Over the last two decades,stochastic optimizationalgorithms have proved to be a very promising approach to solving a variety of complex optimization *** eaglesearchoptimization(BES)as a new stochastic optimization ...
详细信息
Over the last two decades,stochastic optimizationalgorithms have proved to be a very promising approach to solving a variety of complex optimization *** eaglesearchoptimization(BES)as a new stochastic optimizationalgorithm with fast convergence speed has the ability of prominent optimization and the defect of collapsing in the local *** avoid BES collapse at local optima,inspired by the fact that the volume of the sphere is the largest when the surface area is certain,an improved bald eagle search optimization algorithm(INMBES)integrating the random shrinkage mechanism of the sphere is ***,the INMBES embeds spherical coordinates to design a more accurate parameter update method to modify the coverage and dispersion of the ***,the population splits into elite and non-elite groups and the Bernoulli chaos is applied to elite group to tap around potential solutions of the *** non-elite group is redistributed again and the Nelder-Mead simplex strategy is applied to each group to accelerate the evolution of the worst individual and the convergence process of the *** results of Friedman and Wilcoxon rank sum tests of CEC2017 in 10,30,50,and 100 dimensions numerical optimization confirm that the INMBES has superior performance in convergence accuracy and avoiding falling into local optimization compared with other potential improved algorithms but inferior to the champion algorithm and ranking *** three engineering constraint optimization problems and 26 real world problems and the problem of extracting the best feature subset by encapsulated feature selection method verify that the INMBES’s performance ranks first and has achieved satisfactory accuracy in solving practical problems.
For multi-threshold segmentation of grayscale and color images, the computational complexity increases exponentially with the increase of the number of threshold levels. In this paper, we propose a new method to segme...
详细信息
For multi-threshold segmentation of grayscale and color images, the computational complexity increases exponentially with the increase of the number of threshold levels. In this paper, we propose a new method to segment grayscale and color images with Kapur entropy as the objective function. The method introduces four strategies such as horizontal crossover and vertical crossover into the bald eagle search optimization algorithm, forming an advanced bald eagle search optimization algorithm (ABES). In the benchmark function comparison experiments at IEEE CEC 2017, ABES was compared with classical and novel algorithms, and the results proved to have stronger convergence speed, convergence accuracy, and stability than other algorithms. To demonstrate the effectiveness of the method in multi-thresholding segmentation of grayscale and color images, it is applied to low-level and high-level image multi-thresholding experiments, and the experimental results show that ABES outperforms other algorithms in the evaluation of PSNR, MSSIM, and FSIM, and ABES is a high-quality image segmentation method.
The Muskingum model is one of the most frequently used flood routing models. In any reach of a river with hydrometric stations at its beginning and end, model parameters can be determined based on floods measured at t...
详细信息
The Muskingum model is one of the most frequently used flood routing models. In any reach of a river with hydrometric stations at its beginning and end, model parameters can be determined based on floods measured at the inlet and outlet of this reach. Therefore, if another flood enters this reach, the outflow flood can also be predicted using the data of this flood and the Muskingum model parameters without cost. In this research, the type-1 non-linear Muskingum model (NLMM1), the flood data clustering method based on the k-means model, and the combination of NLMM1 with k-means (K-NLMM1) are used. This approach is utilized for four flood data sets, including the Wilson flood, the Wye River flood, the Lewis-Viessman flood and the Sutculer flood. Two optimization problems are defined in this paper. The first problem is to determine the center of clusters and the second one is to specify the parameters of NLMM1 and K-NLMM1. The baldeaglesearch (BES) optimizationalgorithm is used to solve such problems. The objective function for determining the Muskingum model parameters is considered to be minimizing the sum of squares of the difference between the observed outflow hydrograph and the routed hydrograph (SSQ). For the first to fourth case studies, the SSQ of the NLMM1 model is computed to be 245.6, 54,185.6, 74,307.2 and 557.3 (m3/s)2, respectively. Also, the values of SSQ for the K6-NLMM1 superior model in the first to fourth case studies are computed to be 3.1, 1185.7, 40,427, and 115.3 (m3/s)2, respectively. Thus, by clustering the data, the SSQ value decreases in the range of 46-99%.
Missing data imputation is a critical task, as incomplete datasets can negatively impact model performance. Traditional imputation methods often fail to capture the intricate relationships within data, resulting in po...
详细信息
Missing data imputation is a critical task, as incomplete datasets can negatively impact model performance. Traditional imputation methods often fail to capture the intricate relationships within data, resulting in poor representations of complex patterns and reduced predictive accuracy. Generative Adversarial Network (GAN)-based imputation methods frequently encounter challenges, such as hyperparameter tuning difficulties and mode collapse. In this paper, we propose a improved GAN-based imputation framework, termed BESGAIN. To be concrete, BESGAIN introduces the baldeaglesearch (BES) optimizationalgorithm to dynamically adjust hyperparameters, ensuring high-quality data imputation. Additionally, we incorporate an enhanced loss function that leverages cosine similarity to improve data diversity. This helps prevent performance degradation caused by overfitting to a single data mode. As a result, BESGAIN achieves a richer and more accurate set of imputed values. We validate the performance of BESGAIN across seven real-world datasets and compare it against state-of-the-art and widely used imputation methods. The experimental results demonstrate that BESGAIN outperforms existing techniques in terms of accuracy, robustness, and generalizability, particularly in datasets with high missing rates.
Accurate assessment of metrics such as state of health (SOH) is of paramount importance in effec-tive battery management systems (BMS), given the propensity of batteries to experience capacity degradation with aging. ...
详细信息
Accurate assessment of metrics such as state of health (SOH) is of paramount importance in effec-tive battery management systems (BMS), given the propensity of batteries to experience capacity degradation with aging. This research endeavors to significantly enhance the precision of battery capacity estimation by effectively mitigating the inherent uncertainties associated with state of charge (SOC) estimation and measurement. To address this challenge, we introduce an innovative approach leveraging the baldeaglesearchalgorithm (BES), a method inspired by the systematic hunting behavior of baldeagles. BES strategically navigates the search space, identifying and selecting promising solutions through fitness evaluations. Our principal aim, utilizing the inherent capabilities of BES, is to pinpoint the optimal candidate that minimizes a designated cost function, while ensuring real-time cell capacity updates facilitated by the incorporation of a memory forgetting factor. The distinctiveness of this study is twofold: firstly, the strategic integration of the BES algorithm within the context of battery capacity optimization, and secondly, the inclusion of a memory forgetting factor to enhance real-time capacity estimations. The efficacy of our approach is rigorously substantiated through validation using NASA's Prognostic Data, along with three battery scenarios for plug-in hybrid and electric vehicles. BES consistently outperformed four aggressive algorithms, demonstrating heightened accuracy with a peak error rate of only 1.06% in the most demanding scenario. Furthermore, the predictive performance measures remained consistently below 0.41%, underscoring the robustness of our proposed methodology.(c) 2023 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://***/licenses/by/4.0/).
暂无评论