Publications Internationales

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

Browse

Filters

20252

Settings

Subject: search.filters.subject.Applied mathematics

Search Results

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    Optimal conventional and nonconventional machining processes via particle swarm optimization and flower pollination algorithm
    (Springer Science and Business Media, 2025) Mellal, Mohamed Arezki; Tamazirt, Imene; Tiar, Maissa; Williams, Edward J.
    Manufacturing requires various machining processes. Nowadays, machining implies advanced technologies in order to meet more exacting process performance criteria. This paper addresses the optimization of four conventional and nonconventional machining processes: drilling, grinding, water jet machining (WJM), and wire electrical discharge machining (EDM). The input process parameters are: cutting speed, feed rate, cutting environment, depth of cut, grit size, water jet pressure, diameter of water jet nozzle, traverse rate of the nozzle, stand-off-distance, ignition pulse current, pulse-off time, pulse duration, servo reference mean voltage, servo speed variation, wire speed, wire tension, and injection pressure. The multi-objective EDM optimization problem is converted to a single-objective problem using the weighted-sum method. Two nature-inspired algorithms of artificial intelligence (AI) are implemented for solving these problems, namely the particle swarm optimization (PSO) and the flower pollination algorithm (FPA). Penalty functions are introduced to handle the constraints and to enhance the algorithms for better results. The machining outputs, required number of function evaluations, CPU time, and standard deviations are the performance metrics. The results obtained are compared and show better performance than that already documented in the literature.
  • Thumbnail Image
    Item
    Enhancing the transient performances and stability of three-tank liquid level using a modified pid controller
    (Revue roumaine des sciences techniques — Série électrotechnique et énergétique, 2025) Idir, Abdelhakim; Nesri, Mokhtar; Belhouchet, Khaled; Guedida, Sifelislam; Canale, Laurent
    Managing liquid levels in industrial tanks is crucial, especially for precise component mixing. Traditional PID controllers, though widely used, often exhibit slow settling times and excessive overshoot, which can affect system performance. This study proposes a fractionalized order PID (FrOPID) controller optimized using the Modified Artificial Hummingbird Algorithm (MAHA) to enhance stability and response in a three-tank system. The controller’s effectiveness is evaluated under varying valve coefficient (Kv) and tank cross-sectional area conditions. A comparative analysis with advanced metaheuristic-optimized PID controllers confirms the superiority of the MAHA/FrOPID in terms of accuracy, response speed, and robustness, making it a highly efficient solution for liquid level control.