Solar photovoltaic (PV) arrays in portable applications are often subject to partial shading and rapid fluctuations of shading. In the usual series-connected wiring scheme, residual energy generated by partially shaded cells either cannot be collected (if diode bypassed) or, worse, impedes collection power from remaining fully illuminated not bypassed). Rapid fluctuation pattern makes maximum point (MPP) tracking difficult; generally, there will exist multiple local MPPs, their values change...

With the wide application of unmanned aerial vehicles (UAVs), performing search and rescue missions autonomously in unknown environment has become an increasingly concerning issue. In this article, we propose adaptive conversion speed Q-Learning algorithm (ACSQL). Performing UAV is divided into two stages: mission stage optimal path stage. first stage, a can find task points as soon possible, efficiency exploration increased by adaptively adjusting UAV. second to get secure short path,...

Hybrid systems composed of fuel cells and batteries combine the high energy density with power batteries. A dc/dc converter is placed between cell battery to balance flow them greatly increase peak output hybrid. This paper presents an adaptive control strategy for active sharing in hybrid source. can adjust current setpoint according state-of-charge (or voltage) battery, applicable two topologies cell/battery hybrids. The implemented Simulink then tested under arbitrary load conditions...

Multi-UAV is a complete system represented by multiple unmanned aerial vehicles (UAV) through collaborative technology. Besides, the UAV group has advantages in number and functions that single does not have. In complex environment, cannot large of material transportations escort functions. This paper, from perspective cooperative control group, solves problem cruise. Then, an autonomous flocking algorithm for heterogeneous swarm (ACHF) proposed. Firstly, movement factor proposed, deployed...

Hybrid power sources composed of fuel cells and secondary batteries can combine the high energy density with batteries. A dc/dc converter be placed between cell battery to balance flow greatly augment peak output power. This work presents a novel, flexible strategy for multiobjective control in hybrid source. The is able regulate current charging or voltage while limiting discharging battery. It used two different configurations without any change. implemented MATLAB/Simulink tested by...

The objective of this work is to determine what extent the performance a battery/capacitor hybrid power source can be improved by isolating battery through converter. an active (battery/capacitor/converter) compared that passive (battery/capacitor) using both simulation and experiment. Particularly in paper, two cells Sony US18650 lithium-ion Maxwell PC 100 supercapacitor are used as energy storage components construction. Both experimental data results show achieved greater mass volume...

With the further development of Unmanned Aerial Vehicle (UAV) technologies, research on multi-UAV formations have also received more attention. Vehicles (UAVs) cooperate with each other to form a formation group, which can give full play advantages that single UAV does not have, and capable working in multi-task scenarios. Based Sierpinski fractal structure, this paper proposes hybrid control architecture. To address aggregation problem swarms, multi-agent deep reinforcement learning (MADRL)...

The problem of task allocation UAV systems is a hot issue in scientific research. Aiming at overcome the shortcomings existing algorithms terms load balancing and execution efficiency, an algorithm Improved Self-Organizing Mapping (ISOM) proposed this paper. Firstly, considering two factors flight distance mission time, balance designed. sum track lengths multiple UAVs total time required by to complete all tasks area are used evaluate quality cooperative assignment UAVs. Secondly, learning...

This paper introduces a comprehensive approach to the study of energy management for all electric ships. Starting from an analysis characteristics power systems on board future ships, describes some control challenges and role storage in definition overall system stability. The theoretical description is then supported with preliminary simulation results.

To address the issues of insufficient attack capability against black-box face recognition models and poor visual quality adversarial samples, we propose Adv-Makeup++ model. First, introduce CBAM attention mechanism module in encoder. Second, a discriminator momentum enhanced training strategy, which strengthens discriminator's robustness to perturbations, thus forcing generator produce more powerful samples. Finally, overcome problem unbalanced efficiency between during GAN training,...

We describe and validate here an optimum configuration of solar array power converter connections for operation photovoltaic arrays in partially shaded or continuously changing shadow conditions portable mobile applications. Our results demonstrate that, under complex irradiance conditions, the generation capability using proposed is enhanced by a factor 2 without extra component costs. In addition, maximum tracking simplified. By controlling terminal voltage to follow prescribed values,...

In this study, a novel on-line pre-column electrochemical instrument (PECI) coupled with high-performance liquid chromatography (HPLC) was developed, and method based on PEC-HPLC-UV for amplifying the ultraviolet (UV) response of iodate (IO₃⁻) studied. Iodate undergoes reduction in PECI, resulting I(-) injected to an HPLC system detected by UV detector. For IO₃⁻ analysis, conditions that can influence efficiency, including applied potential, pH value salt concentration, were investigated...

Carefully perturbing adversarial inputs degrades the performance of traditional machine learning (ML) models. Adversarial (AML) that takes adversaries into account during training and emerges as a valid technique to defend against attacks. Due complexity uncertainty adversaries’ attack strategies, researchers utilize game theory study interactions between an adversary ML system designer. By configuring different rules analyzing outcomes in game, it is possible effectively predict strategies...

The egg production rate is a crucial metric in animal breeding, subject to biological and environmental influences exhibits characteristics of small sample sizes non-linearity. Currently, prediction research predominantly focuses on single-step point prediction, lacking multistep interval exploration. To bridge these gaps, this study proposes recursive, method for rates, integrating variables attention mechanisms. Initially, employed three gradient boosting tree models (XGBoost, LightGBM,...

FANET (Fly-Adhoc-Network) does not rely on prebuilt infrastructure, and can form a temporary network through wireless links anytime anywhere, which has been widely used in emergency communication disaster relief. In order to solve the problem of signal transmission fading caused by heavy fog, smoke other harsh environments, data packet failure high speed movement unmanned aerial vehicle (UAV), cooperative mode is adopted select relay at link layer for forwarding, effectively improves...

As machine learning models become increasingly integrated into practical applications and are made accessible via public APIs, the risk of model extraction attacks has gained prominence. This study presents an innovative efficient approach to attacks, aimed at reducing query costs enhancing attack effectiveness. The method begins by leveraging a pre-trained identify high-confidence samples from unlabeled datasets. It then employs unsupervised contrastive thoroughly dissect structural nuances...