This paper proposes a symmetrical sine-carrier pulse width modulation (SSCPWM)-based space vector modulation (SVM) strategy for enhancing the performance of a three-phase two-level Wye rectifier. Unlike conventional triangular-carrier PWM (TCPWM) and inverted sine-carrier PWM (ISCPWM), the proposed method employs a symmetrical sinusoidal carrier waveform to reshape the pulse distribution while preserving the simplicity of comparator-based carrier PWM implementation. The modulation signals are generated from a current-sector-based SVM framework. These signals are then directly compared with the proposed carrier to generate the switching signals. Analytical expressions for the switching instants and duty ratios are derived to clarify the nonlinear carrier-crossing characteristics introduced by the sinusoidal carrier. The proposed method was evaluated under both open-loop and voltage-oriented control (VOC) operations using detailed MATLAB/Simulink simulations. The results demonstrated that the proposed SSCPWM achieved improved harmonic performance relative to TCPWM and ISCPWM schemes. The proposed method also reduced the DC-link voltage ripple. Comparable switching-loss characteristics were still preserved. Under VOC operation, the proposed method achieved a phase-current THD of 3.21%. This value corresponded to 16.0% and 16.8% reductions compared with TCPWM and ISCPWM, respectively. The DC-link voltage regulation performance remained comparable to that of the conventional methods. The proposed strategy exhibits improved transient current quality during load variations without increasing the switching frequency or modifying the conventional VOC structure. The results indicated that the proposed SSCPWM can provide a practically feasible solution for improving the power quality of a grid-connected two-level Wye rectifier.
Conventional protection systems for induction motors often fail to detect inter-turn short circuit (ITSC) faults accurately. These faults are usually misclassified as overload or phase imbalance, which may lead to unnecessary tripping and downtime. Early detection of ITSC faults is important, as fault severity increases rapidly due to insulation degradation. This paper presents a diagnostic method based on wavelet transform and artificial neural networks (WTANN). The method uses stator current signals for fault detection without requiring additional sensors. The extracted features are used to classify motor conditions as ITSC, overload (OL), or short circuit fault (SCF). Experimental results on a 1 Hp, 415 V induction motor show that the proposed method achieves 97.4% testing accuracy, compared to 24.44% for the FFT-based method. The method performs well even under noisy conditions (20 dB and 30 dB). The results show that the proposed approach is accurate and suitable for practical applications.
In recent decades, there has been a substantial and dramatic implementation of renewable energy sources globally. Electrical systems should fulfill consumer load requirements while transmitting electricity with reduced loss, increased power quality, and dependability. However, the accessibility of steady and dependable electricity in emerging nations raises concerns about the depletion of energy production and the detrimental effects it has on the ecosystem. The most effective real-world and operational method to meet consumers' growing electricity requirements while upholding uncompromising ecological standards for power generation is to integrate a wind-solar-based microgrid into the distribution system. The microgrid system and a distribution network face severe issues, difficulties, and challenges. The fluctuation of hybrid renewable energies, namely wind and PV resources, causes issues including the occurrence of voltage dips and surges, voltage imbalances, and energy losses. The core objective of this research is to combine a microgrid network with a DSTATCOM controller to guarantee increased power flow and improve the voltage profile. The DSTATCOM is controlled using fuzzy logic controllers (FLC) and proportional integrals. By interconnecting to the primary distribution system, the effectiveness of DSTATCOM is evaluated with microgrid integration. Eventually, it is shown that DSTATCOM increases the distribution line’s capability in the system.
Economic emission dispatch (EED) is an important optimization problem in today's power systems with significant renewable energy integration. This work presents a hybrid particle swarm optimization and grey wolf optimization (PSO-GWO) approach for the multi-objective economic emission dispatch (EED) problem with solar and wind integration. The hybrid algorithm improves the exploration-exploitation trade-off by incorporating the social learning behaviour of particle swarm optimization and the hierarchical hunting behavior of grey wolf optimization. The uncertainty of renewable energy is modeled using probability distributions to enhance dispatch reliability. A weighted multi-objective approach is adopted to optimize fuel cost and emissions. The proposed approach is tested on a benchmark 10-unit and 6 IEEE generating units interconnected power system. The proposed method exhibits better convergence and performs better than the traditional PSO, genetic algorithm and other benchmark methods. The proposed approach reduces cost by 8.3% and emissions by 12.6% and is suitable for sustainable and efficient power system operation.
The analysis of soils and proper prediction of crops are very important to help grow more productive agriculture and sustainable food production. A Hybrid CNN–Machine Learning (CNN-ML) Framework for Soil Classification and Crop Prediction is proposed in this paper to combine deep learning and machine learning approaches for intelligent farming decision-making. The proposed framework uses Convolutional Neural Networks (CNNs) for automatic soil classification and machine learning algorithms for crop recommendation. There are five types of soil which are described by their image, such as: Black Soil, Cinder Soil, Laterite Soil, Peat Soil and Yellow Soil. CNN, MobileNetV2 and ResNet50 were implemented and compared to assess the effectiveness of different deep learning architectures. The CNN model was found to be the most effective soil classification model with 98.71% accuracy, which is better than MobileNetV2 and ResNet50, as it can automatically extract discriminative texture, color features from soil images for classification. After soil classification, soil-specific nutrient characteristics (Nitrogen (N), Phosphorus (P), Potassium (K), pH, temperature, humidity and ranges of rainfall) were gathered from literature. A dataset of crops was then downloaded from Kaggle and categorised into five soil types according to these nutrient profiles found in literature. For soil-specific crop prediction, several machine learning algorithms such as Logistic Regression, K-Nearest Neighbors (KNN), Random Forest and XGBoost were evaluated. Each soil category was analyzed independently with Logistic Regression, K-Nearest Neighbors (KNN), Random Forest and XGBoost algorithms for crop prediction. The final performance was calculated by taking mean of the results across the five soil datasets. Experimental results demonstrated that among the different machine learning models, XGBoost model outperforms others with the highest average accuracy of 88.0% in crop prediction. Experimental results showed that XGBoost model has the highest accuracy of 88.0% in crop prediction with the best predictive capability and generalization performance among the other machine learning models.
Mobile Ad Hoc Networks are highly dynamic and infrastructure-less environments that are inherently vulnerable to a range of routing-based attacks. To address the limitations of traditional defense mechanisms, this paper proposes STAGNet-COpt, a novel spatio-temporal trust-aware defense framework that integrates deep learning-based attack detection, fuzzy clustering, and hybrid meta-heuristic routing optimization. The detection component, STAGNet, leverages Graph Attention Networks in combination with Long Short-Term Memory networks to capture both the topological dependencies and temporal behavior of nodes for accurate intrusion detection. To enhance routing security and efficiency, TrustFuzz employs fuzzy logic for trust-aware clustering, while WOACO-R, a hybrid of Whale Optimization Algorithm and Ant Colony Optimization, is utilized for adaptive and trustworthy route selection. The framework is evaluated using NS-2 simulations under diverse attack scenarios, including blackhole, wormhole, grayhole, denial-of-service, and integrated attacks, across varying node densities. Results show that STAGNet-COpt achieves an average Packet Delivery Ratio of 94.05%, packet loss of 5.45%, throughput of 119.5 kbps, end-to-end delay of 3.63 ms, and routing overhead of 325.25, significantly outperforming existing benchmarks. The proposed model demonstrates high scalability, detection accuracy, and resilience, establishing a robust and intelligent solution for secure MANET communication.
The emergence of wireless communication networks (WCNs) introduces new opportunities for efficient spectrum utilization through wide scanning of the network. Leveraging Software Defined Radios (SDRs), users can conduct wide spectrum sensing and adjust transmission properties dynamically. The concept of opportunistic use of available spectrum requires adaptable antennas with ultra-wide band scanning capabilities. Dynamic Spectrum Access (DSA) emerges as a promising solution to congestion within densely populated networks. In this study, we introduce an innovative compact antenna specifically crafted for wide spectrum sensing. Circular polarization property enhances the antenna’s ability for orientation-free spectrum scanning, facilitated by corner truncation. The antenna design incorporates novel T-fractal slots and electromagnetic band gap structures on a small FR4 substrate measuring 12x18x1.6 mm2. The aim is to identify unused channel opportunities within radio frequency sensor networks (RFSN) through wideband scanning. The impact of insertion of band gap and fractal slots provide a matching of wide range of impedance bandwidth. With an average gain of 2.5 dBi across its operational frequency range, the antenna meets FCC standards, affirming its suitability for spectrum sensing in RFSN. Experimental findings reveal a wide bandwidth from a range of 0.79 to 12.0 GHz in terms of return loss characteristics relative to the feed port.
This paper presents a powerful hybrid localization model of drones to be used in GPS-denied areas through the combination of Long Short-Term Memory (LSTM) networks with an Extended Kalman Filter (EKF). The system network is based on a Wireless Sensor Network (WSN) in which sensors are arranged in equilateral triangle triples to enable the combination of the Angle-of-Arrival (AoA) and Time-Difference-of-Arrival (TDoA) measurements. The traditional EKF only model may not be good with complex residual dynamics as well as motion uncertainties but the LSTM component is specifically used to model these nonlinearities which gives much better state estimation compared to what conventional kinematic models can offer. The work presents a strict theoretical framework of triangular anchor geometry using Dilution of Precision (DOP) analysis and applying the nearest triple geometry to approximate localization using triangulation techniques when a drone goes into sensor field. Wide Monte Carlo simulation and statistical analysis with confidence intervals prove that the framework is very precise, that is, it can attain a distance error of less than 0.48 meters. Moreover, the suggested approach has shown a 42 percent reduction in Root Mean Square Error (RMSE) as compared to conventional EKF benchmarks. The contribution of each component is verified in detailed ablation studies, which prove that the system has the real-time computational performance required to deploy a drone in practice.
The Doubly Fed Induction Generator (DFIG) utilized in wind energy conversion systems (WECS) requires regulation of both active and reactive power to ensure stability and proper functioning. Model Reference Adaptive Control (MRAC) scheme augmented with flux-oriented vector control strategy is proposed for ensuring effective power management without extra sensors. RSC (machine side converter) and GSC (grid side converter) are to be controlled using Lyapunov-based adaptive control for improved power extraction and stability of the grid. Proposed model was simulated in MATLAB/Simulink and tested under dynamic conditions of wind, including gust and ramp variation. Results show that both GSC and RSC axis current components tracks the references. MRAC-based controller has maintained the dc link stability with less than 5% overshoot. The simulation verification indicated an excellent achievement of decoupled control of power. This demonstrates MRAC's robustness for power regulation in DFIG-based WECS, which enhances reliability and stability in energy generation.
The adoption of electric vehicles is increasing rapidly; this EV charging is most uncertain thing. The distribution system losses are high due to uncertain usage of power supply issues and it increases due to EV charging stations integration in the distribution side. This paper proposes the Secretary Bird Optimization Algorithm based optimal OLTC tap positions to minimize the power losses and improvement of voltage profile in the distribution system. In general, constant power loading is considered in distribution system but here in this proposed approach voltage dependent load modelling is adapted and integrated the various DG systems (both PV and wind), capacitor placement, to compensate the power losses before going to apply the OLTC as much as possible to reduce the power loss. The proposed approach is tested on IEEE 15 and IEEE 33 bus standard system and evaluated the proposed Secretary Bird Optimization Algorithm. Compared the proposed approach with existing algorithms like standard PSO algorithm, as well as compared with modern algorithms like PBO, GWO-MLP algorithms, Secretary Bird Optimization Algorithm outperformed and exhibits better loss reduction and improvement in voltage profile.
With the rise of digital content creation and sharing, protecting multimedia assets from illegal use, decoding, and piracy, without a doubt, is becoming more important as time passes. As such, this work proposes a novel digital image watermarking framework that combines Convolutional Neural Networks (CNNS), Discrete Wavelet Transform (DWT), and a dual chaotic encryption technique based on Logistic Map and Tent Map fusion. An adaptive encryption and watermark embedding in the frequency domain of the host image ensures imperceptibility, robustness, and security. Feature statistics are extracted from a normalized watermark using a two-layer CNN, dynamically creating initial conditions for chaotic maps. The high-entropy resultant mask is fused with the generated sequences and used to encrypt the watermark with modular arithmetic. The encrypted watermark is embedded in the LL sub-band of a DWT-transformed host image using alpha blending. The final watermarked image is then reconstructed using inverse DWT. CNN features are used to regenerate identical chaotic sequences to decrypt and retrieve the watermark during extraction. In order to validate the approach, it was tested on images from the COCO (Common Objects in Context) and ImageNet datasets. Initially, average PSNR (Peak Signal-to-Noise Ratio) values were larger than 41 dB, and SSIM (Structural Similarity Index) values were over 0.97, thus having good visual fidelity. The system was found to be resilient to typical attacks (Gaussian noise, cropping, JPEG compression, rotation), with PSNR values between 30.1dB and 37.8 dB and SSIM over 0.94. As such, the PSNR of “Sports Car” remained 37.74 dB in the case of noise and 37.81 dB under cropping for the noise “Laptop”. COCO images like “Dog” and “Person” demonstrated PSNR above 35 dB for most distortions. Chaos parameters were adaptively generated from CNN features using a dual chaotic map fusion, enhancing security and embedding them into the DWT domain to improve robustness. This integrated approach establishes a secure intelligent watermarking framework suitable for real-world applications such as copyright protection, secure transmission of medical images, and forensics.
In this article, a single-phase modified cascaded H-Bridge eleven-level inverter that produces a precise sinusoidal output voltage while significantly reducing THD. The proposed topology enables a reduction in the number of switches relative to the traditional cascaded H-bridge inverters but with a THD of about 3. The designed inverter construction consists of four bidirectional switches, including a conservative H-Bridge that allows adequate voltage control and enhances overall structure performance. The primary goal of this effort is to decrease the number of devices without sacrificing power quality and to provide an affordable solution that is suitably tailored for various applications that demand high-quality AC voltage. By minimizing harmonics and guaranteeing that the production voltage waveform intentionally closes a sinusoidal waveform, a suitable switching policy contributes to the development of the system's performance. The integrity of the system’s design produces flexibility, making it efficiently extendable to deliver higher output voltage levels should ultimate demands desire it. The designed system is modeled and simulated using the MATLAB/SIMULINK and PROTEUS domains to demonstrate its feasibility. The findings of the experiment indicate that they are very close to simulation, confirming the pragmatic practicability of the system. Moreover, experimental outcomes further validate the successful process of the inverter, demonstrating valuable performance in terms of output quality and system performance.
The improvement of power quality (PQ) is the main aim of this paper. Here, the enhancement of PQ through harmonic filters (HFs) is simulated on the IEEE-69 bus radial distribution system (RDS) with nonlinear distributed generation (NLDG). The aim of minimizing harmonic distortion within standard limits is achieved using proper placement of HFs and an optimization algorithm. The optimization problem in this study is characterized by nonlinear constraints. The placement of HFs is accomplished using a newly published method. It is a nonlinear load position-based current injection (NLPCI). To determine an appropriate rating of HFs for reducing total harmonic distortion of voltage (THDv) and meeting the standards set by the IEEE, a novel bat algorithm (NBA) is employed. The NBA is compared with the Bat Algorithm (BA), Particle Swarm Optimization (PSO), Gray Wolf optimization (GWO) and the Firefly Algorithm (FA) in terms of efficacy. The comparative analysis of results shows that in terms of computational efficiency, the NBA performs better than the BA, PSO, GWO and FA.
This study aims to evaluate the capability of the Random Forest model to predict the flashover voltage of polluted insulators, with particular emphasis on the effect of hyperparameter tuning strategies on model accuracy and stability. A two-stage methodology was adopted. In the first stage, Grid Search and Particle Swarm Optimization were compared for tuning the model hyperparameters using a published dataset of cap-and-pin insulators. The results showed close agreement between the two methods in terms of the mean root mean square error, with a slight accuracy advantage for Particle Swarm Optimization, whereas Grid Search provided higher stability and greater computational simplicity. Accordingly, the Grid Search–tuned Random Forest model was adopted in the second stage, where its performance was evaluated after merging the published data with local laboratory measurements obtained at the High Voltage Laboratory of Damascus University, Syria. The model demonstrated high predictive performance under 70/30 and 80/20 training/testing splits. In addition, ten-fold cross-validation confirmed the stability of the model performance across different data partitions. The feature-importance analysis revealed that surface conductivity was the most influential factor affecting flashover voltage, followed by the geometrical characteristics of the insulator. The results confirm that the Grid Search–tuned Random Forest model provides an effective and initially generalizable tool for predicting the flashover voltage of polluted insulators. However, further expansion of the laboratory database is recommended to improve the reliability of practical applications.
Magnetic Resonance Imaging (MRI) is the cornerstone of modern medical diagnosis and research, providing high spatial resolution visualization of the anatomical and functional information of the body’s internal structure in a non-ionizing, non-carcinogenic and non-invasive manner. Despite these superior properties, the MRI data acquisition process is inherently slow, limiting this technique in scenarios where time is crucial. Consequently, accelerating MRI through k-space under-sampling at sub-Nyquist rates and reconstructing high-quality images from incomplete measurements has emerged as an active area of research in the past few decades. This paper proposes a learned three-operator splitting algorithm implemented in an unrolled complex-valued deep neural network architecture for accelerated Magnetic Resonance (MR) image reconstruction. Objective results evaluated using Peak Signal–to–Noise Ratio (PSNR), Structural Similarity Index (SSIM) and Normalized Root–Mean–Squared Error (NRMSE) at ×2 to ×5 acceleration show superior reconstruction performance of the proposed network compared to other state–of–the–art learned algorithms at comparable reconstruction time. Subjective results show that the proposed network reconstructed images visually similar to ground truth images. The proposed network has the potential to enable real-time MRI applications with high-quality images.
Filtered orthogonal frequency division multiplexing (F-OFDM) has emerged as a promising waveform candidate for 5G systems due to its enhanced spectral containment and flexibility. However, conventional F-OFDM implementations rely on single-domain windowing or filtering, which limits the achievable trade-off between spectral efficiency and error-rate performance. This paper proposes a unified hybrid-domain Kaiser windowing framework that jointly applies timedomain and frequency-domain shaping within a single analytical formulation. A weighted hybrid shaping parameter is derived to balance time localization and spectral confinement in a transparent and reproducible manner. The proposed framework is evaluated in terms of power spectral density (PSD), adjacent channel leakage ratio (ACLR), and bit error rate (BER) under AWGN, Rician fading (K = 6 dB, fd = 200 Hz), and Rayleigh fading (fd = 200 Hz) channels. Results demonstrate that the hybrid-domain approach achieves superior out-of-band emission suppression and improved ACLR compared to conventional time-domain and frequency-domain Kaiser windowing, while maintaining robust BER performance under realistic fading and mobility conditions. The findings confirm that joint time–frequency windowing provides a practical and effective solution for meeting 5G spectral emission requirements without introducing excessive computational complexity.
This paper presents a robust control design method for Differential Drive Mobile Robots (DDMR) to address the trajectory tracking problem under conditions of disturbances and uncertainties, specifically focusing on terrain-induced variations. The proposed controller utilizes Sliding Mode Control (SMC) to ensure robustness and fast response. To mitigate the inherent chattering phenomenon of SMC and enhance tracking accuracy, two disturbance observers are designed and integrated into the system: A Nonlinear Disturbance Observer (NDO) and a Neural Network Observer (NNO) using Radial Basis Functions (RBF). These observers are tasked with estimating aggregate disturbances, including friction, model uncertainties, and terrain effects, thereby effectively compensating the control signal. This approach allows for a reduction in the switching gain of the SMC, resulting in improved control quality. Simulation results demonstrate that the SMC-NDO-NNO controller significantly improves control performance even in the presence of uncertain disturbances and substantially reduces chattering.
This work introduces a novel dual-band circular polarized (CP) microstrip patch antenna for wireless communications. The antenna features a square patch with corner truncated and an etched rectangular slot for circular polarization. The geometric modifications are mainly to control resonating frequencies, bandwidth and to optimize the axial ratio (AR) at the resonating frequencies. The proposed dual-band CP metamaterial antenna is designed to function within the wireless frequency ranges i.e., 2.28-2.48 GHz and 4.48-4.64 GHz with a measured gains of 2.62 dBi and 2.8 dBi at 2.4 as well as 4.5 GHz respectively. The CP is achieved using truncated corners at opposite sides and a slot in the square patch to improve the AR bandwidth. The dimensions of the proposed dual band antenna are 50 × 50 × 1.6 mm³.The antenna is realized using a low-cost FR-4 laminate with a εr of 4.4 and a tanδ of 0.02. The experimental performance of modeled antenna closely matches with the simulation results and confirms the effectiveness of the design methodology. Comparative analysis further indicates that the proposed antenna achieves higher gain and occupies less physical area relative to similar designs reported in recent literature
This manuscript provides an analysis of a wind-driven self-excited reluctance generator (WDSERG) performance running under variable load conditions while maintaining a regular output voltage, and designs an Artificial Neural Network (ANN) model to forecast the value of the excitation capacitance required to maintain a WDSERG's generated voltage within desired bounds. The self-excited reluctance generator (SERG) has advantages over the induction generator (IG), which include steady frequency regardless of load or capacitance variation with proper performance. The analysis of steady-state for the reluctance generator (RG) is conducted according to d-q axes transformation. The suitable capacitance is determined for varying operating conditions to meet the main requirements for a loaded RG at constant voltage. To provide guidelines for designers, the variations of the capacitance corresponding to any change in the load impedance, power factor and prime mover speed are determined. The predicted excitation capacitance values that required to keep the generated voltage at a preferred constant level of 1± 0.1 pu under inputs condition. For example, at conditions (ZL (load impedance) =3pu (per unit), PF (power factor) =1, and constant speed) the capacitance 16.5275 µF (micro farad) make voltage constant at 1pu. It investigates how changing machine parameters affects the generator's performance and presents the speed above which excitation is not possible. Simulation results are supported through MATLAB coding analysis. This paper demonstrates the feasibility of steady voltage operation under in a variable load for SERG, presenting insights for practical implementation in standalone wind energy systems.
Acute Myocardial Infarction (AMI) is a vital public health concern, because it is the primary factor of death globally. Therefore, timely identification is crucial, especially in resource-constrained situations without centralized testing. (1) Background: Assessment, risk assessment, and treatment all depend on electrocardiograms (ECGs). ECG segments are artificially corrupted with various noise types (e.g., Gaussian noise, baseline wander) to create noisy training data.; (2) Methods: In this paper, signal denoising with an Optimized Variational Stacked Autoencoder (OVSAE) model which involves training a Neural Network (NN) to reconstruct clean ECGs from noisy versions, effectively learning to separate signal from noises and then decompressing the noise removed signals. OVSAE is introduced to adaptively remove noisy signals from ECG signals. VSAE learns hierarchical latent representations of clean signal structures through multiple nonlinear encoding layers. To get improved parameter initialization and optimum results, layer-wise prior training and modification are implemented. To enhance training stability and reconstruction accuracy, a layer-wise greedy pre-training strategy is adopted, followed by global fine-tuning of the entire network. Self-Attention Long Short-Term Memory (SALSTM) classifier is designed for AMI detection. To adaptively weight the significance of various temporal aspects and numerous ECG signals, the LSTM model incorporates a self-attention mechanism. A gating mechanism is added by LSTM, a gated network, to regulate the NN's information transfer; (3) Results: Clinical Parameters in Risk Stratification, and PTB-XL ECG diagnostic dataset includes of 18885 patients' 10-second clinical 12-lead ECGs, that of 21837.Furthermore, the results are quantified using measures like Mean Square Error (MSE), Mean Absolute Error (MAE), Structured Similarity Index (SSIM), and Peak Signal to Noise Ratio (PSNR). AMI results have been evaluated using metrics such as ERRor (ERR), ACCuracy (ACC), SPEcificity (SPE), and SENsitivity (SEN) against k-fold cross-validation.
