Search

The electricity grids have become a key player in the society. An increased usage of electricity is both a result from the more electrified society, but also as a main solver in reaching the climate goals by reducing emissions. This thesis work explores some of the new features for the electricity grid from integration of electrification from renewable energy resources (RES) and from strategies for energy optimization related to the loads and specifically from thermal heat pumps. These strategies lie in the field of demand response, which takes advantage of the flexibility of loads in terms of willingness to switch or decrease their consumption at a particular moment of the day. This research proposes a three-step framework to harness the flexibility of Thermo-Statically controlled loads (TCLs) based on a simulated grey-box building model that uses historical outside temperature and prices data and learns the thermal parameters such as Thermal Resistance, Thermal Capacitance, but also price responsiveness (pth ) through a Differential Evolution (DE) based optimization algorithm. The price responsiveness is used to provide further insight into the flexibility of the loads and is employed in the last step to propose a price policy estimation algorithm also based on DE that minimizes the gap between supply and demand while preserving supplier and customer profitability. The proposed approach has proven to be accurate for a large number of parameters but also effective with reduced training data (prediction errors around 2.5% on the power average and standard deviation), as well as to be successful in providing a Day-Ahead Real-Time Price policy that maximizes supplier and customer utility. The price policy provides a lower total price for the customer compared with a tariff without demand response (reduction up to 53.63 %), reduces the gap between RES-based energy sources and heating demand, and respects grid technical constraints., Elnätet har blivit en nyckelaktör i samhället. En ökad elanvändning är både ett resultat av det mer elektrifierade samhället, men också som en huvudlösare för att nå klimatmålen genom att minska utsläpp. Detta examensarbete undersöker några av de nya funktionerna för elnätet från integration av elektrifiering från förnybara energiresurser (RES) och från strategier för energioptimering relaterade till ellaster och specifikt från termiska värmepumpar. Dessa strategier ligger inom området för efterfrågerespons, som drar fördel av lasternas flexibilitet när det gäller viljan att byta eller minska sin konsumtion vid en viss tidpunkt på dagen. Den här forskningen föreslår ett ramverk i tre steg för att utnyttja flexibiliteten hos termostatiskt kontrollerade laster (TCL) baserat på en simulerad gråbox-byggnadsmodell som använder historisk utomhustemperatur och prisdata och lär sig de termiska parametrarna som termisk motstånd, termisk kapacitans , men också priskänslighet (pth) genom en Differential Evolution (DE) baserad optimeringsalgoritm. Priskänsligheten används för att ge ytterligare insikt om lasternas flexibilitet och används i det sista steget för att föreslå en prispolitisk uppskattningsalgoritm också baserad på DE som minimerar klyftan mellan utbud och efterfrågan samtidigt som leverantörs- och kundlönsamheten bevaras. Det föreslagna tillvägagångssättet har visat sig vara korrekt för ett stort antal parametrar men också effektivt med reducerad träningsdata (förutsägelsefel runt 2,5% på effektmedelvärde och standardavvikelse), samt vara framgångsrik i att tillhandahålla en Day-Ahead Real -Tidsprispolicy som maximerar leverantörs- och kundnyttan. Prispolicyn ger ett lägre totalpris för kunden jämfört med en tariff utan efterfrågerespons (reduktion upp till 53, 63 %), minskar gapet mellan RES-baserade energikällor och värmebehov samt följer nättekniska begränsningar.

This research proposes a novel Hybrid Success History Intelligent Optimizer with Gaussian Transformation (SHIOGT) for solving different complexity level optimization problems and for Convolutional Neural Network (CNNs) hyperparameter tuning. SHIOGT algorithm is designed to balance exploration and exploitation phases through the addition of Gaussian Transformation to the original Success History Intelligent Optimizer. The inclusion of Gaussian Transformation enhances solution diversity enables SHIO to avoid local optima. SHIOGT also demonstrates robustness and adaptability by dynamically adjusting its search strategy based on problem characteristics. Furthermore, the combination of Gaussian and SHIO facilitates faster convergence, accelerating the discovery of optimal or near-optimal solutions. Moreover, the hybridization of these two techniques brings a synergistic effect, enabling SHIOGT to overcome individual limitations and achieve superior performance in hyperparameter optimization tasks. SHIOGT was thoroughly assessed against an array of benchmark functions of varying complexities, demonstrating its ability to efficiently locate optimal or near-optimal solutions across different problem categories. Its robustness in tackling multimodal and deceptive landscapes and high-dimensional search spaces was particularly notable. SHIOGT has been benchmarked over 43 challenging optimization problems and have been compared with state-of-the art algorithm. Further, SHIOGT algorithm is applied to the domain of deep learning, with a case study focusing on hyperparameter tuning of CNNs. With the intelligent exploration–exploitation balance of SHIOGT, we hypothesized it could effectively optimize the CNN's hyperparameters. We evaluated the performance of SHIOGT across a variety of datasets, including MNIST, Fashion-MNIST, CIFAR-10, and CIFAR-100, with the aim of optimizing CNN model hyperparameters. The results show an impressive accuracy rate of 98% on the MNIST dataset. Simil

Complex fractional-order (CFO) transfer functions, being more generalized versions of their real-order counterparts, lend greater flexibility to system modeling. Due to the absence of commercial complex-order fractance elements, the implementation of CFO models is challenging. To alleviate this issue, a constrained optimization approach that meets the targeted frequency responses is proposed for the rational approximation of CFO systems. The technique generates stable, {minimum-phase, and real-valued coefficients based approximants}, which are not always feasible for the curve-fitting approach reported in the literature. {Stability and performance studies of the CFO proportional-integral-derivative (CFOPID) controllers for the Podlubny's, the internal model control, and the El-Khazali's forms are considered to demonstrate the feasibility of the proposed technique}. Simulation results highlight that, for a practically reasonable order, all the designs achieve good agreement with the theoretical characteristics. {Performance comparisons with the CFOPID controller approximants determined by the Oustaloup's CFO differentiator based substitution method justify the proposed approach.

There is a growing number of network attacks and the data on the network is more exposed than ever with the increased activity on the Internet. Applying Machine Learning (ML) techniques for cyber-security is a popular and effective approach to address this problem. However, the data which is used by ML algorithms have to be protected. In this paper, we present a framework that combines autoencoder, multi-objective optimization algorithms, and different ML algorithms to encode the network data, reduce its size, and detect and classify the network attacks. The novel element used in this framework, with respect to earlier research, is the application of multi-objective optimization algorithms, such as Multi-Objective Differential Evolution or Non-dominated Sorting Genetic Algorithm-II, to handle the different objectives in the fitness function of the autoencoder (autoencoder decoding error and accuracy of ML algorithm). We evaluated six different ML algorithms for attack detection and classification on network dataset UNSWNB15. The performance of the proposed framework is compared with single-objective Differential Evolution. The results showed that Multi-Objective Differential Evolution outperforms the counterparts for attack detection, while all the evaluated algorithms showed similar performance for attack classification.

Trading contracts with future commitments requires posting a collateral, called initial margin requirement, to cover associated risks. Differences in estimating those risks and varying risk appetites can however lead to identical contracts having different initial margin requirements at different market places. This creates a potential for minimizing those requirements by reallocating contracts. The task of minimizing the requirement is identified as a black-box optimization problem with constraints. The aim of this project was to investigate that optimization problem, how it can best be tackled, and comparing different techniques for doing so. Based on the results and obstacles encountered along the way, some guidelines are then outlined to provide assistance for whomever is interested in solving this or similar problems. The project consisted both of a literature study to examine existing knowledge within the subject of optimization, and an implementation phase to empirically test how well that knowledge can be put to use in this case. During the latter various algorithms were tested in a number of different scenarios. Focus was put on practical aspects that could be important in a real situation, such as how much they could decrease the initial margin requirement, execution time, and ease of implementation. As part of the literature study, three algorithms were found which were evaluated further: simulated annealing, differential evolution, and particle swarm optimization. They all work without prior knowledge of the function to be optimized, and are thus suitable for black-box optimization. Results from the implementation part showed largely similar performance between all three algorithms, indicating that other aspects such as ease of implementation or parallelization potential can be more important to consider when choosing which one to use. They were all well able to optimize different portfolios in a number of different cases. However, in more complex situat

Trading contracts with future commitments requires posting a collateral, called initial margin requirement, to cover associated risks. Differences in estimating those risks and varying risk appetites can however lead to identical contracts having different initial margin requirements at different market places. This creates a potential for minimizing those requirements by reallocating contracts. The task of minimizing the requirement is identified as a black-box optimization problem with constraints. The aim of this project was to investigate that optimization problem, how it can best be tackled, and comparing different techniques for doing so. Based on the results and obstacles encountered along the way, some guidelines are then outlined to provide assistance for whomever is interested in solving this or similar problems. The project consisted both of a literature study to examine existing knowledge within the subject of optimization, and an implementation phase to empirically test how well that knowledge can be put to use in this case. During the latter various algorithms were tested in a number of different scenarios. Focus was put on practical aspects that could be important in a real situation, such as how much they could decrease the initial margin requirement, execution time, and ease of implementation. As part of the literature study, three algorithms were found which were evaluated further: simulated annealing, differential evolution, and particle swarm optimization. They all work without prior knowledge of the function to be optimized, and are thus suitable for black-box optimization. Results from the implementation part showed largely similar performance between all three algorithms, indicating that other aspects such as ease of implementation or parallelization potential can be more important to consider when choosing which one to use. They were all well able to optimize different portfolios in a number of different cases. However, in more complex situat

Complex fractional-order (CFO) transfer functions, being more generalized versions of their real-order counterparts, lend greater flexibility to system modeling. Due to the absence of commercial complex-order fractance elements, the implementation of CFO models is challenging. To alleviate this issue, a constrained optimization approach that meets the targeted frequency responses is proposed for the rational approximation of CFO systems. The technique generates stable, {minimum-phase, and real-valued coefficients based approximants}, which are not always feasible for the curve-fitting approach reported in the literature. {Stability and performance studies of the CFO proportional-integral-derivative (CFOPID) controllers for the Podlubny's, the internal model control, and the El-Khazali's forms are considered to demonstrate the feasibility of the proposed technique}. Simulation results highlight that, for a practically reasonable order, all the designs achieve good agreement with the theoretical characteristics. {Performance comparisons with the CFOPID controller approximants determined by the Oustaloup's CFO differentiator based substitution method justify the proposed approach.

Complex fractional-order (CFO) transfer functions, being more generalized versions of their real-order counterparts, lend greater flexibility to system modeling. Due to the absence of commercial complex-order fractance elements, the implementation of CFO models is challenging. To alleviate this issue, a constrained optimization approach that meets the targeted frequency responses is proposed for the rational approximation of CFO systems. The technique generates stable, {minimum-phase, and real-valued coefficients based approximants}, which are not always feasible for the curve-fitting approach reported in the literature. {Stability and performance studies of the CFO proportional-integral-derivative (CFOPID) controllers for the Podlubny's, the internal model control, and the El-Khazali's forms are considered to demonstrate the feasibility of the proposed technique}. Simulation results highlight that, for a practically reasonable order, all the designs achieve good agreement with the theoretical characteristics. {Performance comparisons with the CFOPID controller approximants determined by the Oustaloup's CFO differentiator based substitution method justify the proposed approach.

This paper presents an overview of the history and recent efforts in combining chaos theory and evolutionary computation techniques. Various algorithms from the evolutionary computation domain, also known as metaheuristic algorithms, have been successfully enhanced with chaotic components in the past. Numerous ways to incorporate chaos have been examined, and many impressive results have been reported. Implementations of discrete chaotic maps such as Lozi, Henon, and logistic map as generators of chaotic pseudo-random sequences for controlling evolution operators in metaheuristics have achieved significant popularity. In this survey, we focus on the research field itself and perform a bibliographical analysis to show how broad and active is nowadays the research field of chaos-enhanced metaheuristics and what are some of the most recent works published.

Depuis quelques années, plusieurs entreprises ont commencé à exploiter les données qu’ils ont emmagasinées pour pouvoir extraire des connaissances et faire des prévisions. Une bonne utilisation de ces données permet aux entreprises de comprendre leurs clients et pouvoir améliorer leurs services. Les techniques d’intelligence artificielle se présentent comme des moyens pour révolutionner les prises de décisions. Plus d’entreprises embauchent des data scientists pour concevoir des modèles de prédiction. Le but des data scientists est de concevoir des modèles et de les performer. L’un des défis majeurs rencontré par ces data scientists est le prétraitement des données. Le prétraitement des données impact la qualité du modèle mis en place. Ce problème est dû au fait qu’il n’existe pas une seule bonne manière de prétraité les données. Les techniques utilisées dépendent du problème et du type de donnée. Ce mémoire a pour objectif d’évaluer la possibilité de développer une solution qui permet d’automatiser les opérations de prétraitement des données pour les acteurs de données (data scientistes, chercheurs, etc.). Nous utiliserons le terme pipeline afin de nommer la chaîne de pré-traitement et de traitement des données (de la donnée brute jusqu’à la classification de la donnée). En effet, nous proposons un pipeline qui permet de choisir les meilleures techniques de prétraitements en fonction du problème. Plusieurs techniques sont répertoriées et nous utilisons des techniques d’optimisation métaheuristique pour minimiser le temps de recherche du meilleur modèle. Ces techniques sont par la suite comparées avec des techniques d’optimisation standard comme la recherche par grille. Le pipeline est par la suite testé sur des données d’une entreprise d’assurance automobile du Canada. Les résultats nous montrent que les méthodes d’optimisation métaheuristique sont plus efficaces sur des big data. Elles ont permis de diminuer d’environ 30% le temps nécessaire pour retrouver un bon

The Flying-V is a disruptive flying wing aircraft configuration that was realized at the Airbus Future Projects Office (FPO), with research and development continuing at TU Delft. Preliminary findings have indicated that the Flying-V is 25\% more aerodynamically efficient than the state-of-the-art Airbus A350. To further reap the benefits of this configuration, the aerodynamic design of every element of the Outer Mould Line (OML) must be refined to maximize performance during cruise flight. One such element is the wingtip device, specifically the winglet, which are non-planar wings that can significantly reduce the lift-induced drag of an aircraft. For the Flying-V, the winglets also double as vertical stabilizers. A numerical experiment was conducted to evaluate and augment the lateral-directional stability and aerodynamic performance of the Flying-V, incorporating a contemporary winglet design. To enable the winglet’s equally weighted multi-objective optimization, a dedicated module was developed for the existing Flying-V design framework. This module comprises an automated geometry generator, spatial discretizer, flow solver and a differential evolution-based optimizer. For all purposes, the turbulent flow field around the aircraft was resolved using the Reynolds-averaged Navier-Stokes (RANS) model. The Morris method was employed to analyse the sensitivities of the cost functions to the design parameters, yielding an improved baseline design from the ensuing sampling study. The optimization was performed sequentially with complementary planform and airfoil optimization steps. The cruise efficiency of the Flying-V with the initial winglet is found to be lower than the winglet-off case despite portraying strong stability characteristics during approach conditions. While the baseline design’s efficiency is enhanced relative to the winglet-off scenario, the stability performance drops from the initial design. A considerable improvement is obtained in both objectives, Flying-V, Aerospace Engineering | Flight Performance and Propulsion

Evolutionary algorithms (EA) are efficient population-based stochastic algorithms for solving optimization problems. The performance of EAs largely depends on the configuration of values of parameters that control their search. Previous works studied how to configure EAs, though, there is a lack of a general approach to effectively tune EAs. To fill this gap, this paper presents a consistent, automated approach for tuning and controlling parameterized search of an EA. For this, we propose a deep reinforcement learning (DRL) based approach called ‘DRL-APC-DE’ for online controlling search parameter values for a multi-objective Differential Evolution algorithm. The proposed method is trained and evaluated on widely adopted multi-objective test problems. The experimental results show that the proposed approach performs competitively to a non-adaptive Differential Evolution algorithm, tuned by grid search on the same range of possible parameter values. Subsequently, the trained algorithms have been applied to unseen multi-objective problems for the adaptive control of parameters. Results show the successful ability of DRL-APC-DE to control parameters for solving these problems, which has the potential to significantly reduce the dependency on parameter tuning for the successful application of EAs.

This paper proposes an adaptive heterogeneous particle swarm optimization with a comprehensive learning strategy for solving single-objective constrained optimization problems. In this algorithm, particles can use an exploration strategy and an exploitation strategy to update their positions. The historical success rates of the two strategies are used to adaptively control the adoption rates of strategies in the next iteration. The search strategy in the canonical particle swarm optimization algorithm is based on elite solutions. As a result, when no particles can discover better solutions for several generations, this algorithm is likely to fall into stagnation. To respond to this challenge, a new strategy is proposed to explore the neighbors of the elite solutions in this study. Finally, a constraint handling method is equipped to the proposed algorithm to make it be able to solve constrained optimization problems. The proposed algorithm is compared with the canonical particle swarm optimization, differential evolution, and several recently proposed algorithms on the benchmark test suite. The Wilcoxon signed-rank test results show that the proposed algorithm is significantly better on most of the benchmark problems compared with the competitors. The proposed algorithm is also applied to solve two real-world mechanical engineering problems. The experimental results show that the proposed algorithm performs consistently well on these problems.

The challenge of heat exchanger network retrofit is often addressed using deterministic algorithms. However, the complexity of the retrofit problems, combined with multi-period operation, makes it very difficult to find any feasible solution. In contrast, stochastic algorithms are more likely to find feasible solutions in complex solution spaces. This work presents a customized evolutionary based optimization algorithm to address this challenge. The algorithm has two levels, whereby, a genetic algorithm optimizes the topology of the heat exchanger network on the top level. Based on the resulting topology, a differential evolution algorithm optimizes the heat loads of the heat exchangers in each operating period. The following bullet points highlight the customization of the algorithm: •The advantage of using both algorithms: the genetic algorithm is used for the topology optimization (discrete variables) and the differential evolution for the heat load optimization (continuous variables). •Penalizing and preserving strategies are used for constraint handling •The evaluation of the genetic algorithm is parallelized, meaning the differential evolution algorithm is performed on each chromosome parallel on multiple cores., + ID der Publikation: hslu_93764 + Art des Beitrages: Wissenschaftliche Medien + Jahrgang: 2022 + Sprache: Englisch + Letzte Aktualisierung: 2023-01-24 12:28:10

In the last years, the study of Fractional Optimal Control Problem (FOCP) configures a very interesting challenge due to numerical difficulties inherent in this type of investigation. Traditionally, this problem has been solved by considering three different approaches, namely, Direct and Indirect strategies and Hamilton–Jacobi–Bellman (HJB) equation. The first consists in solving the FOCP through the discretization of state and/or control variables. The resulting nonlinear optimization problem is solved by considering either classical or heuristic methods. On the other hand, the indirect approach consists in obtaining the necessary conditions, i.e., the original FOCP is converted into a two-point boundary value problem. The third strategy considers an extension of the well-known HJB equation for fractional order dynamic systems. In the present contribution, the solution of FOCP with specified final state variable is addressed by using the direct approach. For this purpose, the association involving the Orthogonal Collocation Method (OCM) and the Differential Evolution (DE) algorithm is investigated. In order to evaluate the proposed methodology, a classical mathematical problem and a two degree-of-freedom given by a spring-mass-damper system are considered. As expected, the results indicate that the variation of the fractional order implies different values for the original objective function. Furthermore, depending on the fractional order value, it may not be possible to find a solution that satisfies the boundary conditions for a given application. Finally, it is pointed out that the proposed methodology is considered as a promising strategy to solve FOCPs., En los últimos años, el estudio del Problema de Control Óptimo Fraccionado (FOCP) configura un reto muy interesante debido a las dificultades numéricas inherentes a este tipo de investigación. Tradicionalmente, este problema se ha resuelto considerando tres enfoques diferentes, a saber, estrategias directas e indirectas y la ecuación de Hamilton-Jacobi-Bellman (HJB). La primera consiste en resolver el FOCP mediante la discretización de variables de estado y/o control. El problema de optimización no lineal resultante se resuelve considerando métodos clásicos o heurísticos. Por otro lado, el enfoque indirecto consiste en obtener las condiciones necesarias, es decir, el FOCP original se convierte en un problema de valores en la frontera de dos puntos. La tercera estrategia considera una extensión de la conocida ecuación HJB para sistemas dinámicos de orden fraccionario. En la presente contribución, la solución de FOCP con una variable de estado final especificada se aborda utilizando el enfoque directo. Para ello, se investiga la asociación entre el Método de Colocación Ortogonal (OCM) y el algoritmo de Evolución Diferencial (DE). Para evaluar la metodología propuesta se considera un problema matemático clásico y dos grados de libertad dados por un sistema resorte-masa-amortiguador. Como era de esperar, los resultados indican que la variación del orden fraccionario implica valores diferentes para la función objetivo original. Además, dependiendo del valor del orden fraccionario, puede que no sea posible encontrar una solución que satisfaga las condiciones de contorno para una aplicación determinada. Finalmente, se señala que la metodología propuesta se considera como una estrategia promisoria para solucionar los FOCP.

In the last years, the study of Fractional Optimal Control Problem (FOCP) configures a very interesting challenge due to numerical difficulties inherent in this type of investigation. Traditionally, this problem has been solved by considering three different approaches, namely, Direct and Indirect strategies and Hamilton–Jacobi–Bellman (HJB) equation. The first consists in solving the FOCP through the discretization of state and/or control variables. The resulting nonlinear optimization problem is solved by considering either classical or heuristic methods. On the other hand, the indirect approach consists in obtaining the necessary conditions, i.e., the original FOCP is converted into a two-point boundary value problem. The third strategy considers an extension of the well-known HJB equation for fractional order dynamic systems. In the present contribution, the solution of FOCP with specified final state variable is addressed by using the direct approach. For this purpose, the association involving the Orthogonal Collocation Method (OCM) and the Differential Evolution (DE) algorithm is investigated. In order to evaluate the proposed methodology, a classical mathematical problem and a two degree-of-freedom given by a spring-mass-damper system are considered. As expected, the results indicate that the variation of the fractional order implies different values for the original objective function. Furthermore, depending on the fractional order value, it may not be possible to find a solution that satisfies the boundary conditions for a given application. Finally, it is pointed out that the proposed methodology is considered as a promising strategy to solve FOCPs., En los últimos años, el estudio del Problema de Control Óptimo Fraccionado (FOCP) configura un reto muy interesante debido a las dificultades numéricas inherentes a este tipo de investigación. Tradicionalmente, este problema se ha resuelto considerando tres enfoques diferentes, a saber, estrategias directas e indirectas y la ecuación de Hamilton-Jacobi-Bellman (HJB). La primera consiste en resolver el FOCP mediante la discretización de variables de estado y/o control. El problema de optimización no lineal resultante se resuelve considerando métodos clásicos o heurísticos. Por otro lado, el enfoque indirecto consiste en obtener las condiciones necesarias, es decir, el FOCP original se convierte en un problema de valores en la frontera de dos puntos. La tercera estrategia considera una extensión de la conocida ecuación HJB para sistemas dinámicos de orden fraccionario. En la presente contribución, la solución de FOCP con una variable de estado final especificada se aborda utilizando el enfoque directo. Para ello, se investiga la asociación entre el Método de Colocación Ortogonal (OCM) y el algoritmo de Evolución Diferencial (DE). Para evaluar la metodología propuesta se considera un problema matemático clásico y dos grados de libertad dados por un sistema resorte-masa-amortiguador. Como era de esperar, los resultados indican que la variación del orden fraccionario implica valores diferentes para la función objetivo original. Además, dependiendo del valor del orden fraccionario, puede que no sea posible encontrar una solución que satisfaga las condiciones de contorno para una aplicación determinada. Finalmente, se señala que la metodología propuesta se considera como una estrategia promisoria para solucionar los FOCP.

In this paper, two methodologies are used to speed up the maximization of the breakdown voltage (BV) of a vertical GaN diode that has a theoretical maximum BV of -2100 V. Firstly, we demonstrated a 5X faster accurate simulation method in Technology Computer-Aided-Design (TCAD). This allows us to find 50 % more numbers of high BV (>1400 V) designs at a given simulation time. Secondly, a machine learning (ML) model is developed using TCAD-generated data and used as a surrogate model for differential evolution optimization. It can inversely design an out-of-the-training-range structure with BV as high as 1887 V (89 % of the ideal case) compared to -1100 V designed with human domain expertise.

[Abstract]: A memetic version between an evolutionary algorithm (differential evolution) and the local search provided by protein fragment replacements was defined for protein structure prediction. In this problem, it is intended to find the global minimum in a high-dimensional energy landscape to discover the native structure of the protein. This problem presents a multimodal energy landscape which can additionally present deceptiveness when searching for the protein structure with minimum energy. One strategy is to try to obtain a diverse set of optimized and different protein conformations, which can be located in different local minima of the energy landscape. For this purpose, different niching methods (crowding, fitness sharing and speciation) were integrated into the memetic algorithm. The integration of niching makes it possible to obtain in a straightforward way a diverse set of optimized and structurally different protein conformations. Compared to previous studies, as well as to the widely used Rosetta protein structure prediction method, the potential solutions offered here present a diverse set of folds with different distances (RMSD) from the real native conformation, with wide RMSD distributions, and obtaining conformations closer to the native structure (in RMSD values) in some proteins.

The sovereign credit rating is considered as a quantified assessment of country's economic and political stability. Due to its importance and increasing amount of available information, the sovereign credit rating is considered as a hot topic in the last few years. However, the models that predict the credit ratings used by the several big credit rating agencies are unavailable, and can therefore be considered as the black boxes. In this paper, we are tackling this problem of predicting sovereign credit ratings by proposing a hybrid model based on interpolative Boolean algebra (IBA) and differential evolution (DE). Namely, we aim to obtain a logical/pseudo-logical function in IBA framework using DE metaheuristic that could underline connections of chosen macroeconomic indicators and sovereign credit ratings. Such functions are easy to interpret and able to make a subtle fuzzy gradation among countries. Country's economic indicators together with credit ratings from 2000 to 2016 are used for the model training. Acquired model is further tested on the data for 2017 and 2018.

Nowadays, the sovereign credit rating is not only an index of a country's economic performance and political stability but also an overall indicator of development and growth, as well as the trust factor that is associated with the country. Due to its importance, the vast amount of available information, and the lack of a closed-form solution, prediction models based on machine learning (ML) and computation intelligence (CI) techniques are being increasingly used to complement traditional financial approaches. In this paper, we aim to introduce a novel ML-CI approach for sovereign credit rating prediction based on a differential evolution (DE) algorithm and interpolative Boolean algebra (IBA). In fact, the proposed approach is based on a pseudo-logical function in the IBA framework derived from the historical data of publicly available indicators using the DE algorithm. Such functions are easily interpreted and enable a subtle gradation among countries. It is shown that the IBA-DE approach outperforms back-propagation neural networks on the observed problem while also providing a deeper insight into each of the indicators used for prediction and its respective influence on the prediction rating on the other.

The challenge of heat exchanger network retrofit is often addressed using deterministic algorithms. However, the complexity of the retrofit problems, combined with multi-period operation, makes it very difficult to find any feasible solution. In contrast, stochastic algorithms are more likely to find feasible solutions in complex solution spaces. This work presents a customized evolutionary based optimization algorithm to address this challenge. The algorithm has two levels, whereby, a genetic algorithm optimizes the topology of the heat exchanger network on the top level. Based on the resulting topology, a differential evolution algorithm optimizes the heat loads of the heat exchangers in each operating period. The following bullet points highlight the customization of the algorithm: •The advantage of using both algorithms: the genetic algorithm is used for the topology optimization (discrete variables) and the differential evolution for the heat load optimization (continuous variables). •Penalizing and preserving strategies are used for constraint handling •The evaluation of the genetic algorithm is parallelized, meaning the differential evolution algorithm is performed on each chromosome parallel on multiple cores., + ID der Publikation: hslu_93764 + Art des Beitrages: Wissenschaftliche Medien + Jahrgang: 2022 + Sprache: Englisch + Letzte Aktualisierung: 2023-01-24 12:28:10

The challenge of heat exchanger network retrofit is often addressed using deterministic algorithms. However, the complexity of the retrofit problems, combined with multi-period operation, makes it very difficult to find any feasible solution. In contrast, stochastic algorithms are more likely to find feasible solutions in complex solution spaces. This work presents a customized evolutionary based optimization algorithm to address this challenge. The algorithm has two levels, whereby, a genetic algorithm optimizes the topology of the heat exchanger network on the top level. Based on the resulting topology, a differential evolution algorithm optimizes the heat loads of the heat exchangers in each operating period. The following bullet points highlight the customization of the algorithm: •The advantage of using both algorithms: the genetic algorithm is used for the topology optimization (discrete variables) and the differential evolution for the heat load optimization (continuous variables). •Penalizing and preserving strategies are used for constraint handling •The evaluation of the genetic algorithm is parallelized, meaning the differential evolution algorithm is performed on each chromosome parallel on multiple cores., + ID der Publikation: hslu_93764 + Art des Beitrages: Wissenschaftliche Medien + Jahrgang: 2022 + Sprache: Englisch + Letzte Aktualisierung: 2023-01-24 12:28:10

With the increasing use of computer networks and distributed systems, network security and data privacy are becoming major concerns for our society. In this paper, we present an approach based on an autoencoder trained with differential evolution for feature encoding of network data with the goal of improving security and reducing data transfers. One of the novel elements used in differential evolution for intrusion detection is the enhancements in the fitness function by adding the performance of a machine learning algorithm. We conducted an extensive evaluation of six machine learning algorithms for network intrusion detection using encoded data from well-known publicly available network datasets UNSW-NB15. The experiments clearly showed the supremacy of random forest, support vector machine, and K-nearest neighbors in terms of accuracy, and this was not affected to a high degree by reducing the number of features. Furthermore, the machine learning algorithm that was used during training (Linear Discriminant Analysis classifier) got a 14 percentage points increase in accuracy. Our results also showed clear improvements in execution times in addition to the obvious secure aspects of encoded data. Additionally, the performance of the proposed method outperformed one of the most commonly used feature reduction methods, Principal Component Analysis.

This paper calibrates through loss functions the parameters of Heston’s stochastic volatility model by using two different methods: minimizing a nonlinear objective function (a loss function) with constraints on the values of the parameter and using a differential evolution algorithm. Both methods are applied to implied volatilities on the Mexican Stock Exchange Index with four maturities and twenty-eight strike prices. The selection criterion for the parameters is minimizing the value of the mean square error of the implied volatility. The first method has a better performance with less error and time. However, empirical results show that for both methods the adjustment of implied volatilities is better for options with longterm maturities than for short-term maturities.

It is necessary to optimize the design of the missile aerodynamic shape for better performance while meeting tactical specifications. However, current design methods for aerodynamic shape are based on manual design and physical model simulations, which are very time-consuming. Therefore, we propose an optimization framework based on conditional Wasserstein Gan-GP (CWGAN-GP), convolutional neural network (CNN), multi-task learning with multi-gate mixture-of-Experts-3D (MMoE-3D) and differential evolution (DE). This method consists of four stages. In the first step, CWGAN-GP can learn the relationship between existing missile shape designs and shape conditions, generating diverse missile shapes as required. In the second step, CNN is used for feature extraction of missile design drawing, and the missile shape generated by CWGAN-GP is transformed into missile shape parameters. In the third step, the MMoE-3D model is trained in the subsonic and supersonic ranges to efficiently generate aerodynamic data corresponding to the missile shape. In the fourth step, DE is used to select the optimal missile shape by adjusting the potential variables of CWGAN-GP. The efficiency of the proposed optimization framework is verified by optimizing the rate of change of the center of pressure and the lift-to-drag ratio, with the neural network-based optimization framework achieving almost the same optimization results in a shorter time compared to conventional optimization with DATCOM.

The present study investigates how to apply continuous tow shearing (CTS) in a manufacturable design parameterization to obtain reduced imperfection sensitivity in lightweight, cylindrical shell designs. The asymptotic nonlinear method developed by Koiter is applied to predict the postbuckled stiffness, whose index is constrained to be positive in the optimal design, together with a minimum design load. The performance of three machine learning methods, namely, Support Vector Machine, Kriging, and Random Forest, are compared as drivers to the optimization towards lightweight designs. The new methodology consists of contributions in the areas of problem modeling, the selection of machine learning strategies, and an optimization formulation that results in optimal designs around the compromise frontier between mass and stiffness. The proposed ML-based framework proved to be able to solve the inverse problem for which a target design load is given as input, returning as output lightweight designs with reduced imperfection sensitivity. The results obtained are compatible with the existing literature where hoop-oriented reinforcements were added to obtain reduced imperfection sensitivity in composite cylinders., Aerospace Structures & Computational Mechanics

In Swiss process industry, process integration is often applied to retrofit existing plants with multi-period operation. Such periods may experience a high degree of variation in temperature or mass flow. Some process streams may not exist in every period or are soft streams. The resulting retrofitted network needs to be able to ensure feasible heat transfer in each period by the integration of mixer configurations to control the temperature. These attributes increase the complexity of the solution space. Hence, this work proposes an evolutionary two-level algorithm for heat exchanger network retrofit. Genetic algorithm is used for topology optimization and a differential evolution algorithm handles the heat loads. The algorithm is extended with practical constraints such as a maximum number of heat exchangers. Explicit mixer temperature calculations are implemented using the Lambert W-function. The algorithm was successfully applied to an industrial case study, reducing its total annual cost by approximately 66%., + ID der Publikation: hslu_93469 + Art des Beitrages: Wissenschaftliche Medien + Jahrgang: 2022 + Sprache: Englisch + Letzte Aktualisierung: 2022-10-26 17:30:39

In Swiss process industry, process integration is often applied to retrofit existing plants with multi-period operation. Such periods may experience a high degree of variation in temperature or mass flow. Some process streams may not exist in every period or are soft streams. The resulting retrofitted network needs to be able to ensure feasible heat transfer in each period by the integration of mixer configurations to control the temperature. These attributes increase the complexity of the solution space. Hence, this work proposes an evolutionary two-level algorithm for heat exchanger network retrofit. Genetic algorithm is used for topology optimization and a differential evolution algorithm handles the heat loads. The algorithm is extended with practical constraints such as a maximum number of heat exchangers. Explicit mixer temperature calculations are implemented using the Lambert W-function. The algorithm was successfully applied to an industrial case study, reducing its total annual cost by approximately 66%., + ID der Publikation: hslu_93469 + Art des Beitrages: Wissenschaftliche Medien + Jahrgang: 2022 + Sprache: Englisch + Letzte Aktualisierung: 2022-10-26 17:30:39

Not all generate-and-test search algorithms are created equal. Bayesian Optimization (BO) invests a lot of computation time to generate the candidate solution that best balances the predicted value and the uncertainty given all previous data, taking increasingly more time as the number of evaluations performed grows. Evolutionary Algorithms (EA) on the other hand rely on search heuristics that typically do not depend on all previous data and can be done in constant time. Both BO and EA community typically assess their performance as a function of the number of evaluations, i.e., data efficiency. However, this is unfair once we start to compare the efficiency of these classes of algorithms, as the overhead times to generate candidate solutions are significantly different. We suggest to measure the efficiency of generate-and-test search algorithms as the expected gain in the objective value per unit of computation time spent, i.e., time efficiency. To the time-efficient search algorithm, we therefore propose a new algorithm, a combination of BO and an EA, BEA for short, that starts with BO, then transfers knowledge to an EA, and subsequently runs the EA. We compare the BEA with BO, the EA, Differential Evolution (DE), and Particle Swarm Optimization (PSO). The results show that BEA outperforms BO, the EA, DE and PSO in terms of time efficiency, and ultimately leads to better performance on well-known benchmark objective functions with many local optima. Moreover, we test BEA, BO, and the EA on nine test cases of robot learning problems and here again we find that BEA outperforms the other algorithms.

In Swiss process industry, process integration is often applied to retrofit existing plants with multi-period operation. Such periods may experience a high degree of variation in temperature or mass flow. Some process streams may not exist in every period or are soft streams. The resulting retrofitted network needs to be able to ensure feasible heat transfer in each period by the integration of mixer configurations to control the temperature. These attributes increase the complexity of the solution space. Hence, this work proposes an evolutionary two-level algorithm for heat exchanger network retrofit. Genetic algorithm is used for topology optimization and a differential evolution algorithm handles the heat loads. The algorithm is extended with practical constraints such as a maximum number of heat exchangers. Explicit mixer temperature calculations are implemented using the Lambert W-function. The algorithm was successfully applied to an industrial case study, reducing its total annual cost by approximately 66%., + ID der Publikation: hslu_93469 + Art des Beitrages: Wissenschaftliche Medien + Jahrgang: 2022 + Sprache: Englisch + Letzte Aktualisierung: 2022-10-26 17:30:39

Differential Evolution (DE) has been widely appraised as a simple yet robust population-based, non-convex optimization algorithm primarily designed for continuous optimization. Two important control parameters of DE are the scale factor F, which controls the amplitude of a perturbation step on the current solutions and the crossover rate Cr, which limits the mixing of components of the parent and the mutant individuals during recombination. We propose a very simple, yet effective, nearest spatial neighborhood-based modification to the adaptation process of the aforesaid parameters in the Success-History based adaptive DE (SHADE) algorithm. SHADE uses a historical archive of the successful F and Cr values to update these parameters and stands out as a very competitive DE variant of current interest. Our proposed modifications can be extended to any SHADE-based DE algorithm like L-SHADE (SHADE with linear population size reduction), jSO (L-SHADE with modified mutation) etc. The enhanced performance of the modified SHADE algorithm is showcased on the IEEE CEC (Congress on Evolutionary Computation) 2013, 2014, 2015, and 2017 benchmark suites by comparing against the DE-based winners of the corresponding competitions. Furthermore, the effectiveness of the proposed neighborhood-based parameter adaptation strategy is demonstrated by using the real-life problems from the IEEE CEC 2011 competition on testing evolutionary algorithms on real-world numerical optimization problems. © 2022

The present study investigates how to apply continuous tow shearing (CTS) in a manufacturable design parameterization to obtain reduced imperfection sensitivity in lightweight, cylindrical shell designs. The asymptotic nonlinear method developed by Koiter is applied to predict the postbuckled stiffness, whose index is constrained to be positive in the optimal design, together with a minimum design load. The performance of three machine learning methods, namely, Support Vector Machine, Kriging, and Random Forest, are compared as drivers to the optimization towards lightweight designs. The new methodology consists of contributions in the areas of problem modeling, the selection of machine learning strategies, and an optimization formulation that results in optimal designs around the compromise frontier between mass and stiffness. The proposed ML-based framework proved to be able to solve the inverse problem for which a target design load is given as input, returning as output lightweight designs with reduced imperfection sensitivity. The results obtained are compatible with the existing literature where hoop-oriented reinforcements were added to obtain reduced imperfection sensitivity in composite cylinders., Aerospace Structures & Computational Mechanics

Differential evolution (DE) is widely used for global optimisation problems due to its simplicity and efficiency. L-SHADE is a state-of-the-art variant of DE algorithm that incorporates external archive, success-history-based parameter adaptation, and linear population size reduction. L-SHADE uses a current-to-pbest/1/bin strategy for mutation operator, while all individuals have the same probability to be selected. In this paper, we propose a novel L-SHADE algorithm, RWS-L-SHADE, based on a roulette wheel selection strategy so that better individuals have a higher priority and worse individuals are less likely to be selected. Our extensive experiments on the CEC-2017 benchmark functions and dimensionalities of 30, 50 and 100 indicate that RWS-L-SHADE outperforms L-SHADE.

[Abstract] Protein folding is the dynamic process by which a protein folds into its final native structure. This is different to the traditional problem of the prediction of the final protein structure, since it requires a modeling of how protein components interact over time to obtain the final folded structure. In this study we test whether a model of the folding process can be obtained exclusively through machine learning. To this end, protein folding is considered as an emergent process and the cellular automata tool is used to model the folding process. A neural cellular automaton is defined, using a connectionist model that acts as a cellular automaton through the protein chain to define the dynamic folding. Differential evolution is used to automatically obtain the optimized neural cellular automata that provide protein folding. We tested the methods with the Rosetta coarse-grained atomic model of protein representation, using different proteins to analyze the modeling of folding and the structure refinement that the modeling can provide, showing the potential advantages that such methods offer, but also difficulties that arise.

[Resumen] Los manipuladores robóticos son sistemas altamente no lineales, y es difícil obtener un modelo matemático preciso con técnicas convencionales. Aplicando técnicas de control clásico se puede resolver este tipo de problemas, pero tiene el inconveniente de que se tarda una gran cantidad de tiempo en obtener un resultado satisfactorio para manipuladores donde se tenga una gran cantidad de grados de libertad (GDL). Es por esto, por lo que se requiere una técnica eficiente para tratar con este tipo de sistemas complejos y dinámicos. El algoritmo de Differential Evolution (DE) es una técnica de optimización global muy potente la cual en la actualidad se ha hecho popular debido a su posible aplicación en una gran cantidad de campos dentro de la robótica. En el presente trabajo se muestra un aplicación directa de este método de optimización en un modelo simulado de un brazo de 6 GDL y una posterior aplicación sobre un modelo de bajo coste., [Abstract] Robotic manipulators are highly non-linear systems, and it is difficult to obtain an accurate mathematical model with conventional techniques. Applying classical control techniques can solve this type of problem, but it has the disadvantage that it takes a large amount of time to obtain a satisfactory result for manipulators with a large number of degrees of freedom (GDL). Therefore, an efficient technique is required to deal with such complex and dynamic systems. The Differential Evolution (DE) algorithm is a very powerful global optimisation technique which has nowadays become popular due to its possible application in a large number of fields within robotics. This paper shows a direct application of this optimisation method on a simulated model of a 6 GDL arm and a subsequent application on a low-cost model.

This paper calibrates through loss functions the parameters of Heston’s stochastic volatility model by using two different methods: minimizing a nonlinear objective function (a loss function) with constraints on the values of the parameter and using a differential evolution algorithm. Both methods are applied to implied volatilities on the Mexican Stock Exchange Index with four maturities and twenty-eight strike prices. The selection criterion for the parameters is minimizing the value of the mean square error of the implied volatility. The first method has a better performance with less error and time. However, empirical results show that for both methods the adjustment of implied volatilities is better for options with longterm maturities than for short-term maturities.

Not all generate-and-test search algorithms are created equal. Bayesian Optimization (BO) invests a lot of computation time to generate the candidate solution that best balances the predicted value and the uncertainty given all previous data, taking increasingly more time as the number of evaluations performed grows. Evolutionary Algorithms (EA) on the other hand rely on search heuristics that typically do not depend on all previous data and can be done in constant time. Both BO and EA community typically assess their performance as a function of the number of evaluations, i.e., data efficiency. However, this is unfair once we start to compare the efficiency of these classes of algorithms, as the overhead times to generate candidate solutions are significantly different. We suggest to measure the efficiency of generate-and-test search algorithms as the expected gain in the objective value per unit of computation time spent, i.e., time efficiency. To the time-efficient search algorithm, we therefore propose a new algorithm, a combination of BO and an EA, BEA for short, that starts with BO, then transfers knowledge to an EA, and subsequently runs the EA. We compare the BEA with BO, the EA, Differential Evolution (DE), and Particle Swarm Optimization (PSO). The results show that BEA outperforms BO, the EA, DE and PSO in terms of time efficiency, and ultimately leads to better performance on well-known benchmark objective functions with many local optima. Moreover, we test BEA, BO, and the EA on nine test cases of robot learning problems and here again we find that BEA outperforms the other algorithms.

Differential Evolution (DE) has been widely appraised as a simple yet robust population-based, non-convex optimization algorithm primarily designed for continuous optimization. Two important control parameters of DE are the scale factor F, which controls the amplitude of a perturbation step on the current solutions and the crossover rate Cr, which limits the mixing of components of the parent and the mutant individuals during recombination. We propose a very simple, yet effective, nearest spatial neighborhood-based modification to the adaptation process of the aforesaid parameters in the Success-History based adaptive DE (SHADE) algorithm. SHADE uses a historical archive of the successful F and Cr values to update these parameters and stands out as a very competitive DE variant of current interest. Our proposed modifications can be extended to any SHADE-based DE algorithm like L-SHADE (SHADE with linear population size reduction), jSO (L-SHADE with modified mutation) etc. The enhanced performance of the modified SHADE algorithm is showcased on the IEEE CEC (Congress on Evolutionary Computation) 2013, 2014, 2015, and 2017 benchmark suites by comparing against the DE-based winners of the corresponding competitions. Furthermore, the effectiveness of the proposed neighborhood-based parameter adaptation strategy is demonstrated by using the real-life problems from the IEEE CEC 2011 competition on testing evolutionary algorithms on real-world numerical optimization problems. © 2022

This paper used an artificial jellyfish search (AJFS) optimizer suitable for selective harmonic elimination-based modulation for multilevel inverter (MLI) voltage control application. The main objective was to remove the undesired lower-order harmonics in the output voltage waveform of an MLI. This algorithm was motivated by the behavior of jellyfish in the ocean. Jellyfish have the ability to find the global best position where a large quantity of nutritious food is available. The paper applied AJFS algorithm on five, seven, and nine levels of CHB-MLI. The optimum switching angle was calculated for the entire modulation range for the desired lower-order harmonics elimi-nation. The problem formulated to achieve the objective was solved in a MATLAB environment. The total harmonic distortion (THD) values of five-, seven-, and nine-level inverters for various modulation indexes were computed using AJFS and compared with the powerful differential evolution (DE) algorithm. The comparison of THD results clearly demonstrated superior THD in the output of CHB-MLI of the AJFS algorithm over DE and GA algorithm for low and medium values of modulation index. The experimental results further validated the better performance of the AJFS algorithm.

The allocation of trucks in open pit mines is a field with great potential for optimizing resources and applying advanced computer modeling techniques, mainly because many companies still choose to use manual allocation, which is premised on the decisions made by the operator, being subject to common failures and not reaching the maximum potential that the equipment can provide. Therefore, this work focuses on optimizing the allocation of trucks in order to increase production, reducing queue time and keeping ore grades within proper limits. The proposed algorithm was based on the differential evolution technique, where two types of mutation operators were used: rand/1/bin and best/1/bin, thus verifying the most suitable to solve the problem. The trucks were allocated in the ore loading and unloading process, aiming to improve the production capacity in a virtual mine. The results brought a convergence to the maximum global production, in addition to which, the allocation of unnecessary transport equipment to the planned routes was avoided. The two mutation operators compared had certain advantages and disadvantages, each better adapting to certain types of situations. The proposed technique can still be extended to other areas, for example, in the transport of grain on the road network or in the implementation of an allocation in freight cars that transport grain.

The allocation of trucks in open pit mines is a field with great potential for optimizing resources and applying advanced computer modeling techniques, mainly because many companies still choose to use manual allocation, which is premised on the decisions made by the operator, being subject to common failures and not reaching the maximum potential that the equipment can provide. Therefore, this work focuses on optimizing the allocation of trucks in order to increase production, reducing queue time and keeping ore grades within proper limits. The proposed algorithm was based on the differential evolution technique, where two types of mutation operators were used: rand/1/bin and best/1/bin, thus verifying the most suitable to solve the problem. The trucks were allocated in the ore loading and unloading process, aiming to improve the production capacity in a virtual mine. The results brought a convergence to the maximum global production, in addition to which, the allocation of unnecessary transport equipment to the planned routes was avoided. The two mutation operators compared had certain advantages and disadvantages, each better adapting to certain types of situations. The proposed technique can still be extended to other areas, for example, in the transport of grain on the road network or in the implementation of an allocation in freight cars that transport grain.

Differential evolution (DE) is a simple and efficient population-based stochastic algorithm for solving global numerical optimization problems. DE largely depends on algorithm parameter values and search strategy. Knowledge on how to tune the best values of these parameters is scarce. This paper aims to present a consistent methodology for tuning optimal parameters. At the heart of the methodology is the use of an artificial neural network (ANN) that learns to draw links between the algorithm performance and parameter values. To do so, first, a data-set is generated and normalized, then the ANN approach is performed, and finally, the best parameter values are extracted. The proposed method is evaluated on a set of 24 test problems from the Black-Box Optimization Benchmarking (BBOB) benchmark. Experimental results show that three distinct cases may arise with the application of this method. For each case, specifications about the procedure to follow are given. Finally, a comparison with four tuning rules is performed in order to verify and validate the proposed method’s performance. This study provides a thorough insight into optimal parameter tuning, which may be of great use for users.

The past years there has been a significant increase in the popularity and commercial availability of consumer drones., Aerospace Engineering

With the increase in the use of Solar Electric Propulsion for Earth-centered applications, comes an additional challenge in the field of low-thrust spacecraft trajectory optimization. This MSc project implements a hybrid optimization approach, using control parametrization to reduce the number of optimization parameters and orbital averaging to decreasing computational speeds. Several improvements are made in comparison to previous works, using co-state scaling, and self-adaptive differential evolution to improve convergence. Additionally, variable-step integration of the orbital state averages improved propagation speed. The implemented optimization tool was applied to the GTO to GEO transfer problem, demonstrating improved convergence compared to the previous authors, with the exception of eclipse conditions, requiring further verification. Additionally, a transfer with significant plane changes to a space debris object in LEO was investigated, including oblateness and aerodynamic perturbations. Without any manual tuning, nor a-priori estimates, the optimization tool is able to find time-optimal trajectories. The tool shows remarkable flexibility, allowing additional perturbations or operational constraints, providing a powerful additional asset to Tudat for the preliminary design of low-thrust transfer trajectories., Aerospace Engineering

This paper considers a passive target localization problem based on the noisy time of arrival measurements obtained from multiple receivers and a single transmitter. The maximum likelihood (ML) estimator for this localization problem is formulated as a highly nonlinear and non-convex optimization problem, where conventional optimization methods are not suitable for solving such a problem. Consequently, this paper proposes an improved adaptive hybrid firefly differential evolution (AHFADE) algorithm, based on hybridization of firefly algorithm (FA) and differential evolution (DE) algorithm to estimate the unknown position of the target. The proposed AHFADE algorithm dynamically adjusts the control parameters, thus maintaining high population diversity during the evolution process. This paper aims to improve the accuracy of the global optimal solution by incorporating evolutionary operators of the DE in different searching stages of the FA. In this regard, an adaptive parameter is employed to select an appropriate mutation operator for achieving a proper balance between global exploration and local exploitation. In order to efficiently solve the ML estimation problem, this paper proposes the well-known semidefinite programming (SDP) method to convert the non-convex problem into a convex one. The simulation results obtained from the proposed AHFADE algorithm and well-known algorithms, such as SDP, DE and FA, are compared against Cramer-Rao lower bound (CRLB). The statistical analysis has been performed to compare the performance of the proposed AHFADE algorithm with several well-known algorithms on CEC2014 benchmark problems. The obtained simulation results show that the proposed AHFADE algorithm is more robust in high-noise environments compared to existing algorithms.

This paper considers the problem of constrained multi-objective non-linear optimization of planetary gearbox based on hybrid metaheuristic algorithm. Optimal design of planetary gear trains requires simultaneous minimization of multiple conflicting objectives, such as gearbox volume, center distance, contact ratio, power loss, etc. In this regard, the theoretical formulation and numerical procedure for the calculation of the planetary gearbox power efficiency has been developed. To successfully solve the stated constrained multi-objective optimization problem, in this paper a hybrid algorithm between particle swarm optimization and differential evolution algorithms has been proposed and applied to considered problem. Here, the mutation operators from the differential evolution algorithm have been incorporated into the velocity update equation of the particle swarm optimization algorithm, with the adaptive population spacing parameter employed to select the appropriate mutation operator for the current optimization condition. It has been shown that the proposed algorithm successfully obtains the solutions of the non-convex Pareto set, and reveals key insights in reducing the weight, improving efficiency and preventing premature failure of gears. Compared to other well-known algorithms, the numerical simulation results indicate that the proposed algorithm shows improved optimization performance in terms of the quality of the obtained Pareto solutions.

This work aims to describe the spatial distribution of flow from characteristics of the underlying pore structure in heterogeneous porous media. Thousands of two-dimensional samples of polydispersed granular media are used to (1) obtain the velocity field via direct numerical simulations, and (2) conceptualize the pore network as a graph in each sample. Analysis of the flow field allows us to distinguish preferential from stagnant flow regions and to quantify how channelized the flow is. Then, the graph’s edges are weighted by geometric attributes of their corresponding pores to find the path of minimum resistance of each sample. Overlap between the preferential flow paths and the predicted minimum resistance path determines the accuracy in individual samples. An evolutionary algorithm is employed to determine the “fittest” weighting scheme (here, the channel’s arc length to pore throat ratio) that maximizes accuracy across the entire dataset while minimizing over-parameterization. Finally, the structural similarity of neighboring edges is analyzed to explain the spatial arrangement of preferential flow within the pore network. We find that connected edges within the preferential flow subnetwork are highly similar, while those within the stagnant flow subnetwork are dissimilar. The contrast in similarity between these regions increases with flow channelization, explaining the structural constraints to local flow. The proposed framework may be used for fast characterization of porous media heterogeneity relative to computationally expensive direct numerical simulations. Article Highlights: 1.A quantitative assessment of flow channeling is proposed that distinguishes pore-scale flow fields into preferential and stagnant flow regions.2.Geometry and topology of the pore network are used to predict the spatial distribution of fast flow paths from structural data alone.3.Local disorder of pore networks provides structural constraints for flow separation into pre

Um dos problemas encontrados na área de Engenharia Elétrica é o problema de despacho econômico (PDE), que visa reduzir o custo total da energia calculando a geração de cada unidade vinculada à rede. Assim, o objetivo é gerar energia com o menor custo de combustível possível, atendendo à demanda e respeitando todas as restrições do sistema de geração. Uma variante do PDE é o PDE com ponto de carga da válvula (PDE-PCV). Este, é formulado como um problema de otimização restrito, não convexo e não diferenciável, não sendo possível resolvê-lo diretamente com métodos determinísticos. Portanto, propõe-se neste trabalho, analisar o desempenho do método heurístico Evolução Diferencial (ED) na resolução do PDE-PCV. Foi desenvolvido um programa de computador em Matlab e realizados testes com o PDE-PCV associado aos sistemas elétricos com 3, 13 e 40 geradores., One of the problems found in the area of Electrical Engineering is the problem of economic dispatch (PDE), which aims to reduce the total cost of energy by calculating the generation of each unit linked to the grid. Therefore, the purpose is to generate energy with the lowest possible fuel cost, meeting demand (consumers) and respecting all the restrictions of the generation system. A variant of the PDE is the PDE with valve loading point (PDE-PCV). This is formulated as a restricted, non-convex and non-differentiable optimization problem, and it is not possible to solve it with deterministic methods directly. Therefore, it is proposed in this work, to analyze the performance of the heuristic method Differential Evolution (DE) for the resolution of the PDE-PCV. A computer program was developed in Matlab software and tests with the PDE-PCV associated with electrical systems with 3, 13 and 40 generators were performed.

Differential evolution (DE) is a simple and efficient population-based stochastic algorithm for solving global numerical optimization problems. DE largely depends on algorithm parameter values and search strategy. Knowledge on how to tune the best values of these parameters is scarce. This paper aims to present a consistent methodology for tuning optimal parameters. At the heart of the methodology is the use of an artificial neural network (ANN) that learns to draw links between the algorithm performance and parameter values. To do so, first, a data-set is generated and normalized, then the ANN approach is performed, and finally, the best parameter values are extracted. The proposed method is evaluated on a set of 24 test problems from the Black-Box Optimization Benchmarking (BBOB) benchmark. Experimental results show that three distinct cases may arise with the application of this method. For each case, specifications about the procedure to follow are given. Finally, a comparison with four tuning rules is performed in order to verify and validate the proposed method’s performance. This study provides a thorough insight into optimal parameter tuning, which may be of great use for users.

In this research, we propose two Particle Swarm Optimisation (PSO) variants to undertake feature selection tasks. The aim is to overcome two major shortcomings of the original PSO model, i.e., premature convergence and weak exploitation around the near optimal solutions. The first proposed PSO variant incorporates four key operations, including a modified PSO operation with rectified personal and global best signals, spiral search based local exploitation, Gaussian distribution-based swarm leader enhancement, and mirroring and mutation operations for worst solution improvement. The second proposed PSO model enhances the first one through four new strategies, i.e., an adaptive exemplar breeding mechanism incorporating multiple optimal signals, nonlinear function oriented search coefficients, exponential and scattering schemes for swarm leader, and worst solution enhancement, respectively. In comparison with a set of 15 classical and advanced search methods, the proposed models illustrate statistical superiority for discriminative feature selection for a total of 13 data sets