Publications Internationales

Permanent URI for this collectionhttps://dspace.univ-boumerdes.dz/handle/123456789/13

Browse

Has files: NoSubject: search.filters.subject.Genetic algorithms

Search Results

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    Fuzzy constraint prioritization to solve heavily constrained problems with the genetic algorithm
    (Elsevier, 2023) Alouane, Basma; Boulif, Menouar
    Genetic algorithms (GAs) are approximate solving methods that have been originally proposed to achieve unconstrained optimization. To handle constrained problems, which is the case for the majority of real-life circumstances, GAs must be equipped with a constraint-handling mechanism. Transformation functions (TFs) are among the constraint-handling approaches that intervene in the phenotypic space. In this paper, we study the impact of considering constraint priorities on the GA performance when it deals with heavily constrained problems. Priorities are set by integrating a constraint order into the TF definition. We consider different TF forms enhanced with a fuzzy inference engine to find the best constraint ordering. Finally, we conduct an experimental study to assess the performance of the proposed approach on the semi-supervised graph partitioning problem. The obtained results show with statistical evidence that the proposed fuzzy method is promising
  • No Thumbnail Available
    Item
    Modeling wax disappearance temperature using advanced intelligent frameworks
    (American Chemical Society, 2019) Benamara, Chahrazed; Nait Amar, Menad; Gharbi, Kheira; Hamada, Boudjema
    The deposition of wax is one of the most potential problems that disturbs the flow assurance during production processes of hydrocarbon fluids. In this study, wax disappearance temperature (WDT) that is recognized as a vital parameter in such circumstances is modeled using advanced machine learning techniques, namely, radial basis function neural network (RBFNN) coupled with genetic algorithm (GA) and artificial bee colony (ABC). Besides, an accurate and user-friendly correlation was established by implementing the group method of data handling. Results revealed the high reliability of the proposed hybrid models and the established correlation. Moreover, RBFNN coupled with ABC (RBFNN-ABC) was found to be the best paradigm with an overall average absolute relative error value of 0.5402% and a total coefficient of determination (R2) of 0.9706. Furthermore, the performance comparison showed that RBFNN-ABC and the established explicit correlation outperform the prior intelligent and thermodynamic models. Finally, by performing the outlier detection, the quality of the utilized database was assessed, the applicability realm of the best model was delineated, and only one point was found as doubtful