In view of the enterprise personnel management (EPM), a novel nature-inspired optimization algorithm called personnel rating optimization (PRO) is proposed in this paper. The design of PRO mimics the workflow of manag...
详细信息
ISBN:
(纸本)9781665426053
In view of the enterprise personnel management (EPM), a novel nature-inspired optimization algorithm called personnel rating optimization (PRO) is proposed in this paper. The design of PRO mimics the workflow of management activities from EPM to achieve the aim for bolstering the team's competence level. Built on the mechanisms of PRO, including personal promotion, swarm interaction, and coordinated management, it could quickly lock onto the solution region in high-dimensional space. In this study, we analyzed the essential similarities between PRO and EMP about management strategies, then presented the detailed procedures and mathematical model applied in PRO. A set of various numerical benchmark functions have been conducted on PRO and some other state-of-art natural computation algorithms including whale optimization algorithm ( WOA), differential evolution algorithm (DE), genetic algorithm (GA) and particle swarm optimization (PSO). Experimental results show that PRO has high precision and good stability, and the premature phenomenon is effectively restrained.
optimization and proactive management of energy systems are crucial for achieving sustainability, efficiency and resilience in future smart energy networks. Data-driven approaches offer promising solutions for tacklin...
optimization and proactive management of energy systems are crucial for achieving sustainability, efficiency and resilience in future smart energy networks. Data-driven approaches offer promising solutions for tackling the complex and dynamic challenges of energy systems, such as uncertainty, variability, and heterogeneity. Meanwhile, recent advances in decreasing hardware costs and improving data accessibility have allowed for the collection of high-quality data, leading to the development of more accurate and robust data-driven models of different energy systems. In this study, a comprehensive overview of current and future trends in data-driven optimization for smart energy systems is presented. After introducing the motivation and the background of this research field, the potential applications and benefits of optimization in various domains is discussed, such as electric vehicles charge, district heating networks and energy districts. Subsequently this review focuses on different methods and techniques for data-driven optimization and proactive management, ranging from scientific models to machine learning algorithms. Finally, the novel European project, DigiBUILD, is introduced, where different case studies are tested in several pilots, including electric vehicle charging management for increasing renewable energy source consumption, district heating network operative costs optimization and building energy and comfort management.
Well-designed arrangements of arrays of circular lossy dielectric rods with finite lengths have been exploited to obtain proper radiation patterns for microwave applications. Accelerated three-dimensional (3D) numeric...
Well-designed arrangements of arrays of circular lossy dielectric rods with finite lengths have been exploited to obtain proper radiation patterns for microwave applications. Accelerated three-dimensional (3D) numerical modeling of such radiation mechanisms via the multilevel fast multipole algorithm (MLFMA) is suitable for well-matched results with real-life experiments. Enabling rigorous and accurate analyses of the corresponding problems, MLFMA, has been observed to provide quasi-optimal designs within a reasonable period of time, when this algorithm is efficiently combined with genetic algorithms (GAs). However, as shown in this contribution, the corresponding two-dimensional (2D) multiple scattering modeling via a well-conditioned version of the T-matrix method provides a much faster tool for GAs to optimize radiation patterns for a goal set on the central cross section plane of the rod arrays. Optimized geometries obtained using a combination of the 2D solver and GAs have been validated by using a 3D solver based on MLFMA to demonstrate the feasibility of the approach.
Electricity plays an indispensable role in human lives. Due to the increasing need for electricity in domestic, commercial, and industrial applications and the deletion of conventional sources, the power generation sy...
详细信息
Semantic segmentation (SS) is one of the most powerful tools to extract particular regions from medical images. Several studies succeeded to increase the segmentation accuracy of SS by combining with preprocessing. Th...
详细信息
ISBN:
(数字)9781665490627
ISBN:
(纸本)9781665490627
Semantic segmentation (SS) is one of the most powerful tools to extract particular regions from medical images. Several studies succeeded to increase the segmentation accuracy of SS by combining with preprocessing. The present study proposed an extraction method of cervical intervertebral disks from videofluorograph (VF), which is commonly used for the diagnosis of dysphagia, based on SS combined with 45 linear and nonlinear image filters. The combination of the image filters was optimized by the simulated annealing algorithm. In this study, live fully convolutional networks (FCNs), i.e. U-Net, feature pyramid network, LinkNet, pyramid scene parsing network and M-Net, were applied to the VF dataset of 19 patients and 39 healthy participants. When the image filters were not used, the mean F measures of the live FCNs were 0.660, 0.752, 0.803, 0.750 and 0.768, respectively, whereas hen used, they were increased to 0.747, 0.794, 0.813, 0.765 and 0.799, respectively. This experimental results demonstrated that the optimal combinations of image filters were effective to improve the segmentation accuracy of FCNs.
The article is considered intellectual information systems for monitoring and diagnosing complex technical objects, considering modern information technologies. Computer intellectualization ways of the functioning mod...
详细信息
The proceedings contain 24 papers. The special focus in this conference is on Artificial Life and Evolutionary Computation. The topics include: On the Evolution of Mechanisms for Collective Decision Mak...
ISBN:
(纸本)9783031239281
The proceedings contain 24 papers. The special focus in this conference is on Artificial Life and Evolutionary Computation. The topics include: On the Evolution of Mechanisms for Collective Decision Making in a Swarm of Robots;a Novel Online Adaptation Mechanism in Artificial Systems Provides Phenotypic Plasticity;Exploration of Genetic algorithms and CNN for Melanoma Classification;using Genetic algorithms to Optimize a Deep Learning Based System for the Prediction of Cognitive Impairments;Autonomous Inspections of Power Towers with an UAV;self-organizing Maps of Artificial Neural Classifiers - A Brain-Like Pin Factory;evolutionary Music: Statistical Learning and Novelty for Automatic Improvisation;ARISE: Artificial Intelligence Semantic Search Engine;influence of the Antigen Pattern Vector on the Dynamics in a Perceptron-Based Artificial Immune - Tumour- Ecosystem During and After Radiation Therapy;effective Resistance Based Weight Thresholding for Community Detection;two-Level Detection of Dynamic Organization in Cancer Evolution Models;artificial Chemical Neural Network for Drug Discovery applications;information Flow Simulations in Multi-dimensional and Dynamic Systems;investigation of the Ramsey-Pierce-Bowman Model;ethical Aspects of Computational Modelling in Science, Decision Support and Communication;an Oracle for the optimization of Underconstrained Compositions of Neural Networks - The Tick Hazard Use Case;obstacles on the Pathway Towards Chemical Programmability Using Agglomerations of Droplets;the Good, the Bad and the Ugly: Droplet Recognition by a "Shootout"-Heuristics;exploring the Three-Dimensional Arrangement of Droplets;geometric Restrictions to the Agglomeration of Spherical Particles;effectiveness of Dynamic Load Balancing in Parallel Execution of a Subsurface Flow Cellular Automata Model;Two Possible AI-Related Paths for Bottom-Up Synthetic Cell Research.
This paper presents the implementations of a commutation sequence for a five and seven-level cascaded H-bridge multilevel inverter (CHBMLI) using a Field Programmable Gate Array (FPGA). The switching angles were calcu...
详细信息
ISBN:
(数字)9798331530525
ISBN:
(纸本)9798331530532
This paper presents the implementations of a commutation sequence for a five and seven-level cascaded H-bridge multilevel inverter (CHBMLI) using a Field Programmable Gate Array (FPGA). The switching angles were calculated using a Genetic Algorithm (GA) to minimize the Total Harmonic Distortion (THD) in the output. The versatility and efficiency of the CHBMLI inverters are highlighted due to their modular configuration, which allows for generating high-fidelity output voltages and minimizing harmonics in the output signal. The fundamental frequency switching modulation technique generates the required output voltages. The FPGA implementation, carried out in VHDL, achieved THD values of 15.4% and 10.4% for the 5-level and 7-level inverters respectively, closely matching the theoretical values. The capability of FPGAs to handle the generation of control sequences for each voltage level is demonstrated with a laboratory prototype. These results demonstrate the effectiveness of the proposed approach, offering a viable and efficient solution for applications in renewable energy systems and motor control. Future work could explore further optimizations and extensions of this method to even more complex inverter configurations.
To address the issues of the Aquila optimization (AO) easily falling into local optima, slow convergence speed, and low convergence accuracy, this paper proposes a fusion of adaptive dynamic Aquila optimization incorp...
To address the issues of the Aquila optimization (AO) easily falling into local optima, slow convergence speed, and low convergence accuracy, this paper proposes a fusion of adaptive dynamic Aquila optimization incorporating with improved Sine Cosine (AO-SC). The AO-SC algorithm combines the adaptive weight mechanism and the nonlinearly decreasing search factor of the sine cosine algorithm to perform secondary optimization, effectively improving convergence accuracy and speed, and escaping from local optima. Additionally, the Cauchy-Gaussian mutation operator is also utilized to enhance the proposed AO-SC algorithm's global search capability. Finally, the introduction of nonlinear adaptive weight factors and the combination of archimedean spiral pattern with the variable step size feature of the original AO algorithm generate new local solutions to improve the local search capability and search accuracy of the AO algorithm. Through testing and comparison with seven other algorithms on ten benchmark functions, the final experimental results demonstrate that, the proposed AO-SC algorithm has significant advantages in terms of convergence speed, accuracy, and stability.
China's high-speed railway construction is making remarkable achievements. The increasing EMUs are approaching the overhaul cycle, which has brought great pressure to the existing 7 overhaul bases. With the increa...
详细信息
暂无评论