Gongcheng Kexue Yu Jishu/Advanced Engineering Science

In Wireless Sensor Networks (WSNs), efficient and reliable data transmission is essential for improving network longevity and efficiency. This paper proposes a Cluster Based Fuzzy Decision-Making approach for traffic-free(FDM-FT) data transmission in WSNs. The system employs intelligent clustering to organize nodes into energy-efficient groups and utilizes fuzzy logic to dynamically evaluate factors like remaining energy and proximity to the base station and node congestion level. Based on these inputs, optimal cluster heads and routing paths are selected to minimize traffic congestion and energy consumption. Simulation results demonstrate improved packet delivery ratio, reduced latency, and enhanced network longevity compared to traditional routing protocols.in This model Cluster Based Fuzzy Decision-Making approach for traffic-free(FDM-FT)offers a scalable and adaptive solution for congestion-aware and energy-efficient communication in resource-constrained sensor networks The FDM-FT mechanism aims to reduce data loss, energy loss, and increase throughput. It is primarily focused on avoiding network traffic interruptions and ensuring continuous data flow source from the node (cluster head), to the base station. The performance of this mechanism is evaluated using the Network Simulator (NS2) and demonstrates impressive levels for the indices, indicating its suitability for real-world applications.

Power transformers need to be continuously monitored in order to identify emerging issues before they become catastrophic failures. In order to assess transformer condition in real time, this paper proposes a fuzzy logic-based transformer health monitoring system that combines several sensor inputs (voltage, current, oil temperature, oil level, etc.). A Mamdani fuzzy inference engine is implemented by an ATmega328 microcontroller, which also uses GSM to notify maintenance staff of faults. To tune the fuzzy logic controller and simulate different fault scenarios (overloads, over-temperature, undervoltage, etc.), a comprehensive MATLAB/Simulink model was created. According to simulation results, the fuzzy system can identify the transformer's health status as Normal, Alert, or Critical and initiate the proper control measures (such as load reduction or emergency trip) well before serious faults occur. The behavior of a hardware prototype that was constructed and tested on a laboratory transformer closely matched the predictions of the simulation. When handling ambiguous sensor data and integrating several fault indicators into a single health index, the fuzzy logic approach demonstrates strong capabilities. In general, this work offers a thorough framework for fuzzy logic-based intelligent transformer condition monitoring, enhancing the sensitivity of fault detection and operational dependability of transformers.

The increasing demand for energy-efficient and high-performance digital systems has made the design of low-power arithmetic circuits a critical area of research. Full adders, being core components of arithmetic logic units (ALUs), play a vital role in determining the overall efficiency of digital architectures. This paper presents a novel Low-Power Hybrid 4-bit Full Adder design utilizing 14nm FinFET technology to address the challenges of power consumption and speed in modern computing applications. The proposed architecture combines Gate Diffusion Input (GDI) logic with Energy-Efficient Diode-Connected DC Biased Positive Feedback Adiabatic Logic (EE-DC-DB PFAL), leveraging both logic minimization and adiabatic switching to enhance power performance. FinFET devices are employed to exploit their superior electrostatic control and reduced leakage characteristics at the nanoscale. Comprehensive simulations conducted using industry-standard tools demonstrate significant improvements in power dissipation, delay, and energy efficiency compared to traditional CMOS-based full adder designs. The results validate the effectiveness of the proposed hybrid approach, making it a promising candidate for integration into low-power digital processing units and future ultra-scaled VLSI systems.

A double-sided linear induction motor (DSLIM) optimised for high-thrust and high-efficiency applications in automation and transportation is designed, analysed, and simulated in this work. With its symmetrical double-sided design, the DSLIM offers balanced magnetic flux and increased thrust density, which makes it appropriate for automated material handling and high-speed urban transit systems like Maglev trains. Finite Element Analysis (FEA) is used to analyse performance and calculate important parameters such as thrust force, synchronous speed, and slip. According to simulation results, the DSLIM design is effective. It produces a lot of thrust, is around 70% efficient at the ideal slip levels, and significantly reduces end effects, which are major problems in LIM applications. The design's precision and dependability are confirmed by the high degree of agreement between theoretical calculations and simulation results.

Concentrated Photovoltaic (CPV) systems are gaining interest due to their excessive overall performance and capacity for sustainable sun electricity era. However, their performance is substantially laid low with actual-global running variability, collectively with fluctuations in sun irradiance, temperature versions, and optical misalignment. This examine develops a MATLAB and Simulink-based completely simulation framework to assess CPV tool normal overall performance underneath dynamic environmental conditions. Key overall performance indicators which encompass electricity output, performance, and thermal stability are analyzed to perceive barriers and areas for optimization. Advanced control techniques, which include advanced Maximum Power Point Tracking (MPPT) algorithms and thermal law techniques, are implemented to beautify gadget reliability. The simulation results advocate an performance development of about X percentage, with optimized manage mechanisms mitigating actual-time fluctuations efficiently. Findings endorse that adaptive optimization techniques play a critical function in retaining CPV typical overall performance stability under numerous situations. This research offers valuable insights for the layout and operation of CPV structures, making sure higher strength yield and prolonged-time period overall performance. The proposed optimization strategies contribute to the advancement of CPV era, promoting its integration into large-scale renewable electricity infrastructures.

In this paper characterizes the Ɱ* quasi paranormal composition operator and Ɱ* quasi paranormal weighted composition operator on L^2 spaces and investigates their various properties.

Post-harvest diseases in mangoes significantly impact their quality, market value, and supply chain efficiency. Identifying and tracking defective mangoes caused by these diseases is essential for reducing losses and ensuring food safety. This study presents the development of a database system designed to record, monitor, and analyse defective mangoes affected by post-harvest diseases such as anthracnose, stem-end rot, and soft rot. The database incorporates key features, including disease classification, severity levels, visual symptoms, geographical origins, storage conditions, and time since harvest. Advanced query functionalities allow stakeholderssuch as farmers, distributors, and researchersto access actionable insights and trends, enabling better decision-making in disease management, transportation, and storage protocols. By integrating data visualization and predictive analytics, the database facilitates early detection of disease patterns, contributing to improved post-harvest handling practices. This initiative aims to enhance the efficiency of mango supply chains, minimize waste, and ensure the delivery of high-quality mangoes to consumers while promoting sustainable agricultural practices.

The demand for sustainable and environmentally friendly printing solutions has prompted a shift from conventional petroleum-based inks to alternative options such as soya-based and biodegradable inks. This research focuses on the comparative analysis of colour difference (Delta E) resulting from the use of conventional, soya-based, and biodegradable inks on paperboard substrates through the offset printing process. As environmental concerns grow, the printing industry is increasingly exploring sustainable alternatives to traditional petroleum-based inks. Soya-based and biodegradable inks offer potential benefits in terms of eco-friendliness, but their performance in terms of colour accuracy and consistency requires systematic evaluation. In this study, standard colour patches are printed using each ink type under identical offset printing conditions. Delta E (∆E), a widely accepted metric for measuring perceptible colour variation, is used to assess the colour differences among the samples. The research methodology involves spectrophotometric measurement of printed samples and calculation of ∆E values using standardized formulas. The study aims to provide insights into the suitability of alternative inks for high-quality printing applications, with a particular focus on colour fidelity, process compatibility, and sustainability.

Over the past few years, the quick development in Artificial Intelligence (AI) has devised novel techniques to manipulate multimedia. The misuse of a face swap approach named deepfake has created various cybercrimes like the spreading of fake news, identity theft, and financial crime. One promising countermeasure in opposition to deepfakes is termed as deepfake detection. Still, the detection of deepfake is complex due to the larger dataset. To resolve such issues, this paper develops the Fractional Secretary Bird Skill Optimization Algorithm-enabled Pyramid Deep Belief Network (FSBSOA-PyramidFDBNet)-based multi-face deepfake detection using Federated Learning (FL). The nodes and servers are the major parts of FL. In the training model, the video frames are subjected to face detection. The facial action units are detected with the utilization of Action Unit Network (AUNet). The feature extraction extracts the required features and the deep fake detection is done using the proposed FSBSOA-PyramidFDBNet. The updated weight from the local nodes is aggregated at the server. In addition, the FSBSOA-PyramidFDBNet-based multi face deepfake detection attained the optimal accuracy, loss function, Mean Square Error (MSE), True Negative Rate (TNR), and True Positive Rate (TPR) of 93.91%, 0.064, 0.179, 94.16%, and 92.35%.

This paper presents a detailed study on lung image segmentation with special focus on the proposed ResUNet++ framework. Automated segmentation plays a key role in computer-aided diagnosis, as manual methods are often slow, subjective, and inconsistent. Traditional techniques struggle with low-contrast or irregular lung regions, whereas deep learning models such as U-Net, UNet++, ResUNet, Attention U-Net, PSPNet, DeepLabV3+, HRNet, and transformer-based networks have brought major improvements. Among these, our experiments show that ResUNet++ offers the best balance of accuracy, stability, and efficiency. It achieved the highest Dice, Precision, Recall, and IoU scores across datasets and showed strong adaptability to different domains. The ablation studies confirm that modules like squeeze-and-excitation and atrous spatial pyramid pooling play a vital role in handling complex pathologies. Visual results further support its ability to generate smooth and anatomically correct lung boundaries. Overall, ResUNet++ advances the state of the art in lung image segmentation and provides a reliable base for clinical tasks such as disease detection, lesion measurement, and treatment monitoring.

Ensuring access to safe water is vital for public health and environmental sustainability. However, conventional water quality monitoring methods are limited by high costs, lack of scalability, and delayed analysis. This survey explores recent innovations in IoT-based Water Quality Monitoring Systems (IoT-WQMS) integrated with Machine Learning (ML) and Deep Learning (DL) to enable real-time, automated water quality assessment. The review covers architectures utilizing multi-parameter sensors (e.g., pH, turbidity, TDS, DO, temperature), along with essential data processing techniques such as imputation and normalization. Advanced feature selection methods (RF-MOA, ensemble voting) and hyperparameter tuning techniques (QPSO, Grid Search) are discussed for model optimization. ML models like XGBoost, Random Forest, and ANN, as well as DL models such as CNN-LSTM and MS-CAGRU, demonstrate predictive accuracies up to 99.9%, supporting early contamination detection and regulatory compliance. Applications span urban rivers, aquaculture, and groundwater systems, offering actionable insights for efficient and sustainable water management. The paper also addresses key challenges including sensor calibration, data heterogeneity, and model adaptability, highlighting the role of hybrid AI and Explainable AI (XAI) in enhancing system robustness and transparency. This review provides a comprehensive perspective to guide future research and deployment of intelligent water monitoring solutions.

More and more technologies are being utilized to help modern agriculture be more productive and sustainable through soil health management with precision monitoring in real-time. Soil health indicators, including soil nutrients such as N, P, and K, and related properties like pH, electrical conductivity (EC), moisture, and water quality, can all be critical to crop yield, so their accurate measurements are crucial. Although traditional, lab-based soil tests can provide valuable information, sensor-enabled systems offer real-time and field-level monitoring, which means tests can happen much more quickly and at more timely intervals. This study will assess how fertilizer applications affect soil properties that are important to soil health in three different agro-ecological zones in Maharashtra: Alandi (Pune Division), Akola (Vidarbha Division), and Chhatrapati Sambhajinagar(Marathwada Division). In addition, this study incorporates systematic soil sampling and enhanced soil health indicators using IoT-enabled sensor technology to measure nutrient movement through time and across agro-ecological conditions. Finally, the study will also assess relationships between soil properties and crop response to provide an understanding of fertilizer practices. In general, outcomes and results will be used to formulate and give region-specific management recommendations and demonstrate how digital protocols can contribute to precision agriculture and long-term soil health sustainability.

Parkinson's disease (PD) is a progressive neurodegenerative disorder that primarily affects motor function, leading to tremors, stiffness, and difficulty with movement and coordination. Early detection of PD is critical for timely intervention and improved patient outcomes. This study introduces a novel method for PD detection through the analysis of activity-related data using a Transformer-based model. The Transformer architecture, renowned for its ability to handle sequential data, leverages self-attention mechanisms to capture complex temporal dependencies and patterns. Unlike traditional models such as LSTM (Long Short-Term Memory), which processes data sequentially, Transformers enable parallel processing, enhancing computational efficiency and boosting prediction accuracy. This approach bypasses the need for image-based analysis and instead focuses on extracting meaningful insights from activity-related data, making it a more accessible and cost-effective solution for early PD detection. The proposed method not only enhances prediction accuracy but also provides a non-invasive and user-friendly tool for PD screening. Preliminary results demonstrate that the proposed hybrid framework achieves a classification accuracy of 91.5%, outperforming traditional models such as SVM and Random Forest. The findings emphasize the potential of deep learning in advancing neurodegenerative disease diagnostics. Future research will focus on refining the model and expanding the dataset to improve its robustness and generalizability across diverse populations.

This study aims to determine how different parameters affect the tensile strength of dissimilar materials joined using Friction Stir Welding (FSW), specifically the AA 8090 aluminum alloy and four polymers: PEEK, Nylon 6, HDPE, and PP. The predominant strengthening mechanism is mechanical interlocking, in which plasticized aluminum permeates the softened polymer phase. However, a crucial secondary mechanism, interfacial chemical bonding, is responsible for the stronger bonds seen in polar polymers (PEEK and Nylon 6), which form covalent and hydrogen bonds. This mechanism is absent in non-polar polymers like PP and HDPE, resulting in less efficient joints. The research found that controlling process parameters specifically tool rotation and traverse speed is crucial, as they directly influence the heat generation and material flow that determine the joint's tensile strength. The results indicate that each polymer requires a specific "processing window." This is necessary to balance the heat and material flow to improve the joint's integrity while minimizing defects like voids or excessive melting.

Edge-Aware Federated Learning (EAFL) has emerged as a promising paradigm to address the dual challenges of data privacy and computational limitations in resource-constrained environments. Traditional federated learning (FL) approaches often overlook the heterogeneous nature of edge devices, leading to suboptimal performance and increased communication overhead. This research introduces an Edge-Aware Federated Learning framework that dynamically adapts model training to account for device-specific capabilities, network conditions, and data distribution. By integrating edge-awareness into the aggregation and optimization processes, the proposed method enhances model accuracy while reducing latency and energy consumption. Furthermore, EAFL incorporates lightweight privacy-preserving techniques, such as differential privacy and secure aggregation, tailored to operate efficiently on low-power devices without compromising data confidentiality. Experimental evaluations on benchmark datasets across diverse edge scenarios demonstrate that EAFL achieves up to 15% improvement in model accuracy and a 30% reduction in communication costs compared to conventional FL methods. The results affirm the potential of edge-aware strategies in bridging the performance gap in federated systems, making FL more viable for real-world applications such as smart healthcare, IoT, and autonomous systems. This work lays a foundational framework for future research in intelligent, privacy-preserving learning at the edge.