Browsing by Author "Bendjeghaba, Omar"

Now showing 1 - 16 of 16

Accelerated modified sine cosine algorithm for data clustering
(IEEE, 2021) Boushaki, Saida Ishak; Bendjeghaba, Omar; Brakta, Noureddine
In artificial intelligence, data mining is a process that automatically discover valuable information from huge amounts of data in order to obtain knowledge. The most important unsupervised technique of data mining is the clustering technic, which his main task is dividing the dataset into homogeneous groups. Metaheuristics based clustering is an actual research area where optimization algorithms have demonstrated their efficiencies to provide near optimal solutions to this problem in a reasonable time, including the recent Sine Cosine metaheuristic Algorithm (SCA). However, its convergence rate is still rather slow. In this paper, an upgraded adaptation of SCA is proposed to improve the exhibition capacities of the quest strategy for ideal results for data Clustering problem, named AMSCAC. In this algorithm, both the local and global search procedures are enhanced by additional strategy. The experimental results on five standard datasets are promising and confirm the superiority of AMSCAC, for the clustering results over SCA, cuckoo search algorithm (CS), differential evolution algorithm (DE), and genetic algorithm (GA)
Big data clustering based on spark chaotic improved particle swarm optimization
(Institute of Advanced Engineering and Science (IAES), 2024) Boushaki, Saida Ishak; Mahammed, Brahim Hadj; Bendjeghaba, Omar; Mosbah, Messaoud
In recent years, the surge in continuously accelerating data generation has given rise to the prominence of big data technology. The MapReduce architecture, situated at the core of this technology, provides a robust parallel environment. Spark, a leading framework in the big data landscape, extends the capabilities of the traditional MapReduce model. Coping with big data, especially in the realm of clustering, requires more efficient techniques. Meta-heuristic-based clustering, known for offering global solutions within reasonable time frames, emerges as a promising approach. This paper introduces a parallel-distributed clustering algorithm for big data within the Spark Framework, named Spark, chaotic improved PSO (S-CIPSO). Centered on particle swarm optimization (PSO), the proposed algorithm is enhanced with a chaotic map and an efficient procedure. Test results, conducted on both real and artificial datasets, establish the superior performance and quality of clustering results achieved by the proposed approach. Additionally, the scalability and robustness of S-CIPSO are validated, demonstrating its effectiveness in handling large-scale datasets.
Conception optimale des systèmes de production par une approche bésée sur une système colonie de fourmis artificielles
(2010) Bendjeghaba, Omar
This work is devoted to the dispatching of the reactive power and the maintain of voltage profile in the electrical network power systems in steward state by a hybrid approach. Combining heuristic and numerical technique has been chosen. The considered approach deals with the security problem, mainly the elimination of voltage constraint voilation disturbances. Firstly, the effect of the reactive power transmission upon the voltage drop and the active power losses has been analysed. Two important proprieties have been extracted; these will be used in the devlopment of the approach…
Continuous firefly algorithm for optimal tuning of PID controller in AVR system
(2014) Bendjeghaba, Omar
This paper presents a tuning approach based on Continuous firefly algorithm (CFA) to obtain the proportional-integralderivative (PID) controller parameters in Automatic Voltage Regulator system (AVR). In the tuning processes the CFA is iterated to reach the optimal or the near optimal of PID controller parameters when the main goal is to improve the AVR step response characteristics. Conducted simulations show the effectiveness and the efficiency of the proposed approach. Furthermore the proposed approach can improve the dynamic of the AVR system. Compared with particle swarm optimization (PSO), the new CFA tuning method has better control system performance in terms of time domain specifications and setpoint tracking
Dips detection techniques discussion
(IEEE, 2021) Brakta, Noureddine; Bendjeghaba, Omar; Yazid Zidani, Mohamed Yazid
In today's industrial world, renewable energy represents 16% of the total produced and consumed energy. Wind energy is gaining popularity and imposing bigger footprint in the natural sources of electricity and plays a key building block for economic recovery from the impact of COVID-19 according to the Global wind Energy Council. Yet, because its presence in the energy resource share is below 5%, since most of the wind farms use Doubly Fed Induction Generator (DFIG) it will suffer from grids faults such as symmetric dips. The scary part of the dips is that it will increase the currents at the back-to-back converters and even when dealing with them oscillation might destroy the turbine and generator assembly. In this paper we will review techniques used to detect the dips and override the heavy increase of the current using a crowbar, we will concentrate only on the detection and compare the resulting parameters while keeping the controllers and the value of the crowbar resistance unchanged
Document clustering analysis based on hybrid cuckoo search and K-means algorithm
(IEEE, 2021) Boushaki, Saida Ishak; Bendjeghaba, Omar; Brakta, Noureddine
The clustering is an interesting technique for unsupervised document organization in the World Wide Web (WWW). The most widely used partitioning clustering algorithm is K-means. However, it has an issue with random initialization, which might lead to local optimum situations. In fact, metaheuristics-based clustering has demonstrated their efficiency to reach a global solution instead of local one. The Cuckoo search (CS) has been widely used for the clustering problem. However, the number of iterations grows dramatically when the dataset is high dimensional like the documents. In this study, the hybridization cuckoo search and K-means algorithms for the document clustering are analyzed. So, three hybrid algorithms are investigated and compared. The performance and the efficiency of the proposed algorithms are evaluated using Reuters 21578 Text Categorization Benchmark Dataset. The obtained results show the capability of the new approaches to generate more compact clustering and enhancing purity and F-measure clustering qualities
Dynamic Performance Improvement of DFIM based on Hybrid Computational Technique
(IEEE, 2021) Zidani, Mohamed Yazid; Brakta, Noureddine; Bendjeghaba, Omar
This paper presents a hybrid intelligent nonlinear control, based on particle swarm optimization (PSO) technique and artificial intelligence controller (AI) to improve the dynamic performance of the system. These controllers are destined for the speed control of Doubly Fed Induction Motor (DFIM). The proportional-integral controller for speed regulation of the induction motor is the most extensively used controller. However, given the various operating conditions and the nature of parameter variability, the PI controller has some drawbacks. So, one of the frequently discussed applications of artificial intelligence (AI) in control is the replacement of a proportional integral speed controller with Artificial Neural Network (ANN) speed controller but the choice of the gain’s parameters controller is one of the main problems. So, Particle Swarm Optimization (PSO) technique on optimization performance is added to the PI and ANN controllers to find the best gain values. The simulation results for different scenarios illustrate the high performance of the proposed artificial intelligence controller for DFIM running at variable speeds in terms of consistency and stability
High-Dimensional Text Datasets Clustering Algorithm Based on Cuckoo Search and Latent Semantic Indexing
(World Scientific Publishing Co. Pte Ltd, 2018) Ishak Boushaki, Saida; Kamel, Nadjet; Bendjeghaba, Omar
The clustering is an important data analysis technique. However, clustering high-dimensional data like documents needs more effort in order to extract the richness relevant information hidden in the multidimensionality space. Recently, document clustering algorithms based on metaheuristics have demonstrated their efficiency to explore the search area and to achieve the global best solution rather than the local one. However, most of these algorithms are not practical and suffer from some limitations, including the requirement of the knowledge of the number of clusters in advance, they are neither incremental nor extensible and the documents are indexed by high-dimensional and sparse matrix. In order to overcome these limitations, we propose in this paper, a new dynamic and incremental approach (CS_LSI) for document clustering based on the recent cuckoo search (CS) optimization and latent semantic indexing (LSI). Conducted Experiments on four well-known high-dimensional text datasets show the efficiency of LSI model to reduce the dimensionality space with more precision and less computational time. Also, the proposed CS_LSI determines the number of clusters automatically by employing a new proposed index, focused on significant distance measure. This later is also used in the incremental mode and to detect the outlier documents by maintaining a more coherent clusters. Furthermore, comparison with conventional document clustering algorithms shows the superiority of CS_LSI to achieve a high quality of clustering.
Hybrid approach for redundancy optimization of multi-state power system
(2007) Bendjeghaba, Omar; Ohahdi, Dris; Zablah, Abdelkader
This paper presents an optimize method of the well known Redundancy Optimization Problem (ROP) using a combined ANT Colony System (ACS) and Universal Moment Generating Function (UMGF). The ACS searches for the minimum cost solution by selecting the appropriate components for a series-parallel system, given a minimum system reliability constraint. The UMGF is used to estimate the system reliability value during search. This approach is an example of a computationally efficient method to apply ACS optimization to problems for which repeated calculation of the objective function is impractical
Improved Cuckoo Search Algorithm for Document Clustering
(SPRINGER, 2015) Ishak Boushak, Saida; Kamel, Nadjet; Bendjeghaba, Omar
Efficient document clustering plays an important role in organizing and browsing the information in the World Wide Web. K-means is the most popular clustering algorithms, due to its simplicity and efficiency. However, it may be trapped in local minimum which leads to poor results. Recently, cuckoo search based clustering has proved to reach interesting results. By against, the number of iterations can increase dramatically due to its slowness convergence. In this paper, we propose an improved cuckoo search clustering algorithm in order to overcome the weakness of the conventional cuckoo search clustering. In this algorithm, the global search procedure is enhanced by a local search method. The experiments tests on four text document datasets and one standard dataset extracted from well known collections show the effectiveness and the robustness of the proposed algorithm to improve significantly the clustering quality in term of fitness function, f-measure and purity.
Improved cuckoo search algorithm for document clustering
(Springer, 2015) Boushaki, Saida Ishak; Kamel, Nadjet; Bendjeghaba, Omar
Maintien des tensions et gestion de la puissance de la réactive dans les réseaux de transport par une approche hybride
(2003) Bendjeghaba, Omar
A new hybrid algorithm for document clustering based on cuckoo search and K-means
(Springer, 2014) Ishak Boushaki, Saida; Nadjet, Kamel; Bendjeghaba, Omar
A novel fault location approach for radial power distribution systems
(Institute of Advanced Engineering and Science, 2022) Hamdouche, Tarek; Bendjeghaba, Omar; Brakta, Noureddine; Ahriche, Aimad
Power distribution systems (PDS) are increasingly complex and spread over long distances and in different locations. Given their radial configuration, the loads could be inserted at the same distances from the substation. This leads to multiple estimation of fault location (FL) and yields time consuming for iterative FL algorithms. In this article, we provide a novel practical approach to fault localization in order to defeat these limitations. The central idea of the proposed approach is to divide the multilateral distribution system into a possible number of mono-lateral sub systems (MLS) using a proposed communicating sensor. Next, we apply two different fault locator algorithms only to the defective MLS. The first variant of the approach is based on the impedance method, while the second variant is non-parametric used only when there is lack in the line data. To test the proposed technique in practice, we used the IEEE 13 Node test feeder, and a real Algerian PDS. The results obtained clearly show the contribution of the dedicated method for locating faults in the PDS
Optimal Sizing and Localization of Multiple Distributed Generations in Distribution Systems Using an Improved Grey Wolf Optimization Algorithm
(2024) Benahcour, Souheyla; Bendjeghaba, Omar
This study investigates the impact of the localization and sizing of distributed generations in distribution systems using a combined approach of improved grey wolf optimizer (IGWO) and Newton-Raphson load flow algorithms. The suggested method optimizes the size and position of distributed generation generating both real and reactive power while ensuring power system constraints are not violated. The suggested algorithm optimizes the location and sizing of dis-tributed generations. Nevertheless, investigations show that the proposed method outperforms the PSO optimizer and takes less calculation time. Moreover, in contrast with other meta-heuristic algorithms such as JAYA, PSO, SFO, BO, SMA, GA, and GJO, the proposed approach produces a better voltage profile of the distribution system with smaller distributed generator sizes. To demonstrate the advantages of the suggested approach, the IEEE-13, IEEE-37, and IEEE-123 bus distribution systems are used as test cases, and the outcomes are contrasted with those of other meta-heuristic methods. According to simulation data, IGWO outperforms other meta-heuristic algorithms when it comes to the quality of the solution while satisfying all system constraints.
A Practical Integrated Fault Location Technique for Radial Power Distribution Systems
(IEEE, 2019) Hamadouche, Tarek; Bendjeghaba, Omar; Ahriche, Aimed
power distribution systems (PDS) are becoming more complex and dispersed at long distances and different locations. With its radial and several laterals configuration, loads could be connected at similar distances from the substation which leads to a multi estimation of fault location and consuming more time for iterative fault location algorithms. In order to overcome those difficulties, a practical integrated fault location method for radial PDS is presented in this paper. The basic idea of the proposed approach is to partition a multi lateral distribution system to possible mono lateral system (MLSs) by a proposed communicant sensor (CS). Then an impedance based algorithm is applied only at the faulty MLS. To examine the whole method in the field, a real PDS from the Algerian distribution grid is used. Experimental results present significant benefits compared to a previous method reported in the literature.