Yazar "Chen Y." seçeneğine göre listele

Listeleniyor 1 - 3 / 3
  • Küçük Resim Yok
    Öğe
    Disturbance rejection fopid control of rotor by multi-objective bb-bc optimization algorithm
    (American Society of Mechanical Engineers (ASME), 2017) Ates A.; Alagoz B.B.; Yeroglu C.; Yuan J.; Chen Y.
    This paper presents a FOPID tuning method for disturbance reject control by using multi-objective BB-BC optimization algorithm. Proposed method allows multi-objective optimization of set-point performance and disturbance rejection performances of FOPID control system. The objective function to be minimized is composed of the weighted sum of MSE for set-point performance and RDR for disturbance rejection improvement. The proposed optimization performs maximization of RDR and minimization of MSE and it can deal with the tradeoff between RDR performance and step-point performance. Application of the method is shown for autotuning of FOPID controller that is employed for control of TRMS model. We observed that low-frequency RDR indices can be used to improve disturbance rejection performance in multi-objective controller tuning problems. Particularly, for flight control application, disturbance reject control is very substantial to robust performance of propulsion systems. © Copyright 2017 ASME.
  • Küçük Resim Yok
    Öğe
    Fractional order chaotic model based enhanced equilibrium optimization algorithm for controller design of 3 DOF Hover flight system
    (American Society of Mechanical Engineers (ASME), 2021) Ates A.; Chen Y.
    In this study, the K feedback gain vector parameters that are used for the control of three degree of freedom four rotor quadcopter system (3 DOF Hover) are optimized with the Enhanced Equilibrium Optimization Algorithm (E2O). The E2O algorithm is proposed with using parameters obtained from fractional order chaotic oscillator models instead of random variables. The Basic EO algorithm is inspired by volume-mass balance. In EO algorithm, each particle is called a motion that searches a parameter vector space. However, random coefficients derived from uniform distribution are used in the parameters updating process or in the generation of the initial population. The E2O algorithm was proposed by using vectors obtained from fractional order chaotic oscillators instead of stochastic coefficients in the basic Equilibrium optimization algorithm. Genesio Tesi, Rössler, Lotka Volterra fractional-order chaotic oscillator models were used in the E2O algorithm to optimize K feedback gain vector of 3 DOF Hover. The order and initial conditions the fractional chaotic oscillator models were experimentally adjusted for the control of 3 DOF problem. Thus, suitable fractional-order chaotic models for the problem were obtained. The E2O algorithm results are compared with the Stochastic Multi Parameter Optimization (SMDO) and Discreet Stochastic Optimization (DSO) algorithms for the system's pitch, roll and yaw angles. Copyright © 2021 by ASME
  • Küçük Resim Yok
    Öğe
    Fractional order filter discretization with marine predators algorithm
    (American Society of Mechanical Engineers (ASME), 2021) Ates A.; Chen Y.
    In this study, discrete time models of continuous time fractional order filters are obtained by using the Marine Predators Algorithm (MPA). Marine Predators optimization algorithm is a population-based heuristic method. This method is inspired by the hunting behavior of marine predators. The algorithm works on three basic phases. These phases occur according to the difference or equality of the velocity of the prey and the predator. As it is known, uniform distribution is generally used in stochastic based optimization algorithms. However, in the MPA method, Brownian and Levy distributions are also used as well as uniform distribution. First, continuous time frequency responses of fractional order filters are generated. Then, fourth order discrete time filters are designed that can give similar responses with generated continues time filter frequency responses. Ten parameters were optimized for the design of fourth order discrete time filters numerator and denominator. The Marine Predators method's results are compared with the results of the Fractional order Darwinian Particle Swarm Optimization (FODPSO) algorithm, from which discrete time filters are obtained for two fractional order continuous time filter models. In this way, it has been shown comparatively that the Marine Predators Algorithm can be used in real engineering problems and can do filter discretization better. Copyright © 2021 by ASME