Search

In science and engineering, mathematical modeling serves as a tool to understand processes and systems acting as a testing bed for several hypotheses. The selection of a specific model, as well as its variables and parameters, depends on the nature of the system under analysis and the acceptable simplifying assumptions. Therefore, it must allow for a good fit between both the hypothesis and the available data. Opposite to other design approaches based on experimental data or/and complex models, this work presents a simpler numerical design method for efficiency maximization of an Hydraulic Jet Pump (HJP) for oil-well extraction process, considering its hydraulic and geometric parameters. The design process consists in setting and solving a constrained non-linear optimization problem by taking into account the hydraulic model of the HJP in terms four design variables: throat area, nozzle area, injection flow, and injection pressure to the oil-well. The objective function of this case aims to maximize the HJP's efficiency avoiding to approach cavitation condition as well fulfilling technical constraints. A numerical technique, Differential Evolution Algorithm (DEA), has been implemented to solve the optimization problem. The proposed methodology leads to a solution set by considering only commercial geometries and feasible operating conditions for the HJP, which facilitates its practical implementation. A set of ten oil-wells with land production data, operating in the southeaster of Mexico, is used to compare and validate several Jet pump designs, i. e., through comparison with actual oil-well's operation condition., Peer Reviewed

This paper describes optimisation method for designing sensoric layout for Active Structural Health Monitoring (A-SHM) by Ultrasonic Guided Waves (UGW) on metal and non-metal (composite) materials. The SHM sensors need to be placed optimally in order to detect structural damage with high probability before the damage turns critical. Configuration of used optimisation algorithm for such task is not straightforward. Differential Evolution (DE) has two configuration parameters – the mutation factor F and the crossover rate CR – whose settings largely influence the solution quality the optimisation process can yield. For that matter we describe an elaborated a method to guide this selection towards good results using visual heat maps with the intent to select best DE’s variant and particular configuration to receive the most optimal SHM sensorics layout.

This research paper analyses an external archive of inferior solutions used in Success-History based Adaptive Differential Evolution (SHADE) and its variant with a linear decrease in population size L-SHADE. A novel implementation of an archive is proposed and compared to the original one on CEC2015 benchmark set of test functions for two distinctive dimensionality settings. The proposed archive implementation is referred to as Enhanced Archive (EA) and therefore two Differential Evolution (DE) variants are titled EA-SHADE and EA-L-SHADE. The results on CEC2015 benchmark set are analyzed and discussed. © Springer Nature Switzerland AG 2019.

This research deals with the hybridization of two computational intelligence fields, which are the chaos theory and evolutionary algorithms. Experiments are focused on the extensive investigation on the different randomization schemes for selection of individuals in differential evolution algorithm (DE). This research is focused on the hypothesis whether the different distribution of different pseudo-random numbers or the similar distribution additionally enhanced with hidden complex chaotic dynamics providing the unique sequencing are more beneficial to the heuristic performance. This paper investigates the utilization of the two-dimensional discrete chaotic systems, which are Burgers and Lozi maps, as the chaotic pseudo-random number generators (CPRNGs) embedded into the DE. Through the utilization of either chaotic systems or equal identified pseudo-random number distribution, it is possible to entirely keep or remove the hidden complex chaotic dynamics from the generated pseudo random data series. This research utilizes set of 4 selected simple benchmark functions, and five different randomizations schemes; further results are compared against canonical DE. © Springer Nature Switzerland AG 2019.

This paper provides a closer insight into applicability and performance of the hybridization of symbolic regression open framework, which is Analytical Programming (AP) and Differential Evolution (DE) algorithm in the task of time series regression. AP can be considered as a robust open framework for symbolic regression thanks to its usability in any programming language with arbitrary driving metaheuristic. The motivation behind this research is to explore and investigate the applicability and differences in performance of AP driven by basic canonical entirely random or best solution driven mutation strategies of DE. An experiment with four case studies has been carried out here with the several time series consisting of GBP/USD exchange rate. The differences between regression/prediction models synthesized using AP as a direct consequence of different DE strategies performances are statistically compared and briefly discussed in conclusion section of this paper. © 2019, Springer International Publishing AG, part of Springer Nature.

This work studied a relationship between optimization qualities of Success-History based Adaptive Differential Evolution algorithm (SHADE) and its self-adaptive parameter strategy. Original SHADE with improvement based adaptation is compared to the SHADE with Distance based parameter adaptation (Db_SHADE) on the basis of the CEC2015 benchmark set for continuous optimization and a novel approach combining both distance and improvement adaptation (DIb_SHADE) is presented and tested as a trade-off between both approaches. © 2019, Springer International Publishing AG, part of Springer Nature.

Many key problems in science and engineering can be formulated and solved using global optimization techniques. In the particular case of computational biology, the development of dynamic (kinetic) models is one of the current key issues. In this context, the problem of parameter estimation (model calibration) remains as a very challenging task. The complexity of the underlying models requires the use of efficient solvers to achieve adequate results in reasonable computation times. Metaheuristics have been the focus of great consideration as an efficient way of solving hard global optimization problems. Even so, in most realistic applications, metaheuristics require a very large computation time to obtain an acceptable result. Therefore, several parallel schemes have been proposed, most of them focused on traditional parallel programming interfaces and infrastructures. However, with the emergence of cloud computing, new programming models have been proposed to deal with large-scale data processing on clouds. In this paper we explore the applicability of these new models for global optimization problems using as a case study a set of challenging parameter estimation problems in systems biology. We have developed, using Spark, an island-based parallel version of Differential Evolution. Differential Evolution is a simple population-based metaheuristic that, at the same time, is very popular for being very efficient in real function global optimization. Several experiments were conducted both on a cluster and on the Microsoft Azure public cloud to evaluate the speedup and efficiency of the proposal, concluding that the Spark implementation achieves not only competitive speedup against the serial implementation, but also good scalability when the number of nodes grows. The results can be useful for those interested in using parallel metaheuristics for global optimization problems benefiting from the potential of new cloud programming models.

Currently, an increasing amount of debris is floating in Low Earth Orbit (LEO) endangering space operations. To limit the contribution of non-operational satellites to the space debris problem a semi-controlled end-of-life de-orbit strategy can be employed. In this strategy the satellite is controlled from the end-of-life orbit down to low altitudes (120 to 200 km). From this altitude the satellite follows a ballistic trajectory towards the surface of the Earth. The target of such maneuver is to phase the impact probability track, with the Earth, such that the resulting impact track covers mostly uninhabited areas. The objective of this thesis, is to provide insights into the effects of assumptions made in recent research of the semi-controlled de-orbit strategy. Furthermore, the extent to which the casualty risk can be reduced is investigated. Two optimization strategies are carried out. First, the control profile for the complete semi-controlled de-orbit strategy is optimized for minimum casualty risk using a differential evolution algorithm. Second, the initial state of the ballistic trajectory, at the point where the controls are turned off, is also optimized for minimum casualty risk using a differential evolution algorithm. In case similar impact tracks are achieved for both optimizations, future analysis of the semi-controlled de-orbit strategy can be performed which less computational effort. Employing a semi-controlled low-thrust de-orbit strategy can reduce the casualty risk with 2 to 3 orders of magnitude with respect to an uncontrolled de-orbit strategy. Large ballistic coefficients are beneficial for the reduction of the casualty risk. Furthermore, orbits with high inclinations results in larger reductions of the casualty risk with respect to low inclination orbits. Also the results for low inclination orbits are less sensitive to errors in the re-entry state caused by imperfections in the GNC system of the satellite during the de-or, Aerospace Engineering | Astrodynamics & Space Missions

Comets, the sporadic visitors from the outer edges of the Solar System, are considered to hold the key for understanding the formation of planets and the origin of life on Earth. Having spent the majority of time away from the radiative environment of the inner Solar System, the chemistry of the comets has remained unaltered, making them the pristine samples of the matter from the ancient Solar nebula. A mission to bring cometary particles back to Earth enables the examination of the materials in well equipped laboratories and saves the mass of the instruments to be carried on board. As conventional propulsion methods require a large quantity of propellant for this type of mission, the feasibility of using the novel propulsion technique of solar sailing is explored in this thesis. In order to return the comet samples to Earth within a reasonable time period, the orbit transfer is considered as an optimal control problem with constraints placed on the sailcraft’s position and velocity. The Differential Evolution (DE) algorithm was used to search for time-optimal trajectories that minimize the approach distance and the relative velocity with respect to the comet during sample collection. The optimal trajectory obtained predicts the solar sailcraft to reach the comet, collect the samples and return back to Earth in 6.8 years. The time of arrival at the comet was found to match with the comet's perihelion passage, enabling effective sample collection. The outcome of the trajectory analysis, thus successfully demonstrates the applicability of solar sailing to comet sample return missions in the near future., Aerospace Engineering | Astrodynamics & Space Missions

Block Matching is one of the most efficient techniques for motion estimation for video sequences. Metaheuristic algorithms have been used effectively for motion estimation. In this paper, we propose two hybrid algorithms: Artificial Bee Colony with Differential Evolution and Harmony Search with Differential Evolution based motion estimation algorithms. Extensive experiments are conducted using four standard video sequences. The video sequences utilized for experimentation have all essential features such as different formats, resolutions and number of frames which are generally required in input video sequences. We compare the performance of the proposed algorithms with other algorithms considering various parameters such as Structural Similarity, Peak Signal to Noise Ratio, Average Number of Search Points etc. The comparative results demonstrate that the proposed algorithms outperformed other algorithms.

This paper discusses a set of façade design alternatives for form-finding problem focusing on conceptual phase. In this respect, the aim of the research is to propose a multi-objective optimization approach for a façade design of public pool building. We present a set of solution belonging to Self-adaptive Multi-objective Ensemble Differential Evolution (JE_DEMO) and Self-adaptive Multi-Objective Differential Evolution (JDEMO) algorithm. We focus on maximization of daylight performance and minimization of structural displacement. Based on results, two algorithms presented competitive results. Contributions are presented based on objectives functions as new trade-offs and proposed JE_DEMO algorithm for design problems., Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public., Design Informatics

This research represents a detailed insight into the modern and popular hybridization of deterministic chaotic dynamics and evolutionary computation. It is aimed at the influence of chaotic sequences on the performance of four selected Differential Evolution (DE) variants. The variants of interest were: original DE/Rand/1/ and DE/Best/1/ mutation schemes, simple parameter adaptive jDE, and the recent state of the art version SHADE. Experiments are focused on the extensive investigation of the different randomization schemes for the selection of individuals in DE algorithm driven by the nine different two-dimensional discrete deterministic chaotic systems, as the chaotic pseudorandom number generators. The performances of DE variants and their chaotic/non-chaotic versions are recorded in the one-dimensional settings of 10D and 15 test functions from the CEC 2015 benchmark, further statistically analyzed. © 2018, Brno University of Technology. All rights reserved.

This paper deals with the Success-History based Adaptive Differential Evolution with Linear decrease in population size and Distance based adaptation of control parameters (DbL-SHADE). The algorithm is subject to the analysis of the population clustering via Density-Based Spatial Clustering of Applications with Noise (DBSCAN). The CEC2015 benchmark set in three dimensional settings (10D, 30D and 50D) was used as a testbed for comparison of the population clustering between L-SHADE and DbL-SHADE algorithms and the results are reported and discussed. © 2018 IEEE.

This research focuses on the modern hybridization of the discrete chaotic dynamics and the evolutionary computation. It is aimed at the influence of chaotic sequences on the population diversity as well as at the algorithm performance of the simple parameter adaptive Differential Evolution (DE) strategy: jDE. Experiments are focused on the extensive investigation of totally ten different randomization schemes for the selection of individuals in DE algorithm driven by the default pseudo random generator of Java environment and nine different two-dimensional discrete chaotic systems, as the chaotic pseudo-random number generators. The population diversity and jDE convergence are recorded for 15 test functions from the CEC 2015 benchmark set in 30D. © Springer International Publishing AG, part of Springer Nature 2018.

This research deals with the modern and popular hybridization of chaotic dynamics and evolutionary computation. Unlike many research studies on the combination of chaos and metaheuristics, this paper focuses on the deeper insight into the population dynamics, specifically influence of chaotic sequences on the population diversity. The optimization algorithm performance was recorded as well. Experiments are focused on the extensive investigation of the different randomization schemes for the selection of individuals in a simple parameter adaptive Differential Evolution (DE) strategy: JDE algorithm. The jDE was driven by the nine different two-dimensional discrete chaotic systems, as the chaotic pseudo-random number generators. The population diversity and jDE convergence are recorded on the two-dimensional settings (10D and 30D) and 15 test functions from the CEC 2015 benchmark. © 2018 IEEE.

In this paper, an analysis of a distance based parameter adaptation in Success-History based Differential Evolution (SHADE) is presented in order to show that it can have a beneficial effect on the premature convergence of the algorithm. The premature convergence of SHADE is an issue mainly in higher dimensional decision spaces. Therefore, the tests are done on the basis of CEC2015 benchmark in 10D, 30D and 50D. The results are provided and discussed. © 2018 IEEE.

This paper provides an analysis of the population clustering in a novel Success-History based Adaptive Differential Evolution algorithm with Distance based adaptation (Db_SHADE) in order to analyze the exploration and exploitation abilities of the algorithm. The comparison with the original SHADE algorithm is performed on the CEC2015 benchmark set in two dimensional settings (10D and 30D). The clustering analysis helps to answer the question about prolonged exploration phase of the Db_SHADE algorithm. Possible future research directions are drawn in the discussion and conclusion. © 2018, Brno University of Technology. All rights reserved.

This research deals with a novel approach to classification - pseudo neural networks (PNN). This technique was inspired in classical artificial neural networks (ANN), where a relation between inputs and outputs is based on the mathematical transfer functions and optimised numerical weights. Compared to ANN, the whole structure in PNN, i.e. the relation between inputs and output(s), is fully synthesised by evolutionary symbolic regression tool - analytic programming. Compared to previous synthesised models, the PNN in this paper were synthesised via a new approach to constant estimation inside the analytic programming - direct coding. Iris data was used for the experiments and PNN were used for the synthesis of a complex classifier for more classes. For experimentation, Differential Evolution (de/rand/1/bin) for optimisation in analytic programming (AP) was used. © ECMS Lars Nolle, Alexandra Burger, Christoph Tholen, Jens Werner, Jens Wellhausen

This research deals with the modern and popular hybridization of chaotic dynamics and evolutionary computation. It is aimed at the influence of chaotic sequences on the population diversity as well as the algorithm performance of the simple parameter adaptive Differential Evolution (DE) strategy: jDE. Experiments are focused on the extensive investigation of the different randomization schemes for the selection of individuals in DE algorithm driven by the nine different two-dimensional discrete chaotic systems, as the chaotic pseudo-random number generators. The population diversity and jDE convergence are recorded on the 15 test functions from the CEC 2015 benchmark. © 2018, Springer International Publishing AG, part of Springer Nature.

This article presents a comparative study of the classification of Elliott waves in data. Regarding the methods of classification, the paper deals with three approaches. The first one is a multilayer artificial neural network (ANN) with sigmoid activation function and backpropagation and Levenberg–Marquardt training algorithm. Second approach uses training algorithms of ANN but forms of activation functions of hidden nodes and nodes in output layers have been proposed by analytical programming with the differential evolution. The last approach offers results performed by synthesized pseudo neural networks where the symbolic regression is used for synthesis of a whole structure of the classifier, i.e., the relation between inputs and output(s) similar to ANN. In this case, meta-evolution version of analytic programming with differential evolution is used. In conclusion, all results of this experimental study were evaluated and compared mutually. © 2016, Springer-Verlag Berlin Heidelberg.

In this paper, a comparative study of seven variants of the Success-History based Adaptive Differential Evolution (SHADE) algorithm is presented with the aim to show the influence of improvement and distance based parameter adaptation to its performance. Seven algorithms range from improvement only based SHADE through a balanced improvement and distance based DIb_SHADE to distance only based Db_SHADE. The algorithm set is compared on the CEC2015 benchmark set, results are statistically tested, commented and possible future research directions are provided. © ECMS Lars Nolle, Alexandra Burger, Christoph Tholen, Jens Werner, Jens Wellhausen

This research deals with the constrained industrial optimization task, which is the optimization of technological parameters for the waste processing batch reactor. This paper provides a closer insight into the performance of connection between constrained optimization and different strategies of Differential Evolution (DE). Thus, the motivation behind this research is to explore and investigate the differences in performance of basic canonical strategies of DE as well as by the state of the art strategy, which is Success-History based Adaptive Differential Evolution (SHADE). The simple experiment has been carried out here and 30 times repeated. Consequences of different DE strategies performances are briefly discussed within conclusion section of this paper. © Springer International Publishing AG 2018.

This research deals with the modern and popular hybridization of chaotic dynamics and evolutionary computation. It is aimed at the influence of chaotic sequences on the performance of four selected Differential Evolution (DE) variants. The variants of interest were: original DE/Rand/1/ and DE/Best/1/ mutation schemes, simple parameter adaptive jDE, and the recent state of the art version SHADE. Experiments are focused on the extensive investigation of the different randomization schemes for the selection of individuals in DE algorithm driven by the nine different two-dimensional discrete chaotic systems, as the chaotic pseudo-random number generators. The performances of DE variants and their chaotic/non-chaotic versions are recorded in the one-dimensional settings of 10 $D$ and 15 test functions from the CEC 2015 benchmark. © 2018 IEEE.

The paper deals with the discovery of trigonometric identities of four functions via Analytic Programming and four strategies of Differential evolution (canonical DE/Rand/l/Bin, chaos-based DE/Rand/l/Bin-Lozi, DE/Rand/l/Bin-Burgers and SHADE). The results showed that all four strategies were comparable for this specific task. © 2018 IEEE.

Many real-world control and classification tasks involve a large number of features. When artificial neural networks (ANNs) are used for modeling these tasks, the network architectures tend to be large. Neuroevolution is an effective approach for optimizing ANNs; however, there are two bottlenecks that make their application challenging in case of high-dimensional networks using direct encoding. First, classic evolutionary algorithms tend not to scale well for searching large parameter spaces; second, the network evaluation over a large number of training instances is in general time-consuming. In this work, we propose an approach called the Limited Evaluation Cooperative Co-evolutionary Differential Evolution algorithm (LECCDE) to optimize high-dimensional ANNs. The proposed method aims to optimize the pre-synaptic weights of each post-synaptic neuron in different subpopulations using a Cooperative Co-evolutionary Differential Evolution algorithm, and employs a limited evaluation scheme where fitness evaluation is performed on a relatively small number of training instances based on fitness inheritance. We test LECCDE on three datasets with various sizes, and our results show that cooperative co-evolution significantly improves the test error comparing to standard Differential Evolution, while the limited evaluation scheme facilitates a significant reduction in computing time.

In this work, a new approach for aerodynamic optimization of horizontal axis wind turbine (HAWT) airfoil is presented. This technique combines commercial computational fluid dynamics (CFD) codes with differential evolution (DE), a reliable gradient-free global optimization method. During the optimization process, commercial CFD codes are used to evaluate aerodynamic characteristics of HAWT airfoil and an improved DE algorithm is utilized to find the optimal airfoil design. The objective of this research is to maximize the aerodynamic coefficients of HAWT airfoil at the design angle of attack (AOA) with specific ambient environment. The airfoil shape is modeled by control points which their coordinates are design variables. The reliability of CFD codes is validated by comparing the analytical results of a typical HAWT airfoil with its experimental data. Finally, the optimal design of wind turbine airfoil is evaluated about aerodynamic performance in comparison with existing airfoils and some discussions are performed.

A single-loop deterministic method (SLDM) has previously been proposed for solving reliability-based design optimization (RBDO) problems. In SLDM, probabilistic constraints are converted to approximate deterministic constraints. Consequently, RBDO problems can be transformed into approximate deterministic optimization problems, and hence the computational cost of solving such problems is reduced significantly. However, SLDM is limited to continuous design variables, and the obtained solutions are often trapped into local extrema. To overcome these two disadvantages, a global single-loop deterministic approach is developed in this article, and then it is applied to solve the RBDO problems of truss structures with both continuous and discrete design variables. The proposed approach is a combination of SLDM and improved differential evolution (IDE). The IDE algorithm is an improved version of the original differential evolution (DE) algorithm with two improvements: a roulette wheel selection with stochastic acceptance and an elitist selection technique. These improvements are applied to the mutation and selection phases of DE to enhance its convergence rate and accuracy. To demonstrate the reliability, efficiency and applicability of the proposed method, three numerical examples are executed, and the obtained results are compared with those available in the literature.

[EN] A fuzzy rule-based system combining empirical data on hydraulic preferences and literature information on temperature requirements was used to foresee the brown trout (Salmo trutta L.) spawning habitat degradation induced by climate change. The climatic scenarios for the Cabriel River (Eastern Iberian Peninsula) corresponded to two Representative Concentration Pathways (4.5 and 8.5) for the short (2011¿2040) and mid (2041¿2070) term horizons. The hydraulic and hydrologic modelling were undertaken with process-based numerical models (i.e., River2D© and HBV-light) while the water temperature was modelled by assembling the predictions of three machine learning techniques (M5, Multi-Adaptive Regression Splines and Support Vector Regression). The predicted rise in the water temperature will not be compensated by the more benign lower flows. Consequently, the suitable spawning habitat will be reduced between 15.4¿48.7%. The entire population shall suffer the effects of climate change and will probably be extirpated from the downstream segments of the river.

Mixed Weibull distribution is a probability distribution noted for its wide applicability in many diverse fields. The ability to accurately estimate mixed distribution parameters is essential for data-driven modeling, simulation, and analysis of the phenomena represented by mixed Weibull models. Nature-inspired metaheuristics for continuous parameter optimization have shown good potential for approximating parameters of complex statistical models. Differential evolution is a popular evolutionary real-parameter optimization method with good results in many areas. This work uses differential evolution to fit mixed Weibull distribution to data and analyzes the ability of different differential evolution variants to estimate mixture parameters., Smíšená distribuce Weibull rozdělení je známá pro svou širokou použitelnost v mnoha rozmanitých oblastech. Schopnost přesně odhadnout smíšené distribuční parametry je nezbytná pro modelování, simulaci a analýzu jevů reprezentovaných temito smíšenými Weibullovými modely. Přírodou inspirovaná metaheuristika pro kontinuální optimalizaci parametrů ukázala dobrý potenciál pro aproximaci parametrů složitých statistických modelů. Diferenciální Evoluce (DE) je populární evoluční metoda optimalizace skutečných parametrů s dobrými výsledky v mnoha oblastech. Tato práce využívá metodu DE k přesnému proložení dat směsí Weibull rozdělení a analyzuje schopnost různých DE variant odhadnout směsového modelu.

Currently, an increasing amount of debris is floating in Low Earth Orbit (LEO) endangering space operations. To limit the contribution of non-operational satellites to the space debris problem a semi-controlled end-of-life de-orbit strategy can be employed. In this strategy the satellite is controlled from the end-of-life orbit down to low altitudes (120 to 200 km). From this altitude the satellite follows a ballistic trajectory towards the surface of the Earth. The target of such maneuver is to phase the impact probability track, with the Earth, such that the resulting impact track covers mostly uninhabited areas. The objective of this thesis, is to provide insights into the effects of assumptions made in recent research of the semi-controlled de-orbit strategy. Furthermore, the extent to which the casualty risk can be reduced is investigated. Two optimization strategies are carried out. First, the control profile for the complete semi-controlled de-orbit strategy is optimized for minimum casualty risk using a differential evolution algorithm. Second, the initial state of the ballistic trajectory, at the point where the controls are turned off, is also optimized for minimum casualty risk using a differential evolution algorithm. In case similar impact tracks are achieved for both optimizations, future analysis of the semi-controlled de-orbit strategy can be performed which less computational effort. Employing a semi-controlled low-thrust de-orbit strategy can reduce the casualty risk with 2 to 3 orders of magnitude with respect to an uncontrolled de-orbit strategy. Large ballistic coefficients are beneficial for the reduction of the casualty risk. Furthermore, orbits with high inclinations results in larger reductions of the casualty risk with respect to low inclination orbits. Also the results for low inclination orbits are less sensitive to errors in the re-entry state caused by imperfections in the GNC system of the satellite during the de-or, Aerospace Engineering | Astrodynamics & Space Missions

Block Matching is one of the most efficient techniques for motion estimation for video sequences. Metaheuristic algorithms have been used effectively for motion estimation. In this paper, we propose two hybrid algorithms: Artificial Bee Colony with Differential Evolution and Harmony Search with Differential Evolution based motion estimation algorithms. Extensive experiments are conducted using four standard video sequences. The video sequences utilized for experimentation have all essential features such as different formats, resolutions and number of frames which are generally required in input video sequences. We compare the performance of the proposed algorithms with other algorithms considering various parameters such as Structural Similarity, Peak Signal to Noise Ratio, Average Number of Search Points etc. The comparative results demonstrate that the proposed algorithms outperformed other algorithms.

Мета дослідження: доведення можливості параметричного синтезу оптичних систем світлосильних фотооб’єктивів зі зменшеною дисторсією. Завдання дослідження: 1. Проаналізувати відомі методи глобальної оптимізації та обрати найбільш ефективний для його подальшого використання в процесі параметричного синтезу оптичних систем світлосильних фотооб’єктивів зі зменшеною дисторсією для зоряного орієнтатора. 2. Розробити метод проектування оптичних систем світлосильного фотооб’єктиву зі зменшеною дисторсією 3. Виконати експериментальну перевірку запропонованого методу шляхом розрахунку оптичних систем з світлосильного фотооб’єктиву зі зменшеною дисторсією. Об’єкт дослідження: процес проектування оптичних систем світлосильного об’єктиву зі зменшеною дисторсією. Предмет дослідження: метод автоматизованого розрахунку оптичної системи світлосильного об’єктиву зі зменшеною дисторсією., Research goal is to prove the possibility of parametric synthesis of optical systems of light-sensitive photolights with reduced distortion. Research objectives: 1. Analyze the known methods of global optimization and choose the most effective for its further use in the process of parametric synthesis of optical systems of light-photolithic lenses with reduced distortion for the stellar orientation. 2. To develop a method for the design of optical systems of light-photolens with reduced distortion. 3. Perform an experimental verification of the proposed method by calculating optical systems from a light-photon photobjective with reduced distortion. Object of research is the process of designing optical systems of light-weight lens with reduced distortion. Subject of research is the method of automated calculation of optical system of light-weight lens with diminished distortion.

Розробка простого та ефективного методу проектування оптичних систем з дифракційними поверхнями, доведення можливості параметричного синтезу оптичних систем з дифракційними оптичними поверхнями, Therefore, it would be desirable to develop simple and efficient design method of optical systems with diffractive surfaces. Research goal is to prove the possibility of parametric synthesis of optical systems with diffractive optical surfaces.

JADE is an state-of-the-art adaptive differential evolution algorithm which implements "DE/current-to-pbest" as its mutation strategy, adapts its mutation factor and crossover rate, and uses an optional external archive to keep track of potential removed individuals in previous generations. This paper proposes MPADE, which extends JADE by using a multi-populated approach to solve high dimensional real-parameter optimization problems. This mechanism helps preventing the two well-known problems affecting the differential evolution algorithm performance, which are premature convergence and stagnation. The algorithm was tested using the benchmark functions in IEEE Congress on Evolutionary Computation 2014 test suite. MPADE was compared using Wilcoxon test to JADE algorithm and with other state-of-the-art algorithms that either use a multi-population approach or adapt their parameters. The experimental results show that the proposed new algorithm improves significantly its precursor and it is also suggested that other state-of-the-art algorithms could benefit from the multi-populated based approach.

Differential evolution has many mutation strategies which are problem dependent. Some Adaptive Differential Evolution techniques have been proposed tackling this problem. But therein all individuals are treated equally without taking into account how good these solutions are. In this paper, a new method called Ranked-based Mutation Adaptation (RAM) is proposed, which takes into consideration the ranking of an individual in the whole population. This method will assign different probabilities of choosing different mutation strategies to different groups in which the population is divided. RAM has been integrated into several well-known adaptive differential evolution algorithms and its performance has been tested on the benchmark suit proposed in CEC2014. The experimental results shows the use of RAM can produce generally better quality solutions than the original adaptive algorithms.

Novel automated simulator-independent ESD model characterization method based on Differential evolution and Nelder-Mead Simplex algorithms is presented in this paper. It offers an alternative for time and human-resources demanding manual characterization that is still widely used. The paper also presents stable models of the four most often used snapback-based protection devices in CMOS technologies, i.e., NMOST and three variants of silicon-controlled rectifier structure. These models were used for evaluation of the proposed method and the results are included and discussed.

This paper discusses a set of façade design alternatives for form-finding problem focusing on conceptual phase. In this respect, the aim of the research is to propose a multi-objective optimization approach for a façade design of public pool building. We present a set of solution belonging to Self-adaptive Multi-objective Ensemble Differential Evolution (JE_DEMO) and Self-adaptive Multi-Objective Differential Evolution (JDEMO) algorithm. We focus on maximization of daylight performance and minimization of structural displacement. Based on results, two algorithms presented competitive results. Contributions are presented based on objectives functions as new trade-offs and proposed JE_DEMO algorithm for design problems., Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public., Design Informatics

Significant technological advances in the electric sector have led to complexity increase and created new challenges to transmission expansion planners. On the one hand, the increasing penetration of renewable energy sources impacts the entire network. On the other hand, DC systems have become a favorable and viable option to link some of these sources of energy to the consumer centers. Thus, new models of the network equipment must be incorporated to the Transmission Expansion Planning (TEP) model to take such changes into account in a more accurate way. This research work aims to include AC and multi-terminal DC transmission lines as candidates in the expansion planning process. The AC network model is used and shunt compensation is also taken into account to allow more flexibility to the expansion options. The TEP problem is solved using a combination of nonlinear programming and differential evolution (DE). The results obtained using a 9-bus test network and a modified IEEE-118 bus show the feasibility of this approach.

In this work, a new approach for aerodynamic optimization of horizontal axis wind turbine (HAWT) airfoil is presented. This technique combines commercial computational fluid dynamics (CFD) codes with differential evolution (DE), a reliable gradient-free global optimization method. During the optimization process, commercial CFD codes are used to evaluate aerodynamic characteristics of HAWT airfoil and an improved DE algorithm is utilized to find the optimal airfoil design. The objective of this research is to maximize the aerodynamic coefficients of HAWT airfoil at the design angle of attack (AOA) with specific ambient environment. The airfoil shape is modeled by control points which their coordinates are design variables. The reliability of CFD codes is validated by comparing the analytical results of a typical HAWT airfoil with its experimental data. Finally, the optimal design of wind turbine airfoil is evaluated about aerodynamic performance in comparison with existing airfoils and some discussions are performed.

A single-loop deterministic method (SLDM) has previously been proposed for solving reliability-based design optimization (RBDO) problems. In SLDM, probabilistic constraints are converted to approximate deterministic constraints. Consequently, RBDO problems can be transformed into approximate deterministic optimization problems, and hence the computational cost of solving such problems is reduced significantly. However, SLDM is limited to continuous design variables, and the obtained solutions are often trapped into local extrema. To overcome these two disadvantages, a global single-loop deterministic approach is developed in this article, and then it is applied to solve the RBDO problems of truss structures with both continuous and discrete design variables. The proposed approach is a combination of SLDM and improved differential evolution (IDE). The IDE algorithm is an improved version of the original differential evolution (DE) algorithm with two improvements: a roulette wheel selection with stochastic acceptance and an elitist selection technique. These improvements are applied to the mutation and selection phases of DE to enhance its convergence rate and accuracy. To demonstrate the reliability, efficiency and applicability of the proposed method, three numerical examples are executed, and the obtained results are compared with those available in the literature.

This paper provides an analysis of the population clustering in a novel Success-History based Adaptive Differential Evolution algorithm with Distance based adaptation (Db_SHADE) in order to analyze the exploration and exploitation abilities of the algorithm. The comparison with the original SHADE algorithm is performed on the CEC2015 benchmark set in two dimensional settings (10D and 30D). The clustering analysis helps to answer the question about prolonged exploration phase of the Db_SHADE algorithm. Possible future research directions are drawn in the discussion and conclusion. © 2018, Brno University of Technology. All rights reserved.

This research represents a detailed insight into the modern and popular hybridization of deterministic chaotic dynamics and evolutionary computation. It is aimed at the influence of chaotic sequences on the performance of four selected Differential Evolution (DE) variants. The variants of interest were: original DE/Rand/1/ and DE/Best/1/ mutation schemes, simple parameter adaptive jDE, and the recent state of the art version SHADE. Experiments are focused on the extensive investigation of the different randomization schemes for the selection of individuals in DE algorithm driven by the nine different two-dimensional discrete deterministic chaotic systems, as the chaotic pseudorandom number generators. The performances of DE variants and their chaotic/non-chaotic versions are recorded in the one-dimensional settings of 10D and 15 test functions from the CEC 2015 benchmark, further statistically analyzed. © 2018, Brno University of Technology. All rights reserved.

This research represents a detailed insight into the modern and popular hybridization of deterministic chaotic dynamics and evolutionary computation. It is aimed at the influence of chaotic sequences on the performance of four selected Differential Evolution (DE) variants. The variants of interest were: original DE/Rand/1/ and DE/Best/1/ mutation schemes, simple parameter adaptive jDE, and the recent state of the art version SHADE. Experiments are focused on the extensive investigation of the different randomization schemes for the selection of individuals in DE algorithm driven by the nine different two-dimensional discrete deterministic chaotic systems, as the chaotic pseudorandom number generators. The performances of DE variants and their chaotic/non-chaotic versions are recorded in the one-dimensional settings of 10D and 15 test functions from the CEC 2015 benchmark, further statistically analyzed.

This paper provides an analysis of the population clustering in a novel Success-History based Adaptive Differential Evolution algorithm with Distance based adaptation (Db_SHADE) in order to analyze the exploration and exploitation abilities of the algorithm. The comparison with the original SHADE algorithm is performed on the CEC2015 benchmark set in two dimensional settings (10D and 30D). The clustering analysis helps to answer the question about prolonged exploration phase of the Db_SHADE algorithm. Possible future research directions are drawn in the discussion and conclusion.

In this paper, a comparative study of seven variants of the Success-History based Adaptive Differential Evolution (SHADE) algorithm is presented with the aim to show the influence of improvement and distance based parameter adaptation to its performance. Seven algorithms range from improvement only based SHADE through a balanced improvement and distance based DIb_SHADE to distance only based Db_SHADE. The algorithm set is compared on the CEC2015 benchmark set, results are statistically tested, commented and possible future research directions are provided. © ECMS Lars Nolle, Alexandra Burger, Christoph Tholen, Jens Werner, Jens Wellhausen

This research deals with a novel approach to classification - pseudo neural networks (PNN). This technique was inspired in classical artificial neural networks (ANN), where a relation between inputs and outputs is based on the mathematical transfer functions and optimised numerical weights. Compared to ANN, the whole structure in PNN, i.e. the relation between inputs and output(s), is fully synthesised by evolutionary symbolic regression tool - analytic programming. Compared to previous synthesised models, the PNN in this paper were synthesised via a new approach to constant estimation inside the analytic programming - direct coding. Iris data was used for the experiments and PNN were used for the synthesis of a complex classifier for more classes. For experimentation, Differential Evolution (de/rand/1/bin) for optimisation in analytic programming (AP) was used. © ECMS Lars Nolle, Alexandra Burger, Christoph Tholen, Jens Werner, Jens Wellhausen

Novel automated simulator-independent ESD model characterization method based on Differential evolution and Nelder-Mead Simplex algorithms is presented in this paper. It offers an alternative for time and human-resources demanding manual characterization that is still widely used. The paper also presents stable models of the four most often used snapback-based protection devices in CMOS technologies, i.e., NMOST and three variants of silicon-controlled rectifier structure. These models were used for evaluation of the proposed method and the results are included and discussed.

Recently, emotion recognition using low-cost wearable sensors based on electroencephalogram and blood volume pulse has received much attention. Long short-term memory (LSTM) networks, a special type of recurrent neural networks, have been applied successfully to emotion classification. However, the performance of these sequence classifiers depends heavily on their hyperparameter values, and it is important to adopt an efficient method to ensure the optimal values. To address this problem, we propose a new framework to automatically optimize LSTM hyperparameters using differential evolution (DE). This is the first systematic study of hyperparameter optimization in the context of emotion classification. In this paper, we evaluate and compare the proposed framework with other state-of-the-art hyperparameter optimization methods (particle swarm optimization, simulated annealing, random search, and tree of Parzen estimators) using a new dataset collected from wearable sensors. Experimental results demonstrate that optimizing LSTM hyperparameters significantly improve the recognition rate of four-quadrant dimensional emotions with a 14% increase in accuracy. The best model based on the optimized LSTM classifier using the DE algorithm achieved 77.68% accuracy. The results also showed that evolutionary computation algorithms, particularly DE, are competitive for ensuring optimized LSTM hyperparameter values. Although DE algorithm is computationally expensive, it is less complex and offers higher diversity in finding optimal solutions.