KUFA JOURNAL OF ENGINEERING

In this investigation, the soil-water-retention is described in the soil-water-characteristics-curve (SWCC). SWCC contains the data required to explain the unsaturated soils' mechanical behavior. The reliable findings of unsaturated sand soil tests and the impact of initial density on the hydraulic properties under various matric suction and net normal stresses are examined. The tests are carried out on a sample of soil from the northern region of Al-Najaf city, Iraq. The sample of soil has been disturbed and remolded in the cell of Oedometer device. A modified high-air-entry-disc-based Oedometer is used with accessories to control the air and water pressures to achieve the specific matric suction in the unsaturated test. Two sets of tested specimens are prepared, first, at 90% and second at 95% of the maximum dry unit weight from the standard Proctor experiment. Each set is performed for different matric suction under 200 and 400 kPa net normal stress. (SWCC) The soil water retention curve is affected by the application of net normal stress and initial density. There is a decrease in water volume change with wetting progress in unsaturated soil tests due to an increase in soil density.

As diagnosed, water is an essential component in the preparation of water base mud, but drilling fluids cannot be prepared with high-purity water (practically and economically unavailable), as water may sometimes contain salts and ions such as chlorides and sulfates, negatively affecting the quality and properties of the mud. This research aims to find the effect of impure water on the drilling fluid and then provide the necessary solutions and treatments for that. Samples of water were used and examined, and they contained a percentage of calcium ions at a rate of 800gm/liter, 1000 gm/liter, and 1400 gm./liter, this is unacceptable, so soda ash was used to reduce the above percentages. On the other hand, the water samples had a salinity of 20,000, 30,000, and 40,000 ppm/liter, which is undesirable, so it is preferable to dilute the mud by adding water. The results showed the effect of the presence of those salts on the effectiveness of the added polymers, as well as causing problems with CMC, as it reduces its performance in increasing the viscosity of the mud. The results showed that the acceptable percentage of calcium should be less than 400 mg/liter, and salinity should not exceed 10000 ppm/liter. The results also proved the negative effect of the presence of sulfate in the water on the properties of the drilling fluid, as it reduced its viscosity and pH. This was treated by adding caustic soda to the drilling mud to increase the pH. It was also noted that the addition of caustic soda led to an increase in the viscosity of the drilling mud. What is novel in this research is that it is possible to treat contaminated drilling mud by adding environmentally friendly materials such as micro-silica to reduce the effect of contamination with calcium ions.

A significant number of causes for people killed by automobile accidents on both public and interior roads are the driver's failure to pay attention to the roadway, as well as other factors such as tiredness, sleep during driving, using the phone, or even the driver being intoxicated, etc. This work aims to identify sleepy status, pinpoint the onset of sleep, and alert the driver when their level of alertness drops below a level that does not improve productivity or the vehicle's overall safety protocols to prevent drowsiness-related crashes. In this article, we present a new method for implementing drowsiness detection inexpensively, using a Raspberry Pi as a central processor board that connects all practical measurement components. In addition, the proposed method includes the following activities: tracking the driver's head vibration while driving using tilt sensors. Moreover, a camera was mounted in front of the driver to take photos and record videos of the driver's health. As a result, the results were entirely satisfactory, and if the driver was operating the vehicle while fatigued, the response was 95% reliable and fast.

This paper investigates the potential of deploying flaps on wind turbine blades to enhance their efficiency and energy capture capabilities. The FX66-S-196V, FX63-137 S, and SG6043 supercritical airfoils were used and distributed along the blade radius. Flaps, situated 20% along the trailing edge of the blade chord, offer a means of actively controlling aerodynamic forces and optimizing blade performance under varying wind conditions. Through computational fluid dynamics (CFD) simulations and optimization techniques. The aerodynamic effects of flap deployment on wind turbine blades are analyzed. The study explores the impact of flap angle, position, and deployment strategy on key performance metrics such as power coefficient, lift-to-drag ratio, and energy extraction efficiency. Results demonstrate that judiciously deploying flaps can lead to significant improvements in turbine efficiency, with power output enhancements ranging from 2.5% to 4.6%, depending on operating conditions such as wind speed, tip speed ratio, angle of attack, and flap angle setting. Furthermore, sensitivity analysis reveals optimal flap configurations for different wind regimes, highlighting the importance of adaptive control strategies. This research contributes to the growing body of knowledge on active aerodynamic control techniques for wind turbine optimization and underscores the potential of flaps as a viable means of enhancing wind turbine blade efficiency in practical applications.

The introduction of GPS technology has led to the creation of vast amounts of spatio-temporal data, which captures the movement patterns of different things. Efficient allocation of resources to ensure user satisfaction is a crucial factor in shaping the future of urban planning and development. It is required to comprehend the factors that can contribute to the creation of methods for studying user behaviours using a substantial number of persons within a brief timeframe. It is essential to employ appropriate clustering approaches to analyze this data in order to comprehend spatio-temporal behaviours.Heatmaps offer a graphical display of changes in density across both location and Time, making them a user-friendly tool for initial data analysis and identifying areas of high activity. The Spatio-Temporal Dynamic Graph Neural Network (ST-DGNN) utilizes graph neural networks to represent the intricate connections present in spatio-temporal data, encompassing both spatial interdependencies and temporal changes. Our methodology improves the accuracy and interpretability of trajectory clustering by integrating different methods. The suggested method has been shown to identify relevant clusters effectively and reveal noteworthy spatio-temporal characteristics through experimental analysis on real-world GPS datasets. The research utilizes a dataset comprising 182 users for analysis. Numerous measures are taken to boost the clustering accuracy of the applied techniques, including addressing missing values and outliers. Additionally, this thesis introduces a framework for time estimation based on graph-based deep learning, termed Spatio-Temporal Dual Graph Neural II Networks (STDGNN). The method entails constructing node-level and edge-level graphs that depict the adjacency connections between intersections and road segments. The results showed a number of cluster changes in each period of time dependent on move users and period; for example, the (2592) cluster of period one hour.

The current work was developed under the title of Artificial Neural Network (ANN) Proportional Integral Derivative (PID) for the arm rehabilitation device and included building and designing the simulation model and simulation results for the arm rehabilitation device. A set of tests were proposed to include firstly testing a system that represents the state of the open arm rehabilitation device and secondly It represents the closed arm rehabilitation device, third represents the closed-loop arm rehabilitation device with PID control device, fourth represents the arm rehabilitation device using ANN, and finally the closed-loop arm rehabilitation device can be used with a comparison between PIDC and ANN. To conduct all the proposed test cases, a program can be used MATLAB, which can help simulate a device that represents an attempt to regain movement in the arm, which is called rehabilitation. It can be noted that the target group is some people who suffer from stroke. By representing the system in the proposed simulation model, its effectiveness can be verified. It is possible to conduct tests aimed at improving performance by working on developing the model by adopting the appropriate design for the characteristics that match the required operational behavior of the system with all conditions that suit different situations. The test cases demonstrated through the simulation results the possibility of identifying the system behavior for the proposed cases. The difference between the system behavior for all these cases was also identified. In addition to the possibility of improving the performance of the movement recovery device to rehabilitate the injured arm through the system’s performance in the presence of an expert neural network controller, it is better than the traditional controller.

Permanent magnet synchronous motor (PMSM) is extensively employed in AC servo drives owing to their superior torque-to-inertia ratio, power density, efficiency, and power factor compared to other motors. So, it is a crucial point to regulate the PMSM speed. Conventional proportional, integral, and differential (PID) is a simple controller and easy to implement but it is coefficients are essentially determined by experience when used in PMSM to control the speed. This invariably produces unacceptable outcomes, in addition when it comes to low-power application drives, PID controller gains typically produce adequate results but, when it comes to high-power application drives, an untuned PID does not deliver satisfactory performance. The optimization algorithm offers an effective method to produce optimal PID gains. Therefore, to optimize the PID coefficients to regulate the PMSM speed, this study suggests a mayfly optimization algorithm (MA). Recently, the MA was introduced as a new intelligent optimization method with exceptional optimization capabilities. Nuptial dancing and random flight improve the ability of the algorithm to balance its features of exploitation and exploration while assisting in its escape from local optima. This suggested approach has been verified with MATLAB, and the outcomes are compared with the standard particle swarm optimization technique (PSO) and conventional PID. The outcomes show that compared to the standard PSO or conventional PID, the PID parameters adjusted by the MA method can produce faster speed responses and less overshoot. Furthermore, the system's optimal ITAE index value, as determined by the MA technique, is smaller (0.794) as compared to other techniques 1.503 and 1.906 respectively.

This in-depth review article provides a comprehensive analysis of mitigation strategies for seismic damage in structures, focusing on the importance of understanding seismic properties and implementing advanced analysis techniques. The review underscores the critical role of structural properties, including stiffness, strength, and ductility, in designing earthquake-resistant structures. By exploring a range of seismic analysis methods, such as force-based, displacement-based, and numerical approaches, the review highlights the significance of a holistic approach in assessing structural response to seismic events. Moreover, the review delves into innovative methods like subspace system identification for damage detection and statistical analysis of seismic sequences, offering promising avenues for future research. The adoption of passive energy dissipation systems and base isolation techniques emerges as key strategies in reducing seismic damage and enhancing structural resilience. Additionally, the review discusses the potential of machine learning techniques in predicting seismicity rates and improving risk management in seismic-prone areas. By combining traditional and advanced methods, this review sets the stage for further developments in seismic damage mitigation strategies. The integration of technological expertise, scientific aptitude, and interdisciplinary approaches is highlighted as essential in ensuring the built environment adapts to unforeseeable natural forces. Ultimately, this review provides a foundational resource for academics, policymakers, and practitioners in the field of earthquake mitigation and structural safety.

The importance of deep learning has heralded transforming changes across different technological domains, not least in the enhancement of robotic arm functionalities of object detection’s and grasping. This paper is aimed to review recent and past studies to give a comprehensive insight to focus in exploring cutting-edge deep learning methodologies to surmount the persistent challenges of object detection and precise manipulation by robotic arms. By integrating the iterations of the You Only Look Once (YOLO) algorithm with deep learning models, our study not only advances the innovations in robotic perception but also significantly improves the accuracy of robotic grasping in dynamic environments. Through a comprehensive exploration of various deep learning techniques, we introduce many approaches that enable robotic arms to identify and grasp objects with unprecedented precision, thereby bridging a critical gap in robotic automation. Our findings demonstrate a marked enhancement in the robotic arm’s ability to adapt to and interact with its surroundings, opening new avenues for automation in industrial, medical, and domestic applications. The impact of this research extends lays the groundwork for future developments in robotic autonomy, offering insights into the integration of deep learning algorithms with robotic systems. This also serves as a beacon for future research aimed at fully unleashing the potential of robots as autonomous agents in complex, real-world settings.

The cold rolling effect on the microstructure and mechanical properties of recycled AA6111 alloy with Zr or/and Ni were investigated. All alloys were produced by the stir-casting method. homogenization heat treatment (500 ºC for 6 hours) was performed on AA6111 alloy only. The alloys were solutionized (at 540 ºC for 30 minute) and artificially aged (at 180 ºC for 10 hours). The optical micrograph, scanning-electron microscope (SEM) and X-ray Diffraction (XRD) studies for all samples. The microstructure analysis displayed different fine intermetallic phases distributed in large amounts in alloys containing Zr or/and nickel compared to AA6111 alloy. The combined alloying elements' addition effect with the cold rolling effect and artificial aging led to great improvements in the mechanical properties, where the hardness and tensile properties of all alloys increased. The AA6111+0.41wt%Ni alloy showed the highest hardness value, as it increased to 161 HV after artificial aging, while the AA6111+0.54wt%Zr alloy showed the highest YS and UTS, which increased to 252 MPa and 342 MPa, respectively.

The purpose of the current research is to determine the effect of fiber type, volume ratio, and matrix type on the vibration properties of sandwich beams made of composite face sheet and core. The typical sandwich structure consists of three layers: face sheets, core, and adhesive bonding, and in this research, the adhesive layer between the face sheets and core was abolished by preparing the overall mold with fibers inside and casting the resin to fill the face sheet and core parts. The face sheets of the composite beams are made from a polyester or epoxy matrix reinforced with glass fiber, carbon fiber, and hybrid fiber, and the core is a honeycomb consisting of random glass fibers immersed in a resin matrix (polyester or epoxy). 22 composite sandwich beams were constructed to conduct vibration testing. The vibration results obtained experimentally were compared to the ANSYS R1 2022 software, and the results were in very good agreement. Hybrid fibers and polyester matrix achieved the highest values of natural frequency for (clamped-clamped) boundary conditions, where the natural frequency value of the hybrid fiber and polyester matrix reached (2037 Hz) at a volume fraction of (23.14%) experimentally and the natural frequency reached (1804.5 Hz) at a volume fraction of (21.95%) experimentally for the simply supported boundary condition.

Shape memory alloys (SMAs) are one of the most important types of smart materials. It define as the materials that can undergo large pseudo elastic deformations while having the ability to recover to their original shape once subjected to a memory stimuli such as a specific temperature ,stress ,or strain. This study examines the effects of varying Zirconium (Zr) additive amounts on the microstructure, mechanical, and shape memory effect for all specimens of Cu-Zn-Al SMAs. Powder metallurgy was used to create Cu–22%Zn–4%Al SMAs, both base and with the inclusion of 0.2 and 0.4 weight percent Zr. 650MPa compact pressure was used to create the alloys after the particles had been mixed for five hours. The alloys underwent a three-step sintering procedure in a vacuum tube furnace for one hour at 350 ˚C, then for one hour at 550 ˚C, and for two hours at 850 ˚C. To characterise the microstructural and phase characteristics of alloys, both with and without Zr additions, XRD diffraction analyses, optical microscopy (OM), and scanning electron microscopy (SEM) were performed. With differential scanning calorimetry, the transition temperatures of all alloys were determined (DSC).As well as the shape memory effect test (SME) was measured. Results confirmed that all alloys compositions were found to include the predominant (Cu5Zn8) phase, as proven by XRD and microstructural investigation. The Zr addition dramatically lowered the transition temperatures, according to the transformation temperature data. Hardness experiments reveal that alloys' hardness and volume losses were enhanced by adding up to 0.4% weight percent of Zr reached to (64.8%) and (0.3909)mm3 respectively compared to base alloy.

Currently, sustainability of the construction and building industry taken a priority. This study investigates the feasibility of using a high volume (up to 50%) of blended waste brick powder (BP) and metakaolin (MK) as ordinary Portland cement (OPC) replacements. The binder of the control mixture was a blend of 50% OPC and 50% BP, while the other two mixes were prepared by substituting 10% and 20% of BP with MK. The characteristics of fresh concrete were assessed depending on measuring the mixture temperature, the fresh density, and the workability. The bulk density, and the mechanical properties were investigated and tested at 7 and 28 days. In the line of durability parameters, the void content and drying shrinkage up to 90 days of all mixtures were evaluated. The findings have demonstrated that the control mixture achieved high workability (slump =180 mm), structural compressive strength (34 MPa) at 28 days, low void content (<3%), and acceptable shrinkage strain. The workability of the mixes containing 10%MK:40%BP and 20%MK:30%BP has slightly decreased, while the mechanical properties were increased and the drying shrinkage were declined. However, the inclusion of This study highlighted an ecological technique toward waste management of construction materials and confirmed the possibility of including a high volume of BP as a cementitious material to synthesize more sustainable concrete.

A modified TPE approach to perform finite deflection analysis of slender beam-column elements has been developed. The proposed approach utilizes the energy principal method and takes into account the geometric nonlinearity including the effects of axial force on bending stiffness, the end moments on axial stiffness (bowing), and the initial imperfection. A new equation of the deformation curve that approaches to the exact solution is used in the strain-displacement relation to obtain a more accurate beam-column response. The derived formulation of displacement of the beam-column under axial compressive load with single curvature bowing is presented with initial imperfection and different end eccentricities. The Green strain tensor equation is developed to consider higher-order bowing term. Nonlinear analysis of central finite deflection is carried out using Newton-Raphson iteration that includes high order terms of total potential energy (TPE). The beam-column stability is verified by computing the hessian determinant of the total potential energy. The validity of the new approach is established by comparing the numerical results obtained using the proposed equations against data previously published in the literature. Outputs from the analysis indicate that the proposed approach is capable of capturing the deflection of the beam-column with enhanced accuracy, ranging from 8.5% for e = 0.025 and up to 23.5% when e = 0.125.

A fiber glass -silica fume and silica fume (SF) mix (F-SF) have been employed in this study in order to enhance and relate their effects on the expansive soil. Bentonite (B) was added with natural soil in a lab at a ratio of 50% by soil dry weight to create the soil used in this study. The (SF) was added alone at (8,9,10,11,12, and 13) % and 1 % of Fiber glass (F) was added to the (clay soil - SF) mixture. For both treated and untreated soil, odometer, shear box, and swelling tests were used to examine the behavior of the soil after the addition of (SF) and (F-SF) mixture. From the outcome of the experimental tests, we noticed that (SF) reducing both comparison index (Cc), reloading index (Cr), and swelling percentage (SW)%. Also, it is influenced on shear strength parameters by decreasing cohesion (c) and increasing internal angle friction (φ). As well as adding 1 % (F) to the (clay soil - SF) mixture modify the prepared soil by reducing (Cc), (Cr), (SW) % and increasing (c, φ). Based on all of the above results, we concluded that adding (SF) and (F-SF) mix. can enhance the engineering characteristics of swelling soil.

Sandwich panel structures are composed of three layers: core and two facings. The configuration of the core deeply influences the mechanical properties of the sandwich panel. Numerous innovative core shapes have been proposed to enhance these properties; however, many have not been implemented due to manufacturing difficulties, particularly spherical shapes. This research aims to address this manufacturing challenge by utilizing 3D printing technology to produce sandwich panels with a spherical core. Additionally, the study investigates the effects of varying design parameters like sphere diameter, offset distance and face thickness on the free vibration features of the sandwich panels experimentally and numerically. Experimental tests validated the finite element models with an error margin below 12%. The key parameters explored were spherical diameter (3-12mm), offset distance (10-33mm), and face thickness (1-5mm). The results demonstrate a direct correlation between these parameters and the sandwich beam natural frequency.

Generally, Cold asphalt emulsion mixtures (CAEM) produced and compacted at ambient temperatures. The main objective of this study is to investigate the improvement of the performance for mixtures by the addition of geopolymers based on cementitious material by preparing the asphalt samples using three concentrations of geopolymer utilized as a substitute for limestone filler (LF),The mechanical properties of the mixtures were evaluated using the Marshall test, Indirect Tensile Strength (ITS) test, Kim test and Water sensitivity test TSR. The test results revealed that the concentration of (NaOH) has a substantial influence on the structure and characteristics of geopolymers, enhancing Marshall stability, ITS, and rut resistance. CAEM with geopolymer considerably improved Marshall stability and deformation resistance especially those containing fly ash.CAEM with geopolymer provides larger ITS conditioning and a TSR over 100% better than hot mix asphalt HMA. Furthermore, it is noteworthy to mention that the addition of geopolymer has somewhat improved the mechanical properties and durability characteristics of the CAEM.

The knowledge of the exact thresholds of parameters in the diesel engines, during combustion, is essential to simulate the combustion process, establish parametric values, reduce cost and predict exhaust emissions. Accordingly, the present paper applies the grey wolf optimization method to determine the optimal threshold of parameters and engine responses in a direct ignition engine. Twelve formulated linear equations of engine responses are introduced to the objective function of the grey wolf optimizer. A computer program in C++ was applied successfully using literature data to validate the grey wolf optimization procedure based on the encircling, hunting and attacking of prey by the wolf. The results show that load demand and turbocharge boast air pressure have the least and highest values of engine outputs, respectively. The blend ratio had its highest values when optimized alongside the main injection duration. The responses and parameters greatly improved from initial values to stopping criterion of 200 iterations. Instances reported include brake specific fuel consumption, which improved from 2.6468 to 1.0816 g/kWhr, blend ratio changes from 0.5031 to 0.4760%, speed drop from 0.0031 to 0.0010rpm, and load drop from 0.0017 to 0.0010%. The main contribution of this paper is to establish the optimal thresholds of engine responses using the grey wolf optimizer in a diesel engine combustion chamber. The development of a new method to optimize response and parameters of an internal combustion process using grey wolf optimizer is the novel aspect of this work. The results have essential practical significance to establish new emission profile for biodiesel. The practising engineers and researchers have a holistic insight into the problem’s solution and can utilize the results to enhance their engine responses.

This paper presents an experimental investigation of the performance of two-span reinforced concrete slabs repaired by CFRP composites against flexural loads. Thirteen continuous reinforced concrete (RC) one-way-slabs were cast, with dimensions of 2400 mm length, 500 mm width, and 100 mm thick. The first specimen had no strengthening at all and was considered a reference slab (Control). The rest of the slabs were repaired using different configurations of external bonding CFRP sheets and NSM bars. Before the repairing process, the slabs were subjected to 40% and 65% of the control slab ultimate load. The test was done under two-line loads. Moreover, the effect of different lengths, spacing, repairing techniques, and damage ratios were studied in this work. The finding shows that the repairing methods were effective in improving the ultimate load capacity ranges between 42% to 70% and reducing the deflection by about 11% to 25%.

The work done in this research is dedicated to compute the critical buckling load of combined columns. This is performed using theoretical and experimental analyses. The combined column is made of two materials, namely copper and aluminum alloys. The theoretical approach involves numerical analysis using ANSYS and analytical analysis using the equivalent stiffness method. The study determines the critical buckling load by considering many parameters. These parameters include lengths of copper and aluminum in the combined columns, cross-sectional geometries (square, circle, and hexagon), and boundary conditions. The experimental work is carried out using a Strut Bucking Apparatus on a combined column with circular cross-sections and different end supports. The findings show that the critical buckling loads achieved from numerical and analytical analysis are in good agreement with those obtained from experimental analysis. It is found that the critical buckling load increases with increasing the length of copper in the combined column and with more restrictive end supports. There is a good match of critical buckling load between those achieved from the equivalent stiffness method and ANSYS model for any considered cross-sectional geometries (square, circle, and hexagon) and two types of support (fixed-fixed and fixed-pinned). The results reveal that the combined columns with square cross-sections have higher critical buckling loads than those with circular or hexagonal cross-sections. The highest percentage errors of critical buckling load between the experimental method and theoretical methods (equivalent stiffness method and ANSYS models) are (-13.120, -12.768, and -12.453) % and (-13.083, -18.039, and -19.052) % for pinned-pinned, fixed-pinned, and fixed-fixed columns respectively.

The global use of IoT-enabled devices powered by renewable energy can create a smart, efficient, and eco-friendly network. By leveraging data analytics and machine learning, these systems can predict energy consumption, optimize resources, and maintain renewable energy assets proactively. IoT device adoption is rapidly increasing, with a projected 13% growth in 2024, bringing the total to 18.8 billion devices globally. However, there is a critical need for advanced applications of IoT technology, improved sensors implementation, and up-to-date software development to control IoT devices which is essential to fully harness renewable energy and combat global warming, ultimately achieving the vision of a smart green planet. This review examines the operational and technical needs of IoT devices, their integration with various renewable energy sources using Raspberry Pi or Arduino platforms equipped with sensors (such as motion, vibration, flow, and gyroscope), and the application of machine learning in IoT devices to showcase the potential of these technologies in fostering a smart, green world. To further this investigation, the study thoroughly examined the pros and cons of integrating various renewable energy sources to develop smart systems, such as smart grids, smart cities, smart transportation, and smart waste collection. Additionally, this study also examines the challenges faced by smart IoT systems, providing critical analysis and identifying promising directions for future advancements.

The present study involves strengthening process applied upon brick walls to increase their ultimate load carrying capacity or improving their ductility behavior by using different types of strengthening materials. These materials include either traditional well-known materials or newly proposed materials. The traditional materials include steel wire mesh (SWM) and carbon fiber-reinforced polymer sheet (CFRPS) while the proposed materials include recycled damaged tire strips (DTS). The purpose of using such abundant and polluted material is to satisfy, respectively, the economic considerations and clean environment in the field of enhancing the structural behavior of brick walls. Eight brick walls models were constructed with dimensions of (1200×1200×240) mm, respectively represent width, height and thickness of each model to be tested by 45° compressive loading. Results exhibited that, the SWM increases the ultimate load carrying capacity for the strengthened models to 51.85% compared to the non-strengthened model when placed on one side of the model. Also, the use of CFRPS placed on one side and on both sides of the strengthened models increases the ultimate load capacity respectively to 18.2% and 385% compared to the non-strengthened model. The DTS material has no ability to increase the ultimate load capacity (-18.2 %) but it successfully converts the brittle behavior of the brick walls to a ductile one. It is recommended to use DTS to increase the ductility for brick walls that are likely be damaged or collapsed to avoid the sudden failure.

The uncontrolled friction stir welding heat generation impacts the quality of welds. However, the intuition and experience of the engineer fail to regulate the effects of excessive heat generation on the weld quality and research has not addressed this aspect yet. This paper fills the gap by introducing an integrated CRITIC-BPNN (CRiteria Importance Through Intercriteria Correlation-Back Propagation Neural Network) method to investigate the selection and optimisation characteristics of the friction stir welding process for AA6082-T6 material. In this study, two major performance characteristics i.e. ultimate tensile strength (UTS) and percentage elongation (%EL), were chosen for analysis. The input parameters for the machining were the tool rotational speed, welding speed, tool pin profile and tool shoulder diameter. For the back propagation neural network model, a four-layer network with sigmoid hidden neurons and output neurons was selected. The weight estimates of the friction stir welding parameters are determined by the CRITIC method. For further weight determination between the nodes and edges of the neural networks, the Poisson distribution model was introduced. This stochastic-based method was used to calculate the weights at the edges, between the inputs, hidden layers and outputs of the neural network. The forward pass and backward passes are then used for updating and error minimisation. The welding speed has the highest weight with a contribution of 49.72% using the CRITIC method, implying that welding speed is the best and most influential parameter of the friction stir welding process. For the 4-1-2 neural network architecture, the values of the ultimate tensile strength and percentage elongation at the optimal thresholds are 0.6457 and 0.6019, respectively, for the first forward pass and 0.6123 and 0.6356, respectively, for the second forward pass. The predicted tensile strength is 320.64 MPa and the prediction for the percentage elongation is 18.83%. The results obtained from the proposed method could be useful for planning purposes during the friction welding process.

The need for environmentally friendly and affordable building materials has sparked researches for materials to use in the construction industry. This paper therefore, studied the characterization of ceiling boards manufactured from plaster of Paris (POP) reinforced with banana fibres (BF) and coconut shell (CS) particulates. The samples were produced by adding different ratios of BF and CS to the POP matrix. To evaluate the effectiveness of the samples, several tests, were conducted. The results demonstrated that the addition of reinforcements improved the properties. The sample with a mixture of POP, BF, and CS demonstrated superior properties with the lowest water absorption (2.77%), high compressive strength (7.74 MPa), least thermal conductivity (0.2157 kW/mK) and hardness value of (23.2 HVN), these are within the standard range. The Scanning Electron Microscopy (SEM) analysis confirmed the results obtained. This study established the possibility of using local accessible materials to produce high quality ceiling boards.

Keratoconus is a disease that ML has contributed much in its diagnosis and management. It is not a widely prevalent disease, with a research gap caused by the absence of standardized datasets for model training and evaluation. This work presents a novel dataset, which strengthens the CNN model's resilience and creates standards for assessing keratoconus diagnostic techniques. The research depends on data of patients examined at Jenna Ophthalmic Center in Baghdad. The proposed system works on three stages: pre-processing, feature extraction, and classification with machine learning algorithms including NB, KNN, ADA, DT, and CNN deep learning. The pre-processing stage involves cropping images to retain the relevant maps, which were subjected to contrast enhancement to improve image quality. The pre-processed data is then fed into Machine Learning(ML) algorithms and Convolutional Neural Network(CNN) models, by which the four corneal maps were analyzed. The precision of the ML method was quantified, yielding a precision score of 0.79 for the AdaBoost algorithm and an impressive score of 0.99 for the suggested CNN model exemplifying its high accuracy and ability to surpass all machine learning approaches. Applying PCA for feature extraction before utilizing tradition ML algorithms and CNN helps in achieving high-accuracy results.