Vol. 2, No. 2 (2018): Regular Issue

Fuzzy logic control algorithms are regarded to as a relatively new concept in modern control theory. This paper presents a comparative analysis of two qualitatively different approaches used for angular velocity control of a DC motor subject to chaotic disturbances coming from a gear with a transmission belt carrying a vibrating load. The purpose is to achieve accurate control of speed of the DC motor (a plant), especially, when the motor parameters and some external loading conditions are partially unknown. First, the classical approach based on the PID control is considered, and then a fuzzy logic based alternative is proposed. Two different controllers are developed, i.e. the classical PID controller and a Mamdani type fuzzy logic PI controller. Both control algorithms are implemented on an 8-bit AVR ATmega644PA microcontroller. Based on step responses of the plant, an analysis as well as an interesting comparison of the controllers’ performance has been presented.

A logistic map with parametric perturbation is studied. We confirm the model exhibits self-similar structures in the parameter space known as shrimps. The organization of such structures may be describe through extreme curves, giving exactly the position of each one of them in a complicated behavior in the parameter space. Boundary crisis are also discussed. They lead to an abrupt destruction of the chaotic attractor and how such destruction affects the dynamics of the system is discussed, particularly affecting the time before the orbit leaves to infinite. By a specific choice of the parameters we show the existence of strange non-chaotic attractors in the parameter space.

In this paper, independent periodic motions in the two-degree-of-freedom (2-DOF) van der Pol-Duffing oscillator are investigated. From the semi-analytical method, the 2-DOF van der Pol-Duffing oscillator is discretized to obtain implicit discrete mappings. From the implicit mapping structures, periodic motions varying with excitation frequency are obtained semi-analytically, and the corresponding stability and bifurcation are obtained by eigenvalue analysis. The frequency-amplitude characteristics of periodic motions are also presented. Thus, from the analytical prediction, numerical simulations of periodic motions are performed for comparison of numerical and analytical results. The harmonic amplitude spectrums of periodic motions are also presented for harmonic effects on the periodic motions. Through this study, the order of symmetric period-1 to chaotic motions (i.e., 1(S)⊳1(A)⊳3(S)⊳2(A)⊳· · ·⊳m(A)⊳(2m+1)(S)⊳· · · ) (m→∞) is discovered. Chaotic motions or catastrophe jumping phenomena between the two independent periodic motions exist. The independent periodic motions can be used for specific applications in phase locking, and such results can be useful to develop series of the van der Pol-Duffing circuits for applications.

The local flexible structure (LFS) is used on the airfoil to enhance the lift, and the influences of structure parameters of LFS on the lift enhancement are studied numerically. Coupling with the structure solver, the unsteady flows over airfoil with LFS are simulated by using the CBS-ALE scheme and dual time stepping. The results show that the amplitude oscillation is the key factor to enhance the lift as the self-induced oscillation is used. The large oscillation amplitude of LFS can only be induced in certain frequency range with suitable structure parameters, such as smaller elastic stiffness and damping. The oscillation of LFS can lead to the formation of individual separation bubble and vortex, which are beneficial to the lift enhancement of airfoil. With large oscillation amplitude, vortices with low pressure can be generated early and stay on upper surface of airfoil for a long time, resulting in the significant lift enhancement.

A continuous adaptive controller is designed for a famous example of Brockett system which is an example of a class of non-holonomic systems. The controllability Lie Algebra of the Brockett’s system contains Lie brackets of depth one. It is shown that the closed loop system is globally uniformly asymptotically stable. The advantage of this method is that it does not require the conversion of the system model into a “chained form,” and thus does not rely on any special transformation techniques. The practical effectiveness of the controller is illustrated by numerical simulations.

Gas turbines are widely found in industries, especially in gas compressor stations, where they are used to ensure the transportation of gas in high pressure pipelines over a long distance. Nevertheless, these rotating machines are subject to vibrations due to the influence of several causes, such as mass imbalance and shaft misalignment. These vibrations can affect directly the bearings, where their lifespan can be shorten remarkably. Furthermore they can be damaged causing catastrophic failure in the gas turbine and its installation. The main purpose of this paper is to develop an approach that can provide the estimated remaining life of the gas turbine bearings based on the vibration signals obtained via installed sensors. This approach allows us to estimate the optimal expected life of the studied gas turbine bearings by the prediction of the expected failures times. The obtained result under the proposed approach is satisfactory and shows that the use of the proposed approach can avoid the costly damage caused by the mentioned failures in the gas turbine under the imposed constraints.