Dissertations / Theses on the topic 'Differential Evolution'
To see the other types of publications on this topic, follow the link: Differential Evolution.
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 dissertations / theses for your research on the topic 'Differential Evolution.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
1
Yilmaz, Halis. "Evolution equations for differential invariants." Thesis, University of Glasgow, 2003. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.274288.
Full text
2
Silva, Eduardo Krempser da. "Differential evolution for constrained optimization problems." Laboratório Nacional de Computação Científica, 2009. http://www.lncc.br/tdmc/tde_busca/arquivo.php?codArquivo=188.
Full textAbstract:
Optimization is a large area of knowledge concerned with the need of a better use of resources and activities, becoming indispensable in the solution of several problems which arise from the study and formulation of real-world problems. Furthermore, the constraints that must be respected for each situation introduce in the methodologies of optimization an additional complication. Differential Evolution, which in its original formulation is applied only to unconstrained optimization problems in continuous space, also provides good results when applied to constrained optimization with discrete and continuous variables. This work presents the necessary improvements to Differential Evolution for its proper application to this class of problems, and proposes a new combination of techniques for this application, as well as a mechanism for dynamic selection of the appropriate variant of the technique. The initial proposal is a combination of Differential Evolution with a technique of adaptive penalty (APM) and the second proposal concerns the dynamic selection of variants during the search process. Several computational experiments are carried out confirming the competitiveness of the proposed algorithms.A otimização é uma grande área de conhecimento voltada para a necessidade de um melhor aproveitamento de recursos e atividades, tornando-se indispensável na resolução de grande parte dos problemas oriundos de estudos e formulações de problemas reais. Além disso, as restrições que devem ser respeitadas para cada situação introduzem nas metodologias de otimização um complicador adicional. A Evolução Diferencial, que em sua formulação original é aplicada somente a problemas de otimização irrestrita e em espaços contínuos, apresenta também bons resultados quando aplicada à otimização restrita com variáveis contínuas e discretas. Este trabalho apresenta os aperfeiçoamentos necessários à Evolução Diferencial para sua adequada aplicação sobre essa classe de problemas, além de propor uma nova combinação de técnicas para essa aplicação, bem como um mecanismo de seleção dinâmica da variante adequada da técnica. A proposta inicial é a combinação da Evolução Diferencial com uma técnica adaptativa de penalização (APM) e a segunda proposta visa a seleção dinâmica de variantes durante o processo de busca. Vários experimentos computacionais são executados confirmando a competitividade dos algoritmos propostos.
3
Farah, Abdulkadir. "Differential evolution algorithms for network optimization." Thesis, University of Reading, 2013. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.602400.
Full textAbstract:
Many real world optimization problems are difficult to solve, therefore, when solving such problems it is wise to employ efficient optimization algorithms which are capable of handling the problem complexities and of finding optimal or near optimal solutions within a reasonable time and without using excessive computational resources. The objective of this research is to develop Differential Evolution (DE) algorithms with improved performance capable of solving difficult and challenging global constrained and unconstrained optimization problems, as well as extending the application of the these algorithms to real-world optimization problems, particularly wireless broadband network placement and deployment problems. The adaptation of DE control parameters has also been investigated and a novel method using Mann-Iteration and Tournament scoring is proposed to improve the performance of the algorithm. A novel constraint handling technique called neighborhood constraints handling (NCR) method has been also proposed. A set of experiments are conducted to comprehensively test the performance of the proposed DE algorithms for global optimization. The numerical results for well-known optimization global optimization test problems are shown to prove the performance of the proposed methods. In addition, a novel wireless network test point (TP) reduction algorithm (TPR) has been presented. The TPR algorithm and the proposed DE algorithms have been applied for solving the optimal network placement problem. In order to utilize the value of flexibility a novel value optimization problem formulation integrating the state of the art approaches of cash flow (CF) analysis and real option analysis (ROA) for network deployment has been presented, utilizing the proposed DE algorithms to obtain the optimal roll-out sequence that maximizes the value of the wireless network deployment. A numerical experimentation, based on a case study scenario of an optimal network placement and deployment for wireless broadband access network, has been conducted to confirm the efficiency of these algorithms.
4
Nguyen, Thieu-Huy. "Functional partial differential equations and evolution semigroups." [S.l.] : [s.n.], 2003. http://deposit.ddb.de/cgi-bin/dokserv?idn=973911344.
Full text
5
Cedeño-Maldonado, José R. "Differential evolution based optimal power flow algorithm /." The Ohio State University, 2002. http://rave.ohiolink.edu/etdc/view?acc_num=osu1486402288260595.
Full text
6
Leon, Miguel. "Enhancing Differential Evolution Algorithm for Solving Continuous Optimization Problems." Licentiate thesis, Mälardalens högskola, Inbyggda system, 2016. http://urn.kb.se/resolve?urn=urn:nbn:se:mdh:diva-33466.
Full textAbstract:
Differential Evolution (DE) has become one of the most important metaheuristics during the recent years, obtaining attractive results in solving many engineering optimization problems. However, the performance of DE is not always strong when seeking optimal solutions. It has two major problems in real world applications. First, it can easily get stuck in a local optimum or fail to generate better solutions before the population has converged. Secondly, its performance is significantly influenced by the control parameters, which are problem dependent and which vary in different regions of space under exploration. It usually entails a time consuming trial-and-error procedure to set suitable parameters for DE in a specific problem, particularly for those practioners with limited knowledge and experience of using this technique. This thesis aims to develop new DE algorithms to address the two aforementioned problems. To mitigate the first problem, we studied the hybridization of DE with local search techniques to enhance the efficiency of search. The main idea is to apply a local search mechanism to the best individual in each generation of DE to exploit the most promising regions during the evolutionary processs so as to speed up the convergence or increase the chance to scape from local optima. Four local search strategies have been integrated and tested in the global DE framework, leading to variants of the memetic DE algorithms with different properties concerning diversification and intensification. For tackling the second problem, we propose a greedy adaptation method for dynamic adjustment of the control parameters in DE. It is implemented by conducting greedy search repeatedly during the run of DE to reach better parameter assignments in the neighborhood of a current candidate. The candidates are assessed by considering both, the success rate and also fitness improvement of trial solutions against the target ones. The incorporation of this greedy parameter adaptation method into standard DE has led to a new adaptive DE algorithm, referred to as Greedy Adaptive Differential Evolution (GADE). The methods proposed in this thesis have been tested in different benchmark problems and compared with the state of the art algorithms, obtaining competitive results. Furthermore, the proposed GADE algorithm has been applied in an industrial scenario achieving more accurate results than those obtained by a standard DE algorithm.Differential Evolution (DE) har blivit en av de viktigaste metaheuristikerna under de senaste åren och har uppnått attraktiva resultat för att lösa många optimeringsproblem inom teknik. Dock är prestationen hos DE inte alltid framgångsrik när man söker optimala lösningar. Det finns två huvudsakliga problem för applikationer i den verkliga världen. Det första är att den lätt kan fastna i lokala optimum eller misslyckas att generera bättre lösningar före det att populationen (en grupp av lösningar) har hunnit konvergera. Det andra är att prestandan påverkas märkvärdigt av kontrollparametrar, vilkas optimala värden beror på problem som ska lösas och varierar mellan regioner i sökrymden. Detta innebär oftast ett tidskrävande trial-and-error förfarande för att hitta lämpliga parametrar till ett specifikt DE-problem, framför allt för utövare med begränsad kunskap och erfarenhet av DE. Syftet med denna licentiatavhandling är att utveckla nya DE-algoritmer för att behandla de ovannämnda problemen. För att möta det första problemet så studerades hybridisering av DE och lokala söktekniker för att effektivisera sökningen. Tanken är att använda en lokal sökmekanism på den bästa individen i varje generation i DE-algoritmen och utnyttja de mest lovande regionerna under evolutionsprocessen för att snabba på konvergensen eller öka chansen att undvika lokala optimum. Fyra lokala sökstrategier har integrerats och testats i det globala DE-ramverket vilket har lett till fyra varianter av DE-algoritmerna med olika egenskaper beträffande diversifiering och intensifiering. Till det andra problemet föreslås en greedy adaptation method för dynamisk justering av kontrollparametrarna i DE. Den implementeras genom att utföra greedy search upprepade gånger under körningen av DE för att hitta bättre värden till kontrollparametrarna. Utvärderingen av parameterval baseras på både success rate och fitness improvement av trial lösningar jämfört med target lösningar. Sammanslagningen av DE och denna greedy parameter adaptation har lett till en ny adaptiv DE-algoritm som kallas Greedy Adaptive Differential Evolution (GADE). Den föreslagna metoden i denna licentiatavhandling har testats i olika prestandamätningar och jämförts med state-of-the-art-algoritmer, med goda resultat. Dessutom har den föreslagna GADE-algoritmen använts i ett industriellt scenario och uppnådde då mer exakta resultat än den med en standard DE-algoritm.
7
Enaganti, Srujan Kumar. "Solving correlation matrix completion problems using parallel differential evolution." Thesis, University of British Columbia, 2010. http://hdl.handle.net/2429/30302.
Full textAbstract:
Matrix Completion problems have been receiving increased attention due to their varied applicability in different domains. Correlation matrices arise often in studying multiple streams of time series data like technical analysis of stock market data. Often some of the values in the matrix are unknown and some reasonable replacements have to be found at the earliest opportunity to avert an unwanted consequence or keep up the pace in the business. After looking to background research related to solving this problem, we propose a new parallel technique that can solve general correlation matrix completion problems over a set of computers connected to a high speed network. We present some of our results where we could reduce the execution time.
8
Christoph, Raab. "Reconstruction of PINGU data with a Differential Evolution Minimizer." Thesis, Uppsala universitet, Högenergifysik, 2013. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-214617.
Full textAbstract:
The Precision IceCube Next Generation Upgrade (PINGU) is supposed to have an energy threshold below 10 in order to resolve the neutrino mass hierarchy. In order to reconstruct the energy and direction of neutrinos interacting in this array, producing both a hadronic cascade and a muon track, advanced reconstruction methods need to be employed. A class of these seeks to maximize a complicated likelihood function within an 8-dimensional parameter space describing the event, and requires sophisticated minimizers to achieve the necessary resolution in a reasonable time. In this report, a pre-existing but hitherto unused minimizer which samples that parameter space with several Markov chains at once, based on the Differential Evolution Monte Carlo algorithm, is developed further and its behaviour and performance is tested on simulated data of the IceCube/PINGU array. The tests compare both various configurations of the minimizer and Markov Chain Monte Carlo, a similar previous approach.
9
Dawar, Deepak. "Adaptive Differential Evolution and its Application to Machine Vision." Diss., North Dakota State University, 2016. http://hdl.handle.net/10365/25732.
Full textAbstract:
Over recent years, Evolutionary Algorithms (EA) have emerged as a practical approach for solving hard optimization problems ubiquitously presented in real life. The inherent advantage of EA over other types of numerical optimization methods lies in the fact that they require much less or no prior knowledge of the objective function. Differential Evolution (DE) has emerged as a highly competitive and powerful real parameter optimizer in the diverse community of evolutionary algorithms. The study of this dissertation is focused on two main approaches. The first approach focuses on studying and improving DE by creating its variants that aim at altering/adapting its control parameters and mutation strategies during the course of the search. The performance of DE depends largely upon the mutation strategy used, its control parameters namely the scale factor F, the crossover rate Cr, and the population size NP, and is quite sensitive to their appropriate settings. A simple and effective technique that alters F in stages, first through random perturbations and then through the application of an annealing schedule, is proposed. After that, the impact and efficacy of adapting mutation strategies with or without adapting the control parameters is investigated. The second approach is concerned with the application side of DE which is used as an optimizer either as the primary algorithm or as a surrogate to improve the performance of the overall system. The focus area is video based vehicle classification. A DE based vehicle classification system is proposed. The system in its essence, aims to classify a vehicle, based on the number of circles (axles) in an image using Hough Transform which is a popular parameter based feature detection method. Differential Evolution (DE) is coupled with Hough Transform to improve the overall accuracy of the classification system. DE is further employed as an optimizer in an extension of the previous vehicle detector and classifier. This system has a novel appearance based model utilizing pixel color information and is capable of classifying multi-lane moving vehicles into seven different classes. Five different variants of DE on varied videos are tested, and a performance profile of all the variants is provided.
10
Mulumba, Tshina Fa. "Application of differential evolution to power system stabilizer design." Master's thesis, University of Cape Town, 2012. http://hdl.handle.net/11427/12026.
Full textAbstract:
Includes synopsis.Includes bibliographical references.
In recent years, many Evolutionary Algorithms (EAs) such as Genetic Algorithms (GAs) have been proposed to optimally tune the parameters of the PSS. GAs are population based search methods inspired by the mechanism of evolution and natural genetic. Despite the fact that GAs are robust and have given promising results in many applications, they still have some drawbacks. Some of these drawbacks are related to the problem of genetic drift in GA which restricts the diversity in the population. ... To cope with the above mentioned drawbacks, many variants of GAs have been proposed often tailored to a particular problem. Recently, several simpler and yet effective heuristic algorithms such as Population Based Incremental Learning (PBIL) and Differential Evolution (DE), etc., have received increasing attention.
11
Nguyen, Thi. "On the Evolution of Virulence." CSUSB ScholarWorks, 2014. https://scholarworks.lib.csusb.edu/etd/91.
Full textAbstract:
The goal of this thesis is to study the dynamics behind the evolution of virulence. We examine first the underlying mechanics of linear systems of ordinary differential equations by investigating the classification of fixed points in these systems, then applying these techniques to nonlinear systems. We then seek to establish the validity of a system that models the population dynamics of uninfected and infected hosts---first with one parasite strain, then n strains. We define the basic reproductive ratio of a parasite, and study its relationship to the evolution of virulence. Lastly, we investigate the mathematics behind superinfection.
12
Zangeneh, Bijan Z. "Semilinear stochastic evolution equations." Thesis, University of British Columbia, 1990. http://hdl.handle.net/2429/31117.
Full textAbstract:
Let H be a separable Hilbert space. Suppose (Ω, F, Ft, P) is a complete stochastic basis with a right continuous filtration and {Wt,t ∈ R} is an H-valued cylindrical Brownian motion with respect to {Ω, F, Ft, P). U(t, s) denotes an almost strong evolution operator generated by a family of unbounded closed linear operators on H. Consider the semilinear stochastic integral equation
[formula omitted]
where
• f is of monotone type, i.e., ft(.) = f(t, w,.) : H → H is semimonotone, demicon-tinuous, uniformly bounded, and for each x ∈ H, ft(x) is a stochastic process which satisfies certain measurability conditions.
• gs(.) is a uniformly-Lipschitz predictable functional with values in the space of Hilbert-Schmidt operators on H.
• Vt is a cadlag adapted process with values in H.
• X₀ is a random variable.
We obtain existence, uniqueness, boundedness of the solution of this equation. We show the solution of this equation changes continuously when one or all of X₀, f, g, and V are varied. We apply this result to find stationary solutions of certain equations, and to study the associated large deviation principles.
Let {Zt,t ∈ R} be an H-valued semimartingale. We prove an Ito-type inequality and a Burkholder-type inequality for stochastic convolution [formula omitted]. These are the main tools for our study of the above stochastic integral equation.Science, Faculty of
Mathematics, Department of
Graduate
13
Maizurna, Isna. "Semigroup methods for degenerate cauchy problems and stochastic evolution equations /." Title page, abstract and contents only, 1999. http://web4.library.adelaide.edu.au/theses/09PH/09phm2328.pdf.
Full text
14
Sum-Im, Thanathip. "A novel differential evolution algorithmic approach to transmission expansion planning." Thesis, Brunel University, 2009. http://bura.brunel.ac.uk/handle/2438/3219.
Full textAbstract:
Nowadays modern electric power systems consist of large-scale and highly complex interconnected transmission systems, thus transmission expansion planning (TEP) is now a significant power system optimisation problem. The TEP problem is a large-scale, complex and nonlinear combinatorial problem of mixed integer nature where the number of candidate solutions to be evaluated increases exponentially with system size. The accurate solution of the TEP problem is essential in order to plan power systems in both an economic and efficient manner. Therefore, applied optimisation methods should be sufficiently efficient when solving such problems. In recent years a number of computational techniques have been proposed to solve this efficiency issue. Such methods include algorithms inspired by observations of natural phenomena for solving complex combinatorial optimisation problems. These algorithms have been successfully applied to a wide variety of electrical power system optimisation problems. In recent years differential evolution algorithm (DEA) procedures have been attracting significant attention from the researchers as such procedures have been found to be extremely effective in solving power system optimisation problems. The aim of this research is to develop and apply a novel DEA procedure directly to a DC power flow based model in order to efficiently solve the TEP problem. In this thesis, the TEP problem has been investigated in both static and dynamic form. In addition, two cases of the static TEP problem, with and without generation resizing, have also been investigated. The proposed method has achieved solutions with good accuracy, stable convergence characteristics, simple implementation and satisfactory computation time. The analyses have been performed within the mathematical programming environment of MATLAB using both DEA and conventional genetic algorithm (CGA) procedures and a detailed comparison has also been presented. Finally, the sensitivity of DEA control parameters has also been investigated.
15
Yu, Edmund Po-ning. "Evolution equations for magnetic islands in a reversed field pinch." Access restricted to users with UT Austin EID, 2001. http://wwwlib.umi.com/cr/utexas/fullcit?p3037030.
Full text
16
Wolter, Uwe. "Spot evolution and differential rotation of the ultrafast rotator Speedy Mic." [S.l. : s.n.], 2004. http://deposit.ddb.de/cgi-bin/dokserv?idn=972154787.
Full text
17
Mannakee, Brian K., and Ryan N. Gutenkunst. "Selection on Network Dynamics Drives Differential Rates of Protein Domain Evolution." PUBLIC LIBRARY SCIENCE, 2016. http://hdl.handle.net/10150/621480.
Full textAbstract:
The long-held principle that functionally important proteins evolve slowly has recently been challenged by studies in mice and yeast showing that the severity of a protein knockout only weakly predicts that protein's rate of evolution. However, the relevance of these studies to evolutionary changes within proteins is unknown, because amino acid substitutions, unlike knockouts, often only slightly perturb protein activity. To quantify the phenotypic effect of small biochemical perturbations, we developed an approach to use computational systems biology models to measure the influence of individual reaction rate constants on network dynamics. We show that this dynamical influence is predictive of protein domain evolutionary rate within networks in vertebrates and yeast, even after controlling for expression level and breadth, network topology, and knockout effect. Thus, our results not only demonstrate the importance of protein domain function in determining evolutionary rate, but also the power of systems biology modeling to uncover unanticipated evolutionary forces.
18
Cuomo, Claudia. "Gut patterning in development and evolution : a comparative differential transcriptomics approach." Thesis, Open University, 2017. http://oro.open.ac.uk/50273/.
Full textAbstract:
All bilaterians share a common kit of transcription factors and cis-regulatory elements that compose the Gene Regulatory Network (GRN) essential for the development of the body plan. Modifications within the elements of the GRN determine the evolution of animal forms. In this study, the GRN for embryonic gut specification in two echinoderm species, the sea urchin Strongylocentrotus purpuratus and the sea star Patiria miniata, has been studied with the purpose of acquiring further knowledge on the process of gut patterning in the sea urchin larva and to compare it with the sea star embryo, focusing on the role that Xlox and Cdx transcription factors have in this process in both echinoderm species. Taking advantage of genome-wide approaches and modern high-throughput technologies, a partial reconstruction of the sea urchin GRN around Xlox and Cdx transcription factors, known for their key role to start the gut specification process, has been achieved leading to describe the putative interactions of the TFs in the network. An important novel node of the sea urchin GRN featuring the interaction between Sp-Meis, an homeobox gene, and Sp-Lox protein has been revealed and the occupancy of Sp-Lox protein on Sp-Meis regulatory region has been demonstrated by ChIP-PCR. The comparison of the differentially expressed genes after Xlox and Cdx perturbation in both sea urchin and sea star embryos has revealed that, although the absence of these two proteins affects some digestive functions and developmental processes related to its specification, however many of the regulatory genes involved in these mechanisms are not the same in the two species. This result suggests a phenomenon of rewiring of the gut GRN in S. purpuratus and P. miniata that, although belonging to the same phylum Echinodermata, are distant million years in term of evolutionary time, as recorded by fossil records.
19
Al-Sawwa, Jamil. "Scalable Particle Swarm Optimization and Differential Evolution Approaches Applied to Classification." Diss., North Dakota State University, 2019. https://hdl.handle.net/10365/31538.
Full textAbstract:
Applying the nature-inspired methods in the data mining area has been gaining more attention by researchers. Classification is one of the data mining tasks which aims to analyze historical data by discovering hidden relationships between the input and the output that would help to predict an accurate outcome for an unseen input. The classification algorithms based on nature-inspired methods have been successfully used in numerous applications such as medicine and agriculture. However, the amount of data that has been collected or generated in these areas has been increasing exponentially. Thus, extracting useful information from large data requires computational time and consumes memory space. Besides this, many algorithms suffer from not being able to handle imbalanced data.
Apache Spark is an in-memory computing big data framework that runs on a cluster of nodes. Apache Spark is more efficient for handling iterative and interactive jobs and runs 100 times faster than Hadoop Map-Reduce for various applications. However, the challenge is to find a scalable solution using Apache Spark for the optimization-based classification algorithms that would scale very well with large data.
In this dissertation, we firstly introduce new variants of a centroid-based particle swarm optimization (CPSO) classification algorithm in order to improve its performance in terms of misclassification rate. Furthermore, a scalable particle swarm optimization classification algorithm (SCPSO) is designed and implemented using Apache Spark. Two variants of SCPSO, namely SCPSO-F1 and SCPSO-F2, are proposed based on different fitness functions. The experiments revealed that SCPSO-F1 and SCPSO-F2 utilize the cluster of nodes efficiently and achieve good scalability results.
Moreover, we propose a cost-sensitive differential evolution classification algorithm to improve the performance of the differential evolution classification algorithm when applied to imbalanced data sets. The experimental results demonstrate that the proposed algorithm efficiently handles highly imbalanced binary data sets compared to the current variants of differential evolution classification algorithms.
Finally, we designed and implemented a parallel version of a cost-sensitive differential evolution classifier using the Spark framework. The experiments revealed that the proposed algorithm achieved good speedup and scaleup results and obtained good performance.
20
Zielinski, Karin. "Optimizing real-world problems with differential evolution and particle swarm optimization." Aachen Shaker, 2009. http://d-nb.info/993509398/04.
Full text
21
Wang, He. "Advanced Electromyogram Signal Processing with an Emphasis on Simplified, Near-Optimal Whitening." Digital WPI, 2019. https://digitalcommons.wpi.edu/etd-theses/1338.
Full textAbstract:
Estimates of the time-varying standard deviation of the surface EMG signal (EMGσ) are extensively used in the field of EMG-torque estimation. The use of a whitening filter can substantially improve the accuracy of EMGσ estimation by removing the signal correlation and increasing the statistical bandwidth. However, a subject-specific whitening filter which is calibrated to each subject, is quite complex and inconvenient. To solve this problem, we first calibrated a 60th-order “Universal” FIR whitening filter by using the ensemble mean of the inverse of the square root of the power spectral density (PSD) of the noise-free EMG signal. Pre-existing data from elbow contraction of 64 subjects, providing 512 recording trials were used. The test error on an EMG-torque task based on the “Universal” FIR whitening filter had a mean error of 4.80% maximum voluntary contraction (MVC) with a standard deviation of 2.03% MVC. Meanwhile the subject-specific whitening filter had performance of 4.84±1.98% MVC (both have a whitening band limit at 600 Hz). These two methods had no statistical difference. Furthermore, a 2nd-order IIR whitening filter was designed based on the magnitude response of the “Universal” FIR whitening filter, via the differential evolution algorithm. The performance of this IIR whitening filter was very similar to the FIR filter, with a performance of 4.81±2.12% MVC. A statistical test showed that these two methods had no significant difference either. Additionally, a complete theory of EMG in additive measured noise contraction modeling is described. Results show that subtracting the variance of whitened noise by computing the root difference of the square (RDS) is the correct way to remove noise from the EMG signal.
22
von, Glehn Ingrid. "A closest point penalty method for evolution equations on surfaces." Thesis, University of Oxford, 2014. http://ora.ox.ac.uk/objects/uuid:29385f90-b927-4151-b5df-cf877cef00ef.
Full textAbstract:
This thesis introduces and analyses a numerical method for solving time-dependent partial differential equations (PDEs) on surfaces. This method is based on the closest point method, and solves the surface PDE by solving a suitably chosen equation in a band surrounding the surface. As it uses an implicit closest point representation of the surface, the method has the advantages of being simple to implement for very general surfaces, and amenable to discretization with a broad class of numerical schemes. The method proposed in this work introduces a new equation in the embedding space, which satisfies a key consistency property with the surface PDE. Rather than alternating between explicit time-steps and re-extensions of the surface function as in the original closest point method, we investigate an alternative approach, in which a single equation can be solved throughout the embedding space, without separate extension steps. This is achieved by creating a modified embedding equation with a penalty term, which enforces a constraint on the solution. The resulting equation admits a method of lines discretization, and can therefore be discretized with implicit or explicit time-stepping schemes, and analysed with standard techniques. The method can be formulated in a straightforward way for a large class of problems, including equations featuring variable coefficients, higher-order terms or nonlinearities. The effectiveness of the method is demonstrated with a range of examples, drawing from applications involving curvature-dependent diffusion and systems of reaction-diffusion equations, as well as equations arising in PDE-based image processing on surfaces.
23
Rahnamayan, Shahryar. "Opposition-Based Differential Evolution." Thesis, 2007. http://hdl.handle.net/10012/2784.
Full textAbstract:
Evolutionary algorithms (EAs) are well-established techniques to approach those problems which for the classical optimization methods are difficult to solve. Tackling problems with mixed-type of variables, many local optima, undifferentiable or non-analytical functions are some examples to highlight the outstanding capabilities of the evolutionary algorithms. Among the various kinds of evolutionary algorithms, differential evolution (DE) is well known for its effectiveness and robustness. Many comparative studies confirm that the DE outperforms many other optimizers. Finding more accurate solution(s), in a shorter period of time for complex black-box problems, is still the main goal of all evolutionary algorithms. The opposition concept, on the other hand, has a very old history in philosophy, set theory, politics, sociology, and physics. But, there has not been any opposition-based contribution to optimization. In this thesis, firstly, the opposition-based optimization (OBO) is constituted. Secondly, its advantages are formally supported by establishing mathematical proofs. Thirdly, the opposition-based acceleration schemes, including opposition-based population initialization and generation jumping, are proposed. Fourthly, DE is selected as a parent algorithm to verify the acceleration effects of proposed schemes. Finally, a comprehensive set of well-known complex benchmark functions is employed to experimentally compare and analyze the algorithms. Results confirm that opposition-based DE (ODE) performs better than its parent (DE), in terms of both convergence speed and solution quality. The main claim of this thesis is not defeating DE, its numerous versions, or other optimizers, but to introduce a new notion into nonlinear continuous optimization via innovative metaheuristics, namely the notion of opposition. Although, ODE has been compared with six other optimizers and outperforms them overall. Furthermore, both presented experimental and mathematical results conform with each other and demonstrate that opposite points are more beneficial than pure random points for black-box problems; this fundamental knowledge can serve to accelerate other machine learning approaches as well (such as reinforcement learning and neural networks). And perhaps in future, it could replace the pure randomness with random-opposition model when there is no a priori knowledge about the solution/problem. Although, all conducted experiments utilize DE as a parent algorithm, the proposed schemes are defined at the population level and, hence, have an inherent potential to be utilized for acceleration of other DE extensions or even other population-based algorithms, such as genetic algorithms (GAs). Like many other newly introduced concepts, ODE and the proposed opposition-based schemes still require further studies to fully unravel their benefits, weaknesses, and limitations.
24
Hunag, Shin-chia, and 黃信嘉. "Dynamic Clustering Using Differential Evolution." Thesis, 2008. http://ndltd.ncl.edu.tw/handle/21178749888377219637.
Full textAbstract:
碩士大同大學
資訊工程學系(所)
96
Data clustering is one of the important issues on the data mining techniques which is the process of considering as the number of cluster and the center of cluster. Most of data clustering algorithms are prior known of the number of cluster but dynamic clustering is able to find the optimal number of cluster and center of cluster dynamically by algorithms. This research is used differential evolution algorithm to perform data clustering which is called as dynamic clustering using differential evolution (DCDE). This algorithm is accessed the number of cluster of solution vectors first by normal distribution and then updating the center of cluster of every solution vectors by differential evolution. Finally, we combine cluster validity index to estimate the results of dynamic clustering to make the solution vectors move to the optimal number of cluster of the subspace constantly. This paper uses eight artificial data sets and four real-world data sets to test. The experimental results show that DCDE is able to find the accurate number of cluster and better and more stable center of cluster with unknown the accurate number of cluster.
25
Chien, Wan-Jou, and 簡宛柔. "A Novel Differential Evolution Algorithm with co-evolution strategy." Thesis, 2010. http://ndltd.ncl.edu.tw/handle/42161994725526399905.
Full textAbstract:
碩士中原大學
資訊管理研究所
98
Differential evolution, termed DE, is a novel and rapidly developed evolution computation in recent year. There are some advantages of DE, including simple structure, easy use and rapid convergence speed. Besides, DE can be also applied on complex optimization problem. However, there are some problems, such as premature convergence and stagnation, remaining in DE algorithm. To overcome those disadvantages, a different method was proposed, named CO-DE, by combining with a simple co-evolutionary model and reset mechanism. Thus, CO-DE can maintain appropriate swarm diversity and reduce the premature convergence. On the other hand, a reset mechanism was set to avoid the particle stagnates, which can further improve the performance of differential evolution. The proposed model can be now successfully applied with some well-known benchmark functions.
26
Raghunathan, T. "Differential Evolution Based Interceptor Guidance Law." Thesis, 2011. http://etd.iisc.ernet.in/handle/2005/2014.
Full textAbstract:
Kinematics based guidance laws like the proportional navigation (PN) and many other linear optimal guidance laws perform well in near-collision course conditions. These have been studied thoroughly in the literature from all aspects, ranging from optimality to capturability, for planar or two dimensional interceptor-target engagements, and to a lesser extent, for three dimensional engagements. But guidance in widely off-collision course conditions like high initial heading errors has been relatively less studied. This is probably due to the inherently high nonlinearity of the problem, which makes it a far more difficult problem to solve. However, with increasing speed and agility of interceptors and targets, solutions of such problems have acquired an increased urgency, as has been reflected in the recent literature.
This thesis proposes a guidance law based on differential evolution (DE), a member of the evolutionary algorithms (EA) family. While EAs have been applied extensively to static optimization problems, they have been considered unsuitable for solving dynamic optimization or optimal control problems, due to their computationally intensive nature, and their consequent inability to produce solutions online in real-time, except for systems with very slow dynamics. This thesis proposes an online-implementable optimal control for interceptor guidance, a problem with inherently fast dynamics.
The proposed law is applicable to all initial geometries including those that involve high to very high heading errors. While interception by itself is a challenging task in the presence of high heading errors, an additional requirement of optimality is also imposed. The first part of the thesis considers only the 2-D kinematic model with high heading errors. In the second part, a 3-D realistic dynamic model, which includes a time-varying interceptor speed, thrust, drag and mass, besides gravity in the vertical plane of motion, and upper bound on the lateral acceleration, is considered, in addition to high heading errors. It is shown that the same structure of the law that is proposed for the 2-D kinematic model can also be used for the 3-D realistic model, if the rest of the complexities of moving from 2-D space to 3-D space, and from kinematics to dynamics is duly addressed.
The guidance law proposed does not require time-to-go, the estimation of which can be a difficult problem in high heading error scenarios in which the closing velocity can be negative. Easy to compute and simple to implement in practice, the proposed law does not need any of the techniques or methods from classical optimal control theory, which are complicated and suffer from several limitations. The empirical pure PN (PPN) law is augmented with a term that is a polynomial function of the heading error. The values of the coefficients of the polynomial are found by using the DE. The computational effort required for this low dimensional polynomial optimization problem is shown to be low enough to enable online implementation in real-time.
The performance of the proposed law in nominal and off-nominal conditions is validated through several simulations for the 2-D kinematic model, and the 3-D realistic dynamic model. The results are compared with the PPN, augmented PPN and the all-aspect proportional navigation (AAPN) laws in the literature, as per several criteria like optimality, peak latax required and robustness to off-nominal conditions. A successful online implementation of the proposed law for application in practice is also demonstrated.
An obvious limitation of optimization by soft computation methods like differential evolution is that no rigorous proof of either convergence or optimality exists for such methods. A fallback option in the form of a conventional guidance law is included in the scheme in case of failure of convergence, and an indirect proof of optimality is provided in the third and final part of the thesis. The same guidance problem is solved by direct multiple shooting method, and it is shown that the numerical results of the two methods compare favourably. The solution by the shooting method is optimal, but computationally far more intensive and takes a computation time of an order of magnitude that is at least one or two times that of the simulation time of the plant. It also needs a good initial guess solution that lies within the region of convergence, which can be a difficult task by itself. Moreover, the shooting method solution is only open loop, and hence applicable for the given model and initial conditions only. Whereas, the simplicity of the method proposed in the thesis makes the solution or guidance law computable in a fraction of the flight time of the engagement, thereby making it online implementable. Equally important, is the fact that it is closed loop, and hence robust to off-nominal conditions like variations in the plant model and parameters assumed in its design.
27
Tseng, Ko-Ying, and 曾科穎. "Modified Differential Evolution for Structural Optimization." Thesis, 2010. http://ndltd.ncl.edu.tw/handle/31731599346536819939.
Full textAbstract:
博士大同大學
機械工程學系(所)
98
Differential evolution (DE) is a heuristic optimization method used to solve many optimization problems in real-valued search space. It has the advantage of incorporating a relatively simple and efficient form of mutation and crossover. In this study, two modified differential evolution algorithm, adaptive multi-population differential evolution and modified binary differential evolution, have been developed for dealing with different types of optimization problems. The adaptive multi-population differential evolution (AMPDE), including a proposed penalty-based self-adaptive strategy and multi-population mechanism, is developed in this study to enhance the performance of optimum search in truss structure optimization problems. Although the efficiency of proposed AMPDE is better than original DE and other population-based methods, it still has a difficulty in dealing with binary optimization problems due to the fact that the representation of design variable is a real-value type. In order to develop a differential evolution algorithm which can be suitable for both real-valued and binary optimization problems, a new modified binary differential evolution (MBDE) with a simple and new binary mutation mechanism based on a logical operation is proposed in this study. The developed MBDE is suitable for dealing with binary and real-valued optimization problems. Some numerical optimization problems, including test functions and a uniformity optimization of heat bonder, are first used to validate the correctness of architecture and performance of optimal search of the proposed MBDE algorithm. Different structural topology optimization problems are utilized to illustrate the high viability of the proposed algorithm in binary optimization problems. From the result of this study it is shown that the developed MBDE is suitable for dealing with real-valued and binary optimization problems. Besides, the proposed MBDE was observed to approach solutions better than those found in the references in the field of topology optimization of structures.
28
Rogalsky, Timothy P. "Acceleration of differential evolution for aerodynamic design." 2004. http://hdl.handle.net/1993/15752.
Full text
29
Kajee-Bagdadi, Zaakirah. "Differential evolution algorithms for constrained global optimization." Thesis, 2008. http://hdl.handle.net/10539/4733.
Full textAbstract:
In this thesis we propose four new methods for solving constrained global optimization problems.
The first proposed algorithm is a differential evolution (DE) algorithm using penalty
functions for constraint handling. The second algorithm is based on the first DE algorithm
but also incorporates a filter set as a diversification mechanism. The third algorithm is also
based on DE but includes an additional local refinement process in the form of the pattern
search (PS) technique. The last algorithm incorporates both the filter set and PS into the DE
algorithm for constrained global optimization. The superiority of feasible points (SFP) and
the parameter free penalty (PFP) schemes are used as constraint handling mechanisms.
The new algorithms were numerically tested using two sets of test problems and the
results where compared with those of the genetic algorithm (GA). The comparison shows
that the new algorithms outperformed GA. When the new methods are compared to each
other, the last three methods performed better than the first method i.e. the DE algorithm.
The new algorithms show promising results with potential for further research.
Keywords: constrained global optimization, differential evolution, pattern search, filter
method, penalty function, superiority of feasible points, parameter free penalty.
ii
30
Guo, Ching-Yi, and 郭青沂. "Improved Differential Evolution Algorithm and Its Applications." Thesis, 2008. http://ndltd.ncl.edu.tw/handle/gkkc6z.
Full textAbstract:
碩士國立高雄第一科技大學
系統資訊與控制研究所
96
In this thesis, we proposed two improved differential evolution (DE) algorithms, which are, respectively, called Taguchi-mutation differential evolution (TMDE) algorithm and Taguchi-crossover differential evolution (TCDE) algorithm. The proposed TMDE is a method of combining the tradition DE with the Taguchi method into mutation, and TCDE is a method of using sliding levels of Taguchi method into crossover. On the other hand, the biggest problem in finding optimal solutions is that we cannot know the range of searching. Hence, in this thesis, we incorporate a search space expansion scheme in the TMDE and TCDE, which are called adjustable Taguchi-mutation differential evolution (ATMDE) algorithm and adjustable Taguchi-mutation differential evolution (ATCDE) algorithm. There two methods make search space of TMDE and TCDE dynamical and promote the solution efficiency. The proposed methods are applied to three examples. The first example is to study the problem of model reduction. The second example is to investigate the problem of parameter identification for nonlinear systems. The third example is to research the optimization problem of surface grinding operations. Simulation results of the three examples show the presented of TMDE, TCDE, ATMDE and ATCDE better than traditional DE.
31
Chen, CHIH-YUNG, and 陳智勇. "Using Discrete Differential Evolution for Portfolio Optimization." Thesis, 2014. http://ndltd.ncl.edu.tw/handle/3767z5.
Full textAbstract:
碩士大同大學
資訊經營學系(所)
102
In the portfolio optimization problem, return and risk factors are considered at the same time, thus the problem is a multi-objective problem. When the number of stocks is large, the complexity of the problem becomes high and it consumes much more CPU time to solve it. Recently some researchers have utilized particle swarm optimization (PSO) and differential evolution (DE) to solve the portfolio optimization problem. Unfortunately, their CPU usage becomes high as the sizes of test problems increase. This study tries to overcome this problem. A discrete differential evolution (DDE) algorithm is proposed. The solution string consists of two sections: an integer -number section and a real-number section. The length of solution sections is equal to the cardinality number, rather than the number of stocks. Experimental results show that the design of solution string allows DDE to solve the portfolio optimization problem in a more efficient way.
32
Lee, Kuo-Ming, and 李國銘. "Engineering Optimization Using Improved Differential Evolution Algorithms." Thesis, 2011. http://ndltd.ncl.edu.tw/handle/42379422042200615432.
Full textAbstract:
博士國立高雄第一科技大學
工程科技研究所
99
For many practical engineering optimization problems, the exact design values of machining variables are not easy to be set because of both operating environmental noise and manufacturing variations. To overcome this difficulty, in this paper, two improved differential evolution approaches are proposed to solve several mixed-discrete-continuous problems and the surface grinding process under considering the tolerances of machining variables. The first approach is called as the Taguchi-sliding-based differential evolution algorithm (TSBDEA), where the Taguchi-sliding-level-method (TSLM) is applied to provide a new systematic crossover operation to replace the original crossover operation of differential evolution algorithm. Then, the systematic reasoning ability of the TSLM-based crossover operation with considering the influence of scaling factor is used to breed better individuals in order to generate the representative individuals to be the new potential offspring (or trial vectors). In addition, the second approach is named as Taguchi differential evolution outer array (TDEOA), which is extending the methodology in TSBDEA and introducing the variable tolerances in an outer array, is proposed to reduce the impact of variations of design variables. Furthermore, a distinct way has been introduced to estimate the reliability of a set of measurement from a surface grinding process, and demonstrate that the reliability of the proposed TDEOA approach can perform and maintain its functions in routine circumstances, as well as hostile or unexpected circumstances, where some specific tolerances are considered.
33
Hsu, Song-Lin, and 許松麟. "Model Reduction Using Taguchi-Differential Evolution Algorithm." Thesis, 2007. http://ndltd.ncl.edu.tw/handle/57441144799761042542.
Full textAbstract:
碩士國立高雄第一科技大學
系統與控制工程研究所
95
In this thesis, the problems of model reduction are solved by using two improved differential evolution algorithms (DEA), which are called Taguchi-differential evolution algorithm (TDEA) and adjustable Taguchi-differential evolution algorithm（ATDEA）. The use of a reduced-order model makes it easier to implement analyses, simulations and control designs. The proposed TDEA is modified from the traditional DEA with mutation operation. The systematic reasoning ability of the Taguchi method can promote the mutation efficiency. The Taguchi method applied into mutation operation and replaced the action of perturbed vectors were chosen randomly in the tradition DEA. TDEA can avoid premature convergence with controllable deteriorating probability. However, the biggest problem in optimal solution is that we can’t know the range of searching. Here, we incorporate a search space expansion scheme in the ATDEA approach and let search space dynamical. The efficiency will be promoted. It will also be shown that the TDEA and ATDEA perform batter than traditional DEA by using some benchmark test functions. Finally, we use TDEA and ATDEA to solve examples of model reduction. Simulation results show that the proposed TDEA and ATDEA approaches can obtain better performances than the traditional DEA and the existing DEA reported recently in the literature.
34
Su, Tse, and 蘇哲. "Segment message exchange based Differential Evolution Algorithm." Thesis, 2014. http://ndltd.ncl.edu.tw/handle/16951498417151116336.
Full textAbstract:
碩士亞東技術學院
資訊與通訊工程研究所
102
In general, different type of basic mutation strategies differential evolution will affect evolutionary directions directly. However, it cannot be avoided that all the basic mutation strategies may drive vectors form fall into local optimal solutions. In order to overcome this weakness of differential evolution, in this paper, the segment message exchange based differential evolution algorithm, which is with featuring sustained convergence, is proposed for solving single-objective optimization problems. In order to address the weaknesses of differential evolution, in this study, the mutation and crossover strategies of the basic-type differential evolution were modified to prevent premature convergence of the algorithm and from fall into local optimal solution. Furthermore, the dissolution mechanism is proposed. After several iterations, the solutions may become very similar. The dissolution mechanism will be activated to regenerate all the vectors except the best vectors will be retained. It will force algorithm to re-search the solution space. In the experiments, the CEC2005 test functions were adopted to test the proposed method and compare it with related methods which proposed in recent years. From the results, it can be observed that the proposed performs better in most test functions than other DE approaches, and exhibit signification improvement in solving specific problems.
35
HUANG, PO-JUNG, and 黃柏融. "Differential Evolution Algorithm with Winner Mutation Strategy." Thesis, 2013. http://ndltd.ncl.edu.tw/handle/46903997150840027616.
Full textAbstract:
碩士龍華科技大學
電機工程系碩士班
101
Inspired by 2-opt algorithm, this thesis proposes a new mutation strategy, namely winner mutation strategy, for differential evolution (DE) algorithm. The proposed winner mutation strategy could provide a different chance to find a better solution and avoid the algorithm trapping into local minimum. Such a DE algorithm is termed DE/winner hereafter. In addition, the linearly decreasing mutation factor and crossover rate are also applied to DE/winner so as to maintain both local and global search ability throughout the entire evolution process. DE/winner with linearly decreasing mutation factor and crossover rate could further improve the solution accuracy of DE algorithm and prevent the premature convergence in the evolution process. The proposed DE/winner is applied to solve the optimization problems of twelve unimodal and multimodal benchmark functions for demonstrating its search performance. Besides, DE/winner algorithm is also applied to optimize the parameters of proportional-integral-derivative (PID) controller. Simulation and experiment results demonstrate the search ability and control performance of DE/winner algorithm.
36
Xia, Yu-Ting, and 夏毓婷. "Digital System Design by Differential Evolution Algorithm." Thesis, 2018. http://ndltd.ncl.edu.tw/handle/c3n6j3.
Full textAbstract:
碩士樹德科技大學
電腦與通訊系碩士班
106
This thesis is to employ the differential evolution (DE) algorithm to solve for the parameter estimation of the digital systems. The DE algorithm is one of evolutionary computations, which imitates the evolutionary behavior of natural biology. It begins with generating many random parameter vectors, and then a fitness function corresponded to each of parameter vectors is evaluated. Based on these obtained values, the mutation, crossover, and selection operations are performed to derive better parameter vectors. Moreover, the advantage of the proposed algorithm is a real-value manipulation during the evolution, and it is very convenient and intuitional. In this thesis, the DE algorithm is utilized to deal with the parameter estimations for three different kinds of digital systems, including the infinite impulse response (IIR) system, nonlinear rational system, and bilinear system. All of numerical simulations confirm the applicability and effectiveness of the proposed DE algorithm.
37
Hasani, Shoreh Maryam. "Differential Evolution for Dynamic Constrained Continuous Optimisation." Thesis, 2020. http://hdl.handle.net/2440/129596.
Full textAbstract:
In this thesis, we choose the evolutionary dynamic optimisation methodology
to tackle dynamic constrained problems. Dynamic constrained problems
represent a common class of optimisation that occur in many real-world
scenarios. Evolutionary algorithms are often considered very general search
heuristics. Their main advantages (in comparison to problem-specific search
methods) are their robustness, flexibility and extensibility, as well as the fact
that almost no domain knowledge is required for their implementation and
application.
Our research is focused on the following areas. In the first part of the thesis,
we modify common constraint handling techniques from static domains to
suit dynamic environments. We investigate the deficiencies of such techniques
and the potential of each method based on the change characteristics of the environment.
In the second part, we propose a framework to create benchmarks,
since we have observed a lack of benchmarks to evaluate algorithms in dynamic
continuous optimisation. Third, we carry out an exhaustive empirical
study of diversity mechanisms applied to solve dynamic constrained optimisation
problems. Finally, we investigate the integration of a neural network
into the evolution process and analyse it’s effectiveness compared to that of
popular diversity mechanisms. We address the possibility of integrating such
mechanisms with a neural network approach in order to improve the results.Thesis (Ph.D.) -- University of Adelaide, School of Computer Science, 2020
38
Hsu, Zhan-Rong, and 許展榮. "A Modified Differential Evolution Algorithm with Activation Strategy." Thesis, 2009. http://ndltd.ncl.edu.tw/handle/24511459914337920395.
Full textAbstract:
碩士中原大學
資訊管理研究所
97
Evolutionary Computation (EC) provides high performance on real world optimization problems such as scheduling, resource distribution, portfolio optimization etc. Differential Evolution (DE) algorithm was first reported in 1996. Based on the characteristics of simple structure, high accuracy and efficiency, and the requirement of fewer parameters, DE has received significant attention from researchers. It has been applied to numerous fields and performs much better than other evolutionary computation. Although DE represents powerful performance, but it has attach importance to the drawbacks of unstable convergence, breakaway the solution space, and the common defect of evolution computation “dropping into regional optimum”. In this study, we attempt to improve the traditional differential evolution algorithm and propose a novel algorithm “Activated Strategy Differential Evolution” (ASDE). Based on import the Activated Strategy (AS) intensified the structure of traditional DE for balancing the solution accuracy and stability.
39
Adeyemo, Josiah Adetayo. "Application of differential evolution of water resources management." 2009. http://encore.tut.ac.za/iii/cpro/DigitalItemViewPage.external?sp=1001078.
Full textAbstract:
D.Tech. Civil EngineeringWater resources management is very important since the building of new water resources structures mainly large dams is very expensive and highly opposed. In this thesis, water resources management in the agricultural sector being the largest consumer of water through irrigation is studied. A water scarce country like South Africa should learn how to manage water efficiently to make every cubic meter of water used on the farm accountable. Profitability of irrigation water should be quantified in the same way as other economic products. There should be a change of attitude in the management of irrigation water. In South Africa, farmers are unlike other users in managing water efficiently. Studies show that other water users are more economical. Farmers need to plant more economical crops to justify the irrigation water supplied to them from the reservoirs since farmers are given priorities in the water allocation. In single objective optimization problem of water resources in crop planning, irrigation water use and optimization of planting areas are addressed in this study using differential evolution (DE) algorithm which is an evolutionary based algorithm. The algorithm proved to be computationally effective in solving these problems. The different areas of land where the crops are to be planted to maximize the total net benefit (TNB) in monetary terms are optimized. The ten strategies of differential evolution are tested with this problem. DE performs excellently for this problem. The results are further compared with those obtained by linear programming (LP). Both LP and DE obtain an income of ZAR 46.06 bn. The convergence speed of DE is effective and efficient. The effect of different combination of population size (NP), crossover constant (CR) and weighting factor (F) on the 10 different strategies of DE is studied. It is observed that Strategy 1, DE/rand-1-bin performs best for the problem with combination of NP, CR and F of 160, 0.95 and 0.5 respectively. An irrigation area divided into farmlands of 60 ha, 70 ha, 75 ha and 135 ha of land with irrigation water amounting to 9 143 m3/ha/annum is studied to find the optimum cropping pattern that can derive the highest TNB from farming. Three cropping patterns, A, B and C are suggested and tested with the model to find the best cropping pattern that will generate the highest total net benefit (TNB). The TNBs of ZAR 2 204 200, ZAR 2 645 100, ZAR 2 871 600 and ZAR 6 050 000 can be derived from planting on farmlands of 60 ha, 70 ha, 75 ha and 135 ha respectively using cropping pattern B in farmlands of 60 ha, 70 ha, 75 ha and cropping pattern C in farmland 135 ha. The net benefits per hectare are ZAR 36 737/ha, ZAR 37 787/ha, ZAR 38 288/ha and ZAR 44 815/ha for the cropping patterns with highest TNB. In this study, a new multi-objective evolutionary algorithm named multiobjective differential evolutionary algorithm (MDEA) is developed. Differential evolution (DE), an evolutionary algorithm (EA), known to be fast and robust in numerical optimization is extended to multi-objective problems. The new algorithm adjusts the selection scheme of traditional DE to solve multiobjective problems. The algorithm also modifies the domination criteria for the population. The algorithm produces more Pareto optimal solutions than the previous algorithms and retains the fast convergence and diversity exhibited by DE in global optimization. MDEA is applied to multi-objective water ix resources management namely, cropping pattern determination, crop planning, irrigation planning and reservoir operation. Multi-objective differential evolution algorithm (MDEA) technique is adapted to crop planning in a farmland in the Vaalharts irrigation scheme (VIS) in South Africa. The three objectives of the model are to maximize the total net profit (NF) in monetary terms (South African Rand, ZAR) generated on the farm by planting different crops, maximize total planting area (m2) and minimize the irrigation water use (m3). Vanderkloof dam along the Orange River is a multipurpose reservoir for flood control, irrigation, hydropower and recreation activities in that order of importance. The model in this study is adapted to the dam to determine the optimal monthly release for irrigation release requirements. From the study, an average cultivated area of 29 086 ha is suitable for the irrigation water available from the Vanderkloof dam through the canal. The total benefit that can be generated on the cultivated area is ZAR 947 000 000. It is sugested that 3 958 m3/ha/annum should be released to the famers in the area for optimum irrigation planning in the water deficient area against the 3 600 m3/ha/annum supplied presently. The model provides many alternative Pareto optimal solutions. MDEA proved exceptionally useful in water resources management especially in the Lower Orange catchments.
40
Chang, Ting-Kang, and 張庭綱. "FUZZY NEURAL NETWORK STUDY USING DIFFERENTIAL EVOLUTION ALGORITHM." Thesis, 2010. http://ndltd.ncl.edu.tw/handle/73101702481839796098.
Full textAbstract:
碩士大同大學
電機工程學系(所)
98
A differential evolution (DE) algorithm based fuzzy neural network (FNN) (DEFNN) controller is proposed in this thesis. DEFNN controller is composed of an FNN identifier, a DE estimator, a computation controller, and a hitting controller. There are two main learning phases in DEFNN controller – the training phase and the online phase. The training phase is utilized to find the best preset parameters of DEFNN controller. In this thesis, several parameters such as the learning rates of the back-propagation (BP) algorithm, the parameters of error term which are used in BP algorithm, the initial values of the FNN identifier and some preset parameters of DEFNN controller are provided by DE estimator. After the best preset parameters are obtained, DEFNN controller will be active online. In the online phase, the FNN identifier is used to identify the unknown terms of the nonlinear system dynamic. The BP algorithm is adopted to update the parameters of the FNN identifier to achieve favorable approximation performance. Then the computation controller is designed to calculate the outputs of the FNN identifier. Finally, the hitting control which is utilized to eliminate the uncertainties and external disturbances of the nonlinear system combine with the output of computation controller to form the main control effort. The results of the simulations are implemented to verify the effectiveness of the proposed controller.
41
Wu, Jian-Kuan, and 吳建寬. "A Differential Evolution Approach for Machine Cell Formation." Thesis, 2008. http://ndltd.ncl.edu.tw/handle/94803362160942084975.
Full textAbstract:
碩士大同大學
資訊工程學系(所)
96
This paper presents a new approach based on differential evolution algorithms to solve cell formation problems. The proposed approach handles the problem in a way of data bi-clustering and can form machine cells and part families concurrently. Differential evolution is simple to implement and has fewer parameters needed to set. The proposed approach applies differential evolution to find machine cluster centers and part cluster centers at the same time. Thus the approach can form machine cells and their corresponding part families automatically. A number of test problems had been selected from literature and the experimental results reveal that the proposed approach is able to solve cell formation problems effectively.
42
Wu, Huang-Lyu, and 吳皇履. "Adaptive Differential Evolution Algorithm with High Diversity Population." Thesis, 2014. http://ndltd.ncl.edu.tw/handle/53047488535929297577.
Full textAbstract:
碩士亞東技術學院
資訊與通訊工程研究所
102
This paper proposed an adaptive differential evolution algorithm with high diversity population (ADE-HP). The proposed method can increase diversity of population and increase vectors’ searching ability for solving single-objective numerical optimization problems. In order to increase diversity of population in original DE, several mechanisms are proposed. First, Elitist mechanism can avoid vectors are guided to the same position (global best particle) and can prevent vectors form fall into local optimum even early convergence. Second, Real rand mechanism can give higher ability to jump out from local optimum and provide varied information to help particles toward to potential unsearched solution space for solution exploration. Finally, in order to increase vectors’ explore probability, the partial crossover mechanism is proposed. 25 test functions of CEC 2005 were adopted for experiments through a reasonable average and fitness evaluations. From the results, it can be observed that the proposed method can efficiently find better solutions than recent DE works for solving optimization problems.
43
Hsueh, Feng, and 薛凡. "MapReduce-based Discrete Differential Evolution for Portfolio Optimization." Thesis, 2015. http://ndltd.ncl.edu.tw/handle/39225599523554477678.
Full textAbstract:
碩士大同大學
資訊經營學系(所)
103
The portfolio optimization (PO) problem is a multi-objective combinatorial optimization which needs to consider expected return and risk at the same time. Thus it needs an efficient computing way to find optimal solutions. Differential Evolution is a meta-heuristic algorithm and can be applied to the PO problem for finding optimal solutions. When we think of the PO problem, the more choices of stocks are, the more complicated the problem is. Therefore, the computation time on a single machine surprisingly increases. The solution to this computational problem is to use Hadoop. MapReduce is a key component of Hadoop and can be applied to solving many optimization problems. This study proposes a discrete differential evolution algorithm for solving the PO problem, and the proposed algorithm is implemented by using the MapReduce framework run on multiple machines in order to obtain optimal solutions within a shorter time. Experimental results show that using the MapReduce framework to implement the discrete differential algorithm can solve larger PO problems more quickly than on a single machine.
44
Pang-HanHsu and 徐邦瀚. "Improving Constraint-activated Differential Evolution with Escape Vectors." Thesis, 2015. http://ndltd.ncl.edu.tw/handle/17573304194808320028.
Full textAbstract:
碩士國立成功大學
資訊工程學系
103
In system design, the best system designed under a simple experimental environment may not be suitable for application in real world if dramatic changes caused by uncertainties contained in the real world are considered. To deal with the problem caused by uncertainties, designers should try their best to get the most robust solution. The most robust solution can be obtained by constrained min-max optimization algorithms. In this thesis, the scheme of generating escape vectors has been proposed to solve the problem of premature convergence of differential evolution. After applying the proposed scheme to the constrained min-max optimization algorithm, the performance of the algorithm could be greatly improved. To evaluate the performance of constrained min-max optimization algorithms, more complex test problems have also been proposed in this thesis. Experimental results show that the improved constrained min-max optimization algorithm is able to achieve a 100% success rate on all considered test problems under limited accuracy.
45
Thuy, N. T. P., R. Pendyala, Nejat Rahmanian, and N. Marneni. "Heat exchanger network optimization by differential evolution method." 2014. http://hdl.handle.net/10454/18554.
Full textAbstract:
NoThe synthesis of heat exchanger network (HEN) is a comprehensive approach to optimize energy utilization in process industry. Recent developments in HEN synthesis (HENS) present several heuristic methods, such as Simulated Annealing (SA), Genetic Algorithm (GA), and Differential Evolution (DE). In this work, DE method for synthesis and optimization of HEN has been presented. Using DE combined with the concept of super-targeting, the ΔTmin optimization is determined. Then DE algorithm is employed to optimize the global cost function including the constraints, such as heat balance, the temperatures of process streams. A case study has been optimized using DE, generated structure of HEN and compared with networks obtained by other methods such as pinch technology or mathematical programming. Through the result, the proposed method has been illustrated that DE is able to apply in HEN optimization, with 16.7% increase in capital cost and 56.4%, 18.9% decrease in energy, global costs respectively.
46
Lee, Yi-Fan, and 李翼帆. "Differential Evolution Algorithm with Mean-best Mutation Strategy." Thesis, 2015. http://ndltd.ncl.edu.tw/handle/75484722005426600118.
Full textAbstract:
碩士龍華科技大學
電機工程系碩士班
103
This thesis proposes a novel mutation strategy, termed mean-best mutation strategy to preserve the diversity of the population and prevent the premature convergence of differential evolution (DE) algorithm, where “mean-best” represents the mean of some top-best vectors. The DE variant with the proposed mutation strategy is termed differential evolution with mean-best mutation strategy and denoted by DE/mbest/1 or DE/mbest/2 hereafter. The results of global optimization problems, PID controller designs and system identifications show that DE/mbest/1 and DE/mbest/2 is comparable to or better than the other evolution algorithms in terms of accuracy reliability and efficiency.
47
Du, Plessis M. C. (Mathys Cornelius). "Adaptive multi-population differential evolution for dynamic environments." Thesis, 2012. http://hdl.handle.net/2263/28211.
Full textAbstract:
Dynamic optimisation problems are problems where the search space does not remain constant over time. Evolutionary algorithms aimed at static optimisation problems often fail to effectively optimise dynamic problems. The main reason for this is that the algorithms converge to a single optimum in the search space, and then lack the necessary diversity to locate new optima once the environment changes. Many approaches to adapting traditional evolutionary algorithms to dynamic environments are available in the literature, but differential evolution (DE) has been investigated as a base algorithm by only a few researchers. This thesis reports on adaptations of existing DE-based optimisation algorithms for dynamic environments. A novel approach, which evolves DE sub-populations based on performance in order to discover optima in an dynamic environment earlier, is proposed. It is shown that this approach reduces the average error in a wide range of benchmark instances. A second approach, which is shown to improve the location of individual optima in the search space, is combined with the first approach to form a new DE-based algorithm for dynamic optimisation problems. The algorithm is further adapted to dynamically spawn and remove sub-populations, which is shown to be an effective strategy on benchmark problems where the number of optima is unknown or fluctuates over time. Finally, approaches to self-adapting DE control parameters are incorporated into the newly created algorithms. Experimental evidence is presented to show that, apart from reducing the number of parameters to fine-tune, a benefit in terms of lower error values is found when employing self-adaptive control parameters.Thesis (PhD)--University of Pretoria, 2012.
Computer Science
unrestricted
48
Chen, Ya-Chi, and 陳雅祺. "A Differential Evolution Approach to Portfolio Optimization Problems." Thesis, 2012. http://ndltd.ncl.edu.tw/handle/17848569889518895663.
Full textAbstract:
碩士大同大學
資訊經營學系(所)
100
Because of the impact of the financial tsunami and economic recession, it is easy for the investors to get huge losses in the stock market. Portfolio optimization is a multi-objective problem in which we expect to get high expected returns and low risks. To solve the problem an efficient heuristic algorithm must be used to obtain optimal solution. The objective of this paper is to use differential evolution (DE) for portfolio optimization. Compared to other heuristic algorithms, differential evolution has three advantages: fewer parameters, fast convergence and easy implementation. The experimental results show that differential evolution has good performance in solution quality and computation time. We also show that differential evolution is an effective approach to solve portfolio optimization problems.
49
Lin, Shu-Yan, and 林書延. "Adjustable Proportional Distribution for Multi-Objective Differential Evolution." Thesis, 2012. http://ndltd.ncl.edu.tw/handle/81815457574789297260.
Full textAbstract:
碩士國立東華大學
電機工程學系
100
Recently, Multi-Objective Differential Evolution (MODE), powerful and efficient population-based stochastic processing, has become an indispensable algorithm for solving numerical optimization problems widely. It is found in various benchmark functions that traditional MODE is unable to search global optimal completely, falling into local optimal because only using one strategy to search global optimal. This paper proposes “Adjustable Proportional Distribution” (APD) for this problem. The proposed APD can combine several strategies to search Global optimal. It calculates proportions of each strategy in external archive and then uses Taguchi method to select the best proportion in evolution. In next iteration, it selects the best proportion to adjust particles size and scale factor F used in each strategy according to Taguchi method. Experience proves that APD can improve the external archive diversity of solutions and find global optimal more effectively.
50
LU, WEN-CHIA, and 呂旻珈. "Optimal Chiller Loading Using Modified Differential Evolution Algorithm." Thesis, 2018. http://ndltd.ncl.edu.tw/handle/5ht9k8.
Full textAbstract:
碩士大同大學
機械工程學系(所)
106
The aims of this study is to develop an algorithmic approach for solving the problem of optimal chiller loading by using a modified differential evolution algorithm. The objective is to minimize the power consumption of multiple chiller system with specific cooling loads. In order to balance exploration and exploitation of optimization search, three mutation mechanisms with different limits of mutation rate F and crossover rate CR are used in modified differential evolution algorithm of this study. The robustness of developed algorithm will be improved to avoid the influence of selection values of constant F and CR. For minimization of power consumption, some parametric adjustment mechanisms are also applied in the algorithm to modify parametric values according to the results of optimization search. Finally optimal solutions of reference papers for different cooling loads are used to analyze effects of adjustment mechanisms and modification of parameters of developed algorithm applied in optimal chiller loading problem.