Dijlah Journal of Engineering Sciences

This study addresses the critical issue of risk management in infrastructure
projects within developing countries, where traditional global frameworks
often fall short of addressing local challenges. A significant research gap
exists due to the lack of tailored frameworks that consider the specific
political, financial, environmental, and technical risks inherent in these
contexts and to bridge this gap, a mixed-methods approach was employed,
incorporating case studies, semi-structured interviews, and thematic analysis
to capture insights from projects in Nigeria, Nepal, and Bangladesh and the
resulting framework integrates local context evaluation, cost-benefit-based
risk prioritization, and active stakeholder engagement, demonstrating robust
applicability in mitigating both immediate and long-term risks and the study
makes a dual contribution: it offers practical recommendations for enhancing
project sustainability and provides theoretical insights that advance the field
of risk management. Future research directions include broadening the
study’s geographical and sectoral coverage and incorporating quantitative
methods, such as artificial intelligence-based risk modeling, to further
optimize risk assessment and decision-making processes.

In the (electricity plants, water desalination plants) whose parts are
connected in a row, there are no fixed maintenance procedures for them
because of the happening of faults at different times. To fulfill the research
aims is developed a predictive maintenance program, using A.I. techniques.
Al-fifth Dora's unit data. Two stages of research were conducted. First,
missing data without temporal sequences were processed. The data was
filled using time series hour after hour and consideration filling times as
system working hours, making the volume of data relatively high for 2015
2016-2017, and 2018 was utilized as a test year to assess the modelling work
and validate experimental results. In the second step, the artificial neural
networks approach employed the python program as an A.I., and the affinity
ratio to real data using the performance measurement (Root Mean Square
Error) R.M.S.E (0.507). 2019-2030 showed the program's long-term
predictability.

The present work aims to synthesize and characterize Zeolite using Iraqi
metakaolin and to study their performance in the removal of dye from
wastewater. Techniques such as scanning electron microscope SEM, X-ray
diffractometer XRD, and FTIR Fourier transmission infrared spectrometer
were used to synthesize zeolite characterization. A UV spectrometer
determined zeolite's efficiency in removing RB dye from wastewater. The
X-ray diffraction result for the zeolite prepared showed that the zeolite was
identical in structure to the commercial zeolite Type A. The prepared zeolite
was examined as an adsorbent to remove RB dye from wastewater. The
adsorption Experiment revealed that approximately 13 ppm of dye could be
removed per 2 grams of zeolite, which was deemed unsatisfactory and
compatible with similar studies. Additionally, adsorption isotherms
modelling indicated that the Langmuir model best described the adsorption
behavior, suggesting a monolayer adsorption process.

This paper presents the unified power flow controller (UPFC) in 500
kV / 100 MVA transmission line which is a converter of voltage
sources depend on supple AC systems for series and shunt
improvement between multi substation transmission. Full model
comprises the 48 pulses gate thyristors which is constructed to
become digitalized forms of simulations system to investigate the
dynamics operations of control design. Two series and shunt voltage
source controller for reactive and active power improvements with
voltage stabilizing in electric grids networks. The comprehensive
digital simulations of voltage source operation in shunt condition is
statically synchronizes the compensator of STATCOM controller of
the voltage at the buses and series operation as static series
capacitors SSSC control inject voltages. Kept injected voltages in
quadrature and current inside power systems is achieved in
MATLAB/SIMULINK block sets by means of power systems
blocks. The structure of UPFC controller and electric grids networks
has been modeled through specific block from power systems. The
controlling of shunt with 2-series voltage sources remain in this work
depend on the decouple currents controller scenarios. The model
scheme behavioral connects to 500 kV grid to examine and evaluated
the suggested techniques. The results of simulation provide
important increasing in power flow speed compared with other
approaches which reveal the UPFC devices has excellent capability
to enhance the reactive and real power flows.

In this paper, the study and analysis of universal bridge converter
Based Insulated Gate Bipolar Transistors (IGBT) and pulse width
modulation (PWM) Techniques in SIMULIK Environments is
presents. To realize a PWM as controller with three phase AC/DC
converter, the basic of space vector modulation is investigated and
developed. The progression of this type of conversion in many active
conditions effectively is depending on the applied methods. The
power of proposed procedure approach is located in the current
distortion and switching frequency which is inspected in this work.
The off line calculation in the pulse width first cycle section is
depending and save these data in specific tables. For all cycle life,
the residual pulses are created base on the value of initial quarter
cycle since there are situations of quarter and half wave reliability.
The results of simulation design show a significant economy in
microcontroller time and memory increasing which will support all
converter duty.

With the development of the world the technology of Face recognition with
Artificial Intelligence became used everywhere nowadays in several
applications. This technology analyses and identifies the faces of humans
into footage of videos or digital images through algorithms by using
machine learning techniques. The first step always involves the face
recognition technology, such as facial detection and alignment. It also entails
allocation of relevant facial features and face matching. Although facial
recognition systems offer benefits such as improved security and efficiency
of the system processes, there are ethical issues regarding their potential
abuse, prejudice, and privacy invasion. As a result, there is a need to apply
facial recognition technology while those unchecked factors are mitigated.
Similarly, the use of face recognition technology powered with Artificial
Intelligence (AI) is now commonplace in the sectors of revenue collection,
security, medicine, and even movies. Such major improvements have been
attributed to the enhanced capturing and monitoring of individual’s facial
features to enhance operational productivity and security. The broad
adoption of AI technology has, however, raised serious issues of privacy,
ethical and biased treatment. This research focuses on the advancement of
the Artificial Intelligence -Based Face Recognition Systems, their
application areas, and existing challenges. The results draw attention to the
need for using such technologies responsibly and so that privacy and bias
issues are reduced.

Electric Discharge Machining (EDM) is a non-contact and non-conventional
tool used in industry for machining hard surfaces and high precision
products. After being machined, the EDM process is commonly assessed
based on surface roughness and recast layer (the white layer) on the
product's surface. The electrical discharge phenomenon has been the subject
of numerous investigations and methods, which have been suggested
through advancements and process optimization. The thickness of the white
layer is one of the most important parameters to consider in the
contemporary electrical discharge machining process as it greatly impacts
the surface quality of the specimens that are milling with it during the
process. White layer development is the primary characteristic of EDM, and
it has been examined in the current paper with respect to open circuit
voltage. The desired white layer thickness (WLT) is determined by using
strict control parameters over the cutting process, such as voltage,
wavelength, and other parameters, to ensure uniformity, cracklessness, and
white layer thickness in smoothing. The research focuses on the relationship
between voltage and the thickness of welded layers (WLT) in AISI 444
stainless steel. The tests showed that WLT increased because the EDM
parameters of voltage (140, 240) V, current Ip (12, 24, 50) A, pulse on time
Ton (100, 200, 400) µs, and pulse off time Toff (3, 6.5, 12) µs were
increased. An analysis of accuracy involved four adjustable parameters
consisting of voltage (140, 240) V together with current Ip (12, 24, 50) A
and pulse on time Ton (100, 200) µs. Higher values of voltage, current and
pulse on time result in increased WLT according to test results.

Skin cancer is one of the most common types of cancer, often caused by
prolonged exposure to ultraviolet (UV) radiation. While visual inspection
can provide a preliminary diagnosis, accurate identification typically
requires a biopsy, an invasive, costly, and time-consuming procedure.
Convolutional neural networks (CNNs) offer a promising alternative by
achieving high diagnostic accuracy with reduced time and cost. Despite
numerous studies in this area, the influence of training hyperparameters,
particularly batch size, has not been extensively examined in the context of
skin cancer detection. This study explores the effect of batch size on the
performance of the ResNet18 network, known for its balance between
accuracy and computational efficiency in medical imaging tasks. Batch size,
defined as the number of training samples processed before weight updates,
plays a critical role in model performance. Seven batch sizes (4, 8, 16, 32,
64, 128, and 256) were evaluated. Results revealed that smaller batch sizes
led to lower performance, while a batch size of 32 achieved the highest
multi-class accuracy. Batch sizes of 64 and 128 improved precision and
recall. However, increasing the batch size beyond 128 did not yield
additional benefits. Therefore, batch sizes between 32 and 128 are
recommended for optimal results.

Since heart attacks continue to be a leading cause of mortality globally, these
numbers should encourage scientists to develop more effective methods of
prevention and early detection. Using clinical data as a starting point, we
train and verify a model to forecast the likelihood of a heart attack using
machine learning techniques. Indicators of a patient's health form the basis
of this concept. The best classification model was chosen after extensive
testing and evaluation of many models, such as logistic regression, decision
trees, random forest, support vector machines (SVM), k-nearest neighbours
(kNN), Naïve Bayes, and Extreme Gradient Boosting (XGBoost). Feature
scaling and encoding were applied to all 303 patients in the sample, who
possessed a total of 14 distinct characteristics.
Two sets of data were
created: one for training and one for testing, so that the models could be
tested. For every model, we determined the following metrics: ROC-AUC,
F1-score, recall, accuracy, precision**, and precision. By a wide margin,
XGboost outperformed kNN and SVM classifiers with its 90.2% prediction
accuracy. Furthermore, we were able to train an ensemble voting classifier
that achieved somewhat better results than the top individual model as well
as its individual components. According to feature significance analysis,
the two most essential criteria in determining the likelihood of a heart attack
are the "which type" of chest pain and the "what kind" of exercise-induced
angina.
We go deeply into the models' inner workings, discussing the
consequences of our discoveries and possible future enhancements. By
utilizing machine learning algorithms, medical professionals may improve
their ability to foretell the likelihood of a cardiac event, which might result
in more timely and effective treatment. Future efforts to improve prediction
performance could make advantage of more complete ensemble techniques,
bigger datasets, and other characteristics.

Due to the importance of Sustainable Construction, a comprehensive assessment of the
sustainability of the built environment is required. Studying the sustainability
performance indicators of current buildings is a crucial first step in evaluating a
building's sustainability. This study aims to understand the main performance of
sustainability indicators for public buildings in Iraq as a result of relying on these
buildings intensively for the development and welfare of society in Iraq. Based on
previous studies sustainability indicators were collected and identified, and 71 indicators
were obtained, distributed among a group of categories within the three dimensions of
environmental, social, and economic sustainability. The relative importance of the
indicators was determined by organizing a questionnaire that was distributed to a group
of engineers with experience in sustainable buildings and relied on 103 forms in the
analysis. The results of the survey were analyzed using the statistical program (SPSS-V
26) to give priority to the indicators identified using the relative importance index (RII)to
point out the relative importance of environmental, social, and economic indicators. The
Pareto principle was also applied to reduce the number of indicators and choose the most
relatively important ones.14 key performance indicators for the sustainability of public
buildings were obtained.

Management has emerged as one of the pillars of this field. Given that the
construction sector is one of the main pillars and levers of the economy, the
application of this process plays a significant and broad role. Through this
research, we studied aspects of knowledge management in the construction
sector in Iraq, examining aspects in terms of knowledge creation,
organization, and distribution, as well as the process of knowledge
application. This was conducted in four government-owned companies in
Iraq and we used a questionnaire as a means of gathering information, which
included 44 specialists. We concluded that the strengths of these
government-owned companies lie in the area of knowledge application as
well as the process of knowledge creation and distribution. The weaknesses
that pose a challenge to them are the process of knowledge distribution. We
came up with recommendations to enhance these strengths and address
weaknesses through the use of a fuzzy synthetic assessment.

The “switched reluctance motor” SRM can be operated in many modes
based on the desired application that employs such kinds of motor. The SRM
speed in each mode can be controlled. Thus, it’s better to select the fast
response speed control method with the lowest torque ripples and speed
variations. The aim of the current research is to enhance the SRM speed
control by that works in “hysteresis current control” HCC mode. The SRM
speed is regulated by a “fuzzy logic controller” FLC. The role of the speed
control signal by FLC is compared with the SRM position sensor to produce
a reference current signal. This signal is fed to a hysteresis band to control
the speed of the motor. This method of speed is referred to as HFLC. In this
paper, another approach is made by replacing the hysteresis band with a
hyperbolic tan function for more smooth in system control. This method is
called TanHFLC. The results showed that the speed control in TanHFLC is
faster by 6.2 %, and current and torque ripples are less by 10.9% and 20.8%,
respectively. This work is simulated by the “MATLAB Simulink” software
successfully at several operating conditions.

Friction Stir Welding (FSW) demonstrates superior corrosion resistance in
aluminum alloys compared to conventional techniques like arc and laser
welding, owing to reduced heat-affected zones (3–5 mm) and refined
microstructures. Environmental conditions significantly influence corrosion
behavior, with marine (3.5% NaCl) and acidic (pH 3) environments
accelerating pitting and intergranular corrosion in AA5000 and AA2000
alloys, respectively. Alloy composition plays a critical role: AA6061-T6
(AA6000 series) exhibits lower pitting rates (0.18 mm/year) than copper-rich
AA2024 (0.34 mm/year), while optimized Li/Cu ratios in AA8090 (AA8000
series) reduce exfoliation corrosion by 25%. Mechanical-corrosion synergy
reveals residual stresses and microstructural features, such as low-angle
grain boundaries, directly impact stress corrosion cracking (SCC) and pitting
resistance,Industrial applications in aerospace, marine, and infrastructure
highlight FSW’s advantages, including 30% lower SCC risk in AA7000
fuselage panels and 50% extended service life for AA5083 ship hulls.
Enhanced predictive modeling, integrating environmental variables, aligns
with experimental data, supporting optimized welding parameters and post
weld treatments like laser peening and anodizing,This study underscores the
interplay between alloy design, welding parameters, and environmental
adaptation to enhance durability in critical engineering applications.

The network computing device that work and communicated without human
interference is referred to term of “Internet of Things” (IoT). Currently, this
technology is the utmost excite area of computing with its application in
numerous areas like city, home, infrastructures, hospital, and transportations.
The security issue surrounds IoTs device increased as they develop. In order
to addressed this issue, this paper present a new idea for enhance the security
of IoTs system by use machines learning (MLs) classifier. The suggested
methods analyze latest technology, security, intelligent solution, and
vulnerability in MLs IoTs-base intelligent system as an vital technologies to
develop IoTs security. This paper illustrate the benefit and limitation of
apply the MLs in an IoTs environments and provide a security models
depend on MLs that manage originally the increasing numbers of security
issue associated to the IoT domains. In addition, this approach suggests an
ML-base security models that independently handle the rising numbers of
security issue related to the IoT area. This investigation introduced a
significant contributions by developed a cyber-attacks recognition solutions
for IoTs device by use machine learning algorithms. Many ML algorithms
has been used to classify the greatest accurate classifier for their AI-base
reactions agent implementations stage, which could recognize attacks
activity and pattern in network connecting to the IoTs. The suggested
approach realized 99.98% accuracies, 99.97% detections, and 99.93 F1
scores, compare to the current methods. Also, this paper highlight the
outperforming previous ML-based model in term of implementation speeds
and accuracies and proves that the proposed method outperform preceding
ML-based model in performance accuracy and time.

This work is concerned to the welding process for joining pipes made
from alloy steel which is widely used in heat exchangers structures.
The chemical composition analysis of this steel type close to 213 Gr
T5 according to ASTM standard. This work shows that this metal has
a good weldability and high level of weld metal integrity. Weld filler
type ER 502 of 2.4 mm diameter is used in this welding method to
join sections of pipes by using Tungsten Inert Gas welding (TIG)
process. This investigation is covered the approving of weld integrity
assessments and mechanical prosperities by doing a number of tests
for weld joint, for example, tensile testing, bending, hardness, x-ray
and Liquid penetrant tests). Eventually it was confirmed that, the weld
metal has a reasonable joint mechanical property and no series weld
defects.