Search

This paper presents a novel approach to product design optimization through the synergy of Genetic Algorithms (GA) and specific AI techniques. By iteratively evolving solutions inspired by natural selection principles and utilizing AI for intelligent analysis of vast datasets, designers can efficiently navigate design spaces to identify optimal solutions meeting multiple constraints and objectives. The integration of Convolutional Neural Networks (CNNs) with GA resulted in a 30% faster convergence rate and improved solution quality, facilitating enhanced exploration-exploitation trade-offs. Case studies ranging from automotive engineering to consumer electronics illustrate the effectiveness of these methods in enhancing design efficiency by 25%, reducing costs by 20%, and accelerating time-to-market by 15%. The proposed approach demonstrated superior performance metrics, including improved efficiency, durability, operational safety, and minimized environmental impact. Comparative analysis revealed the advantages of this method over traditional techniques, validating its transformative potential in advancing product design paradigms. The approach is validated through various case studies, highlighting significant improvements in the design process and underscoring its advantages over traditional techniques.

A fundamental challenge in virology research lies in effectively detecting malicious code. N-gram analysis has become a cornerstone technique, but selecting the most informative features, especially for longer n-grams, remains crucial for efficient detection. This paper addresses this challenge by introducing a novel feature extraction method that leverages both adjacent and non-adjacent bi-grams, providing a richer set of information for malicious code identification. Additionally, we propose a computationally efficient feature selection approach that utilizes a genetic algorithm combined with Boosting principles. Our experimental results show that this detection system significantly outperforms existing methods in virus detection accuracy. The system improves detection accuracy by 15% and reduces false positives by 20% compared to traditional n-gram techniques. Additionally, it cuts computational overhead by about 30%, making it suitable for real-time applications. These advancements demonstrate the effectiveness and practicality of our approach. Future research will focus on applying our methods to polymorphic viruses and other malware to further enhance their robustness and applicability.

The numerical derivation of high-order Extended Stability Runge-Kutta (ERSK) integration schemes presents significant challenges due to the high-dimensional, non-convex, non-linear optimization problems with hard equality constraints. Traditional optimization techniques often struggle to converge on valid solutions efficiently. This paper proposes a hybrid approach that integrates Genetic Algorithm (GA) techniques with an interior-point optimizer to address these challenges. GA-generated values are used as starting points for the interior-point method, which refines candidate solutions to meet the stringent order conditions. Experimental results demonstrate a significant improvement in efficiency when GA-generated values are used. The average number of iterations required for convergence is reduced from 477 (with random initialization) to 143 when GA values are used. The median number of iterations drops from 280 to 105. The standard deviation is also reduced from 361 to 112, indicating more consistent performance. These reductions highlight the efficiency of the GA-plus-polish approach, which outperforms traditional random initialization in terms of convergence time and consistency. Despite some remaining variance, this hybrid method significantly reduces the computational cost of deriving stable Butcher tableaux, particularly in high-dimensional, non-convex optimization problems.

The development and the upgrade of railway networks is one of the strategies to reach decarbonization targets in the transportation field, thanks to the considerably lower energy consumption of electric trains with respect to other vehicles, typically fossil fuel powered. The design process of electric railway power systems is complex, requiring advanced simulation tools. The paper proposes a novel methodology for the design of the electrical power system of railway tracks, using genetic optimization. For this purpose, the authors developed ROAR, a flexible simulation and optimization software that generates optimized railway power system designs, helping engineers find the most efficient design solutions from a technical and economic feasibility perspective. After validating the simulation engine and comparing it with well-established software, the proposed method was applied to an operational electrified railway line in Italy to assess the effectiveness of the optimization algorithm. The results demonstrate excellent convergence properties, finding a different infrastructure design that achieves the same electrical performance, reducing costs with respect to the existing design.

In this study, a method that automatically generates regular expressions using genetic algorithms is designed to extract relevant images on web pages. Data extraction, which is usually done with web scrapers, can also be done with regular expressions. The complexity of regular expressions and the fact that they require expert knowledge make their writing difficult. With this study, a regular expression is automatically created to obtain relevant images of news content on websites. With the principle of genetic algorithms, the survival of the good and the elimination of the bad, a regular expression that can reach the most relevant image is produced. Thus, instead of a time-consuming and error-prone method such as creating the appropriate pattern for each site with web scraper tools, automatic regular expression generation using genetic algorithm methods can be used as a better method. A data set containing text-based related and irrelevant images from 200 websites collected from 58 countries was used in the study. There are 22,682 relevant images among 635,015 image data in the dataset. With the method developed using the genetic algorithm, the rate of accessing the relevant images by regular expressions produced by only looking at the relevant image data is approximately 98.49%.

IoT has recently witnessed a boom in AI deployment at the edge as a result of the newly developed small size Machine Learning (ML) models and integrated hardware accelerators. Although it brings huge benefits such as privacy-preserving and low-latency applications, it still suffers from typical resource limitations of edge devices. A new approach aims to deploy multiple inference models varying in size and accuracy onboard the edge device which could alleviate some of these limitations. This dynamic system can be leveraged to provide real-time energy efficient application by smartly allocating inference tasks to inference local models or offload to edge servers based on current constraints. In this work, we tackle the problem of efficiently allocating inference models for a given set of inference tasks between local inference models and edge server models in parallel under given time and energy constraints. This problem is considered strongly NP-hard and therefore we propose LITOSS, a 2-stage framework in which we use a lightweight Genetic Algorithm-based schemer for task scheduling along with a Reinforcement Learning (RL) agent for improving edge server selection. We perform experiments using a raspberry pi with a set of edge servers. Results show that our framework performed relatively faster compared to other meta-heuristic schemes such as LGSTO, Ant Colony Optimization (ACO) and Particle Swarm Optimization (PSO) while providing higher average accuracy. We also show that using an RL agent to select the best subset of available edge servers increased, or maintained in worst cases, the average accuracy while reducing the average scheduling times.

In recent years, considerable research efforts have been invested in airborne wind energy systems (AWES) due to traditional wind turbine technologies’ technical and economic limitations. This study describes the design process of a multimode electrical machine for a Fly-Gen AWES, employing a multi-physics approach analytical model with a multi-objective genetic algorithm. The design process integrates an analytical electromagnetic model for a permanent magnet synchronous machine (PMSM) within a multiphysics framework, incorporating thermal and mechanical constraints. To enhance the power density, the design utilizes a cobalt-iron magnetic material. The proposed method is evaluated with various constraints to validate its versatility successfully. Subsequently, a 2.5 kW motor/generator unit is optimized under a 600 V DC voltage condition and verified with 2D finite element method (FEM) analysis. The optimized machine achieves a power density of 1.8 kW/kg with an efficiency of 96%. Thermal analysis confirms that the winding and magnet temperatures remained below the critical temperature thresholds in successive motor and generator operations. Similarly, mechanical strength analysis results in satisfying mechanical limits. These findings highlight the potential of the proposed efficient and reliable machine design method.

An exponential growth and complexity in diverse distribution systems have contributed to protection coordination challenges. Initially, protection coordination schemes were achieved by means of conventional techniques; however, the utilisation of such methods is based on trial-and-error principles and laborious. Consequently, current studies have adopted the utilisation of particle swarm optimization, artificial intelligence models, and genetic algorithms to optimise overcurrent relay selectivity and operational speed. Particle swarm optimization, artificial intelligence, and genetic algorithms are optimization techniques that at times converges prematurely due to poor selection of control parameters and lack of optimal values, which results in increased computational time. Therefore, this paper presents a comprehensive review of recent developments in terms of parametric sensitivity analysis, selection of artificial intelligence models based on data availability, and the likelihood of solving overcurrent relay coordination problems. The reviewed literature shows that particle swarm optimization performance is greatly influenced by inertia weight and swarm size, while the number of iterations has insignificant effect. The findings also indicate that crossover rate, mutation probability, and population size affect genetic algorithms behaviour. Artificial intelligence models lack sensitivity study for parametric tuning, that is, number of hidden layers, membership functions, epsilon in support vector machine, and number of fuzzy rules affects the models’ performance.

In the link prediction problem a relevant algorithm running over a network attempts to determine whether a link between two nodes will exist in the future, given that it is not present at the moment. Most link prediction algorithms take into account the structure of the network on which they are applied and based on this, they attempt to predict the existence or not of future new edges in the network. However, many of them are quite standardized, applying the same concept and parametrization to all networks, thus not always achieving good results in every different network structure. Algorithms based on Graph Neural Networks (GNNs) are more adaptive to any network structure but they do not give appreciable results when the only information available is the network structure. In this paper, we propose a new approach to this problem that approximates the structure of a complex network by allowing adjusted weight to this network structure to create additional information, which we can embed into effective algorithms such as local and superposed random walk link prediction. To achieve this goal, we use well-known kernel functions such as Sigmoids, in which we fit their parameters appropriately by a genetic algorithm to achieve the best possible approximation. To demonstrate the effectiveness of our proposed method we have compared our prediction method results based on precision, AUC and AUPR on eleven selected networks of different structures and properties with seven well-known link prediction algorithms and one more utilizing GNNs. In every case, we have improved the results of random walk algorithms and in most cases we achieved better results from all employed benchmark algorithms.

The aim of this article is to investigate the optimization of electric vehicle charging and swapping for green logistics, as well as path planning considering urban impedance. We improved a road impedance function model suitable for urban road traffic in China to calculate the actual traffic time based on real-time traffic data and the intricate urban road environment. This model is then applied to address the delivery optimization problem. Furthermore, a robust computational approach is introduced to estimate battery degradation costs, accounting for environmental temperature and depth of discharge. The logistics delivery model is effectively tackled using genetic algorithms, and simulation results demonstrate the considerable advantages of electric vehicle swapping, effectively mitigating energy wastage and environmental pollution. Additionally, the integration of road impedance modeling for path optimization proves to significantly reduce logistics costs, time expenditures, and enhance logistics efficiency. A comprehensive sensitivity analysis is also conducted to elucidate the factors influencing electric vehicle battery degradation, revealing a direct correlation between higher temperatures, deeper discharge depth, and increased battery loss. The study underscores the paramount significance of this research for the development and optimization of urban green logistics systems.

In the rapidly evolving domain of wireless networks, adaptive beamforming stands as a cornerstone for achieving higher data rates, enhanced network capacity, and reduced latency. This study introduces a novel integration of deep neural networks (NNs) into adaptive beamforming, specifically for uniform planar arrays (UPAs). We embark on an exploration of different NN architectures (a deep feedforward NN, a gated recurrent unit, and a long short-term memory based recurrent NN) and hyperparameter tuning techniques (grid search, Bayesian optimization, and genetic algorithm) to assemble a model that yields the most promising zero forcing beamforming performance. This thorough examination not only unveils the most prominent model but also highlights the most efficient hyperparameter tuning method. A comprehensive dataset generated using the null steering beamforming algorithm applied to an $8\times 8$ UPA, serves as the bedrock for training these networks. Our evaluation benchmarks the NN-based beamformers against traditional methods, assessing them on crucial metrics such as beamforming competence, response time, and adaptability under various noise and interference scenarios. A significant innovation of this research is the introduction of a continual learning approach, enhancing the NN beamformers’ performance in dynamically changing environments. This self-improving algorithm represents a stride towards more adaptable and efficient beamforming, showcasing improvement on underperforming scenarios without compromising accuracy. Our findings demonstrate the potential of neural networks in reshaping the future of adaptive beamforming, offering a blend of speed and precision that is paramount in modern wireless networks.

Various mobile and edge devices have significantly different processing capabilities, making it challenging to develop a single universal architecture of a neural network to extract facial embeddings. In this paper, we study the automated machine learning techniques to design a neural network with the best performance on a concrete device. The novel procedure is proposed to choose the better subnetwork of the Supernet based on a genetic algorithm with a surrogate binary classifier to compare the expected accuracy of two subnetworks. The latter uses only encoding of a candidate subnetwork and does not require directly estimating its accuracy on a validation set. As a result, the most computationally efficient and accurate model in TensorFlow Lite format is obtained in less than 10 minutes for a specific device and latency constraint. An Android demo application has been developed to demonstrate the potential of designed neural networks. It is experimentally shown that the proposed approach is universal: it can extract deep embeddings for tasks such as face verification and facial expression recognition and for various types of devices, including smartphones and Raspberry Pi single-board mini-computers. Our models process one facial image in real-time and achieve much higher accuracy when compared to the best-known lightweight networks.

There are many reasons for traffic congestions such as the “stop-and-go wave effect”, periodic increase in intensity of vehicle and pedestrian traffic (“rush hours”), frequent maneuvering, uncontrolled pedestrians’ movement on road crossings and other factors. This paper considers the problem of biobjective optimization and rebalancing of vehicles’ and pedestrians’ flows with the use of Manhattan road networks (MRNs) with smart traffic lights (STLs) as the case study of intelligent transportation system (ITS). For this purpose, we have studied the possibilities of applying STLs to control of traffic in large-scale road networks providing a speed harmonization and traffic prioritization between vehicles and pedestrians. The considered multiagent system (MAS) includes agent vehicles, agent pedestrians and agent lights that interact with each other according with given rules (e.g., V2V, V2P, V2I). Such STLs use information on the traffic structure and its density to switch signals at each moment in time. In the non-stationary mode with a periodic traffic intensity to provide the analysis of traffic flows done by STLs it has been suggested to use the fuzzy clustering algorithm aggregated with the density-based spatial clustering algorithm (FCA-DBSCAN). At the uniform motion fixed durations of phases set up for STLs that computed individually with use of the suggested parallel hybrid genetic algorithm (BORCGA-BOPSO). The proposed algorithm combine the use of the Biobjective Real-Coded Genetic Algorithm (BORCGA) for an evolutionary search based on heuristic operators and the Biobjective Particle Swarm Optimization Algorithm (BOPSO) based on simple way to search for optimal solutions. The approach allows to improve significantly the time-efficiency of seeking optimal individualised STLs’ characteristics while keeping up their quality. Moreover, the ITS based on STLs with parameters optimized with the BORCGA-BOPSO provides significant traffic improvement (i.e., a maximization of the vehicle and pedestrian outflows) in MRNs in contrast to the case of uncontrolled pedestrian crossings and using usual (i.e., non-smart) traffic lights.

Designing Convolutional Neural Networks (CNNs) for a specific task requires not only Deep Learning expertise but also knowledge of the problem. The goal of Neuroevolution is to find CNN architectures automatically through evolution. The search time, however, is a critical problem in Neuroevolution since multiple CNNs must be trained in the evolutionary process. In this work, we propose a Neural Architecture Classifier (NAC) to avoid training architectures that would not have good performance, based on knowledge of previously trained architectures. The NAC evaluates each CNN using the CNN architecture encoding as the input. A genetic algorithm (GA) is used for evolution, and a search space with few restrictions; hence, CNN architectures can have any width, depth or shape as well as the suitable number of skip connections. We applied this methodology to solving pattern recognition problems on six well-known datasets: five subsets of the MNIST-Variations (MNIST-V) dataset and the CIFAR-10 dataset. Experiments demonstrated that integrating the NAC in the GA reduces the search time by up to 44% compared to conventional GAs. Additionally, the architectures found by evolution with our classifier achieved a better performance (4.6% on average) than those found by traditional evolution. Our results improved the state-of-the-art by 5.2% (0.33% error reduction) on the most difficult dataset of the MNIST-V (MRDBI), and an average of 4% (0.24% error reduction) on the MNIST-V datasets. Additionally, our results on CIFAR-10 are close to those of the state-of-the-art but with significantly reduced search time. The application of a classifier can be relevant and useful not only for genetic algorithms, but also for other evolutionary algorithms.

As a method to optimize the power frequency of dual-frequency plasma, we propose an optimization method that combines genetic algorithm and plasma simulation. A two-dimensional plasma simulation model of Ar plasma was constructed with a fluid model. Combining this simulation model with a genetic algorithm, 300 cases of plasma conditions with high plasma density and small variation in electron density were calculated. As a result, the optimum conditions were 175 to 210 MHz for the high-frequency generator and 1.0 to 2.5 MHz for the low-frequency generator. Furthermore, two types of single objective functions that combine the electron density and the variation rate were proposed. The single objective functions with emphasis on electron density and with emphasis on variation rate were introduced, and each objective function was optimized. [ABSTRACT FROM AUTHOR]

Software testing (ST) is one of the most important software development life cycle (SDLC) phases and ST effort is often expressed as a percentage of SDLC effort. Unfortunately, in the literature ST effort percentage ranges from 10% to 60%. In the literature most of the machine learning algorithms and metaheuristics for optimizing them have looked at predicting overall SDLC effort without focusing on any specific SDLC phase, including testing. Therefore, this study investigates the application of the Software Testing Effort Prediction (STEP) of Gradient Boosting (GB) machine learning regression algorithm optimized through Differential Evolution (DE). Its prediction accuracy is compared with those obtained when the GB is also optimized through Particle Swarm Optimization (PSO) and Genetic Algorithms (GA). The performance of GB-DE, GB-PSO, and GB-GA was also compared to that of statistical regression (SR). Seven data sets of actual projects were selected from an international public repository for software projects. The results showed that GB-DE was statistically better than SR in all seven data sets at 95% confidence, whereas GB-PSO and GB-GA were better than SR in four and three data sets, respectively. Thus, we can conclude that GB-DE can be used for STEP of either new projects or enhancement projects developed in either the third or fourth programming language generation.

The swift and automated design of antennas remains a challenging aspect in research due to the specific design needs for individual applications. Alterations in resonance frequency or boundary conditions necessitate time-consuming re-designs. Though the application of evolutionary optimization and generative methods in general to antenna design has seen success, it has been mostly restricted to two-dimensional structures. In this work, we present an approach for designing three-dimensional antennas using a genetic algorithm coupled with a region-growing algorithm - to ensure manufacturability - implemented in Matlab manufactured via laser powder bed fusion (LPBF). As a simulation tool for optimization CST is used. The antenna has been optimized in a completely automated manner and was produced using the metal 3D printing technology LPBF and aluminium based AlSi10Mg powder. The presented concept, which builds upon previous two-dimensional techniques, allows for significant flexibility in design, adapting to changing boundary conditions, and avoiding the geometric restrictions seen in prior methods. The optimized antenna has a size of $3.01 \text {cm} \times 3.43 \text {cm} \times 1.67 \text {cm}$ and was measured in an anechoic chamber. According to measurements a minimum reflection coefficient of $\mathrm {-19.95\,\, \text {dB}}$ at $\mathrm {2.462~ \text {G} \text { Hz} }$ and a bandwidth of $\mathrm {308.8~ \text {M} \text { Hz} }$ are observed. CST simulation results predict an efficiency of $\mathrm {98.91~\%}$ and the maximum antenna gain is measured at $\mathrm {2.45~ \text {G} \text { Hz} }$ to be $\mathrm {3.27~ \text {dB} i}$ . Simulations made with CST and Ansys HFSS and measurements are in excellent agreement with a deviation of the resonance frequency of only $\mathrm {0.13~\%}$ , thus further establishing genetic algorithms as a highly viable option for the design of novel antenna structures.

The study of fatigue load reduction in wind turbines has gained significant interest in recent years. In monopile offshore wind turbines, vibrations induced by misalignment between waves and wind directions can considerably shorten the tower’s lifespan due to lateral loads. To cope with this issue, this work focuses on the design and evaluation of control strategies to mitigate these loads. An adaptive active generator torque control (AGTC) scheme is proposed to mitigate lateral tower fatigue within the nominal region. Two variants of AGTC are suggested: with or without using a bandpass filter. A multi-objective optimization using genetic algorithms is employed to determine the optimal AGTC parameters based on wind speed and wave height. Objective functions related to tower and low-speed shaft fatigue load reduction, as well as power fluctuation minimization, are employed. The performance of the proposed AGTC controllers is assessed in comparison to a baseline controller. The results highlight that the AGTC strategy without a filter effectively reduces lateral tower fatigue load at the expense of higher power signal oscillations and increased fatigue in the low-speed shaft. Using a filter in the AGTC mitigates these adverse effects, smoothing power oscillations to achieve similar values as the baseline control, and significantly reducing the load experienced in the low-speed shaft. The proposed methodology is robust and adaptable to various wind and wave conditions, demonstrating its ability to enhance the structural integrity of wind turbines without compromising power generation efficiency.

In this work, we present a novel clustering approach which is inheriting the best characteristics of Unsupervised Functional Link Artificial Neural Network (UFLANN) and Genetic Algorithms (GAs) for uncovering clusters embedded in dataset represented through $(X)_{Nxd}$ , where X consists of $N$ data points with $d$ -dimensions. With an aim to realize natural clusters in a linear space UFLANN mapped the input vectors from lower dimension to higher dimension with a greater hope to achieve linearity in higher dimensional space. As a result, UFLANN introduces the problem of curse of dimensionality in the given datasets. However, it has been realized that the problems like sparse data and distance concentration associated with curse of dimensionality cast this problem to again a very complex problem. Hence to address some of the issues of curse of dimensionality, we have used GAs for selecting optimal number of features in the higher dimension for UFLANN to discover clusters embedded in the dataset. The proposed approach herein after named as GAUFLANN has been experimentally evaluated by using the metrics like (i) Davies-Bouldin Index, ii) Silhouette Score, and iii) Completeness score on different synthetic and real datasets. Our experimental study confirms that GAUFLANN is evidently scoring better in DB-index, Silhouette score, and Completeness score than the clustering methods like K-means, Hierarchical-Agglomerative (Average Linkage), and UFLANN across the datasets like Circles, Moons, Iris, and CORD-19.

Manufacturing is a broad field with different types of production processes. Therefore, specific processes can accommodate multiple parallel machines operating simultaneously in some production environments. This assumption is particularly crucial in factory scheduling for industries such as textile, aircraft, and semiconductor manufacturing. The assumption proposed in this paper is differ from FJSP in that they are closer to the real world by allowing different machines to perform the same multiple processes. In this paper, a new approach for solving the flexible job shop scheduling problem has been proposed, which is referred to as the flexible job shop scheduling problem in a parallel machine environment with a variable group. We proposed a hybrid genetic algorithm-based variable neighborhood search algorithm (GA-VNS) to solve this problem, where the multiple parallel machines can operate the same operations simultaneously and are grouped as a variable group. The GA-VNS algorithm combines the global searching ability of GA with the local searching ability of VNS to fully reflect the condition of the parallel machines with variable groups. The objective functions of the algorithm are to minimize the production makespan and to improve facility usage. Multiple simulation experiments are conducted using forty-seven test instances to assess the feasibility and effectiveness of the new approach. This study is expected to reduce processing time and production costs for Small and Medium Enterprises (SMEs) with low cost and high efficiency compared to existing systems, such as the manufacturing execution system, enterprise resource planning, and job shop scheduling system.

Face Recognition (FR) has been a widely used biometric technique for identity authentication in various domains. Despite the remarkable progress in the field of FR during the past few years, there are still challenges that need to be addressed, including recognition of faces with large pose variation. Following the idea of organic face patches found in and used by human and macaque monkey brains, in this paper we report a set of convolutional neural networks (CNNs) that we have designed to represent face patches specialized in a particular face pose orientation range. We use neuroevolution, employing genetic algorithms (GAs), to define the structure of three CNNs. Each CNN is evolved through a genetic algorithm, to a particular face pose orientation range for small, medium, and large face rotations. Each CNN was evolved to improve FR accuracy using the training partition of the VGGFace2 dataset, split into three sets for small, medium, and large face rotations. The GAs generated new specialized CNNs tuned to the three face orientation ranges: small, medium, and large face rotations. Following standard procedures, using extended training for a longer number of epochs, the best CNN individuals were trained with VGGFace2 (training set), and MS1M datasets. The performance of our proposed method was assessed on several datasets that contain a significant number of faces with large pose variation: VGGFace2 (test set), CPFLW, and CFP_FP. The results reached accuracies greater than those of the state-of-the-art of 95.91%, 95.73%, 94.60% and 99.18% on VGGFace2(test), VGGFace2_FP, CPFLW and CFP_FP, respectively.

Accurate velocity information is often essential to the control of robot manipulators, especially for precise tracking of fast trajectories. However, joint velocities are rarely directly measured and instead estimated to save costs. While many approaches have been proposed for the velocity estimation of robot joints, no comprehensive experimental evaluation exists, making it difficult to choose the appropriate method. This paper compares multiple estimation methods running on a six degrees-of-freedom manipulator. We evaluate: 1) the estimation error using a ground-truth signal, 2) the closed-loop tracking error, 3) convergence behavior, 4) sensor fault tolerance, 5) implementation and tuning effort. To ensure a fair comparison, we optimally tune the estimators using a genetic algorithm. All estimation methods have a similar estimation error and similar closed-loop tracking performance, except for the nonlinear high-gain observer, which is not accurate enough. Sliding-mode observers can provide a precise velocity estimation despite sensor faults.

A fitness function is a type of objective function that quantifies the optimality of a solution; the correct formulation of this function is relevant, in evolutionary-based ATS systems, because it must indicate the quality of the summaries. Several unsupervised evolutionary methods for the automatic text summarization (ATS) task proposed in current standards require authors to manually construct an objective function that guides the algorithms to create good-quality summaries. In this sense, it is necessary to test each fitness function created to measure its performance; however, this process is time consuming and only a few functions are analyzed. This study proposes the automatic generation of heuristic functions, through genetic programming (GP), to be applied in the ATS task. Therefore, our proposed method for ATS provides an automatically generated fitness function for cluster-based unsupervised approaches. The results of this study, using two standard collections, demonstrate to automatically obtain an orientation function that leads to good quality abstracts.

Hyperparameter optimization is a fundamental part of Auto Machine Learning (AutoML) and it has been widely researched in recent years; however, it still remains as one of the main challenges in this area. Motivated by the need of faster and more accurate hyperparameter optimization algorithms we developed HyperBRKGA, a new population-based approach for hyperparameter optimization. HyperBRKGA combines the Biased Random Key Genetic Algorithm with an Exploitation Method in order to search the hyperparameter space more efficiently than other commonly used hyperparameter optimization algorithms, such as Grid Search, Random Search, CMA-ES or Bayesian Optimization. We develop and test two different alternatives for this Exploitation Method: Random Walk and Bayesian Walk. We also discuss and implement other schemes, such as a Training Data Reduction Strategy and a Diversity Control strategy, in order to further improve the efficacy of our method. We performed several computational experiments on 8 different datasets to assess the effectiveness of the proposed approach. Results showed that HyperBRKGA could find hyperparameter configurations that outperformed in terms of predictive quality the baseline methods in 6 out of 8 datasets while showing a reasonable execution time. Lastly, we conducted an ablation study and showed that the addition of every component was relevant to achieving high quality results.

Multi-level DC-DC converters have been widely used in automotive and other high-power applications. Thus, the control of these text multi-level converters is an emerging thematic in power electronics to ensure their proper functioning. This paper provides a novel nonlinear control of a text DC-DC three level boost converter (T-LBC) based on a backstepping (BS) technique with an integral action and is optimized using genetic algorithms (GA). Firstly, the average state model of the text T-LBC is described. Then, this model is used to design an integral BS controller; nevertheless, the controller parameters are often determined manually, which may degrade the control quality. A genetic algorithm-based optimization method is applied to establish the best controller gains and improve the proposed controller efficiency. The asymptotic stability converter is verified using the Lyapunov method criteria. In order to validate the introduced controller under different scenarios, the Matlab/Simulink environment is used. In addition, it is compared with different controllers such as conventional backstepping, fuzzy logic, and proportional–integral–derivate (PID) controllers under varying references to highlight its performance further. Finally, the designed controller is verified experimentally by implementing it using a dSPACE 1104 control board. The simulation and experimental results show that the optimized integral BS controller presents the best performances in terms of settling time, overshoot and steady-state error.

Heuristic search algorithms are informed search strategies that use heuristics to estimate the minimal cost of the path from the current state to the goal. Using this additional knowledge, this type of algorithms can distinguish non-goal states, and then directs its search towards those that look more promising. This makes these algorithms faster than uninformed search algorithms. The A* algorithm is a well-known example of heuristic-based algorithms that is guaranteed to find the least-cost path to a goal state if the heuristic used is admissible, which means that it never overestimates the real cost from the current state to the goal. In this article, we are going to present a neural network architecture that makes it possible to predict admissible heuristics for a given graph and destination node, then we are going to refine our models using genetic algorithms in order to obtain better results.

Monitoring and surveillance of buildings, especially in critical infrastructure is crucial to increase the lifetime of such structures. This is due to the fact, that concrete buildings, like bridges or road tunnels are prone to corrosion due to their exposure to the elements. For this task, wireless sensor nodes can be employed in structural health monitoring. In the last few years passive RFID sensors have been proposed to gather data from inside concrete structures. However, designing antennas which are operational inside concrete to transmit data from the sensor node to the outside world is extremely challenging, due to the fact that the performance of antennas is extremely depended on their surrounding material. In this work, the fully automatized design of an antenna for the UHF band in concrete with varying electrical properties is demonstrated. Using a novel multi-objective optimization method, the antennas are operational over a wide range of dielectric properties of the surrounding materials. In contrast to conventional approaches, no air spacers or boxes are needed around the antenna. The antenna parameters have been carefully obtained during the curing and drying process, in an outdoor like environment with 65% relative humidity. After 157 days of drying, a reflection coefficient of −13 dB and an antenna gain of −84 dB is achieved, at a water content of the concrete of 4.2% by mass.

Genetic algorithms (GAs) have been used to evolve optimal/sub-optimal solutions of many problems. When using GAs for evolving solutions, often fitness evaluation is the most computationally expensive, and this discourages researchers from applying GAs for computationally challenging problems. This paper presents an approach for generating offspring based on a local fitness landscape exploration to increase the speed of the search for optimal/sub-optimal solutions and to evolve better fitness solutions. The proposed algorithm, “Fitness Landscape Exploration based Genetic Algorithm” (FLEX-GA) can be applied to single and multi-objective optimization problems. Experiments were conducted on several single and multi-objective benchmark problems with and without constraints. The performance of the FLEX-based algorithm on single-objective problems is compared with a canonical GA and other algorithms. For multi-objective benchmark problems, the comparison is made with NSGA-II, and other multi-objective optimization algorithms. Lastly, Pareto solutions are evolved on eight real-world multi-objective optimization problems, and a comparative performance is presented with NSGA-II. Experimental results show that using FLEX on most of the single and multi-objective problems, the speed of the search improves up to 50% and the quality of solutions also improves. These results provide sufficient evidence of the applicability of fitness landscape approximation-based algorithms for solving real-world optimization problems.

Grasp stability recognition based on tactile perception has attracted increasing attention in the robotics community. In this article, we extract tactile features from multimodal tactile signals and propose a novel ensemble approach named genetic algorithm-based ensemble hybrid sparse extreme learning machine (GA-EHSELM) for the grasp stability recognition task. In contrast to traditional ensemble extreme learning machines (ELMs), the proposed approach takes the sparsity of the random weights assigned to the connection between the input layer and the hidden layer for each base learner into consideration, in order to avoid the problem of overfitting. In addition, The diversity of base learners is increased by constructing two types of sparse ELM (SELM). The random weights of the base SELMs are sampled from two different distributions with a certain probability. Furthermore, we utilize the genetic algorithm (GA) for the optimization of tactile features and the base sparse ELMs. To be specific, GA is employed for feature selection as well as the optimization of the sparsity of the random weights associated with different types of base learners and the number of each type of base learner. The effectiveness of the proposed approach has been demonstrated on a public tactile grasp stability dataset. [ABSTRACT FROM AUTHOR]

This paper studies an unrelated parallel machine photolithography scheduling problem with dual resource constraints (DRC). This is an emerging problem in the semiconductor industry, where the masks can be transferred from one working station to another, increasing the flexibility of production through sharing the auxiliary resource. It gives rise to a complex DRC scheduling problem involving jobs, machines, and masks, restricted by sequence-dependent setup time and station-dependent mask transfer time. This study proposes two mixed-integer programming (MIP) models (Model 1 and Model 2) and an improved naked mole-rat algorithm that hybrids with a genetic algorithm (GA) and variable neighborhood search algorithm (VNS). Numerical experiments show that Model 2 performs better compared with Model 1, and the improved naked mole-rat algorithm brings a significant improvement over GA and VNS. [ABSTRACT FROM AUTHOR]

Color classification of polycrystalline silicon solar cells is really challenging for performing the task of production quality control during the manufacturing due to the non-Gaussian color distribution and random texture background. The motivation of this work is to present a robust color classification technique by designing a novel tiny color difference feature descriptor. Thus, a genetic algorithm based color difference histograms (GACDH) is proposed. First, the optimal color space of color difference histogram (CDH) to represent tiny color changes is designed. It counts the perceptually uniform color difference in a small local neighborhood in the L*a*b* color space, which reduces the false classification due to small color variations and illumination variation. Second, the genetic algorithm based color quantization for CDH is proposed to select the optimal quantization bins in the L* component, then we make some comparative experiments in a* and b* color components to select optimal quantization bins. The optimization of feature dimension not only reduces the large dimensionality of histogram bins in the computation but also improves the following classification performance. Finally, the proposed algorithm is validated with color dataset of solar cells with distance measure method. Some experimental results and analysis show that the overall performance of the proposed method achieves 98.6% and outperforms other techniques available in the literature in terms of weak discriminative color difference classification. [ABSTRACT FROM AUTHOR]

Splitting and placing interfacing inductance on both sides of the transformer has been proven to be an effective method, which extends the zero-voltage switching region for all the switching devices in the dual active bridge (DAB) converter. With the trend toward operating in higher frequency, achieving higher power density and higher efficiency, the converter model becomes more complex due to the non-negligible parasitic components that brings new challenges to DAB converter design. Traditional analytical methods have made it hard to imitate the proposed converter easily and precisely. Thus, artificial intelligence techniques are able to be utilized to assist the design process. When considering the converter system as a gray-box model, the metaheuristic algorithm can be implemented for the targeted design inside such a gray-box. In this article, a genetic algorithm (GA) is employed in the DAB converter parametric design with an explicit fitness desire to help in discovering the high frequency oscillation (HFO) problem. Consequently, the splitting inductance tuning method is proposed for eliminating the HFO problem and minimizing inductors’ loss. The methodology of implementing GA into converter parametric design, and the proposed splitting inductance tuning method are introduced and verified with a 1 MHz gallium nitride high-electron-mobility transistor based DAB converter prototype. The comparitive experimental results prove the effectiveness of the splitting inductance tuning method and achieve 4% efficiency enhancement with 200 W power delivering. [ABSTRACT FROM AUTHOR]

Accurate land use and land cover (LULC) maps are effective tools to help achieve sound urban planning and precision agriculture. As an intelligent optimization technology, genetic algorithm (GA) has been successfully applied to various image classification tasks in recent years. However, simple GA faces challenges, such as complex calculation, poor noise immunity, and slow convergence. This research proposes a two-step ensemble protocol for LULC classification using a grayscale-spatial-based GA model. The first ensemble framework uses fuzzy c-means to classify pixels into those that are difficult to cluster and those that are easy to cluster, which aids in reducing the search space for evolutionary computation. The second ensemble framework uses neighborhood windows as heuristic information to adaptively modify the objective function and mutation probability of the GA, which brings valuable benefits to the discrimination and decision of GA. In this study, three research areas in Dangyang, China, are utilized to validate the effectiveness of the proposed method. The experiments show that the proposed method can effectively maintain the image details, restrain noise, and achieve rapid algorithm convergence. Compared with the reference methods, the best overall accuracy obtained by the proposed algorithm is 88.72%. [ABSTRACT FROM AUTHOR]

With the development of online real estate trading platforms, multi-modal housing trading data, including structural information, location, and interior image data, are being accumulated. The accurate appraisal of real estate makes sense for government officials, urban policymakers, real estate sellers, and personal purchasers. In this study, we propose an interpretable multi-modal stacking-based ensemble learning (IMSEL) method that deals with various modalities for real estate appraisals. We crawl the structural and image data of real estate in Chengdu city, China from the nation's largest real estate transaction platform with the location information, including public services, within 2 km of the real estate using Baidu map. We then compare the predictive results from IMSEL with those from previous state-of-art methods in the literature in terms of the root mean square error, mean absolute percentage error, mean absolute error, and coefficient of determination (R2). The comparison results show that IMSEL outperformed the other methods. We verified the improvement of introducing a data transformation strategy and deep visual features through a 10-fold cross-validation. We also discuss the managerial implications of our research findings. [ABSTRACT FROM AUTHOR]

We propose a novel unified frameork for automated distributed active learning (AutoDAL) to address multiple challenging problems in active learning such as limited labeled data, imbalanced datasets, automatic hyperparameter selection as well as scalability to big data. First, automated graph-based semi-supervised learning is conducted by aggregating the proposed cost functions from different compute nodes and jointly optimizing hyperparameters in both the classification and query selection stages. For dense datasets, clustering-based uncertainty sampling with maximum entropy (CME) loss is applied in the optimization. For sparse and imbalanced datasets, shrinkage optimized KL-divergence regularization and local selection based active learning (SOAR) loss are further naturally adapted in AutoDAL. The optimization is efficiently resolved by iteratively executing a genetic algorithm (GA) refined with a local generating set search (GSS) and solving an integer linear programming (ILP) problem. Moreover, we propose an efficient distributed active learning algorithm which is scalable for big data. The proposed AutoDAL algorithm is applied to multiple benchmark datasets and two real-world datasets including an electrocardiogram (ECG) dataset and a credit fraud detection dataset for classification. We demonstrate that the proposed AutoDAL algorithm is capable of achieving significantly better performance compared to several state-of-the-art AutoML approaches and active learning algorithms. [ABSTRACT FROM AUTHOR]

This article demonstrates the application of a novel multiagent modeling approach to support supply network configuration (SNC) decisions toward addressing several challenges reported in the literature. These challenges include: enhancing supply network (SN)-level performance in alignment with the goals of individual SN entities; addressing the issue of limited information sharing between SN entities; and sustaining competitiveness of SNs in dynamic business environments. To this end, a multistage, multiechelon SN consisting of geographically dispersed SN entities catering to distinct product-market profiles was modeled. In modeling the SNC decision problem, two types of agents, each having distinct attributes and functions, were used. The modeling approach incorporated a reverse-auctioning process to simulate the behavior of SN entities with differing individual goals collectively contributing to enhance SN-level performance, by means of setting reserve values generated through the application of a genetic algorithm. A set of Pareto-optimal SNCs catering to distinct product-market profiles was generated using Nondominated Sorting Genetic Algorithm II. Further evaluation of these SNCs against additional criteria, using a rule-based approach, allowed the selection of the most appropriate SNC to meet a broader set of conditions. The model was tested using a refrigerator SN case study drawn from the literature. The results reveal that a number of SNC decisions can be supported by the proposed model, in particular, identifying and evaluating robust SNs to suit varied product-market profiles, enhancing SC capabilities to withstand disruptions and developing contingencies to recover from disruptions. [ABSTRACT FROM AUTHOR]

The developing optimization algorithms provide promising solutions for speeding up analog integrated circuit sizing. However, the optimization of complicated circuits whose solution regions are narrow remains to be a challenge. With a limited number of sampling points due to the restriction of computational resources, it is difficult for traditional algorithms to achieve satisfactory results for such circuits. To solve this problem, this article proposes a rule-guided genetic algorithm (RG-GA) for analog circuit optimization. Different from the random mutation approach in the traditional genetic algorithm (GA), the RG-GA introduces a design rule-guided mutation (RGM) mechanism which helps to find the solution region in a more straightforward fashion. Instead of handing over circuit optimization tasks to pure mathematical algorithms, the proposed method takes advantages of valuable design knowledge to improve searching efficiency. This novel algorithm is implemented and deployed to design a two-stage rail-to-rail operational amplifier (OPA), an LC voltage controlled oscillator (LC-VCO) and a four-stage OPA. Experimental results show that compared to the traditional GA method, the RG-GA achieves about 1.5 and 3.3 times speed enhancement for the two-stage rail-to-rail OPA and the LC-VCO, respectively. For the four-stage OPA, the RG-GA method can find an acceptable point within the given number of iterations while the traditional GA could not. [ABSTRACT FROM AUTHOR]

The evaluation of the state of transformer oil-immersed insulation based on the traditional methods becomes unreliable due to the nonuniform aging effect. To address this issue, a novel model for accessing the aging information of transformer oil-immersed insulation at the hotspot is proposed. In this article, frequency-domain spectroscopy (FDS) is selected as the carrier for characterizing the aging states of oil-immersed insulation. The multiobjective function for reversing the FDS of the hotspot is constructed based on the dielectric response equivalent circuit. Then, the nondominated sorting genetic algorithm with the multiconstraint scheme is proposed to solve the multiobjective function, by which the FDS data of the transformer hotspot region are acquired. The verification results indicate that the average standard deviation between the computed FDS of the hotspot region and the measured data is less than 0.26. The contribution of this article is in the exploration of the proposed model as a potential tool to extract the FDS data of the hotspot region. This will promote a more reliable aging evaluation of the transformer oil-immersed paper insulation at the hotspot. [ABSTRACT FROM AUTHOR]

In the decoupling control process of the outer rotor coreless bearingless permanent magnet synchronous motor using a least squares support vector machine, the kernel width σ2 and the penalty factor γ of the least squares support vector machine are difficult to tune, resulting in high kernel space complexity and low fitting accuracy. A new improved genetic algorithm is used to optimize the above parameters. First, the generalized inverse system model of the original nonlinear multiple-input multiple-output system is fitted online with the optimized least squares support vector machine, and the generalized inverse system model is connected in series with the original system to form a pseudolinear system. Then, a linear closed-loop controller is used for control. Simulations and experiments show that the robustness and wide applicability of the original system are guaranteed. The dynamic and static performances of the system are improved, verifying the effectiveness of the scheme. [ABSTRACT FROM AUTHOR]

Constant on-time current mode (COTCM) Buck converters are widely used in power management integrated circuit (ICs) for microprocessors. Small-signal models are necessary tools to design high-quality controllers. Existing models neglect output voltage ripples (OVRs) and spectrum extension caused by switching, which makes significant errors under high control bandwidth. This article adopts Fourier series to describe OVRs precisely. Then, influences of OVRs on COTCM are analyzed. Finally, spectrum coupling from current loop and voltage loop is incorporated to obtain a high-frequency model, and four universal analytical expressions in frequency domain are obtained with the aid of matrix calculation. Based on the proposed model and genetic algorithm, an optimal controller design method is developed. Simulation and experiment validate the accuracy of proposed model and the effectiveness of proposed design method. [ABSTRACT FROM AUTHOR]

Nonlinear equation systems (NESs) are ubiquitous, and solving them is an important yet challenging task in numerical computation. Most of the NESs usually contain multiple roots. To locate multiple roots, the use of evolutionary algorithms attracts more attention recently. For the purpose of finding different roots within a limited computational budget in a single run simultaneously, in this article, a hybrid niching-based differential evolution with two archives, referred to as HNDE/2A, is proposed. It can be featured as: 1) two niching techniques, i.e., crowding and speciation, are combined to balance the diversity and convergence; 2) a root archive is used to save the found roots during the run. Additionally, if a root is saved into the root archive, it will be reinitialized immediately to further promote the population diversity; and 3) an inferior offspring archive is presented to utilize the useful information of the inferior offspring. To evaluate the performance of our proposal, 30 problems are chosen as the test suite. Moreover, HNDE/2A is also used to solve two real-world problems. Compared with other algorithms, HNDE/2A yields better results in terms of the root ratio and the success rate with a less computational budget. [ABSTRACT FROM AUTHOR]

The job-shop scheduling problem (JSP) is NP hard, which has very important practical significance. Because of many uncontrollable factors, such as machine delay or human factors, it is difficult to use a single real-number to express the processing and completion time of the jobs. JSP with fuzzy processing time and completion time (FJSP) can model the scheduling more comprehensively, which benefits from the developments of fuzzy sets. Fuzzy relative entropy leads to a method that can evaluate the quality of a feasible solution following the comparison between the actual value and the ideal value (the due date). Therefore, the multiobjective FJSP can be transformed into a single-objective optimization problem and solved by a hybrid adaptive differential evolution (HADE) algorithm. The maximum completion time, the total delay time, and the total energy consumption of jobs will be considered. HADE adopts a mutation strategy based on DE-current-to-best. Its parameters (CR and F) are all made adaptive and normally distributed. The new individuals are selected according to the fitness value (FRE) obtained from a population consisting of N parents and N children in HADE. The algorithm is analyzed from different viewpoints. As the experimental results demonstrate, the performance of the HADE algorithm is better than those of some other state-of-the-art algorithms (namely, ant colony optimization, artificial bee colony, and particle swarm optimization). [ABSTRACT FROM AUTHOR]

Finding the best configuration of algorithms’ hyperparameters for a given optimization problem is an important task in evolutionary computation. We compare in this work the results of four different hyperparameter optimization (HPO) approaches for a family of genetic algorithms (GAs) on 25 diverse pseudo-Boolean optimization (PBO) problems. More precisely, we compare previously obtained results from a grid search with those obtained from three automated configuration techniques: 1) iterated racing; 2) mixed-integer parallel-efficient global optimization (MIP-EGO); and 3) mixed-integer evolutionary strategies. Using two different cost metrics: 1) expected running time (ERT) and 2) the area under the empirical cumulative distribution function (ECDF) curve, we find that in several cases the best configurations with respect to ERT are obtained when using the area under the ECDF curve as the cost metric during the configuration process. Our results suggest that even when interested in ERT performance, it might be preferable to use anytime performance measures for the configuration task. We also observe that tuning for ERT is much more sensitive with respect to the budget that is allocated to the target algorithms. [ABSTRACT FROM AUTHOR]

Neural architecture search (NAS), which automatically designs the architectures of deep neural networks, has achieved breakthrough success over many applications in the past few years. Among different classes of NAS methods, evolutionary computation-based NAS (ENAS) methods have recently gained much attention. Unfortunately, the development of ENAS is hindered by unfair comparison between different ENAS algorithms due to different training conditions and high computational cost caused by expensive performance evaluation. This article develops a platform named BenchENAS, in short for benchmarking evolutionary NAS, to address these issues. BenchENAS makes it easy to achieve fair comparisons between different algorithms by keeping them under the same settings. To accelerate the performance evaluation in a common lab environment, BenchENAS designs a novel and generic efficient evaluation method for the population characteristics of evolutionary computation. This method has greatly improved the efficiency of the evaluation. Furthermore, BenchENAS is easy to install and highly configurable and modular, which brings benefits in good usability and easy extensibility. This article conducts efficient comparison experiments on eight ENAS algorithms with high GPU utilization on this platform. The experiments validate that the fair comparison issue does exist in the current ENAS algorithms, and BenchENAS can alleviate this issue. A Website has been built to promote BenchENAS at https://benchenas.com , where interested researchers can obtain the source code and document of BenchENAS for free. [ABSTRACT FROM AUTHOR]

Compared with traditional observation satellites, agile earth observation satellites are capable of prolonging observation time windows (OTWs) for targets, which significantly alleviates observation conflicts, thereby facilitating imaging data collection. However, it also leads to more uncertainties in determining the start time to image targets within these longer OTWs for an agile satellite network (ASN) to collect imaging data. Furthermore, these collected data are offloaded only within short transmission time windows between data collectors and data sinks, thus resulting in a transmission scheduling problem. Toward this end, this paper investigates joint observation and transmission scheduling in ASNs, aiming at accommodating more imaging data to be collected and offloaded successfully. Specifically, we formulate the studied problem as integer linear programming (ILP) to maximize the weighted sum of scheduled imaging tasks. Then, we explore the hidden structure of this ILP and transform it into a special framework, which can be solved efficiently through semidefinite relaxation (SDR). To reduce computation complexity, we further propose a fast yet efficient algorithm by combining the advantages of the devised SDR method and a genetic algorithm with special population initialization. Finally, simulation results demonstrate that the proposed algorithm can significantly increase the weighted sum of scheduled tasks. [ABSTRACT FROM AUTHOR]

In real-time systems, analyzing the worst-case execution time (WCET) of a task in the presence of data caches is hard. The ACDC is a data cache that provides predictability, facilitating WCET analysis. It works by granting data cache replacement permission to specific load/store instructions. Nonetheless, knowing how to select these instructions to minimize the WCET, i.e., configuring the ACDC, is not trivial. In this paper, we propose four new methods to configure the ACDC, and compare them with existing methods. Unlike those in previous studies, our proposed methods provide specific ACDC configurations for the different phases of a given task, instead of a single ACDC configuration per task. We evaluate the WCET bounds obtained when using different ACDC configuration methods on the TACLeBench benchmark suite. Our results show that the most complex benchmarks work better with multiple-content configurations, which indicates that realistic tasks may also benefit from this kind of configuration. The methods proposed in this study improve the WCET in more than 60% of cases, with an average WCET improvement of nearly 5% and up to 50% in some cases.

Multiple parameters with large nonlinear characteristics must be considered simultaneously to design the coil dimensions of static inductive power transfer (SIPT) systems. The design of dynamic inductive power transfer (DIPT) systems is more challenging due to the large number of parameters needed to be considered. In the conventional artificial neural network (ANN)-based design approach, optimal coil dimensions are found using ANN that has learned the nonlinear characteristics between coil dimensions and magnetic characteristics using the finite element method (FEM). However, this approach requires a large amount of training data, and it is difficult to reach an optimum design if there are many design criteria. In order to overcome these challenges, this paper proposes a design optimization method using two approaches: improving the time efficiency of ANN training data collection by superposing the magnetic fields from the coils and improving the input value of ANN using a genetic algorithm. Design results predicted by the ANN are compared with FEM simulation, circuit simulations, and experimental results to verify the validity of the proposed algorithm. The FEM and circuit simulation results and the ANN prediction results match with errors of 10.2% or less for all design requirements. Experimental results are provided for a 3 kW DIPT system with four transmitter coils and an automated test rail. Comparison results between ANN predicted values and experimental values match with an error of less than 12.7%.

Robotic systems are essential to technological development in the industrial, medical, and aerospace sectors. Nevertheless, their use in different applications requires that the robot have the best possible execution efficiency. For this, a robot with specific optimized characteristics is necessary to impact the performance of the specific task. Different techniques have been used in robot optimization, the most widely used being genetic algorithms (GA) and particle swarm optimization (PSO). However, there are optimization algorithms with high convergence speeds inspired by animal behavior whose application in robot optimization has not been reported. In this work, bio-inspired algorithms Harris hawks optimization (HHO) and Grey wolf optimizer (GWO) are applied to a six-degree-of-freedom (6DOF) robot arm design through kinematic optimization. The lengths of the main robot links are optimized to improve the workspace volume and obtain a better-conditioned robot using the structural length index (SLI) and global condition index (GCI) as objective functions. A comparison is made between the proposed algorithms and GA and PSO regarding convergence speed, computational load, and optimality. Similar behavior has been found for HHO, GWO, and PSO compared to GA for both indexes. For the GCI problem, an average improvement of 14% was found when optimizing an industrial robot arm. Furthermore, multiple runs experiment is performed to test the robustness of the algorithms. The results show that HHO is the best technique to obtain an optimal robot design because it needs less than ten iterations to provide a better result despite its computational load.

Microgrids are promising to enhance power distribution systems’ efficiency, quality, sustainability, and reliability. However, microgrids operation can impose several challenges to traditional protection schemes, like changes in the power flow direction and an increase in short-circuit currents. Microgrids can include several distributed generation technologies with different behaviours during short-circuit conditions, requiring additional protection schemes and devices. In this way, the optimized coordination of reclosers and fuses in distribution networks with directional overcurrent relays, which operate as interconnection devices, might overcome many imposed protection challenges. Regarding different generation technologies, voltage-restrained overcurrent relays and frequency relays are presented as local microgrid protection for rotative and inverter-based distributed generators, respectively. The optimized coordination of these protection devices maximizes microgrid benefits and minimizes operation drawbacks by reducing interruptions impacts and energy not supplied to consumers. This work proposes, thus, a mathematical model for the optimal coordination of protection devices in distribution networks with distributed energy resources operating in grid-connected and islanded modes. The minimization technique of operating times using an elitist genetic algorithm with variable crossover and mutation processes is proposed, as well. The results show adequate coordination using passive and low-cost protection devices.

The use of unmanned aerial vehicles (UAVs) in the collection of data from wireless devices, sensor nodes, and the Internet of Things (IoT) devices has recently received significant attention. In this paper, we investigate the data collection process from a set of smart meters in advanced metering infrastructure (AMI) enabled by UAVs. The objective is to minimize the total annual cost of the electric utility by jointly optimizing the number of UAVs, their power source sizing, the charging locations as well as the data collection trip planning. This is achieved while considering the energy budgets of batteries of UAVs and the required amount of collected data. The problem is formulated as a mixed-integer nonlinear programming (MIN-LP), which is decoupled into two sub-problems where a candidate UAV and a number of buildings are first grouped into trips via genetic algorithms (GAs), and then the optimum trip path is found using a traveling salesman problem (TSP) branch and bound algorithm. Simulation results show that the battery capacity or the number of UAVs increases as the coverage area or the density increases.