Search

Electric vehicles are considered as an emerging solution to mitigate the environmental footprint of transportation sector. Therefore, researchers and automotive developers devote significant efforts to enhance the performance of electric vehicles to promote broader adoption of such technology. One of the critical challenges of the electric vehicle is limited battery lifetime and entailed range anxiety. In his context, development of counter-aging control strategies based on precise battery modeling is regarded as an emerging approach that has a significant potential to address battery degradation challenges. This paper presents a combined trade-off strategy to minimize battery degradation while maintaining acceptable driving performance and charge retention in electric vehicles. A battery aging model has been developed and integrated into a full vehicle model. An optimal control problem has been formulated to tackle the afore-mentioned challenges. Non-dominant sorting genetic algorithms have been implemented to yield the optimal solution through the Pareto-front of three contending objectives, based upon which an online simulation has been conducted considering three standard driving cycles. The results reveal the ability of the proposed strategy to prolong the life cycle of the battery and extend the driving range by 25 % and 8 % respectively with minimal influence of 0.6 % on the driveability. [ABSTRACT FROM AUTHOR]

Nowadays, Convolutional neural networks (CNNs) have achieved relevant results on several data sciences-related tasks, such as image processing and pattern recognition. However, CNNs contain an immense number of parameters which often leads to a huge redundancy, overfitting, and a significant amount of memory. In this paper, we aim to present a multi-objective optimization model for kernels redundancy reduction in convolutional neural networks. In fact, the suggested approach, named MOFM-CNN, allows to minimize redundant feature maps using a set of decision control variables. MOFM-CNN is composed of two objectives where in the first one, the decision variables are technically introduced in the cross-entropy function in order to evaluate the impact of each feature map on the CNNs training. In the second one, the control parameters are used to calculate the proportion of active feature maps, that is related to the complexity of the model. The resultant problem is manipulated and solved using non dominated sorting genetic algorithm (NSGA-II). The performance of our proposal is demonstrated visually and numerically for both classification and features maps optimization. [ABSTRACT FROM AUTHOR]

In this study, we redefine FCM algorithm by integrating fuzzy set theory, fuzzy metrics, and Sugeno negation principles. This innovative approach overcomes the limitations inherent in conventional machine learning models, especially in situations characterized by uncertainty, noise, and ambiguity. Our model utilizes the membership degrees from fuzzy set theory, and transforms the concept of proximity defined by fuzzy metrics into a minimization problem. This transformation is achieved using a linguistic negation operator, which is crucial for optimizing FCM algorithm's objective function. A significant innovation in our research is the use of GA for optimizing parameters within the contexts of fuzzy metrics and Sugeno negation. The precise optimization capabilities of GA greatly enhance the sensitivity and adaptability of FCM algorithm, thereby improving overall performance. By leveraging the meticulous parameter adjustments provided by GA, our approach has shown superior results in practical applications, such as brain MRI image segmentation, surpassing traditional methods. Experimental results highlight the considerable enhancements our proposed FCM algorithms bring over existing methods across various performance metrics. In conclusion, this study makes a valuable addition to the field of fuzzy-based machine learning methodologies. It combines the optimization strength of GA with the flexible classification capabilities of fuzzy logic. The integration of Sugeno negation and fuzzy metrics not only improves the accuracy and precision of FCM algorithm but also provides significant benefits in handling complex and ambiguous datasets. This research signifies a major advance in machine learning and fuzzy logic, setting the stage for future applications and studies. [ABSTRACT FROM AUTHOR]

Hyperparameters play a critical role in analyzing predictive performance in machine learning models. They serve to strike a balance between overfitting and underfitting of research-independent features to prevent extremes. Manual tuning and automated techniques are employed to identify the optimal combination and permutation to achieve the best model performance. This study explores the pursuit of the best fit through various hyperparameters. Following Logistic Regression analysis, this research compared Random Forest, Randomized search, Grid search, Genetic, Bayesian, and Optuna machine learning model tuning for the best accuracy of prediction the student The model accuracy was further assessed using confusion matrices and Receiver Operating Characteristic—Area Under the Curve (ROC-AUC) curves for student grade classification. The genetic algorithm's recommended hyperparameter tuning yielded the highest accuracy (82.5%) and AUC-ROC score (90%) for student result classification. Manual tuning with an estimator of 300, criterion entropy, max features of sqrt, and a minimum sample leaf of 10 achieved an accuracy of 81.1%, which closely resembled the performance randomized search cross-validation algorithm. The default random forest model scored the least accuracy (78%). However, this manual tuning process took a lesser time (3.66 s) to fit the model while grid search CV tuned 941.5 s. Hence, this research made significant contributions to optimizing various machine learning models using a range of hyperparameters for grade classification. [ABSTRACT FROM AUTHOR]

Since digital media has become increasingly popular, video processing has expanded in recent years. Video processing systems require high levels of processing, which is one of the challenges in this field. Various approaches, such as hardware upgrades, algorithmic optimizations, and removing unnecessary information, have been suggested to solve this problem. This study proposes a video saliency map based method that identifies the critical parts of the video and improves the system's overall performance. Using an image registration algorithm, the proposed method first removes the camera's motion. Subsequently, each video frame's color, edge, and gradient information are used to obtain a spatial saliency map. Combining spatial saliency with motion information derived from optical flow and color-based segmentation can produce a saliency map containing both motion and spatial data. A nonlinear function is suggested to properly combine the temporal and spatial saliency maps, which was optimized using a multi-objective genetic algorithm. The proposed saliency map method was added as a preprocessing step in several Human Action Recognition (HAR) systems based on deep learning, and its performance was evaluated. Furthermore, the proposed method was compared with similar methods based on saliency maps, and the superiority of the proposed method was confirmed. The results show that the proposed method can improve HAR efficiency by up to 6.5% relative to HAR methods with no preprocessing step and 3.9% compared to the HAR method containing a temporal saliency map. [ABSTRACT FROM AUTHOR]

Internet of Things (IoT) devices generate and transfer a vast volume of data in interaction with various objects. These data rely on cloud computing (CC) or fog computing (FC) for processing and calculations. Unlike CC, FC delivers task processing closer to IoT devices. Consequently, FC leads to reduced latency, increased efficiency, ease of deployment, and flexibility. The primary concern between the service requester and provider when agreeing to task offloading is trust. While several types of research are based on intelligent models for trust management, trust management in FC requires direct and indirect trust computation due to the presence of malicious nodes. To ensure security and maintain the quality of service when interacting with fog servers, trust is essential. In this paper, a new trust model for FC is proposed using a combination of the Genetic Algorithm (GA) and Learning Automaton (LA). The crossover operator in GA by LA is improved. The changes in the search space and finding optimal solutions are dependent on the crossover rate. Thus, determining an optimal value for the crossover operator impacts the problem's solutions. The proposed methods evaluation was done in a MATLAB 2020 simulation environment. The results show that the performance of the proposed method with 200 iterations and 1000 tasks is superior in terms of reliability rate, energy consumption, and delay compared to GA and particle swarm optimization (PSO) algorithm. [ABSTRACT FROM AUTHOR]

The issue of categorizing imbalanced data is becoming increasingly prevalent. While existing methodologies have demonstrated notable advancements in handling imbalanced data, the challenges of extensive data size and low-quality data in the dataset persist. We propose an innovative hybrid approach that combines game mixed sampling and improved genetic algorithms to address the issues of excessive dataset size and low data quality in imbalanced data classification problems. Specifically, in the game hybrid sampling module, we identify the optimal hybrid sampling method and sampling ratio for the current dataset through the game idea, intending to obtain a diverse dataset to ensure comprehensive coverage of the dataset. Additionally, in the module on improving genetic algorithms, we integrate the classifier group, encode the performance metrics of the sampled data, and perform a comprehensive evaluation of the fitness of each data point. Preserve population data of many excellent individuals through selection operations. The population data is subjected to crossover and mutation operations to explore the search space, and the minimum stable population size is determined by sliding standard deviation. In the real world, where credit card fraud data are highly imbalanced, our combined approach achieves small dataset sizes and high evaluation indices, outperforms existing methods, and demonstrates the effectiveness of game mixed sampling and improved genetic algorithms. [ABSTRACT FROM AUTHOR]

Earthquake magnitude prediction is an exceptionally challenging task that has been explored through various machine learning approaches. However, the development of prediction models is hindered by redundant features and time series properties. The Elite Genetic Algorithm (EGA) excels in searching for optimal feature subsets; while, Long Short-Term Memory (LSTM) is specialized in processing time series and complex data. Hence, we introduce an EGA feature selection and LSTM model (EGA-LSTM) for time series earthquake prediction. Initially, acoustic and electromagnetic data from our developed AETA system are fused and preprocessed using a roulette-based EGA to search strong correlation indicators. Subsequently, LSTM is employed to perform magnitude prediction with the selected features. Specially, the Root Mean Square Error (RMSE) of LSTM and the ratio of selected features are chosen as fitness components of EGA. Finally, we evaluate the proposed EGA-LSTM on AETA data from four regions from China, considering the influence of data in different time periods (timePeriod) and fitness function weights ( ω a and ω F ) on prediction results. Furthermore, our proposed approach outperforms state-of-the-art methods in evaluation metrics such as EV, MAE, MSE, RMSE, and R2. Nonparametric tests reveal that EGA-LSTM is significantly different from others and outperforms the standard LSTM. [ABSTRACT FROM AUTHOR]

To provide better care for the elderly, health trustees establish home health care centers in cities. These centers offer services at homes, and a logistics network is essential for providing these services. An integrated framework is developed to design such a network by which various logistical decisions are made under inherent epistemic uncertainty of the travel and service times. A new mixed integer linear programming model is developed, and its equivalent deterministic formulation is obtained using a well-applied possibilistic programming approach. An improved genetic algorithm is also devised to solve large-sized instances and various sensitivity analyses are performed on key factors. Finally, a case study is presented to validate the developed model and its solution method. The numerical results demonstrate 20% reduction in the total time of employing care givers at almost 3.5% cost savings. [ABSTRACT FROM AUTHOR]

The miniaturization of antennas is crucial as it improves the integration of wireless communication system. In order to achieve miniaturization of antennas, a performance-constrained multi-objective optimization method (PCMOM) considering the size and return loss is proposed. In the PCMOM method, an optimization strategy based on a multi-port network model is introduced, enabling the formation of antennas with various structures. Furthermore, we integrate the constraint of return loss performance into the non-dominated sorting genetic algorithm II (NSGA-II), eliminating solutions that do not meet performance requirements. Three pixel antennas are designed using the PCMOM method and two of them are fabricated. Experimental results demonstrate that the proposed PCMOM method can effectively address the complex trade-off issues in antenna miniaturization design. [ABSTRACT FROM AUTHOR]

At present, the availability of nonrenewable sources and their usage for electric vehicle technology is reducing gradually because of their disadvantages are more environmental pollution, direct effect on human health, less reliability, and taking more time to start functioning. So, in this article, the proton exchange membrane fuel cell (PEMFC) is considered for the automotive application because of its advantages quick startup, more power density, more safety to handle, high efficiency, and capability of operating at very low operational temperature conditions. However, the drawback of PEMFC is very difficult to identify the accurate MPP position of the fuel system. Here, the improved variable step genetic algorithm is added with the adaptive neuro-fuzzy inference system for tracking the operational point of the proposed system with high efficiency. These hybrid MPPT controller features are easy to understand, more accurate, have a better dynamic response, and have low design complexity. The evaluated proposed MPPT controller operational efficiency, and settling time of the converter voltage at different fuel stack temperature conditions are 98.7402%, and 0.01607 s respectively. Finally, the boost converter is used in this work to enhance the voltage supply capability of the entire system. The proposed system is investigated by applying the MATLAB tool. [ABSTRACT FROM AUTHOR]

Artificial selection is a promising way to improve microbial community functions, but previous experiments have only shown moderate success. Here, we experimentally evaluate a new method that was inspired by genetic algorithms to artificially select small bacterial communities of known species composition based on their degradation of an industrial pollutant. Starting from 29 randomly generated four-species communities, we repeatedly grew communities for four days, selected the 10 best-degrading communities, and rearranged them into 29 new communities composed of four species of equal ratios whose species compositions resembled those of the most successful communities from the previous round. The best community after 18 such rounds of selection degraded the pollutant better than the best community in the first round. It featured member species that degrade well, species that degrade badly alone but improve community degradation, and free-rider species that did not contribute to community degradation. Most species in the evolved communities did not differ significantly from their ancestors in their phenotype, suggesting that genetic evolution plays a small role at this time scale. These experiments show that artificial selection on microbial communities can work in principle, and inform on how to improve future experiments. Artificial selection is a promising way to improve microbial community functions. Here, Arias-Sánchez et al. evaluate a method inspired by genetic algorithms to select small bacterial communities of known species composition based on their degradation of an industrial pollutant. [ABSTRACT FROM AUTHOR]

Atomistic simulations are crucial for predicting material properties and understanding phase stability, essential for materials selection and development. However, the high computational cost of density functional theory calculations challenges the design of materials with complex structures and composition. This study introduces new data acquisition strategies using Bayesian-Gaussian optimization that efficiently integrate the geometry of the convex hull to optimize the yield of batch experiments. We developed uncertainty-based acquisition functions to prioritize the computation tasks of configurations of multi-component alloys, enhancing our ability to identify the ground-state line. Our methods were validated across diverse materials systems including Co-Ni alloys, Zr-O compounds, Ni-Al-Cr ternary alloys, and a planar defect system in intermetallic (Ni1−x, Cox)3Al. Compared to traditional genetic algorithms, our strategies reduce training parameters and user interaction, cutting the number of experiments needed to accurately determine the ground-state line by over 30%. These approaches can be expanded to multi-component systems and integrated with cost functions to further optimize experimental designs. [ABSTRACT FROM AUTHOR]

To classify and detect the type and content of petroleum hydrocarbon contaminants in the soil surface layer, fluorescence spectrometry is commonly used. The experimental oils were selected from three common engine oils available in the market: Loxson L-CKC220 gear oil, APSIN 10W-40 engine oil and Jaguar 200 SF MA 15W-40 motorcycle oil. The fluorescence spectra of the oils were obtained using the fluorescence-induced technique, the spectral wavelengths were selected using a genetic algorithm (GA), and the detection models were constructed by combining RF (Random Forest), AdaBoost, and Gradient Enhanced Decision Tree (GBDT) regression algorithms for classification, identification, and concentration prediction analysis. The experimental results show that the average accuracy of classification and identification of gear oil, engine oil and motorcycle oil reach 83.9, 97.8, and 92.2%, respectively. Comparative analysis of the prediction results of the three concentration regression models shows that while all algorithms have high model prediction accuracy, GA combined with GBDT regression model has the best prediction performance for gear oils, engine oils and motorcycle oils, and improves the prediction accuracies by 62.7, 42.3, and 48.3% compared to the prediction accuracies of the wavelength selection without the use of GA, respectively. In summary, GA-based spectral wavelength selection combined with machine learning algorithms has high prediction accuracy and precision for the classification and prediction of motor oil contaminants in selected specific soils, and can be used as an effective detection method. [ABSTRACT FROM AUTHOR]

The background differences in water content of different samples have a very strong influence on the robustness of near-infrared spectroscopy (NIRS). For this reason, this study simulated typical biological water matrix samples with formulated hemoglobin (Hb), glucose (Glc), and distilled water, and attempted to use four different intelligent spectral variable selection algorithms [Competitive Adaptive Reweighted Sampling (CARS), Randomized Frog Hopping Algorithm (RF), Genetic Algorithm (GA), and Variable Projection Importance Algorithm (VIP)] to perform the Hb water interference-resistant feature band preferences, while combining partial least squares (PLS) in parallel to build a robust quantitative model of Hb. In addition, the applicability and validity of the model were validated using three prediction sets P1, P2, P3 with different water backgrounds (the formulation method and composition were kept the same, and only the water content increased sequentially). The results showed that RF, GA, and VIP could effectively screen out the characteristic wavelengths of Hb with low sensitivity to water changes and successfully correct the water effect, but due to the large number of characteristic variables they screened out and the existence of a large number of redundant and water interference variables, this ultimately made the model's robustness less than ideal. The CARS algorithm performed the best, and the RMSEP of the three prediction sets were 0.016, 0.017, and 0.038, which is closer to the RMSECV of the calibration set. Therefore, NIRS combined with the variable selection can reduce the effect of water on model robustness and improve the prediction accuracy of the model by the method of selecting effective wave number intervals, and CARS may be one of the ideal algorithms to solve such problems. [ABSTRACT FROM AUTHOR]

A multi-fidelity optimisation strategy has been developed in the present work, and its performance is illustrated through a series of test cases. The strategy is based on hybrid methods such that two genetic optimisation algorithms are employed, each coupled to a different fidelity level with transfer of information between them. The aim is that the low fidelity model, being less accurate but with a lower computational cost, performs a comprehensive search along the design space guiding the high fidelity model to the optimum region. This strategy has been shown to reduce the computational time of an optimisation through analytical test cases as well as numerical cases. The analytical cases have been used to tune the parameters that define the multi-fidelity strategy, while the numerical cases are employed to apply the method to engineering problems, focusing on the aerodynamic performance of an airfoil. The speed-up shows a certain dependency to the models relation, both regarding their similarity level as well as the relative computational cost. For cases exhibiting a significant dissimilarity between models, wherein the low fidelity model is notably inaccurate, the attained speed-up diminishes, and numerous instances demonstrate an absence of speed-up. However, for most cases, even with poor model similarity the optimisations are accelerated by an order of 2, while values up to 3–5 were found for higher similarity levels. Hence, the developed strategy shows a relevant decrease of computational cost of an optimisation procedure although its performance is affected by the models relative accuracy. [ABSTRACT FROM AUTHOR]

Cuckoo search (CS) algorithm is a classical swarm intelligence algorithm widely used in a variety of engineering optimization problems. However, its search accuracy and convergence speed still have a lot of room for improvement. In this paper, an improved version of the CS algorithm based on intelligent perception strategy, adaptive invasive weed optimization (AIWO), and elite cross strategy, called IIC-CS is proposed. Firstly, the intelligent perception strategy can update the value according to the searching state. Moreover, the CS is hybridized with the AIWO to improve the searching performance of the algorithm. Additionally, the elite cross strategy is employed to enhance the exploration capability and exploitation capability of the algorithm. Combining the improvements of these three methods, the performance of the CS algorithm is significantly improved. Meanwhile, 23 classical benchmark functions, some CEC2014 and CEC2018 benchmark functions are used to test the search accuracy and convergence rate of the IIC-CS. Furthermore, some classical or state-of-the-art algorithms such as the genetic algorithm (GA), particle swarm optimization (PSO), bat algorithm (BA), ant lion optimizer (ALO) and cuckoo search (CS) algorithm, invasive weed optimization (IWO), integrated cuckoo search optimizer (ICSO) and improved island cuckoo search (iCSPM2) are used to make comparisons. Through the statistical results of the experiments, we find that the IIC-CS algorithm can achieve better results on most benchmark functions compared to other algorithms, thus demonstrating the effectiveness of the improvements and the superiority of the IIC-CS algorithm. [ABSTRACT FROM AUTHOR]

Mobile edge computing (MEC) is considered one of the key technologies for large-scale network services. Task scheduling helps to improve the task completion rate of MEC, by properly mapping tasks generated by devices onto MEC resources. However, the mobility of devices introduces complexities, potentially resulting in failed task offloading or unavailable task results. To tackle this issue, we propose a mobile-aware task scheduling scheme. We first model the trajectory of mobile devices and introduce a strategy for the fastest task offloading, coupled with an efficient result return method. Subsequently, to improve the task completion rate, we present a task scheduling model based on task migration and formulate the relevant problem as a Mixed Integer Non-linear Programming (MINLP) problem. To achieve a solution within a reasonable time complexity, we propose a Particle Swarm Optimization and Genetic Algorithm with a Rescheduling operator (PSOGAR). In PSOGAR, particles update their positions using a mating operator, while maintaining diversity by a mutation operator. In addition, a rescheduling operator is used to further improve the task completion rate. Finally, through simulation experiments, compare PSOGAR with state-of-the-art and classic algorithms. The experimental results show that PSOGAR can improve the task completion rate by 18–31% and can be applied to scenarios with tight task deadlines. [ABSTRACT FROM AUTHOR]

Fingerprint recognition systems hold a pivotal role across various modern applications, such as criminal investigation, civil identification, and access control. Fingerprint segmentation is commonly the first stage of fingerprint recognition systems, focusing on extracting the foreground from captured fingerprint images. This helps to reduce extracting false minutiae, speed up the extraction process, and hence, improve the overall system performance. In this paper, we introduce a novel fingerprint segmentation method by first proposing new frequency and intensity features with employing fuzzy-c-means and genetic algorithm, accompanied by proposing a new cluster assignment strategy that involves features weighting and cluster assignment probabilities thresholding which helps to improve the accuracy compared to the standard cluster assignment method. Finally, we introduce a new morphological post-processing technique in order to minimize the misclassified pixels and obtain the final mask. Furthermore, we carry out a comprehensive performance evaluation against leading fingerprint segmentation methods on various known databases. As a result, the proposed method achieves an average error rate of 2.52%, which is much lower than the corresponding errors obtained in the literature. By these experimental results, we show that the proposed method outperforms the best existing methods. [ABSTRACT FROM AUTHOR]

Machine learning-based IDSs have demonstrated promising outcomes in identifying and mitigating security threats within IoT networks. However, the efficacy of such systems is contingent on various hyperparameters, necessitating optimization to elevate their performance. This paper introduces a comprehensive empirical and quantitative exploration aimed at enhancing intrusion detection systems (IDSs). The study capitalizes on a genetic algorithm-based hyperparameter tuning mechanism and a pioneering hybrid feature selection approach to systematically investigate incremental performance improvements in IDS. Specifically, our work proposes a machine learning-based IDS approach tailored for detecting attacks in IoT environments. To achieve this, we introduce a hybrid feature selection method designed to identify the most salient features for the task. Additionally, we employed the genetic algorithm (GA) to fine-tune hyperparameters of multiple machine learning models, ensuring their accuracy in detecting attacks. We commence by evaluating the default hyperparameters of these models on the CICIDS2017 dataset, followed by rigorous testing of the same algorithms post-optimization through GA. Through a series of experiments, we scrutinize the impact of combining feature selection methods with hyperparameter tuning approaches. The outcomes unequivocally demonstrate the potential of hyperparameter optimization in enhancing the accuracy and efficiency of machine learning-based IDS systems for IoT networks. The empirical nature of our research method provides a meticulous analysis of the efficacy of the proposed techniques through systematic experimentation and quantitative evaluation. Consolidated in a unified manner, the results underscore the step-by-step enhancement of IDS performance, especially in terms of detection time, substantiating the efficacy of our approach in real-world scenarios. [ABSTRACT FROM AUTHOR]

Accurate diagnosis and effective treatment of thyroid conditions, such as hypothyroidism and hyperthyroidism, are crucial due to their wide-ranging symptoms and consequences. However, conventional back-propagation neural networks have limitations, including slow convergence and susceptibility to local minima. To deal with these problems, the paper suggests an integrated strategy that includes back-propagation (BP) and genetic algorithms (GA). The suggested method uses evolutionary algorithms to investigate various weight combinations and back-propagation to modify weights in response to the discrepancy between expected and actual results. The study evaluated the method using three steps of experimentation, including network building, local search and optimization, and evaluation, on a real-world dataset of thyroid illnesses. The obtained results are very satisfying and promising, indicating that the MLP-GA/BP model is robust enough to detect and categorize thyroid disorders effectively and efficiently. This makes it a reliable diagnostic tool for medical practitioners, allowing them to effectively diagnose and treat patients with thyroid diseases. [ABSTRACT FROM AUTHOR]

The allocation of resources and scheduling of tasks, specifically mapping, in multicore systems on-chip (MCSoC), poses significant challenges. Tasks have diverse resource requirements and interact with each other, while network-on-chip (NoC)-based MCSoC consists of heterogeneous cores and communication networks. The heterogeneity of resources in MCSoC, along with the varying computational and communication demands of applications, makes mapping a complex optimization problem. We have mathematically modeled the mapping problem in NoC-based MCSoC using mixed-integer linear programming (MILP) with the objective of minimizing the execution time of applications. This model incorporates computation and communication capacity, energy budget constraints of MCSoC, and execution time requirements of applications. We further propose heuristics, including simulated annealing (SA) and genetic algorithm (GA), considering the capacity and budget constraints of MCSoC systems to accelerate applications and provide quick mapping solutions. Simulation results demonstrate that the performance from SA and GA heuristics is very close (within 10%) to the optimal solutions from MILP across various applications for 2D-Mesh NoCs with 16–100 cores. The energy consumption of SA and GA heuristics is also very close to the optimal solutions from MILP, with a few exceptions on small-scale 16-core NoC. Additionally, SA outperforms GA in most cases for all applications. [ABSTRACT FROM AUTHOR]

The accuracy of pollution source identification significantly depends on the amount of effective information derived from monitoring data. Currently, most of the comprehensive studies on groundwater contamination source identification and optimal design of monitoring solutions are based on hypothetical cases, whereas relevant studies on actual cases only consider one characteristic of the pollution source (either locations or fluxes). An optimal monitoring network (OMN)-based pollution source characterisation framework that takes source locations and source fluxes into account is presented to enhance the accuracy of pollution source identification. The genetic algorithm (GA) and particle swarm optimization (PSO) were used to solve the optimization model of pollution source characteristics identification. The framework is applied to a landfill for waste located in Baicheng City, China. The results showed that the identification results based on OMN has a smaller mean relative error and higher accuracy, compared with those based on random monitoring network (RMN). This study shows that OMNs can provide more effective information for pollution source identification and effectively enhance the accuracy of the groundwater sources characteristics identification. [ABSTRACT FROM AUTHOR]

Machine configuration is a crucial strategic decision in designing a flow shop system (FSS) and directly affects its performance. This involves selecting device suppliers and determining the number of machines to be configured. This study addresses a bi-objective optimization problem for an FSS that considers repair actions and aims to determine the most suitable machine configuration that balances the production reliability and purchase cost. A nondominated sorting genetic algorithm II (NSGA-II) is used to determine all the Pareto solutions. The technique for order preference by similarity to an ideal solution is then used to identify a compromise alternative. It is necessary to assess the production reliability of any machine configuration identified by the NSGA-II. The FSS under the machine configuration is modeled as a multistate flow shop network, and Absorbing Markov Chain and Recursive Sum of Disjoint Products are integrated into the NSGA-II for reliability evaluation. The experimental results of solar cell manufacturing demonstrate the applicability of the proposed hybrid method and validate the efficiency of the NSGA-II compared with an improved strength Pareto evolutionary algorithm. [ABSTRACT FROM AUTHOR]

Wireless power transfer (WPT) technology has been extensively developed for unmanned aerial vehicle (UAV) application in recent years. However, misalignment tolerance has always been an unsolved problem for UAV wireless charging. In this paper, a coupling system attached to UAV landing gear is demonstrated. In the design process, a flexible curved coil has been designed for structure utilization. First, the magnetic induction strength of the arch bending coil is calculated. Then, an unequal turn spaced coil structure is proposed based on the genetic algorithm for parameter optimization. In addition, a 3D EM simulation is implemented for comparison before and after optimization. A resonance frequency of 75 kHz with an SS tuning topology is adopted in accordance with the Qi standard. Finally, the experimental setup is made and the results show that the UAV charging system has better misalignment tolerance. [ABSTRACT FROM AUTHOR]

Permanent magnet synchronous linear motor (PMSLM) parameters are susceptible to temperature changes and motor aging, which can result in inaccurate models and reduced control accuracy. Therefore, PMSLM parameters must be monitored accurately. In this study, a back-propagation (BP) neural network is used to identify the resistance, inductance, and flux linkage parameters of a PMSLM, and a genetic algorithm (GA) is employed to optimize the identification system. A full-rank model is established, and the BP neural network parameter identifier is used to realize the multiparameter discrimination, and the GA is employed to optimize the network parameters of the identifier model to improve its training speed and accuracy. A GABP-based multiparameter identification system for a PMSLM is built, and the experiment results show that the GABP parameter identifier can accurately identify the four parameter values of the PMSLM simultaneously. [ABSTRACT FROM AUTHOR]

A sandwich plate with a corrugation and auxetic honeycomb hybrid core is constructed, and its sound insulation and optimization are investigated. First, the motion governing equation of the sandwich plate is established by the third-order shear deformation theory (TSDT), and then combined with the fluid-structure coupling conditions, and the sound insulation is solved. The theoretical results are validated by COMSOL simulation results, and the effects of the structural parameter on the sound insulation are analyzed. Finally, the standard genetic algorithm is adopted to optimize the sound insulation of the sandwich plate. [ABSTRACT FROM AUTHOR]

In this study, we introduce the pied kingfisher optimizer (PKO), a novel swarm-based meta-heuristic algorithm that draws inspiration from the distinctive hunting behavior and symbiotic relationships observed in pied kingfishers in the natural world. The PKO algorithm is structured around three distinct phases: perching/hovering for prey (exploration/diversification), diving for prey (exploitation/intensification), and fostering symbiotic relations. These behavioral aspects are translated into mathematical models capable of effectively addressing a wide array of optimization challenges across diverse search spaces. The algorithm's performance is rigorously evaluated across thirty-nine test functions, which encompass various unimodal, multimodal, composite, and hybrid ones. Additionally, eight real-world engineering optimization problems, including both constrained and unconstrained scenarios, are considered in the assessment. To gauge PKO's efficacy, it is subjected to a comparative analysis against 3 categories of rival optimizers. The 1st category comprises well-established and widely-cited optimizers such as particle swarm optimization and genetic algorithm. The 2nd category encompasses recently published algorithms, including Harris Hawks optimization, Whale optimization algorithm, sine cosine algorithm, Grey Wolf optimizer, gravitational search algorithm, and moth-flame optimization. The 3rd category includes advanced algorithms, such as covariance matrix adaptation evolution strategy and Ensemble Sinusoidal Differential Covariance Matrix Adaptation with Euclidean Neighborhood (LSHADE-cnEpSin). The comparative analysis employs various performance metrics, including the Friedman mean rank and the Wilcoxon rank-sum test, to reveal PKO's effectiveness and efficiency. The overall results highlight PKO's exceptional ability to tackle intricate optimization problems characterized by challenging search spaces. PKO demonstrates superior exploration and exploitation tendencies while effectively avoiding local optima. The source code for the PKO algorithm is publicly accessible at https://www.mathworks.com/matlabcentral/fileexchange/160043-pied-kingfisher-optimizer-pko. [ABSTRACT FROM AUTHOR]

As the world's population is expected to increase, so is the global demand for food. Sustainable intensification via precision agriculture of existing farms can increase crop production. Agricultural spray drones have recently taken a physical role within precision agriculture, such as aerial application of fluids, solids, and biological control agents but have difficulties spraying in uncontrolled environments caused by wind shifting spray material away from intended target areas. This work proposes an efficient physics-based framework to provide drone operators with trajectory and spray nozzle configuration for optimal target crop-dusting to mitigate spray drifts while providing quantitative approximations of spray particle trajectory and ground concentration. The framework is coupled with a machine-learning algorithm (MLA) to aid users in their search for optimal results and includes two decoupled models that simulate wind and spray particle trajectories. In the model problem, a genetic algorithm (GA) is used to optimize the system where the optimal trajectory and spray nozzle configuration resulted in 64% of crop targets hit while only losing minimal spray material from spray drifts. [ABSTRACT FROM AUTHOR]

The signal strength in 5G mobile communication systems is significantly influenced by the surroundings, with key factors including the path difference, operating frequency, and obstructions at specific locations. Consequently, planning a communication system that can deliver improved signal strength becomes highly challenging. To address this issue, indoor path loss models are employed to estimate signal loss in different environments, frequencies, and distances. This paper introduces an intelligent multi-objective genetic algorithm aimed at enhancing path loss and received signal power. A comparative analysis is conducted to evaluate the performance of the proposed intelligent optimization algorithm against the traditional approach. The path loss and received power of various scenarios are estimated using various path loss models. The 5GCM indoor officce, 5GCM InH shopping mall, 3GPP TR 38.91 InH office, mmMAGIC InH office, METIS InH shopping mall, and IEEE 802.11 ad InH office indoor path loss models estimates the path loss of 62.37 dB, 62.15 dB, 63.12 dB, 50 dB, 55.18 dB, and 52.89 dB in traditional approach and 36.87 dB, 35.86 dB, 36.84 dB, 68.80 dB, 36.23 dB and 33.94 dB using GA approach and received powers of - 12.17 d B m , - 11.37 d B m , - 12.17 d B m , - 5.80 d B m , - 12.24 d B m and - 8.68 d B m in traditional approach and 26.13 dBm, 27.14 dBm, 26.15 dBm, - 5.80 d B m , 26.75 dBm and 29.05 dBm using GA approach repectively. The 5GCM and 3GPP models produces the path loss difference above 25 dB and mmMAGIC, METIS and IEEE models produces a path loss below 19 dB. Except mmMAGIC model, all models produces the recceiver power difference above 37 dBm. Therefore, the highest path loss difference of 26 dB is observed in 5GCM InH shopping mall model and the highest reccieved power difference of 39.01 dBm is observed in METIS InH shopping mall model. The results clearly demonstrate that the proposed intelligent optimization approach outperforms the traditional approach across various indoor scenarios. [ABSTRACT FROM AUTHOR]

A vehicular ad hoc network (VANET) includes groups of stationary or moving vehicles linked by a wireless network. The significant usage of VANET is to offer comfort and safety to drivers in road environments. VANET provides a communication framework that aids in minimizing accidents. Also, sharing data in VANET is time-sensitive and necessitates vigorous and quick network link formation. However, appropriate routing is critical to avoid the streaming issues that occur in VANET applications. This research area is of great concern to the researchers, and this work intends to propose a new Cluster-based VANET routing model with steps like (a) optimal cluster head selection (CHS) and (b) appropriate gateway selection for data transmission. Here, the optimal CHS takes place considering the Packet delivery ratio (PDR), mean routing load, end-to-end (e2e) Delay, Throughput, and packet overhead. This paper introduced a new Customized hunger's foraging honey badger with a Dynamic multi-objective non-sorted genetic algorithm (CHFHB-DMNSGA). Also, the gateway selection is done under two objectives: Vehicular gateway mobility and receiver signal strength, respectively. Finally, the paper presented a comparative assessment to validate the effectiveness of the adopted algorithm CHFHB-DMNSGA over the other algorithms like R-NSGA-II, NSGA-III, NSGA-II, HBA, HGS, ABC, FF, CSO, ACO, LA, MHB-GA, GA-FF, and HFCHBO. [ABSTRACT FROM AUTHOR]

Soluble solid content (SSC) and firmness are significant indexes to evaluate the quality of wolfberry. This study employed hyperspectral imaging (HSI) technology for the rapid detection and visualization of the distribution of SSC and firmness in mature wolfberries. The hyperspectral images of Ningqi 1 and Ningqi 7 were collected in the range of 400–1000 nm. The image segmentation method was used to determine the region of interest (ROI) of the wolfberry samples and extract the mean spectra, and the performance of the four preprocessing techniques was evaluated based on the partial least squares (PLSR) model, which concluded that the standard normal variable transformation (SNV) and multiple scattering correction (MSC) preprocessing methods were able to achieve the optimal results. Principal component analysis (PCA), successive projection algorithm (SPA), competitive adaptive reweighted sampling method (CARS) and their combination were used to select the characteristic wavelength, with CARS-SPA being more accurate. PLSR, support vector machine regression (SVR) and backpropagation genetic algorithm (BPNN-GA) models were used to predict the soluble solid content and firmness of wolfberry by full wavelength and characteristic wavelength, respectively. The optimal model for SSC and firmness of Ningqi 1 was identified as MSC-CARS-SPA-BPNN-GA, with Rp2 of 0.949 and 0.913, RMSEP of 0.365 and 0.524, and RPD of 4.104 and 3.422, respectively. For Ningqi 7, the optimal model was SNV-CARS-SPA-BPNN-GA, with Rp2 of 0.936 and 0.880, RMSEP of 0.364 and 0.537, and RPD of 3.860 and 2.706, respectively. Finally, these optimal models were utilized to visualize the distribution of SSC and firmness in the ROI. The findings underscore the rapid and precise nature of hyperspectral imaging in detecting the SSC and firmness of wolfberry, thereby establishing a technological and theoretical foundation for expedited wolfberry quality assessment. [ABSTRACT FROM AUTHOR]

The major problem facing users of Hopfield neural networks is the automatic choice of hyperparameters depending on the optimisation problem. This work introduces an automatic method to overcome this problem based on an original mathematical model minimizing the energy function. This methods ensures the feasibility of optimal solution obtained by decomposing the set of the feasible solutions. We illustrate the proposed model in the context of six well-known NP-hard problems: meeting scheduling problem, Kohonen network problem, portfolio selection problem, traveling salesman problem, task assignment problem, and max-stable problem. To show the effectiveness of our model, we use particle swarm and genetic algorithms to solve several instances of the last three problems. Numerical results show the good performance of the proposed approach compared to random tuning hyperparameters methods. Indeed, our approach permits an improvement of 49.75% for traveling salesman problem, 5.92% for task assignment problem, and 29.41% for max-stable problem. [ABSTRACT FROM AUTHOR]

Metaheuristic methods have emerged as powerful tools for solving complex optimization problems in various domains, including the sustainability of water distribution systems. They provide efficient and effective solutions by mimicking natural processes and searching for the optimal option within a large solution space. Despite the existence of these methods in the water distribution field for several years, a direct comparison between the various proposed solutions often proves challenging, due to the different parameter definitions used by the authors. The present review presents the solutions proposed by a total of 36 research papers taken from the Web of Science, Scopus and Google Scholar databases focusing on the application of metaheuristic methods for leakage reduction and energy saving in water distribution networks. The review is intended to facilitate comparative analysis among the solutions proposed by authors concerning key aspects of the optimization process. These aspects include the definition of the algorithm, the specification of the objective function, and the strategies employed for reducing the search space. The characteristics of the networks used as case studies by the reviewed papers are also presented to allow the reader to evaluate the applicability of the solutions to specific networks. [ABSTRACT FROM AUTHOR]

Recent years have witnessed the rise of accurate but obscure classification models that hide the logic of their internal decision processes. Explaining the decision taken by a black-box classifier on a specific input instance is therefore of striking interest. We propose a local rule-based model-agnostic explanation method providing stable and actionable explanations. An explanation consists of a factual logic rule, stating the reasons for the black-box decision, and a set of actionable counterfactual logic rules, proactively suggesting the changes in the instance that lead to a different outcome. Explanations are computed from a decision tree that mimics the behavior of the black-box locally to the instance to explain. The decision tree is obtained through a bagging-like approach that favors stability and fidelity: first, an ensemble of decision trees is learned from neighborhoods of the instance under investigation; then, the ensemble is merged into a single decision tree. Neighbor instances are synthetically generated through a genetic algorithm whose fitness function is driven by the black-box behavior. Experiments show that the proposed method advances the state-of-the-art towards a comprehensive approach that successfully covers stability and actionability of factual and counterfactual explanations. [ABSTRACT FROM AUTHOR]

Time series analysis and prediction have attained significant attention from the research community in the past few decades. However, the prediction accuracy of the models highly depends on the models' learning process. In order to optimize resource usage, a better learning methodology, in terms of accuracy and learning time, is needed. In this context, the current research work proposes EvoLearn, a novel method to improve and optimize the learning process of neural-based models. The presented technique integrates the genetic algorithm with back-propagation to train model weights during the learning process. The fundamental idea behind the proposed work is to select the best components from multiple models during the training process to obtain an adequate model. To demonstrate the applicability of EvoLearn, the method is tested on the state-of-the-art neural models (namely MLP, DNN, CNN, RNN, and GRU), and performances are compared. Furthermore, the presented study aims to forecast two types of time series, i.e. air pollution and energy consumption time series, using the developed framework. In addition, the considered neural models are tested on two datasets of each time series type. From the performance comparison and evaluation of EvoLearn using a one-tailed paired T-test against the conventional back-propagation-based learning approach, it was found that the proposed method significantly improves the prediction accuracy. [ABSTRACT FROM AUTHOR]

Offline signature verification (OSV) and identification systems have been used to identify individuals. It has become a significant task to validate the authenticity of forensic and commercial transactions. The accuracy of signature authentication greatly depends on feature extraction and feature selection (FS). Regarding features selection, the less discriminatory features in the global and local features approach result in a reduced recognition rate. Therefore, stable and apropos features have a significant influence on signature verification. A genetic algorithm (GA) was used to select the best features, and it is an effective and efficient technique for optimization and machine learning applications. However, GA has a specific limitation when the dimension of the search increases, such as lack of extensive exploitation and, difficult obtaining good convergence. Thus, the computation increases exponentially, and the complexity of the solution increases. This work proposes a novel Quantum Inspired Genetic Algorithm (QIGA) FS approach based on GA and Quantum Rotation Computing (QRC) to increase the diversity of populations and overcome computational cost and early convergence problems. It employs a superposition in both mutation (divergence) and crossover (convergence) operations to enhance diversity and address population selection. Three public signature databases (SID Arabic handwriting signatures, CEDAR, and UTSIG) are employed to validate the proposed method. The experimental results show that the proposed model enhanced the performance of the classical GA by adopting quantum-inspired computing concepts. [ABSTRACT FROM AUTHOR]

Highlights: The types, working principles, advantages and limitations of pattern recognition methods based on chemiresistive gas sensor array are reviewed and discussed comprehensively. Outstanding and novel advancements in the application of machine learning methods for gas recognition in different important areas are compared, summarized and evaluated. The current challenges and future prospects of machine learning methods in artificial olfactory systems are discussed and justified. As information acquisition terminals for artificial olfaction, chemiresistive gas sensors are often troubled by their cross-sensitivity, and reducing their cross-response to ambient gases has always been a difficult and important point in the gas sensing area. Pattern recognition based on sensor array is the most conspicuous way to overcome the cross-sensitivity of gas sensors. It is crucial to choose an appropriate pattern recognition method for enhancing data analysis, reducing errors and improving system reliability, obtaining better classification or gas concentration prediction results. In this review, we analyze the sensing mechanism of cross-sensitivity for chemiresistive gas sensors. We further examine the types, working principles, characteristics, and applicable gas detection range of pattern recognition algorithms utilized in gas-sensing arrays. Additionally, we report, summarize, and evaluate the outstanding and novel advancements in pattern recognition methods for gas identification. At the same time, this work showcases the recent advancements in utilizing these methods for gas identification, particularly within three crucial domains: ensuring food safety, monitoring the environment, and aiding in medical diagnosis. In conclusion, this study anticipates future research prospects by considering the existing landscape and challenges. It is hoped that this work will make a positive contribution towards mitigating cross-sensitivity in gas-sensitive devices and offer valuable insights for algorithm selection in gas recognition applications. [ABSTRACT FROM AUTHOR]

This research paper introduces the Double Stator (DS) Hybrid Excitation (HE) Halbach Permanent Magnet (HPM) Flux Switching (FS) machine. The machine construction and its optimization specifically designed for electric vehicle (EV)/hybrid electric vehicle (HEV) traction applications are investigated. The optimization using a multi-objective Genetic Algorithm is conducted following a sensitivity analysis-based identification of key optimization parameters and constraints. The finite element results are compared with the performance of a state-of-the-art benchmark FSPM machine having identical PM volume and winding current densities. The proposed design is shown to outperform the benchmark with 16.2% increase in back-electromotive force and 14.7% reduction in cogging torque. Furthermore, the average torque is improved at flux-enhancing operation by 20.8%, and the torque ripple is reduced by 9.9%. Notably, the proposed machine also is capable of flux regulation thereby having the ability to operate in a wide speed range. A detailed explanation of the reasons for the significant improvements in the proposed machine structure is provided to offer a comprehensive understanding of its rationale. These research findings indicate that this innovative DS-HE-HPM-FS machine can enhance the performance of EVs and HEVs. [ABSTRACT FROM AUTHOR]

Over the years, prediction techniques for the highly variable angular velocity of the Earth represented by Earth's rotation (UT1-UTC) and length-of-day (LOD) have been continuously improved. This is because many applications like navigation, astronomy, space exploration, climate studies, timekeeping, disaster monitoring, and geodynamic studies, all rely on predictions of these Earth rotation parameters. They provide early warning of changes in the Earth's rotation, allowing various industries and scientific fields to operate more precisely and efficiently. Thus, in our study, we focused on short-term prediction for UT1-UTC (dUT1) and LOD. Our prediction approach is to combine machine learning (ML) technique with efficient evolutionary computation (EC) algorithms to achieve reliable and improved predictions. Gaussian process regression (GPR) is used as the ML technique with genetic algorithm (GA) as the EC algorithm. GA is used for hyperparameter optimization of GPR model as selecting appropriate values for hyperparameter are essential to ensure that the prediction model can accurately capture the underlying patterns in the data. We conducted some experiments with our prediction approach to thoroughly test its capabilities. Moreover, two forecasting strategies were used to assess the performance in both hindcast and operational settings. In most of the experiments, the data used are the multi-technique combinations (C04) generated by International Earth Rotation and Reference Systems Service (IERS). In one of the experiments, we also investigated the performance of our prediction model on dUT1 and LOD from four different products obtained from IERS EOP 20 C04, DTRF20, JTRF20 and USNO. The prediction products are evaluated with real estimates of the EOP product with which the model is trained. The combined excitations of the atmosphere, oceans, hydrology, and sea level (AAM + OAM + HAM + SLAM) are used as predictors because they are highly correlated to the input data. The results depict the highest performance of 0.412 ms in dUT1 and 0.092 ms/day in LOD, on day 10 of predictions. It is worth noting that the later predictions were obtained by incorporating the uncertainty of the input data as weights in the prediction model, which was a novel approach tested in this study. [ABSTRACT FROM AUTHOR]

As the significance and complexity of solar panel performance, particularly at their maximum power point (MPP), continue to grow, there is a demand for improved monitoring systems. The presence of variable weather conditions in Maroua, including potential partial shadowing caused by cloud cover or urban buildings, poses challenges to the efficiency of solar systems. This study introduces a new approach to tracking the Global Maximum Power Point (GMPP) in photovoltaic systems within the context of solar research conducted in Cameroon. The system utilizes Genetic Algorithm (GA) and Backstepping Controller (BSC) methodologies. The Backstepping Controller (BSC) dynamically adjusts the duty cycle of the Single Ended Primary Inductor Converter (SEPIC) to align with the reference voltage of the Genetic Algorithm (GA) in Maroua's dynamic environment. This environment, characterized by intermittent sunlight and the impact of local factors and urban shadowing, affects the production of energy. The Genetic Algorithm is employed to enhance the efficiency of BSC gains in Maroua's solar environment. This optimization technique expedites the tracking process and minimizes oscillations in the GMPP. The adaptability of the learning algorithm to specific conditions improves energy generation, even in the challenging environment of Maroua. This study introduces a novel approach to enhance the efficiency of photovoltaic systems in Maroua, Cameroon, by tailoring them to the specific solar dynamics of the region. In terms of performance, our approach surpasses the INC-BSC, P&O-BSC, GA-BSC, and PSO-BSC methodologies. In practice, the stabilization period following shadowing typically requires fewer than three iterations. Additionally, our Maximum Power Point Tracking (MPPT) technology is based on the Global Maximum Power Point (GMPP) methodology, contrasting with alternative technologies that prioritize the Local Maximum Power Point (LMPP). This differentiation is particularly relevant in areas with partial shading, such as Maroua, where the use of LMPP-based technologies can result in power losses. The proposed method demonstrates significant performance by achieving a minimum 33% reduction in power losses. [ABSTRACT FROM AUTHOR]

Ballistic impacts on human thorax without penetration can produce severe injuries or even death of the carrier. Soft tissue finite element models must capture the non-linear elasticity and strain-rate dependence to accurately estimate the dynamic human mechanical response. The objective of this work is the calibration of a visco-hyperelastic model for soft tissue simulants. Material model parameters have been calculated by fitting experimental stress–strain relations obtained from the literature using genetic algorithms. Several parametric analyses have been carried out during the definition of the optimization algorithm. In this way, we were able to study different optimization strategies to improve the convergence and accuracy of the final result. Finally, the genetic algorithm has been applied to calibrate two different soft tissue simulants: ballistic gelatin and styrene–ethylene–butylene–styrene. The algorithm is able to calculate the constants for visco-hyperelastic constitutive equations with high accuracy. Regarding synthetic stress–strain curves, a short computational time has been shown when using the semi-free strategy, leading to high precision results in stress–strain curves. The algorithm developed in this work, whose code is included as supplementary material for the reader use, can be applied to calibrate visco-hyperelastic parameters from stress–strain relations under different strain rates. The semi-free relaxation time strategy has shown to obtain more accurate results and shorter convergence times than the other strategies studied. It has been also shown that the understanding of the constitutive models and the complexity of the stress–strain objective curves is crucial for the accuracy of the method. [ABSTRACT FROM AUTHOR]

This paper applies deep learning models to predict Bitcoin price directions and the subsequent profitability of trading strategies based on these predictions. The study compares the performance of the convolutional neural network–long short-term memory (CNN–LSTM), long- and short-term time-series network, temporal convolutional network, and ARIMA (benchmark) models for predicting Bitcoin prices using on-chain data. Feature-selection methods—i.e., Boruta, genetic algorithm, and light gradient boosting machine—are applied to address the curse of dimensionality that could result from a large feature set. Results indicate that combining Boruta feature selection with the CNN–LSTM model consistently outperforms other combinations, achieving an accuracy of 82.44%. Three trading strategies and three investment positions are examined through backtesting. The long-and-short buy-and-sell investment approach generated an extraordinary annual return of 6654% when informed by higher-accuracy price-direction predictions. This study provides evidence of the potential profitability of predictive models in Bitcoin trading. [ABSTRACT FROM AUTHOR]

Groundwater modeling is essential for effective water resource management. However, the heterogeneous distribution of hydrogeological parameters, such as hydraulic conductivity (K), significantly impacts the accuracy and efficiency of simulations. This study investigates the accuracy and computational efficiency of K parameter inversion across varying dimensions. Heterogeneous logarithmic random K (lnK) fields were generated using the Karhunen-Loève Expansion. Three inversion algorithms—Genetic Algorithm, Markov Chain Monte Carlo, and Ensemble Smoother (ES)—were evaluated in conjunction with a Kriging surrogate model. These algorithms were used to invert parameters using different numbers of Inversion Target Features (ITFs) as prediction targets. Results indicate that ES consistently outperformed GA and MCMC in terms of inversion accuracy and computational efficiency across all ITF scenarios (7, 12, 34, 48, and 71 ITFs). Sensitivity analysis revealed that selecting parameters with higher total-order indices for inversion is crucial, as a small subset of ITFs often accounts for a majority of the sensitivity. While increasing the number of ITFs captures greater spatial variability in lnK, high-dimensional cases can lead to less responsive output heads. Furthermore, inadequate grid density may not adequately represent fine-scale lnK distributions. This study provides valuable insights into parameter inversion within heterogeneous groundwater models, furthering our understanding of these complex systems. [ABSTRACT FROM AUTHOR]

The precise estimation of the shear strength of reinforced concrete walls is critical for structural engineers. This projection, nevertheless, is exceedingly complicated because of the varied structural geometries, plethora of load cases, and highly nonlinear relationships between the design requirements and the shear strength. Recent related design code regulations mostly depend on experimental formulations, which have a variety of constraints and establish low prediction accuracy. Hence, different soft computing techniques are used in this study to evaluate the shear capacity of reinforced concrete walls. In particular, developed models for estimating the shear capacity of concrete walls have been investigated, based on experimental test data accessible in the relevant literature. Adaptive neuro-fuzzy inference system, the integrated genetic algorithms, and the integrated particle swarm optimization methods were used to optimize the fuzzy model's membership function range and the results were compared to the outcomes of random forests (RF) model. To determine the accuracy of the models, the results were assessed using several indices. Outliers in the anticipated data were identified and replaced with appropriate values to ensure prediction accuracy. The comparison of the resulting findings with the relevant experimental data demonstrates the potential of hybrid models to determine the shear capacity of reinforced concrete walls reliably and effectively. The findings revealed that the RF model with RMSE = 151.89, MAE = 111.52, and R 2 = 0.9351 has the best prediction accuracy. Integrated GAFIS and PSOFIS performed virtually identically and had fewer errors than ANFIS. The sensitivity analysis shows that the thickness of the wall (b w) and concrete compressive strength (f c) have the most and the least effects on shear strength, respectively. [ABSTRACT FROM AUTHOR]

A newly developed and hand layup hybrid FRP of Basalt, Glass and Kevlar-29 was considered for the characterisation of Nd: YAG laser drilling and optimisation using stochastic methods. Laser current (LC), pulse width (PW), stand-off distance (SoD), and compressed air pressure (CAP) were the laser parameters. The Box-Behnken designed laser drilling experiments were completed to characterise through hole taper angle and average overcut. The grey relational grade (GRG) of the responses was considered for optimisation using a genetic algorithm (GA), teaching learning-based optimisation (TLBO) and golden jackal optimisation (GJO). The predicted optimum setting was LC of 180 A, SoD of 1 mm and CAP of 8 kg/cm2. The GA & TLBO and GJO predicted PW of 3 ms with GRG of 0.9948 and 2.625 ms with GRG of 0.9203, respectively. The experimental verification of the predicted conditions acknowledges the predicted values of the responses with acceptable errors. [ABSTRACT FROM AUTHOR]

Conceptual design plays an important role in determining the basic characteristics and final product performances. However, there are many uncertainties in the conceptual design, such as the vagueness of customer requirements, uncertainty of design parameters, and diversity of the decision-making, which will lead to fluctuations in the performance or even failure of the scheme in many cases. To solve this problem, a conceptual design model was constructed by considering multi-uncertainties. First, taking satisfaction, performance, and production cost as design objectives, a conceptual design optimization model was established. Secondly, an improved non-dominated sorting genetic algorithm-II (NSGA-II) based on the expectation-possibility-probability hybrid model was proposed to search the Pareto solutions of conceptual design. Finally, the optimal conceptual design scheme was selected from the Pareto set by using the intuitionistic fuzzy λ-Shapely Choquet integral method. The effectiveness and efficiency of the proposed method were validated by the polishing machine conceptual design. [ABSTRACT FROM AUTHOR]

Double suction centrifugal pumps, which feature large flow and head, are applied in water utility and transportation sectors. The efficiency, sound, and vibration levels are key performance indexes for double suction centrifugal pumps. This study aims to improve the performance of double suction pumps using a multi-objective optimization method. The Latin hypercube sampling (LHS) method is used to randomly generate sample data considering five key geometric parameters of the impeller, and the agent model training samples are generated using numerical computation. Then, the multi-objective optimization design of the impeller, focusing on the head, efficiency, and pressure pulsation energy as the objectives, was carried out by combining the Gaussian process regression (GPR) and non-dominated sorting genetic algorithm II (NSGA-II) algorithms. Results show that the head is increased by 3.91 m, the efficiency is increased by 0.2 %, and the pressure pulsation energy is reduced by 24 % compared with the original model. Meanwhile, the detailed information of energy loss and pressure pulsation in the pump was analyzed to understand the influence of impeller geometry parameters. This study provides a certain reference for the optimized design of double-suction pumps. [ABSTRACT FROM AUTHOR]

This paper presents a comparative study between four techniques recently used to improve the wind energy conversion system (WECS) to water pumping systems. The WECS is a renewable energy source which has developed rapidly in recent years. The use of the WECS in the water pumping field is a free solution (economically) compared to the use of the electricity grid supply. The control of WECS, equipped with a permanent magnet synchronous generator, has the objective of carefully maximising power generation. A comparative study between the proposed Fuzzy Logic Control, optimised using a genetic algorithm and particle swarm optimisation algorithm, and the conventional Perturb and Observe MPPT method using Matlab/Simulink, is presented. The performance of the proposed system has been verified against the generated output voltage, current and power waveforms, intermediate circuit voltage waveform, and generator speed. The presented results demonstrate the effectiveness of the control strategy applied in this work. [ABSTRACT FROM AUTHOR]

The deep operator network (DeepONet) structure has shown great potential in approximating complex solution operators with low generalization errors. Recently, a sequential DeepONet (S-DeepONet) was proposed to use sequential learning models in the branch of DeepONet to predict final solutions given time-dependent inputs. In the current work, the S-DeepONet architecture is extended by modifying the information combination mechanism between the branch and trunk networks to simultaneously predict vector solutions with multiple components at multiple time steps of the evolution history, which is the first in the literature using DeepONets. Two example problems, one on transient fluid flow and the other on path-dependent plastic loading, were shown to demonstrate the capabilities of the model to handle different physics problems. The use of a trained S-DeepONet model in inverse parameter identification via the genetic algorithm is shown to demonstrate the application of the model. In almost all cases, the trained model achieved an R 2 value of above 0.99 and a relative L 2 error of less than 10% with only 3200 training data points, indicating superior accuracy. The vector S-DeepONet model, having only 0.4% more parameters than a scalar model, can predict two output components simultaneously at an accuracy similar to the two independently trained scalar models with a 20.8% faster training time. The S-DeepONet inference is at least three orders of magnitude faster than direct numerical simulations, and inverse parameter identifications using the trained model are highly efficient and accurate. [ABSTRACT FROM AUTHOR]