Bilad Alrafidain Journal for Engineering Science and Technology (BAJEST)

In the event of an earthquake, pile foundations help to keep a structure stable. Batter piles, which are a sort of angled foundation, can resist both vertical and horizontal loads better than a typical vertical pile. The primary objective of this research is to address the seismic response of negatively battered piles in 2 × 2 pile groups embedded in loose sand through an experimental approach. The experimental study used 2 × 2 pile groups embedded in a loose sandy soil with 31.2% relative density, and seismic loading was replicated using a shaking table. Ground motion in the El Centro and Kobe earthquakes was applied to piles with batter angles of -5°, 0°, and +5°. The purpose was to assess and contrast the case of a negatively battered pile with vertical and positively battered piles, about the lateral and vertical displacements, and the acceleration. Experimental findings demonstrated that negative battering notably amplifies pile group displacements under seismic excitation. Specifically, for the El Centro earthquake, changing the batter angle from 0° to 5° increased the maximum lateral displacement by 24.287%, while adjusting it from 0° to +5° reduced it by 3.877%. Similarly, vertical displacement rose by 19.923% for the negatively battered piles, whereas positive battering resulted in a 7.511% reduction in lateral displacement for the same event.

The combination of renewable power with microgrids also requires intelligent control systems that get the most power being utilized and regulate the power flow in an even manner. We have developed a new paradigm that comprises high-performance DC generators coupled with artificial intelligence-based maximum power point tracking (AI-MPPT) for hybrid solar-wind microgrids. Our system possesses an embedded 36-kW solar photovoltaic panel array, a 10-kW wind turbine system, and 50 kWh battery storage with dynamic grid interaction ability. We developed a deep Q-network (DQN) energy management system to harvest maximum real-time power from renewable sources. It controls charge and discharge cycles of batteries and grid interaction according to time-of-use electricity rates. The AI-MPPT algorithm performs better than traditional algorithms in tracking optimal operating points under changing environmental parameters. It learns and dynamically adapts. Simulation efficiency for 24-hour operation cycles proves the AI-based system has a daily operating cost reduction of 67% compared to traditional rule-based operating practices. The system proves energy autonomy with successful peak load shaving at 57% and grid autonomy at 57.3%. Battery life was enhanced through intelligent scheduling of dynamic electricity prices. The model provides an expandable platform for residential and commercial microgrids. It shows real economic benefit while promoting the integration of renewable power and grid stability. Our approach provides meaningful enhancement in economic efficiency and system reliability for distributed energy consumption

This study examines the behavior and load-bearing capacity of under-reamed piles embedded in cohesive soil at different degrees of saturation (100%, 80%, 60%, and 40%). The major objective is to examine the impact of matric suction (MS), cohesion (Cu), and saturation ratio (Sr) on the axial response of both conventional and under-reamed foundations under unsaturated conditions. Three different pile configurations were tested in laboratory models: a single under-reamed pile (SURP), a double under-reamed pile (DURP), and a straight pile without under-reaming (NP). The ultimate bearing capacity and settlement behavior were evaluated using load-displacement curves and axial compression tests conducted inside a steel soil chamber. The results show that decreasing the degree of saturation significantly increases matric suction, thereby improving the bearing capacity of all pile configurations. For the straight pile (L/D = 22.5), the load capacity increased by 259%, 348%, and 48% at saturation rates of 80%, 60%, and 40%, respectively. The increases observed for the single under-reamed pile were 260%, 346%, and 64%, respectively. Under the same conditions, the double under-reamed pile showed increases of 210.5%, 238%, and 24%. When comparing the different types of piles, it was found that the single under-reamed pile achieved capacity improvements of 42-57% compared to the straight pile. On the other hand, the double under-reamed pile provided even more significant improvements, which ranged from 58% to 109% depending on the degree of saturation. The study highlights the strong interaction between saturation, suction, and pile geometry, offering useful insights for foundation design in partially saturated cohesive soils

Controlling construction project costs is a critical factor in achieving project success, particularly in complex construction environments. Previous studies have shown that weaknesses in supply chain management are among the primary causes of cost overruns. Accordingly, this study investigates the impact of supply chain attributes (SCA) on construction project costs by identifying key characteristics through a comprehensive review of prior literature and developing a framework to evaluate their influence on cost outcomes.


Data were collected from construction projects and analyzed using statistical techniques, including correlation analysis and Structural Equation Modeling (SEM). The results reveal a statistically significant negative correlation between project costs and SCA, indicating that improved supply chain performance contributes to cost reduction. The SEM results validate the proposed framework, with a chi-square value of χ² = 9.865 and a probability value of p = 0.275, which exceeds the acceptable threshold of 0.05 and confirms an adequate model fit.


Further findings demonstrate that supply chain functions, quality management, and human resource management (HRM) significantly enhance project cost efficiency. Among these factors, HRM plays a pivotal role in strengthening supply chain functions and improving overall performance, highlighting its strategic importance within construction organizations. By statistically modeling the relationships and pathways between SCA and project costs, this study provides a systematic approach to understanding how effective supply chain management can improve cost efficiency in construction projects

Electroencephalography (EEG) provides rich information and a representation of brain activity for intelligent
healthcare applications. This study proposes a framework for automated emotion recognition and epileptic
seizure classification using machine learning (ML) and deep learning (DL) techniques. The framework was
tested on three EEG datasets to cover areas of anomaly detection, multi-class emotion recognition, and binary
seizure prediction. EEG signals were first normalized and then segmented into fixed-length windows
depending on each participant's recording traits, thus maintaining subject-specific temporal patterns. Models’
evaluation was done with properly separated training and testing data to guarantee a trustworthy performance
assessment. For emotion recognition, the model based on Gated Recurrent Units (GRU) obtained 96%
accuracy on the test data, whereas ensemble learning with Random Forest got 98%, thereby proving its
excellent discriminative power on structured EEG features. Anomaly detection without supervision through
Histogram-Based Outlier Score (HBOS) was able to detect the abnormal single-channel EEG segments
accurately. In seizure classification, a convolutional neural network (CNN) trained on log, scaled time,
frequency spectrograms yielded 95. 75% accuracy with an AUC of 0. 996 on the test dataset, thus successfully
differentiating interictal and preictal states. The findings confirm that ML models provide robust and
computationally efficient performance on engineered EEG features, whereas DL models effectively capture
complex temporal and spectral representations across multiple EEG analysis tasks.

Road infrastructure in dry and semi-dry climates has become more susceptible to rapid deterioration because of climate pressures and changing pavement performance. This research proposes a data-based predictive model to help assess road conditions and inform maintenance strategies by combining long-term climatic variables (ex, effective rainfall, evapotranspiration, climatic water balance) with pavement performance data collected from vibration sensors and static structural assessments; both datasets are analyzed together using a Random Forest classification model. The results show that incorporating climate data into the analysis produces improvement in the estimated timeliness of construction activity and could promote proactive maintenance through adaptation to climate conditions. Methodologically, this finding is significant in that it creates a single machine-learning framework by combining several independent data sources (i.e., dynamic measure of sensing) into one value using a quantifiable climate index. Practically, this research provides decision-making support for maintenance timing, decreases uncertainty in deterioration estimates, and improves the sustainability of road networks in dry climates.

The rapid development of generative models has resulted in the widespread synthesis of realistic deepfake media, and thus raises fundamental threats to digital trust and media verification. The majority of current deepfake detection methods are based on supervised convolutional neural networks (CNNs) and work with global image presentations, which may easily have a weak generalization ability and lack interpretive power. This paper presents a different paradigm for the detection of deepfakes: to cast the problem as a sequential decision and adopt reinforcement learning to solve it. For that purpose, the paper proposes a patch-based Deep Q-Learning (DQN) approach that enables the agent to selectively explore local face regions and detect manipulation artifacts. Unlike the typical feedforward classifiers, this approach is capable of fine-grained spatial exploration as well as making the model interpretable: it highlights regions that are informative for the decision. This research is offered as a pilot study to examine feasibility and interpretability rather than sample size and benchmarking. Experiments were performed on balanced partial deepfake image datasets released for public use (2400 images), and each image was considered as a single RL episode. Experimental results on a practical deepfake dataset show that the proposed approach has good performance with an AUC of 0.92. Remarkably, despite processing only 18.2% pixels on average per image, the proposed approach achieves this level of performance, confirming the efficiency and forensic potential of the proposed active perception framework

To keep pace with rapid developments in infrastructure, an AI-based damage detection tool is crucial for ensuring safety. However, the rapid deterioration of civil infrastructure necessitates intelligent, autonomous structural health monitoring (SHM) systems capable of real-time damage detection, localization, and quantification. In this article, a novel method for smart structural damage identification is proposed. a novel hybrid architecture integrating electromechanical impedance ( ) sensing with multi-scale attention-based deep learning. Unlike conventional approaches reliant on hand-crafted features, our method employs a Squeeze-and-Excitation Residual Convolutional Network ( ) coupled with multi-head self-attention mechanisms for automatic feature extraction from raw PZT-EMI signals. The model was designed and trained using an effective approach for optimal feature learning from the raw PZT-EMI response signals. Further, Bayesian optimization has increased the network’s reliability. It substantially increases the network performance and knowledge acquisition process. To validate the proposed approach experimentally, PZT-EMI signals have been obtained from a beam with various structural conditions. The model demonstrates 100% classification accuracy across four damage severity levels (healthy, 25%, 50%, 75% stiffness reduction), with an RMSE of 0.137 cm for damage localization. Comparative study against state-of-the-art methodologies, such as GNN reviles the superior efficiency of the introduced approach. This research establishes a new paradigm for intelligent SHM, bridging the gap between data-driven pattern recognition and physics-aware structural assessment

The secure transmission of satellite imagery is essential for contemporary remote sensing, surveillance, and defense applications. Exchanging large, high-resolution datasets over insecure channels presents significant challenges. Conventional cryptographic algorithms are frequently inefficient for image data because of inherent redundancy and pixel correlations. This study introduces an enhanced encryption framework that combines adaptive hash-driven key generation, multi-chaotic synchronization, and reversible Fredkin logic to achieve both high security and computational efficiency. Dynamic encryption keys are generated using the SHA-256 hash of the input image. Unlike some existing encryption methods that prioritize pixels based on spatial frequency and local contrast features, the proposed framework generates dynamic image-dependent keys using a tri-chaotic system combined with SHA-256 hashing. This approach enhances randomness and key sensitivity while improving resistance to statistical and differential attacks in satellite image encryption.


. Fredkin reversible logic gates facilitate bit-level swapping, ensuring complete reversibility of the encryption process. Reed–Solomon error correction, a coding technique for detecting and correcting data errors, is incorporated into the pipeline to enhance robustness in satellite communication and enable recovery from transmission errors. Experimental results on grayscale images demonstrate that the proposed scheme approaches ideal entropy (7.999), resists differential attacks (with NPCR ≈ 99.6% indicating the percentage of pixels changing between encrypted images after a single pixel change in the original, and UACI ≈ 33.2% measuring the average intensity change), achieves low pixel correlation, and passes all NIST SP 800-22 randomness tests. The framework offers a large key space exceeding 2²⁵⁶ and achieves real-time encryption at 0.19 seconds per frame using GPU-based parallel processing. These results confirm the scheme's security, efficiency, and robustness for real-time satellite image transmission.

The increasing density of IoT devices has intensified spectrum scarcity and security challenges in modern wireless communication systems. Conventional spectrum allocation mechanisms are centralized, static, and vulnerable to privacy leakage and malicious attacks. This paper proposes a federated learning–based intelligent spectrum management framework for cognitive IoT networks. Unlike traditional centralized learning, the proposed approach enables distributed IoT nodes to collaboratively train a global spectrum decision model without sharing raw data, thereby preserving privacy and enhancing security. Each node locally learns spectrum availability patterns and interference characteristics, while a lightweight aggregation mechanism updates the global model. Simulation results show that the proposed framework improves spectrum utilization efficiency, reduces interference, and enhances communication security compared to centralized and non-intelligent spectrum management schemes.