Misan Journal of Engineering Sciences

This paper relates to the investigation of the amount of seepage discharge and the water
level height out of the central core in the zoned earth-fill dam using the finite element procedure by
Geo Studio-SEEP/W software. The investigation models were carried out in two shapes of the central
core, the rectangular and the wedge shapes, by 200 miscellaneous models. For each model, seepage
discharge and height of water level out were determined. Investigation results were examined using
the artificial neural network technique (ANN) to demonstrate the influence weight of each independent
variable (geometry and characteristics) on the output-dependent variables results using the statistical
software SPSS. Core material hydraulic conductivity had the highest impact on seepage discharge and
water level height while dam upstream slope had the least impact on them. Two statistical empirical
equations were developed using multiple nonlinear regression for both the seepage discharge and the
height of water level out using the SPSS software. The recommended equations were also verified by
comparing their results with the results of 25% of the models analyzed by Geo Studio software. They
showed great agreement that the coefficient of determination (r2
) was 97.3% for the seepage discharge
equation and 96.6% for the height of the water level-out equation.

When designing structures to withstand explosions, the main goals are to minimize the number and extent
of occupant injuries and to reduce the chance of catastrophic damage to structures. Although there is uncertainty in the
source, extent, and location of explosions, the assessment of blast loading and structural performance is important when
designing blast-resistant structures. This study is a review of the literature on the prediction of blast loads, structural
modeling and analysis, and design criteria for structures to resist explosions. The paper provides in one concise
document the general guidelines, references, and tools that structural engineers and researchers need to analyze and
design structures subjected to blast loading. References on the topics discussed in this work are provided for more detail.

The prediction of a photovoltaic (PV) energy production over time is considered the major challenge to
integrate it with the utilized grid. This is because it is affected by many factors, including geographical locations and
weather conditions. Hence, accurately forecasting PV generation is crucial for ensuring grid stability. In this paper,
several studies are discussed between 2015 to 2023 based on various term forecasting conditions. Then, a neural
network (NN) is employed to forecast a medium-term PV power generation by gathering meteorological data
specific to the MisanGovernorate in the southern region of Iraq. The proposed NN model based on Python language
is trained to find the PV power prediction for various seasons of the year using solar radiation and surrounding
temperature of weather condition tests. At the same time, the MATLAB Simulink model is designed to address the
PV actual power for the same tests. Finally, various statistical measures were introduced to assess the network's
performance, including RMSE, MSE, R2, and MAE. The results demonstrate that the model can predict accurately,
with an RMSE value of 0.092785 kW in September, which is 39.1% lower than in January, 34.14% lower than in
March, and 22.022% lower than in July. Overall, this study contributes to the advancement of solar power forecasting
techniques and highlights the potential of NN in optimizing renewable energy integration into the grid. The average
power generated by the NN in the ninth month was approximately 565 watts, closely resembling the actual value of
574 watts.

This study presents a comprehensive review of the flexural behaviour of corroded Reinforced Concrete
(RC) beams subjected to sustained loads. The investigation synthesizes extant literature to elucidate the complex
interaction between concurrent stress and steel corrosion in RC members. Emphasis is placed on the critical necessity of
conducting corrosion tests under continuous stress conditions to accurately simulate in-situ structural behaviour. The
review encompasses previous numerical studies on deteriorated RC beam performance and provides a critical analysis
of historical loading regimes designed to mitigate corrosion in loaded RC beams. Notably, the literature reveals
conflicting findings regarding the influence of loading on corrosion rates and crack propagation, highlighting areas
necessitating further research. The review also considers the effects of creep, shrinkage, and stress, with particular
emphasis on long-term deflection characteristics. This comprehensive analysis aims to consolidate current knowledge
and identify critical research gaps in understanding the flexural behaviour of corroded RC beams under sustained loads,
thereby providing a foundation for future investigations in this domain.

Classification of sensor data is applied in several domains and enables application tasks
such as fault detection, event recognition, and predictive maintenance. This paper proposes an
application of Multi-Layer Perceptron networks for the classification of noisy and imbalanced sensor
data. In contrast to conventional methods, the proposed approach addresses imbalanced classes, noisy
signals, and high-dimensional data using state-of-the-art preprocessing techniques and thorough
hyperparameter tuning. Using a dataset collected from a SCITOS G5 mobile robot fitted with 24
ultrasound sensors, this research discusses in depth the intricacies of Multi-Layer Perceptron (MLP)
neural networks for the optimal performance of their architecture and training process.
Experimentation and analysis show that the proposed system yielded a good accuracy of 93.04% on
the test dataset, surpassing traditional techniques such as Support Vector Machines and Logistic
Regression. In addition, thorough evaluation metrics on accuracy, precision, recall, and F1 score
demonstrate the model's MLP efficacy across different classes of movements. In this respect, the study
shows not only the effectiveness of MLP networks in sensor data classification but also best practices
in handling challenges that come with the handling of sensor data.

In automation, reliability in robotic grasping in dynamic environment is still a problem
encountered. Further, there is the need to consider deep learning methods, as traditional approaches are
not easily flexible in dealing with different objects and situations. In this work, we aim to analyze how
well deep neural network models perform in predicting grasp strength based on data collected from the
Smart Grasping Sandbox simulation trials. Hence, the proposed approach for analyzing the joint
positions, velocities and efforts led to the design of a deep neural network for improving robot grasp
performance. The question here could is if deep learning could help better predict grasp stability. To
implement the method, we underwent rigorous data pre-processing such as outlier detection, and feature
normalization and employed a structured neural network model for training. Our model got a training
accuracy of nearly 99% and the test accuracy of nearly 96% demonstrating significant promise. These
results were also better than CNN and LSTM where their accuracy rate was 94.12% and 91.81%,
respectively. High deep learning performance was proven in robotic grasping, which can contribute to
the creation of more flexible and sophisticated robotic platforms. This work also provides the foundation
for subsequent research in robotics and automation, with emphasis on the role of data-driven methods
in robotic grasping tasks.

The design of electricity systems is all about ensuring the stable and effective operation
of electrical grids, the capacity to discover and diagnose faults in power transmission lines is the
most important component. to preserve the stability of the power system, save downtime, and avoid
cascading failures. The process of precisely locating the issue and expediting the restoration of the
power supply is known as problem localization, which also helps to minimize the disturbance to
customers. The automatic fault localization and detection using adaptive neural fuzzy inference
systems (ANFIS) method for transmission lines that is measurement-based is presented in this work.
The development and application of ANFIS enables high-speed processing of real-time error
localization and detection. Digital distance protection systems should have the ability to identify
problems in addition to detecting them, according to ANFIS recommendations. After a transmission
encounters a three-phase issue As a result, the recommended approach can identify the affected
stages with accuracy. Numerous types of field data have been used to train and evaluate ANFIS.
Using computer software based on Matlab, the field data are gathered by simulating faults in the
Simulink Matlab model that illustrates the transmission line at various places between Misan – Kut
station 400 Kv along 200 Km. Phase current and voltage data are provided at the busses and are
used as ANFIS inputs. In terms of defect type and detection, the output will show fault and location
identification using simulated processes with a very low error percentage; the results also show that
the approach's selectivity and speed are fairly dependable and offer sufficient performance for
applications involving distribution and transmission monitoring; as well as safety. The study
highlights how crucial it is to identify power transmission line defects quickly and accurately to
maintain the stability and security of the power system. The experimental results show that the two
approaches improve the detection with an accuracy of 100% and the localization reached 99.9% for
( 35 Km) the result of the test program after comparing it with the ANFIS file is (35.00173 Km).

The subject of study known as radio propagation prediction refers to predicting the
behaviour and characteristics of radio waves as they propagate through the atmosphere. It is a basic
element of all wireless communication systems, including satellite communications, broadcasting and
cellular networks. While there is no study covers all the techniques used to predict the radio signal
prediction, this study presents a review of the propagation prediction models between 2018-2023 used
for terrestrial wireless communication systems; the classic empirical models were briefly explained,
followed by the deterministic propagation models that have been developed using ray-tracing with
deep and machine learning techniques. Recent studies on an improvement of the computational
efficiency and accuracy of propagation prediction models were also reviewed, in addition to an
overview of some of the traditional statistical models. Furthermore, some of the new techniques in
propagation prediction were described. The results of these studies explain that techniques using ray
tracing produce better results than other techniques.

Container berths are an essential component of the marine wharf that sustains loads
from various sources, such as self-weight, operation loads, and environment loads. However, the
shipload has more impact on the structure. The container berth in Um Qaser Port is chosen as a
case study. This paper numerically investigates the response of container berth structures under
the impact of dynamic load. Abaqus software is applied to simulate the components of the
structures. The study has dealt with the choice of eight different diameters of the piles, starting
from 800 mm to 1500 mm. The study included three cases of pile diameter. The first case is the
effect of changing the spacing between the piles. The second case is the effect of the pile depth
embedded in the soil. The third case is the change in the thickness of the pile diameter. This
study aims to explore the behavior and performance of the pavement structure by changing some
parameters, including the diameter of the pile, which is considered the prime and significant
portion of the formation of the structure. The changing of pile spacing from 4D to 5D and 6D
leads to an increase in deck displacement by ratio between 44% - 76% and between 100% -
349% respectively, while changing the pile depth from 15 to 18 and 21m leads to decrease deck
displacement between 0.1% - 2% and between 3% - 10% respectively. Also when it changes the
pile thickness from 17 to 20 and 23mm lead to decrease deck displacement range between 9% -
13% and 17% - 24% respectively.