حصاد البحث العلمي

Dr. Abdelhameed Fawzy Ibrahim

Associate Professor
Computer Engineering

afibrahim@uom.edu.sa
EXT: 3100

Sunshine duration is an important atmospheric indicator used in many agricultural, architectural, and solar energy applications (photovoltaics, thermal systems, and passive building design). Hence, it should be estimated accurately for areas with low-quality data or unavailable precise measurements. This paper aimed to obtain a sunshine duration measurement database in Algeria’s south region and also to study the applicability of computational models to predict them. This work develops ensemble learning models for assessing daily sunshine duration with meteorological datasets that include daily mean relative humidity, daily mean air temperature, daily maximum air temperature, daily minimum air temperature, and daily temperature range as input. The study proposes a unique hybrid model, combining grey wolf and stochastic fractal search (GWO-SFS) optimization algorithms with the random forest regressor ensemble. A pre-feature selection process improved the newly suggested model. Various commonly adopted algorithms in relevant studies have been considered as references for evaluating the new hybrid algorithm. The accuracy of models was examined as a function of some frequently used statistical pointers, as well as the Wilcoxon rank-sum test. Besides, the models were evaluated according to the several input combinations. The numerical experiments show that the proposed optimization ensemble with feature preprocessing outperforms stand-alone models in terms of prediction accuracy and robustness, where relative root mean square errors are reduced by over 20% for all considered locations. In addition, all correlation coefficients are higher than 0.999. Moreover, the proposed model, with RMSEs lower than 0.4884 hours, shows significantly superior performances compared to previously proposed models in the literature.

Journal: Theoretical and Applied Climatology

ISI, Q2, Scopus

The development and deployment of an effective wind speed forecasting technology can improve the safety and stability of power systems with significant wind penetration. Due to the wind’s unpredictable and unstable qualities, accurate forecasting of wind speed and power is extremely challenging. Several algorithms were proposed for this purpose to improve the level of forecasting reliability. The Long Short-Term Memory (LSTM) network is a common method for making predictions based on time series data. This paper proposed a machine learning algorithm, called Adaptive Dynamic Particle Swarm Algorithm (AD-PSO) combined with Guided Whale Optimization Algorithm (Guided WOA), for wind speed ensemble forecasting. The AD-PSO-Guided WOA algorithm selects the optimal hyperparameters value of the LSTM deep learning model for forecasting of wind speed. In experiments, a wind power forecasting dataset is employed to predict hourly power generation up to forty-eight hours ahead at seven wind farms. This case study is taken from the Kaggle Global Energy Forecasting Competition 2012 in wind forecasting. The results demonstrated that the AD-PSO-Guided WOA algorithm provides high accuracy and outperforms several comparative optimization and deep learning algorithms. Different tests’ statistical analysis, including Wilcoxon’s rank-sum and one-way analysis of variance (ANOVA), confirms the accuracy of the presented algorithm.

Journal: IEEE Access

ISI, QI, Scopus

As research in alternate energy sources is growing, solar radiation is catching the eyes of the research community immensely. Since solar energy generation depends on uncontrollable natural variables, without proper forecasting, this energy source cannot be trusted. For this forecasting, the use of machine learning algorithms is one of the best choices. This paper proposed an optimized solar radiation forecasting ensemble model consisting of pre-processing and training ensemble phases. The training ensemble phase works on an advanced sine cosine algorithm (ASCA) using Newton’s laws of gravity and motion for objects (agents). ASCA uses sine and cosine functions to update the agent’s position/velocity components by considering its mass. The training ensemble model is then developed using the k-nearest neighbors (KNN) regression. The performance of the proposed ensemble model is measured using a dataset from Kaggle (Solar Radiation Prediction, Task from NASA Hackathon). The proposed ASCA algorithm is evaluated in comparison with the Particle Swarm Optimizer (PSO), Whale Optimization Algorithm (WOA), Genetic Algorithm (GA), Grey Wolf Optimizer (GWO), Squirrel Search Algorithm (SSA), Harris Hawks Optimization (HHO), Hybrid Greedy Sine Cosine Algorithm with Differential Evolution (HGSCADE), Hybrid Modified Sine Cosine Algorithm with Cuckoo Search Algorithm (HMSCACSA), Marine Predators Algorithm (MPA), Chimp Optimization Algorithm (ChOA), and Slime Mould Algorithm (SMA). Obtained results of the proposed ensemble model are compared with those of state-of-the-art models, and significant superiority of the proposed ensemble model is confirmed using statistical analysis such as ANOVA and Wilcoxon’s rank-sum tests.

Journal: IEEE Access

ISI, QI, Scopus

Dr. Zeeshan Shafi Khan

Assistant Professor
Computer Science

zskhan@uom.edu.sa
EXT: 3100

As research in alternate energy sources is growing, solar radiation is catching the eyes of the research community immensely. Since solar energy generation depends on uncontrollable natural variables, without proper forecasting, this energy source cannot be trusted. For this forecasting, the use of machine learning algorithms is one of the best choices. This paper proposed an optimized solar radiation forecasting ensemble model consisting of pre-processing and training ensemble phases. The training ensemble phase works on an advanced sine cosine algorithm (ASCA) using Newton’s laws of gravity and motion for objects (agents). ASCA uses sine and cosine functions to update the agent’s position/velocity components by considering its mass. The training ensemble model is then developed using the k-nearest neighbors (KNN) regression. The performance of the proposed ensemble model is measured using a dataset from Kaggle (Solar Radiation Prediction, Task from NASA Hackathon). The proposed ASCA algorithm is evaluated in comparison with the Particle Swarm Optimizer (PSO), Whale Optimization Algorithm (WOA), Genetic Algorithm (GA), Grey Wolf Optimizer (GWO), Squirrel Search Algorithm (SSA), Harris Hawks Optimization (HHO), Hybrid Greedy Sine Cosine Algorithm with Differential Evolution (HGSCADE), Hybrid Modified Sine Cosine Algorithm with Cuckoo Search Algorithm (HMSCACSA), Marine Predators Algorithm (MPA), Chimp Optimization Algorithm (ChOA), and Slime Mould Algorithm (SMA). Obtained results of the proposed ensemble model are compared with those of state-of-the-art models, and significant superiority of the proposed ensemble model is confirmed using statistical analysis such as ANOVA and Wilcoxon’s rank-sum tests.

Journal: IEEE Access

ISI, Q2, Scopus

Dr. Nacereddine Hammami

Assistant Professor
Computer Engineering

nshammami-t@uom.edu.sa
3119

“This paper discusses and provides some analytical studies for a modified fractional-order SIRD mathematical model of the COVID-19 epidemic in the sense of the Caputo–Katugampola fractional derivative that allows treating of the biological models of infectious diseases and unifies the Hadamard and Caputo fractional derivatives into a single form. By considering the vaccine parameter of the suspected population, we compute and derive several stability results based on some symmetrical parameters that satisfy some conditions that prevent the pandemic. The paper also investigates the problem of the existence and uniqueness of solutions for the modified SIRD model. It does so by applying the properties of Schauder’s and Banach’s fixed point theorems. View Full-Text
Keywords: pandemic; COVID-19; SIRD model; fractional derivative; system; existence; uniqueness”

Journal: Symmetry

ISI, QI, Scopus

Dr. Omar Alaqeeli

Assistant Professor
Computer Science

omalaqeeli@uom..edu.sa
3120

Classification tree is a widely used machine learning method. It has multiple implementations as R packages; rpart, ctree, evtree, tree and C5.0. The details of these implementations are not the same, and hence their performances differ from one application to another. We are interested in their performance in the classification of cells using the single-cell RNA-Sequencing data. In this paper, we conducted a benchmark study using 22 Single-Cell RNA-sequencing data sets. Using cross-validation, we compare packages’ prediction performances based on their Precision, Recall, F1-score, Area Under the Curve (AUC). We also compared the Complexity and Run-time of these R packages. Our study shows that rpart and evtree have the best Precision; evtree is the best in Recall, F1-score and AUC; C5.0 prefers more complex trees; tree is consistently much faster than others, although its complexity is often higher than others.
Keywords: classification tree; single-cell RNA-Sequencing; benchmark; precision; recall; F1-score; complexity; Area Under the Curve; run-time.

Journal: Microbiology Research

ISI, Scopus

Dr. Mohammed Khodja

Assistant Professor
Electronics engineering and Communication

makhodja-t@uom.edu.sa
3105

In this paper, a new optimization algorithm called motion-encoded electric charged particles optimization (ECPO-ME) is developed to find moving targets using unmanned aerial vehicles (UAV). The algorithm is based on the combination of the ECPO (i.e., the base algorithm) with the ME mechanism. This study is directly applicable to a real-world scenario, for instance the movement of a misplaced animal can be detected and subsequently its location can be transmitted to its caretaker. Using Bayesian theory, finding the location of a moving target is formulated as an optimization problem wherein the objective function is to maximize the probability of detecting the target. In the proposed ECPO-ME algorithm, the search trajectory is encoded as a series of UAV motion paths. These paths evolve in each iteration of the ECPO-ME algorithm. The performance of the algorithm is tested for six different scenarios with different characteristics. A statistical analysis is carried out to compare the results obtained from ECPO-ME with other well-known metaheuristics, widely used for benchmarking studies. The results found show that the ECPO-ME has great potential in finding moving targets, since it outperforms the base algorithm (i.e., ECPO) by as much as 2.16%, 5.26%, 7.17%, 14.72%, 0.79% and 3.38% for the investigated scenarios, respectively.

Journal: Sensors

ISI, Q2

In this article, a type-2 fuzzy interval controller is proposed to solve the nonlinear control problems of semi-active suspension system. A suspension model with two degrees of freedom and A fuzzy approach for controller synthesis were proposed. The performance of the IT2FLC-based semi active vehicle suspension system in terms of sprung mass displacement, suspension deflection and tire deflection are compared to the homologous fuzzy type-1 controller (T1FLC), and to the passive suspension system conventional using MATLAB / SIMULINK software for simulation and controller design. The vehicle parameters, called suspension deflection and speed of suspended mass are given as inputs for both controllers. The C semi control signal is the variable damping coefficient. Inputs and outputs are presented by triangular membership functions. Mamdani inference system is used, along with a Karnik-Mendel algorithm to locate the center of gravity in reduction type for IT2FLC controller. Simulation results show that IT2FLC-based semi-active suspension system outperforms T1FLC and passive suspension system. Thus, they show a major improvement in control signal i.e. IT2FLC controller generates a lower damping coefficient than T1FLC controller. In addition, a remarkable reduction in signal energy by IT2FLC compared to
same semi-active suspension system with T1FLC.

Journal: Turkish Journal of Computer and Mathematics Education (TURCOMAT)

Q3, Scopus

Fuzzy gain-scheduled PID (FuGSPID) controllers have attracted significant interest in contemporary research. This paper provides mathematical description help to reduce the code program for a Fuzzy gain-scheduled controller FuGSPID that is subsequently used to stabilize the altitude of a home-constructed Quadcopter. The subject particle swarm optimization approach was employed to optimize the fuzzy output sets of the controller. MATLAB was used to generate the dynamic model, the FuGSPID, and the optimization. A simulation exercise revealed that the new parameters employed in the FuGSPID that were generated as a result of the particle swarm optimization produced fewer trajectory tracking errors. Fuzzy systems are required to process large volumes of tasks and processes. This
subsequently impedes the performance of the microcontroller. In light of this, this paper outlines a mathematical description that can potentially reduce the algorithm code employed in the FuGSPID controller and, thereby, making it more intuitive and reducing the processing speed of the microcontroller and reducing the sampling time of the controller and the whole flight controller. The findings of this study revealed that the mathematical description it be
useful to reduce instruction of fuzzy controller program to implement it in low cost microcontroller and tested effectively for a quadcopter altitude stabilization.

Journal: International Journal of Intelligent Engineering and Systems

Q3, Scopus

Dr. Tarig Ibrahim Ahmed

Assistant Professor
Computer Engineering

tiahmad@uom.edu.sa
114

To summarize, there are basic functions / techniques that have been implemented to ensure that the prototype is used to achieve its primary objectives of reducing the rates of false alarms and their sub-targets being shipped and making fire detection devices “smart”. They are as follows:
1. Use of smoke, temperature and light intensity as fire detection parameters.
2. Use GSM instead of Wi-Fi to send text notifications.

Journal: International Journal of Engineering Research and Applications (IJERA)

Dr. Yunus Ibrahim Gali

Assistant Professor Computer Science

yigali-t@uom.edu.sa

The study amid to analyze the time series in forecasting the movement of the close price of Khartoum stock exchange (KSE), this is done by using Box-Jenkins models and Artificial Neural Networks models, As the comparisons between the two methods and select the best method which are sufficient to represent the statement of the time series and apply it for forecasting.
ARIMA (1, 1,0) was used to build Box-Jenkins model, it confirmed that this model is very good and gives accurate and real forecasts by calculating the Q=Stat, which turned out to be insignificant.
Forecasts were made for the close price in Khartoum stock exchange, where the model was estimated for the period from (3/1/2018 – 31/12/2021).
The graphs showed that there is a congruence between the actual values and estimated values, which is lead to the efficiency of the Box- Jenkins model.
Also the neural networks models were used to forecast the movement of the close price in Khartoum stock exchange, where the multi layers Alberseptron network method was used to build network models for the data study.
The process of partitioning was done randomly into (Preprocessing- Design – training – testing and query).
The logistic function was used as a motivating function in the hidden layer and output layer, a rapid propagation algorithm was used to train this network and it turned out that the resulting network was good and give accurate and real forecasts for the period (2018- 2021).

Journal: Islamic World reseach & Studies Institute 

Mr. Jawad Fathi Abusalama

Lecturer
Computer Engineering

jfabusalama@uom.edu.sa
3116

When a disaster occurs, the single agent does not have complete knowledge about the situation of the disaster. Therefore, the rescue agents should coordinate with each other to perform their allocated tasks efficiently. However, the task allocation process among rescue agents is a complex problem. It is NP-complete, and determining the rescue agents that will perform the tasks efficiently is the main problem, called the Winner Determination Problem (WDP). This paper proposed a new approach to improve rescue agents’ tasks allocation processes for WDP in reverse combinatorial auctions. The main objective of this paper is to propose a rescue agents task allocation algorithm by investigating a new approach to determine the winning bids that will perform the corresponding tasks with minimum cost. The main contribution of the proposed algorithm is to apply a recursion method to shorten many complex steps when determining the winners and allocates the corresponding tasks. The proposed algorithm was compared to Andrea’s algorithm regarding complexity and running time, and the results showed that the proposed algorithm performs slightly better than Andrea’s algorithm.

Journal: The Third International Conference on Computer and Information Sciences 2021 (ICCIS 2021)

IEEE conference

Mrs. Naseem Ahmad Alrobah

Lecturer
Computer Engineering

Naalrobah-t@uom.edu.sa
3214

In recent times, many research projects and experiments target machines that automatically recognize handwritten characters, but most of them are done in Latin. Recognizing handwritten Arabic characters is a complicated process compared to English and other languages as a nature of Arabic words. In the past few years, deep learning approaches have been increasingly used in the field of Arabic recognition. This paper aims to categorize, analyze and presents a comprehensive survey in Arabic handwritten recognition literature, focusing on state-of-the-art methods for deep learning in feature extraction. The paper focuses on offline text recognition, with a detailed discussion of the systematic analysis of the literature. Additionally, the paper is critically analyzing the current literature and identifying the problem areas and challenges faced by the previous studies. After investigating the studies, a new classification of the literature is proposed. Besides, an analysis is performed based on the findings, and several issues and challenges related to the recognition of Arabic scripts are discussed.

Journal: Arabian journal for science and engineering

ISI, Q2

Mrs. Atheer Fahad Almansour

Demonstrator
Computer Science

Afalmansour@uom.edu.sa

Machine learning, images classification

Journal: ICCIT

ISI, Scopus