Mekatronik Mühendisliği Bölümü Yayın Koleksiyonu

Permanent URI for this collectionhttps://hdl.handle.net/20.500.12416/255

Browse

Filters

20213

Settings

Start date: 2020Subject: search.filters.subject.Optimization

Search Results

Now showing 1 - 3 of 3
  • Article
    Citation - WoS: 7
    Citation - Scopus: 10
    Online Path Planning for Unmanned Aerial Vehicles To Maximize Instantaneous Information
    (Sage Publications inc, 2021) Leblebicioglu, Kemal; Ergezer, Halit
    In this article, an online path planning algorithm for multiple unmanned aerial vehicles (UAVs) has been proposed. The aim is to gather information from target areas (desired regions) while avoiding forbidden regions in a fixed time window starting from the present time. Vehicles should not violate forbidden zones during a mission. Additionally, the significance and reliability of the information collected about a target are assumed to decrease with time. The proposed solution finds each vehicle's path by solving an optimization problem over a planning horizon while obeying specific rules. The basic structure in our solution is the centralized task assignment problem, and it produces near-optimal solutions. The solution can handle moving, pop-up targets, and UAV loss. It is a complicated optimization problem, and its solution is to be produced in a very short time. To simplify the optimization problem and obtain the solution in nearly real time, we have developed some rules. Among these rules, there is one that involves the kinematic constraints in the construction of paths. There is another which tackles the real-time decision-making problem using heuristics imitating human- like intelligence. Simulations are realized in MATLAB environment. The planning algorithm has been tested on various scenarios, and the results are presented.
  • Article
    Citation - WoS: 5
    Citation - Scopus: 6
    Multi-Objective Trajectory Planning for Slung-Load Quadrotor System
    (Ieee-inst Electrical Electronics Engineers inc, 2021) Ergezer, Halit
    In this article, multi-objective trajectory planning has been carried out for a quadrotor carrying a slung load. The goal is to obtain non-dominated solutions for path length, mission duration, and dissipated energy cost functions. These costs are optimized by imposing constraints on the slung-load quadrotor system's endpoints, borders, obstacles, and dynamical equations. The dynamic model of a slung-load quadrotor system is used in the Euler-Lagrange formulation. Although the differential flatness feature is mostly used in this system's trajectory planning, a fully dynamic model has been used in our study. A new multi-objective Genetic Algorithm has been developed to solve path planning, aiming to optimize trajectory length, mission time, and energy consumed during the mission. The solution process has a three-phase algorithm: Phase-1 is about randomly generating waypoints, Phase-2 is about constructing the initial non-dominated pool, and the final phase, Phase-3, is obtaining the solution. In addition to conventional genetic operators, simple genetic operators are proposed to improve the trajectories locally. Pareto Fronts have been obtained corresponding to exciting scenarios. The method has been tested, and results have been presented at the end. A comparison of the solutions obtained with MOGA operators and MOPSO over hypervolume values is also presented.
  • Article
    Citation - WoS: 17
    Citation - Scopus: 17
    Developing and Implementation of an Optimization Technique for Solar Chimney Power Plant With Machine Learning
    (Asme, 2021) Kocak, Eyup; Bayer, Ozgur; Beldek, Ulas; Yapic, Ekin Ozgirgin; Ayli, Ece; Ulucak, Oguzhan; Yapici, Ekin Özgirgin
    Green energy has seen a huge surge of interest recently due to various environmental and financial reasons. To extract the most out of a renewable system and to go greener, new approaches are evolving. In this paper, the capability of Artificial Neural Network and Adaptive Neuro-Fuzzy Inference System in geometrical optimization of a solar chimney power plant (SCPP) to enhance generated power is investigated to reduce the time cost and errors when optimization is performed with numerical or experimental methods. It is seen that both properly constructed artificial neural networks (ANN) and adaptive-network-based fuzzy inference system (ANFIS) optimized geometries give higher performance than the numerical results. Also, to validate the accuracy of the ANN and ANFIS predictions, the obtained results are compared with the numerical results. Both soft computing methods over predict the power output values with MRE values of 12.36% and 7.25% for ANN and ANFIS, respectively. It is seen that by utilizing ANN and ANFIS algorithms, more power can be extracted from the SCPP system compared to conventional computational fluid dynamics (CFD) optimized geometry with trying a lot more geometries in a notably less time when it is compared with the numerical technique. It is worth mentioning that the optimization method that is developed can be implemented to all engineering problems that need geometric optimization to maximize or minimize the objective function.