An evaluation is presented of biodegradable cutting fluids, namely lard oil cutting fluid (LOCF), tallow oil cutting fluid (TOCF), palm oil cutting fluid (POCF), and palm kernel oil cutting fluid (PKOCF), in comparison to a mineral oil cutting fluid (MCF), during turning of AISI 1028 steel. Tests were performed at cutting speeds from 73 to 165 rpm, feeds ranging from 0.11 to 0.44 mm/rev, and depths of cut of 1 to 3 mm. Performance was assessed using tool wear, temperature, spindle power consumption (SPC), surface roughness (Ra), material removal rate (MRR), and machining time.The MRR was the highest for POCF (5000 mm/min) and the lowest machining time was 4-5 min. However, PKOCF was better for the tool wear (160 m) and surface roughness (Ra 1.25 m). The palm-based fluids decreased the temperature (135-138 °C) and SPC (0.9-1.2 kW), unlike TOCF, which was the worst since there was greater friction. ANOVA showed that the cutting fluid significantly impacted all the responses (p < 0.05). Taylor's tool life equation VTn = C, n = 0.25 for the HSS-AISI 1028 steel was employed to estimate the tool life at optimum cutting speed of 8.29 m/min (165 rpm). Estimated tool life ranged from 0.58 h (MCF) to 1.08 h (PKOCF), with PKOCF and POCF achieving 1.08 h (65 min) and 1.00 h (60 min), respectively, compared with 0.58–0.75 h for mineral and animal-based fluids. Machining constants were determined through regression method with results from wear experiments data. It shows good agreement of R = 0.92 (92% variation in tool-life) and prediction was consistent with the experimental data because wear decreased, temperatures were lower and steady boundary lubrication had maintained. From these results, palm based cutting fluids showed best performance from the aspect of tribological behaviour and thermal behavior among tested cutting fluid.
This paper includes an extensive examination of existing design processes for 3D printed orthopaedic implants. We considered many different parts of the design process including creating models from imaging data, designing lattice and porous structures, as well as issues of design constraints with metal additive manufacturing (AM). We put particular focus on how numerical techniques including finite element analysis (FEA) were used to analyse the distribution of stresses, deformations, fatigue, and bone-implant interaction prior to production. The findings of the review showed that patient-specific geometries and customized porosities improved the anatomical fit of the implants and reduced stiffness mismatch, thus enhancing performance. However, design errors due to lack of verification or verification processes can cause localized overstressed areas resulting in failure to perform over the long term. The review also identified challenges to improving the current state of the industry including the need for regulatory standards for designs, the establishment of protocols for validating designs with experimental or clinical data and developing methods for managing defects resulting from additive manufacturing processes. Finally, the review offers suggestions for directions of future research including developing automated optimization procedures for designs, creating multi-material print processes, and providing more accurate biomechanical simulations in the development of safer and more effective 3D printed orthopaedic implants.
Video streaming often faces challenges due to data loss and poor network reliability. Packet loss during transmission leads to degraded video quality, making error resilience a critical requirement. Traditional error resilience methods have limitations, including high computational overhead and inefficiency in recovering data under adverse conditions. To address these issues, this research proposes an Error Resilience Video Streaming Technique Based on Hybrid Multiple Description Coding and Yamanaka pattern Color Filter Array algorithm (HMDC-YCFA). The proposed method aims to enhance the robustness of image transmission over error-prone networks, ensuring high-quality reconstruction even under adverse conditions. The approach transforms video sequences into color mosaic images and applies compression techniques to mitigate packet loss, reduce reconstruction errors, and overcome the inefficiencies of traditional error-resilience methods. Using a color mosaic image has the benefit of using one-third less storage space than an RGB image. The presented error-resilience framework is used to improve the transmission performance of multiple video frames. If all descriptor data are successfully received without loss, the restored color mosaic image is 100% identical to the original. The original color mosaic image and reconstructed color mosaic image peak signal-to-noise ratio (PSNR) value is 50 dB, and Similarity Index Measurement (SSIM) value is 1 because there is no difference between images. Simulation results demonstrate that the proposed error-resilient scheme outperforms in terms of PSNR, SSIM, processing time, and compression efficiency.
Clouds are the primary agents of weather systems, climate regulation, and renewable energy forecasting. In the past, traditional methods which relied on handcrafted features and statistical-based modeling have struggled with consistency, scalability, and generalizability in complex atmospheric conditions. Recent developments in deep learning methodologies have revolutionized cloud analysis, facilitating accurate cloud segmentation, classification, and forecasting through advanced data-driven representation learning frameworks. A systematic review of fifty-five peer-reviewed studies published between 2017 and 2025 is conducted, with the literature organized into six methodological research streams: semantic segmentation for clouds, classification of clouds, forecasting rainfall, forecasting solar irradiance, hybrid optimization models for cloud analysis, and surveying contributions to research in this field. Semantic segmentation methods have used convolutions neural networks (CNNs), semi-supervised learning, and CNN-physics hybrid approaches, which typically utilized evaluations based on the metrics of intersection over union (IoU), dice, and pixel accuracy. Classification methods employed in this stream utilized CNN, deep neural networks (DNN), transfer learning, and transformers utilized accuracy, precision, recall, and F1-scores, but these methods have important limitations when working with datasets that lack balance/varied real-time application. Methods for predicting rain and peak sunlight rely on sequential base prediction methods (CNN-RNN) with enhanced predictability while continuing to meet computing resource performance limitations and geographic area restrictions. Evaluation highlights show the standard measurement for segmentation were IoU and Dice metrics, while RMSE and R² were common in Cloud Analysis, rainfall, and solar data analysis. The majority of studies relied on datasets captured from ground-based all-sky imagers or historical datasets, while some studies shared methods of synthetic data or data augmentation methods to avoid issues of data scarcity. Assessment for future directions and needs for work in the cloud analysis domain identified the need for efficient, interpretability, and robust hybrid analysis models.
The concern for environmental issues has brought about the incorporation of ceramic tile powder (CTP) as an additive on the mechanical properties of self-compacting concrete (SCC) in this research. The optimization method explored in this study is Taguchi L9 technique, which was employed to optimize the applied process parameters that involved CTP, water-to-binder (W/B) and superplasticizers (SP). The percentages of CTP added were 0.0%, 10.0 %, 12.5% and 15.0 % of the cement mass. Three levels were assigned to three parameters of CTP, W/B ratio and SP each in the L9 ( ), Taguchi mixed level orthogonal array, as the design of experiment (DoE). The flowability of the fresh SCC properties was mostly VS2 class, with compressive strength (CS) value of 28.27N/mm2 and flexural strength (FS) value of 5.64N/mm2 obtained from Taguchi optimization method. Gene expression programming GEP was used to obtain model for the compressive strength (CS) and flexural strength (FS) parameters. The study concludes that CTP has a significant effect on the mechanical properties of SCC at 12.5% and SP at 1.5% replacement
In the current digital age , security awareness should be taken into account in banking industry involving banking vocational students as future bank employees. That’s huge because the industry relies heavily on digital services and has a high level of sensitivity to financial data. The objective of this research is to measure 135 vocational banking students' level of cybersecurity awareness by using a structured test on the extent that individuals tend toward good security habits. The competition includes categories such as compliance regulations and laws, access and password management, security settings on devices, software updates, data backup practices, use of social media on mobile and browsing safely. The findings indicate high levels of awareness inadequacy, as the majority of students fall within 'Poor' (58.5%) and 'Sufficient' (31.1%) categories for a number of indicators. There were major shortfalls in the scores on the themes of social media safety (mean = 43), rules and laws (mean = 50) and software updates (mean = 53), these will benefit from immediate educational intervention. . Normality and homogeneity tests confirmed that data were compliant with the assumptions of parametric analysis, indicating that the results are robust. These findings are consistent with earlier literature, and suggest that practical, experiential cybersecurity training should be integrated into vocational banking courses to best prepare students for the sector’s digital demands. This paper provides empirical evidence to support those involved in curriculum development and public policy when making decisions regarding the options for enhancing cybersecurity preparedness for future banking employees by identifying specific areas of vulnerability.
This paper reviews the role of Artificial Intelligence (AI) in protecting the elderly against fraud and proposes interdisciplinary research needed to address the resulting technological and human issues.. A type of fraud prevention that involves the use of AI, i.e., when anomaly detection and biometric validation is present, is also functioning, although little is known about its psychological impact on the elderly. The three problems that demonstrated why computer science and psychology should be combined are the enhanced state of anxiety, dependence, and a lack of confidence, which make it essential to understand the role AI has on the psychological stability of elderly people and their confidence. In addition, the paper also identifies morality and law-related aspects of the privacy, informed consent, and the expectations of the long-term effects of AI-based surveillance. As a high percentage of people of the older generation demonstrate deficiencies in digital literacy, the paper suggests turning to policy-level measures that are used facilitate the ethical utilization of AI, not to mention privacy issues. The research with the help of such disciplines as geriatrics, sociology, and digital ethics should be conducted in order to create effective and global AI applications. The application of AI into the medical system, the social welfare system and the financial system that enables complete security is also a subject of the paper. In conclusion, the paper makes the recommendation that the developers, caregivers, policymakers, and communities must convene with an objective of regaining trust with AI, and in doing so, this would allow the elderly to operate within the new online space with no interface with security concerns.
This research explores the impact of dual split injection strategies on the performance, combustion, and emission characteristics of a Common Rail Direct Injection (CRDI) engine fueled with cottonseed oil biodiesel (COTSEDOB B20) and its graphene nanoplatelet (CNP) infused nano-biodiesel blend (COTSEDOB B20 GNP100). The purpose of this work is to explore the higher injection pressures (1000 bar) and precise injection of pilot fuels addressing the higher smoke, particulate matter and NOx emissions encountered with conventional diesel engines where the injection pressures are few hundreds (200-260 bar). Experiments were conducted by varying injection timings (IT), injector pressures (IP), and nozzle geometry (NG), with a focus on comparing single and split injection modes. The optimized parameters of IT, IP and NG coupled with split injection ensures efficient burning of the pilot fuels (biodiesel and nano biodiesel blends) injected into the engine cylinder. The optimized split injection configuration of main injection at 14° BTDC and pilot injection at 7° BTDC with a fuel mass ratio of 90:10 yielded significant enhancements in brake thermal efficiency (BTE) and reductions in smoke, HC, CO, and NOx emissions compared to single injection. Graphene nanoparticle doping contributed to improved air–fuel mixing, reduced ignition delay (ID), combustion duration (CD), and higher peak pressure (PP). At 80% load, split injection of COTSEDOB B20 GNP100 achieved up to 8.64% increase in BTE and reductions of 23.07% in smoke, 18% in ID, and 12% in CD compared to baseline parameters. This study confirms that combining multi-injection strategies with nanoparticle-enriched biodiesel can substantially optimize CRDI engine performance and emission control, making it a viable path toward sustainable diesel engine technology.
The promising characteristic features of Multiple-Input Multiple-Output (MIMO) systems rely on the knowledge of the channel state information (CSI) for coherent signal data detection. The determination of channel state information is achieved using various conventional estimation techniques such as pilot-aided, blind, and semi-blind channel estimation techniques. Obtaining accurate channel state information in MIMO systems is significant tasks upon which system performance depends. This paper presents a comprehensive review of various MIMO channel estimation techniques presented in literature from conventional techniques to more recent deep neural network-based techniques. Various ways of pilot arrangement and complexity reduction techniques are discussed. Furthermore, the key performance indicators in MIMO channel estimation, various algorithms applied in channel estimation and the impact of outdated CSI with its causes are also presented. The recent improvements on the conventional techniques with its impact on the key performance indicators in communication systems such as 5G and beyond Networks, Millimeter-wave Communications, and Massive MIMO system were also reviewed. Accurately estimating wireless channel condition makes signal transmission adaptive leading to optimal performance in transmission and decoding of signals.
The output performance of a conventional transfer field (TF) machine is low in comparison with those of other asynchronous machines. In this study, a reconfigured transfer field machine with simultaneous rotor induced currents and capacitance injection for enhancement is presented. The machine comprises two identical salient-pole machine elements that are coupled mechanically and wound integrally for the same pole number. The salient-pole half axes are displaced in space quadrature in the machine elements comprising the machine. Each stator has dual sets of identical poly-phase windings regarded as primary and secondary that are sinusoidally distributed in the stator slots. Primary windings are interconnected in series between the machine elements and the terminals connected to public utility source while the secondary windings are swapped betwixt the machine elements and then terminated on a balanced variable capacitor bank. Windings are also placed on the rotors. The mathematical model of the machine is derived, the resulting equations therefrom are simulated in MATLAB/Simulink environment. It is shown that with an optimized value of 8,300μF tuned capacitor, a remarkable improvement in the performance characteristics of the machine over the traditional TF machine is obtained when compared. The starting torque, power factor and maximum torque increased by 384.2% (1.9N-m to 9.2N-m), 87.5% (0.32 to 0.6) and 374.1% (6N-m to 28.45N-m) respectively, which confirm superior performance characteristics.
This study presents Ritz variational method for the free transverse harmonic vibration solutions of slender beams on two-parameter elastic foundations (SBo2PEFs). The studied problem is a soil-structure interaction problem of dynamics that is important in the dynamic design of foundations and buried pipelines. The domain equation is derived using variational calculus, and the total energy functional was found for harmonic vibrations in terms of the modal displacement W(x) and the derivatives Minimization criteria with respect to the generalized parameter of the displacement is used to find the characteristic frequency equation. The obtained Ritz equation is an eigenvalue problem. It was found that for simply supported SBo2PEF, the exact sinusoidal shape function used gave the exact eigenfrequency for any mode of vibration. For clamped-clamped SBo2PEF, a one-parameter shape function gave accurate fundamental frequency. For cantilever SBo2PEF, a one-parameter shape function gave accurate fundamental frequency solutions.
Electric energy is one of the most widely used form of energy around the globe and as such has the most dynamic means of impacting positively on economic development of any nation in the world. Therefore, to further grow the economy through increased agricultural productivity and rural development, there is an urgent need to address the issue of poor and ineffective rural electrification strategy for sustainable farm operations. Consequently, this paper presents a framework that uses intelligent load forecast, geospatial and cost-effective metric for the analysis of the economic optimality of grid extension, diesel and hybrid photovoltaic (PV)/diesel renewable power generation systems for rural farm operations. The historic national load demand data for 20 years obtained from the national bureau of statistics and central bank was used for the training and validation of the forecast model. The historic electric load data for the case study farm cluster (Adani Enugu Nigeria) was taken to be the electric energy equivalent of the contribution of Adani Enugu to the gross domestic product (GDP) of Enugu state. Input parameters to the neuro-genetic forecast model are the contribution of rice production to the national GDP, contribution of Adani Enugu farm cluster to the national GDP, electric energy consumption (EEC) per ton of rice produced at Adani Enugu and the annual population growth rate. From the simulations carried out, the economic viabilities of the generation options were assessed in terms of capital expenditure (CAPEX) and operational expenditure (OPEX). However, with a 93.84% decrease in CAPEX per kWh if massive investment and expansion of rice processing capacity were made over the forecast horizon, grid extension was found to have the lowest CAPEX. The OPEX for this generation option remained relatively steady for the mentioned condition. However, the approach presented in this paper can be integrated as core components in any generation analysis tool for driving support in optimal generation planning.
To increase product yield, percentage conversion, and catalyst recovery ease, the reaction parameters required for a successful Suzuki Miyaura cross-coupling process were optimized. To form new carbon-carbon bonds, 3-Indotoluene and Phenylboronic acid were cross-coupled using synthesized Ag-Pd Alloy Nanoparticle photocatalysts. Reaction conditions namely reaction base, solvent, wavelength, light intensity and reaction atmosphere were individually optimized by comparative analysis to find out the best set of parameters that will result in a cross-coupling product with high yield and high percentage conversion. Mix solvent of Dimethylformamide and water in the ratio of 3:1 was determined as the best solvent for the cross-coupling carried out in this research as a percentage conversion of 96% was achieved. Potassium carbonate (K2CO3) was the reaction base that gave better reaction yield and percentage conversion than the other bases that were tested in this research, also Argon as the reaction atmosphere gave better results than another reaction environment, while light intensity and light sources with shorter wavelength (less than 500nm) are favorable as the gave better percentage conversion, due to their ability in activating the Alloy Nano-particle Photo-catalysts that was used in this research. Findings from this research work suggest that for an effective and efficient Suzuki-Miyaura cross-coupling reaction, K2CO3, Dimethylformamide and water in the ratio of 3:1, argon, and a light source with high intensity and shorter wavelength are the appropriate reaction conditions.
This study employs a combination of Vlasov’s thin-walled beam theory and a multi-variable power series approach to analyze the elastic stability of mono-symmetric box girders, a class of thin-walled structural elements widely used in bridge engineering, subjected to eccentric transverse loading. The primary objective is to investigate the discrepancy between the shear center and the center of gravity, which induces complex coupled deformation modes, particularly flexural and distortional effects. Using Varbanov’s modified generalized displacement functions, the governing differential equation of equilibrium were derived based on section properties evaluated at the pole and shear center, through a unit displacement approach. Essential cross-sectional parameters were obtained using enhanced product integrals (diagram multiplications). Given the complexity of the governing equation and boundary conditions, exact closed-form solutions were not attainable. To address this, three analytical methods, power series, trigonometric series, and Taylor-Maclaurin series, were applied to solve the reduced equations, enabling a comprehensive evaluation of flexural and distortional behaviors. Among these, the power series method proved most effective, accurately capturing the multi-variable interactions required to model realistic deformation patterns. Under eccentric loading, maximum flexural deformation occurred at 10 and 40 meters, while distortional deformation peaked at 40 meters and diminished near 45 meters. The Taylor-Maclaurin series showed maximum flexural deformation at 30 meters and distortional deformation at 9 meters. The trigonometric series revealed cyclic deformation patterns indicative of fluctuating load effects but lacked the precision needed for complex geometries. This study addresses a notable gap in the literature by providing a robust analytical framework for mono-symmetric girders and emphasizes the importance of advanced multi-variable analytical techniques in structural design and engineering education.
The proliferation of counterfeit items has hurt the economic growth, public health, and safety. This work aims to develop an innovative system that can counter and mitigate the threat posed by local and global counterfeiters whose activities have caused untold health and economic hardship to society. This paper proposes a novel blockchain-based anti-counterfeiting system that makes use of a product's unique characteristics and its geographical location. Prototype system modelling in this study was accomplished using object-oriented software analysis and design techniques, Rapid Unified Process (RUP) and, Rapid Application Development (RAD) methodologies for QR Code and Blockchain applications respectively. Ganache, a private Ethereum blockchain network, was set up to serve as the backend platform. Open-source software such as the Truffe suite and the Solidity compiler were utilised in setting up the Ganache network as well as in compiling and deploying smart contracts written in Solidity. Results proved that the system, when tested on 50 products, shows low energy consumption, high speed of execution at 38.4s on average, QR code scanning time of 9.5ms on average, very high data integrity, and 100% accuracy record when validating whether or not a product is a counterfeit. This work provides a solution for cost-effective and comprehensive anti-counterfeiting measures, featuring key elements such as traceability, immutability, and transparency. The developed system is unparalleled as it combines blockchain technology, unique product inherent features, location information (GPS coordinates), and Track and Trace technologies, to offer a reliable and secure solution to counterfeit trading. This work, therefore, represents a potentially innovative approach to curbing the proliferation of counterfeit products.
Digital Construction Management (DCM) has the potential to revolutionise Nigeria's construction industry by boosting productivity, quality, and safety. It is important that construction professionals utilise DCM in their various projects. However, adoption has been slow due to a significant skills gap among professionals. To leverage DCM benefits, companies must ensure their employees possess these crucial skills. This study aims to identify and analyse the top ten essential skills for digital construction management across various professions and company sizes. Adopting a quantitative research design, the study adapts 18 skills from previous literature review and uses a questionnaire to achieve the aim. The skills were ranked using Mean Item Scores and Relative Importance Index and subjected to further analysis using non-parametric tests and regression. The most important skill identified was continuous learning. It was also discovered that there was no significant difference in the perception of the skills, both technical and soft skills, among various professional groups and company sizes. This supported the conclusion that the identified skills were universally relevant. Construction professionals should prioritise acquisition and promotion of these skills to improve project delivery with the aid of digital tools.
This study employed a Computational Fluid Dynamics (CFD) approach based on STAR-CD code to investigate the effect of mechanical ventilation on hydrogen gas leaks and diffusion in partially enclosed space. It is a case study of a homogenous charged compression ignition engine (HCCI) laboratory of the Mechanical Engineering Department, University College London (UCL). The 3-D modelling was based on the geometry as well as airflow designed for the test laboratory. Two turbulence models and three differencing schemes were employed on two grid refinement levels. All the differencing Schemes predicted a similar velocity profile and hydrogen concentration below 25% of the lower flammability limit (LFL) in most parts of the test laboratory. Although the predicted hydrogen mass fraction from the steady state simulation does not resolve the buoyant shape of the gas, the time-dependent solution captures the buoyant characteristic of hydrogen. It revealed that the hydrogen gas initially rises to a height 0.55cm above the exit towards the ceiling, from where it gradually diffuses in a radial pattern to a homogenous non-flammable concentration in the room. This predicted pattern of hydrogen gas dispersion is consistent with experimental data. Therefore, a small hydrogen leak of the type and at the airflow rates investigated in this study does not pose a risk of fire in most parts of the Engine Test laboratory; except in the region very close to the leak source.
Structural cracks in buildings present significant safety and integrity challenges, necessitating thorough investigation into causes, effects, and repair strategies. This study explored factors contributing to structural cracks, their implications on the structural members and building as a whole, and various repair methods from the literature. The factors were populated in questionnaires to acquire feedback from respondents with and without knowledge of crack propagation in buildings. Through analysis and observation, including survey insights and visual measurements, diverse causes emerged, from material deficiencies to environmental influences. 81% of respondents understood structural cracks, with 68% aware of common causes that are sometimes neglected. Initial thin cracks, if ignored, can escalate into serious issues and jeopardize safety. 80% of respondents would not inhabit aesthetically compromised buildings due to cracks. Repair strategies are critical to address the underlying causes and factors affecting the cracking of structural members. This will involve efforts from architects, engineers, contractors, and occupants through educational initiatives and awareness campaigns to enhance collective understanding of the impending problems and prevention measures to avoid incessant building collapse.
Road traffic crash prediction (RTCP) is a critical aspect of transportation safety, enabling the identification of high-risk locations and informing the implementation of proactive measures. This study explores the comparative performance of Machine Learning (ML) algorithms and traditional Safety Performance Functions (SPFs) to predict road traffic crashes along the Lagos-Ibadan Expressway, a major highway in Nigeria known for its high crash rates. To achieve the objective, SPFs estimated using Negative Binomial Regression (NBR) and ML regression models mainly Support Vector Machine (SVM), Random Forest (RF) and Extreme Gradient Boosting (XGBoost) were developed using historical crash data collected from Federal Road Safety Commission (FRSC) of Nigeria for 10years duration between 2014 and 2023, traffic components and geometric design features as input variables. The study's findings indicate that ML algorithms outperform SPFs in terms of predictive accuracy and sensitivity to complex, non-linear relationships among crash-contributing factors with R2 of 0.99, 097 and 0.84 for training and 0.93,0.9 and 0.76 for testing dataset in the three ML models. However, SPFs remain advantageous in interpretation and ease of implementation. The analysis also highlights the importance of feature selection, with variables such as traffic volume, traffic speed, road curvature and pavement width emerging as significant predictors. Furthermore, this study offers insights for policymakers, traffic engineers, and researchers seeking to improve road safety outcomes through data-driven crash prediction methods. The results emphasize the potential of integrating ML techniques with traditional methods to develop hybrid frameworks for enhanced crash prediction and prevention strategies on high-risk roadways.
