Search Results (25,089)

We use an array of nine elliptical Planar Hall Effect (PHE) sensors and machine learning algorithms to map the magnetic signal generated by a magnetic source. Based on the obtained mapping, the location and nature of the magnetic source can be determined. The sensors are positioned at the vertices of a symmetrical and evenly spaced 3 × 3 grid. The main electronic card orchestrates their measurement by supplying the required driving current and amplifying and sampling their output in a synchronized manner. A two-dimensional interpolation of the data collected from the nine sensors fails to yield a satisfactory mapping. To address this, we employed the Levenberg–Marquardt Algorithm (LMA) as a deterministic optimization method to estimate the magnetic source’s position and parameters, as well as machine earning (ML) algorithms, which consist of a Fully Connected Neural Network (FCNN). While LMA provided reasonable results, its reliance on a sparse sensor network and initial guesses for variables limited its accuracy. We show that the mapping is significantly improved if the data are processed with an FCNN that undergoes training and testing. Using simulations, we demonstrate that achieving similar improvement without ML would require increasing the number of sensors to more than 50. Full article

A peak shaving approach in selected industrial loads helps minimize power usage during high demand hours, decreasing total energy expenses while improving grid stability. A battery energy storage system (BESS) can reduce peak electricity demand in distribution networks. Quasi-dynamic load flow analysis (QLFA) accurately assesses the maximum loading conditions in distribution networks by considering factors such as load profiles, system topology, and network constraints. Achieving maximum peak shaving requires optimizing battery charging and discharging cycles based on real-time energy generation and consumption patterns. Seamless integration of battery storage with solar photovoltaic (PV) systems and industrial processes is essential for effective peak shaving strategies. This paper proposes a model predictive control (MPC) scheme that can effectively perform peak shaving of the total industrial load. Adopting an MPC-based algorithm design framework enables the development of an effective control strategy for complex systems. The proposed MPC methodology was implemented and tested on the Indian Utility 29 Node Distribution Network (IU29NDN) using the DIgSILENT Power Factory environment. Additionally, the analysis encompasses technical and economic results derived from a simulated storage operation and, taking Puducherry State Electricity Department tariff details, provides significant insights into the application of this method. Full article

This study investigates the impact of environmental conditions on the structural integrity and energy dynamics of historical masonry buildings using an IoT (Internet of Things) LoRaWAN-based (Long Range Wide Area Network) wireless sensor system. Over a six-month period, sensors were used to monitor wall temperature, wall humidity, air temperature, air humidity, crack width, and crack displacement. The data revealed significant correlations between environmental parameters and structural changes. Higher temperatures were associated with increased crack width, while elevated humidity levels correlated with greater crack displacement, showing the potential weakening of the masonry structure. Seasonal variations highlighted the cyclical nature of these changes, emphasizing the need for seasonal maintenance. Additionally, the findings suggest that managing temperature and humidity levels can optimize the building’s energy efficiency by reducing the need for additional heating or cooling. The use of LoRaWAN sensors provided real-time, remote monitoring capabilities, offering a cost-effective and scalable solution for preserving historical buildings. This study underscores the importance of continuous environmental and structural monitoring for the preservation of heritage sites. It also highlights the potential for integrating proactive maintenance strategies and energy optimization, ensuring long-term sustainability. By leveraging this IoT-based approach, this research contributes to the broader field of heritage conservation, offering a universal framework that can be applied to historical buildings worldwide, enhancing both their structural integrity and energy performance. Full article

Psychedelics, historically celebrated for their cultural and spiritual significance, have emerged as potential breakthrough therapeutic agents due to their profound effects on consciousness, emotional processing, mood, and neural plasticity. This review explores the mechanisms underlying psychedelics’ effects, focusing on their ability to modulate brain connectivity and neural circuit activity, including the default mode network (DMN), cortico-striatal thalamo-cortical (CSTC) loops, and the relaxed beliefs under psychedelics (REBUS) model. Advanced neuroimaging techniques reveal psychedelics’ capacity to enhance functional connectivity between sensory cerebral areas while reducing the connections between associative brain areas, decreasing the rigidity and rendering the brain more plastic and susceptible to external changings, offering insights into their therapeutic outcome. The most relevant clinical trials of 3,4-methylenedioxymethamphetamine (MDMA), psilocybin, and lysergic acid diethylamide (LSD) demonstrate significant efficacy in treating treatment-resistant psychiatric conditions such as post-traumatic stress disorder (PTSD), depression, and anxiety, with favorable safety profiles. Despite these advancements, critical gaps remain in linking psychedelics’ molecular actions to their clinical efficacy. This review highlights the need for further research to integrate mechanistic insights and optimize psychedelics as tools for both therapy and understanding human cognition. Full article

The global transition to sustainable energy systems has placed the use of electric vehicles (EVs) among the areas that might contribute to reducing carbon emissions and optimizing energy usage. This paper presents a bibliometric analysis of the interconnected domains of EVs, artificial intelligence (AI), machine learning (ML), and deep learning (DL), revealing a significant annual growth rate of 56.4% in research activity. Key findings include the identification of influential journals, authors, countries, and collaborative networks that have driven advancements in this domain. This study highlights emerging trends, such as the integration of renewable energy sources, vehicle-to-grid (V2G) schemes, and the application of AI in EV battery optimization, charging infrastructure, and energy consumption prediction. The analysis also uncovers challenges in addressing information security concerns. By reviewing the top-cited papers, this research underlines the transformative potential of AI-driven solutions in enhancing EV performance and scalability. The results of this study can be useful for practitioners, academics, and policymakers. Full article

The mobility and openness of smart cities characterize them as particularly complex networks, necessitating the resilience enhancement of smart city regions from a network structure perspective. Taking the Chengdu–Chongqing urban agglomeration as a case study, this research constructs economic, information, population, and technological intercity networks based on the complex network theory and gravity model to evaluate their spatial structure and resilience over five years. The main conclusions are as follows: (1) subnetworks exhibit a ‘core/periphery’ structure with a significant evolution trend, particularly the metropolitan area integration degree of capital cities has significantly improved; (2) the technology network is the most resilient but was the most affected by COVID-19, while the population and information networks are the least resilient, resulting from poor hierarchy, disassortativity, and agglomeration; (3) network resilience can be improved through system optimization and node enhancement. System optimization should focus more on improving the coordinated development of population, information, and technology networks due to their low synergistic level of resilience, while node optimization should adjust strategies according to the dominance, redundancy, and network role of nodes. This study provides a reference framework to assess the resilience of smart cities, and the assessment results and enhancement strategies can provide valuable regional planning information for resilience building in smart city regions. Full article

As Unmanned Aerial Vehicle (UAV) technology advances, UAVs have attracted widespread attention across military and civilian fields due to their low cost and flexibility. In unknown environments, UAVs can significantly reduce the risk of casualties and improve the safety and covertness when performing missions. Reinforcement Learning allows agents to learn optimal policies through trials in the environment, enabling UAVs to respond autonomously according to the real-time conditions. Due to the limitation of the observation range of UAV sensors, UAV target search missions face the challenge of partial observation. Based on this, Partially Observable Deep Q-Network (PODQN), which is a DQN-based algorithm is proposed. The PODQN algorithm utilizes the Gated Recurrent Unit (GRU) to remember the past observation information. It integrates the target network and decomposes the action value for better evaluation. In addition, the artificial potential field is introduced to solve the potential collision problem. The simulation environment for UAV target search is constructed through the custom Markov Decision Process. By comparing the PODQN algorithm with random strategy, DQN, Double DQN, Dueling DQN, VDN, QMIX, it is demonstrated that the proposed PODQN algorithm has the best performance under different agent configurations. Full article

Background: Prone positioning is a standard intervention in managing patients with severe acute respiratory distress syndrome (ARDS) and is known to improve oxygenation. However, its effects on other organs, particularly the kidneys, are less well understood. This study aimed to assess the association between prone positioning and the development of acute kidney injury (AKI), specifically in overweight and obese patients. Methods: A retrospective pre–post study was conducted on a cohort of 60 critically ill ARDS patients who were placed in the prone position during hospitalization. The development of AKI was assessed using the Acute Kidney Injury Network (AKIN) criteria, with AKI measured by both creatinine levels (AKIN_Cr) and urine output (AKIN_UO). Patients were divided into two groups based on body mass index (BMI): overweight/obese (BMI ≥ 25) and non-obese (BMI < 25). Data were collected before and after prone positioning. Results: In overweight/obese patients (n = 39, 57 cases), both the median AKIN_Cr and AKIN_UO scores increased significantly following prone positioning (from 0 to 1, median p < 0.01, and from 0 to 2, median p < 0.01, respectively). No statistically significant changes in AKIN scores were observed in non-obese patients nor were significant differences found in either group after repositioning to supine. Conclusions: Prone positioning is associated with an increased risk of acute kidney injury in overweight and obese ARDS patients. This may be due to the kidneys’ susceptibility to intra-abdominal hypertension in these patients. Further research is needed to explore optimal proning strategies for overweight and obese populations. Full article

Pre-trained neural networks like GPT-4 and Llama2 have revolutionized intelligent information processing, but their deployment in industrial applications faces challenges, particularly in harsh environments. To address these related issues, model offloading, which involves distributing the computational load of pre-trained models across edge devices, has emerged as a promising solution. While this approach enables the utilization of more powerful models, it faces significant challenges in harsh environments, where reliability, connectivity, and resilience are critical. This paper introduces failure-resilient inference in mobile networks (FRIM), a framework that ensures robust offloading and inference without the need for model retraining or reconstruction. FRIM leverages graph theory to optimize partition redundancy and incorporates an adaptive failure detection mechanism for mobile inference with efficient fault tolerance. Experimental results on DNN models (AlexNet, ResNet, VGG-16) show that FRIM improves inference performance and resilience, enabling more reliable mobile applications in harsh operating environments. Full article

According to the physical characteristics of cotton and the work characteristics of cotton pickers in the field, during the picking process, there is a risk of cotton combustion. The cotton picker working environment is complex, cotton ignition can be hidden, and fire is difficult to detect. Therefore, in this study, we designed an improved algorithm for multi-sensor data fusion; built a cotton picker fire detection system by using infrared temperature sensors, CO sensors, and the upper computer; and proposed a BP neural network model based on improved mutation operator hybrid gray wolf optimizer and particle swarm optimization (MGWO-PSO) algorithm based on the BP neural network model. This algorithm includes the introduction of a mutation operator in the gray wolf algorithm to improve the search ability of the algorithm, and, at the same time, we introduce the PSO algorithm idea. The improved fusion algorithm is used as a learning algorithm to optimize the BP neural network, and the optimized network is used to process and predict the data collected from temperature and gas sensors, which effectively improves the accuracy of fire prediction. The sensor measurements were compared with the actual values to verify the effectiveness of the GWO-PSO-optimized BP neural network model. Once experimentally verified, the improved GWO-PSO algorithm achieves a correlation coefficient R of 0.96929, a prediction accuracy rate of 96.10%, and a prediction error rate of only 3.9%, while the system monitors an accurate early warning rate of 96.07%, and the false alarm and omission rates are both less than 5%. This study can detect cotton picker fires in real time and provide timely warnings, which provides a new method for the accurate detection of fires during the field operation of cotton pickers. Full article

Early detection of plant diseases is crucial for agro-holdings, farmers, and smallholders. Various neural network architectures and training methods have been employed to identify optimal solutions for plant disease classification. However, research applying one-shot or few-shot learning approaches, based on similarity determination, to the plantdisease classification domain remains limited. This study evaluates different loss functions used in similarity learning, including Contrastive, Triplet, Quadruplet, SphereFace, CosFace, and ArcFace, alongside various backbone networks, such as MobileNet, EfficientNet, ConvNeXt, and ResNeXt. Custom datasets of real-life images, comprising over 4000 samples across 68 classes of plant diseases, pests, and their effects, were utilized. The experiments evaluate standard transfer learning approaches alongside similarity learning methods based on two classes of loss function. Results demonstrate the superiority of cosine-based methods over Siamese networks in embedding extraction for disease classification. Effective approaches for model organization and training are determined. Additionally, the impact of data normalization is tested, and the generalization ability of the models is assessed using a special dataset consisting of 400 images of difficult-to-identify plant disease cases. Full article

In recent years, social networks have become increasingly popular as platforms for personal expression, commercial transactions, and government management. The way information propagates on these networks influences the quality and expenses of social network activities, garnering substantial interest. This study addresses the enhancement of information spread in large-scale social networks constrained by resources, by framing the issue as a unique weighted k-vertex cover problem. To tackle this complex NP-hard optimization problem, a rapid local search algorithm named FastIM is introduced. A fast constructive heuristic is initially used to quickly find a starting solution, while a sampling selection method is incorporated to minimize complexity during the local search. When the algorithm stalls in local optima, a random walk operator reorients the search towards unexplored regions. Comparative tests highlight the proposed method’s robustness, scalability, and efficacy in maximizing information distribution across social networks. Moreover, strategy validation trials confirm that each element of the framework enhances its overall performance. Full article

This study examines the regional co-funding network as a novel framework for advancing high-quality fundamental research amid systemic reforms in science funding. Based on provincial joint funding data from Mainland China retrieved via the WoS-SCIE and SSCI databases (2013–2022), an interprovincial co-funding network was constructed. Social network analysis, kernel density estimation, and fixed-effects regression analysis were employed to explore the evolution of regional location advantages and their impact on technological innovation. The findings reveal that the co-funding network has become increasingly balanced over time, significantly enhancing the location-based innovation advantages of individual provinces and strengthening the network’s overall capacity to foster innovation. This improved equilibrium has positively influenced regional scientific output, demonstrating that a province’s position within the co-funding network—particularly its individual location advantage—plays a pivotal role in advancing technological progress. However, persistent disparities in regional collaboration and development remain, underscoring the need for more coordinated strategies to address uneven growth dynamics. By introducing the co-funding network as an analytical lens, this study uncovers the hidden channels of resource synergy and their influence on regional innovation. The results provide actionable insights for optimizing co-funding mechanisms and enhancing interprovincial collaboration to maximize innovation potential in China. Full article

With the increasing frequency of extreme weather events such as heavy rainstorm disasters, the stable operation of power systems is facing significant challenges. This paper proposes a two-stage restoration strategy for the distribution networks (DNs). First, a grid-based modeling approach is developed for urban DNs and transportation networks (TNs), capturing the dynamic evolution of heavy rainstorm disasters and more accurately modeling the impact on TNs and DNs. Then, a two-stage restoration strategy is designed for the DN by coordinating soft open points (SOPs) and mobile energy storage systems (MESSs). In the disaster progression stage, SOPs are utilized to enable the flexible reconfiguration and islanding of the DN, minimizing load loss. In the post-disaster recovery stage, the MESS and repair crew are optimally dispatched, taking into account the state of the TN to expedite power restoration. Finally, the experimental results demonstrate that the proposed method reduces load loss during restoration by 8.09% compared to approaches without precise TN and DN modeling. Full article

The biological methanation process has emerged as a promising alternative to thermo-catalytic methods due to its ability to operate under milder conditions. However, challenges such as low hydrogen solubility and the need for precise trace element supplementation (Fe(II), Ni(II), Co(II)) constrain methane production yield. This study investigates the combined effects of trace element concentrations and applied pressure on biological methanation, addressing their synergistic interactions. Using a face-centered composite design, batch mode experiments were conducted to optimize methane production. Response Surface Methodology (RSM) and Artificial Neural Network (ANN)—Genetic Algorithm (GA) approaches were employed to model and optimize the process. RSM identified optimal ranges for trace elements and pressure, while ANN-GA demonstrated superior predictive accuracy, capturing nonlinear relationships with a high R² (>0.99) and minimal prediction errors. ANN-GA optimization indicated 97.9% methane production efficiency with a reduced conversion time of 15.9 h under conditions of 1.5 bar pressure and trace metal concentrations of 25.0 mg/L Fe(II), 0.20 mg/L Ni(II), and 0.02 mg/L Co(II). Validation experiments confirmed these predictions with deviations below 5%, underscoring the robustness of the models. The results highlight the synergistic effects of pressure and trace metals in enhancing gas–liquid mass transfer and enzymatic pathways, demonstrating the potential of computational modeling and experimental validation to optimize biological methanation systems, contributing to sustainable methane production. Full article