages over the “classical” methods of op-timization. In particular, it was the global rather than the local extremum of the objective function that could be found. This method is also easy and powerful when using modern software tools. Differential evolution method is organized in the library SciPy of Python programming language, so calculation program was developed in this language to solve the problem. The work considers algorithm and structure of the developed program, as well as the procedure for preparing initial data and calculation process using example of a specific condensing power plant. Modules used in the program to populate the data arrays, as well as to output the results in the form of high-quality graphs are mentioned. With the help of the program, diagram of the optimal redistribution of capacities between power units for any total capacity of the power station is constructed. Also, for entire power range of the power plant, nominal fuel consumption and fuel economy are calculated when implementing the optimal redistribution of capacity in comparison with an even distribution. Obtained software product, available to everyone on the website of the authors, allows not only to study the practical application of differential evolution method, but also to create programs based on it in order to solve other optimization problems, some of which are mentioned in the article.

scholarly journals SOLVING THE NONLINEAR TRANSPORTATION PROBLEM BY GLOBAL OPTIMIZATION

Transport ◽  
2010 ◽  
Vol 25 (3) ◽  
pp. 314-324 ◽  
Author(s):  
Uroš Klanšek ◽  
Mirko Pšunder
Keyword(s):  
Global Optimization ◽  
Transportation Problem ◽  
Optimization Problems ◽  
Optimization Methods ◽  
Branch And Cut ◽  
Cut Method ◽  
Global Optimization Methods ◽  
Optimum Solution ◽  
Global And Local ◽  
Branch And Cut Method

The aim of this paper is to present the suitability of three different global optimization methods for specifically the exact optimum solution of the nonlinear transportation problem (NTP). The evaluated global optimization methods include the branch and reduce method, the branch and cut method and the combination of global and local search strategies. The considered global optimization methods were applied to solve NTPs with reference to literature. NTPs were formulated as nonlinear programming (NLP) optimization problems. The obtained optimal results were compared with those got from literature. A comparative evaluation of global optimization methods is presented at the end of the paper to show their suitability for solving NTPs.

scholarly journals Global Evolution Commended by Localized Search for Unconstrained Single Objective Optimization

Processes ◽  
2019 ◽  
Vol 7 (6) ◽  
pp. 362 ◽  
Author(s):  
Rashida Khanum ◽  
Muhammad Jan ◽  
Nasser Tairan ◽  
Wali Mashwani ◽  
Muhammad Sulaiman ◽  
...  
Keyword(s):  
Differential Evolution ◽  
Optimization Problems ◽  
Local Area ◽  
Initial Population ◽  
Search Techniques ◽  
Adaptive Differential Evolution ◽  
Complex Functions ◽  
Global Optimizer ◽  
Mutation Strategies ◽  
Global And Local

Differential Evolution (DE) is one of the prevailing search techniques in the present era to solve global optimization problems. However, it shows weakness in performing a localized search, since it is based on mutation strategies that take large steps while searching a local area. Thus, DE is not a good option for solving local optimization problems. On the other hand, there are traditional local search (LS) methods, such as Steepest Decent and Davidon–Fletcher–Powell (DFP) that are good at local searching, but poor in searching global regions. Hence, motivated by the short comings of existing search techniques, we propose a hybrid algorithm of a DE version, reflected adaptive differential evolution with two external archives (RJADE/TA) with DFP to benefit from both search techniques and to alleviate their search disadvantages. In the novel hybrid design, the initial population is explored by global optimizer, RJADE/TA, and then a few comparatively best solutions are shifted to the archive and refined there by DFP. Thus, both kinds of searches, global and local, are incorporated alternatively. Furthermore, a population minimization approach is also proposed. At each call of DFP, the population is decreased. The algorithm starts with a maximum population and ends up with a minimum. The proposed technique was tested on a test suite of 28 complex functions selected from literature to evaluate its merit. The results achieved demonstrate that DE complemented with LS can further enhance the performance of RJADE/TA.

Software suite for determining the tilts of tower-type structures according to terrestrial laser scanning

2020 ◽  
Vol 961 (7) ◽  
pp. 2-7
Author(s):  
A.V. Zubov ◽  
N.N. Eliseeva
Keyword(s):  
Laser Scanning ◽  
Optimization Problems ◽  
Optimization Methods ◽  
User Participation ◽  
Software Suite ◽  
Current State ◽  
Visual Basic For Applications ◽  
Operational Data ◽  
Theoretical Developments

The authors describe a software suite for determining tilt degrees of tower-type structures according to ground laser scanning indication. Defining the tilt of the pipe is carried out with a set of measured data through approximating the sections by circumferences. They are constructed using one of the simplest search engine optimization methods (evolutionary algorithm). Automatic filtering the scan of the current section from distorting data is performed by the method of assessing the quality of models constructed with that of least squares. The software was designed using Visual Basic for Applications. It contains several blocks (subprograms), with each of them performing a specific task. The developed complex enables obtaining operational data on the current state of the object with minimal user participation in the calculation process. The software suite is the result of practical implementing theoretical developments on the possibilities of using search methods at solving optimization problems in geodetic practice.

Sign in / Sign up

Export Citation Format

Share Document