Survey on fuzzy logic methods in control systems of electromechanical plants

Abstract

The paper reviews control algorithms of electromechanical systems using the theory of fuzzy logic, describes basic principles of synthesis of these systems, discusses methods for analyzing their stability based on fuzzy Lyapunov functions. These algorithms are most often implemented in the form of various controllers, which application is justified in the systems with unknown mathematical model, or not determined, or strictly nonlinear, with nonlinear disturbances in their structure or in the external forces. We describe the basic methods of inference used in the design of various types of fuzzy logic controllers proposed by Zadeh, Mamdani, Takagi, Sugeno, and Mendel. A typical structural scheme of such controllers is given. The considered applications of these controllers in the control of various technical plants give the possibility to classify them according to various criteria: topologies of structures, inference mechanism, methods of defuzzification, types of membership functions. We review methods for adjustment of such controllers by genetic algorithms and neural networks with a description of the most commonly used criteria for optimality estimation. It is shown that the use of an expert approach based on fuzzy logic is applicable both in control of various coordinates of information subsystems of robotic complexes, and in control of the power switches of their energy subsystems. In the review of publications, the main attention was given to sources containing comparison with traditional approaches to control, as well as sources in which theoretical studies are supported by experiments involving various electromechanical plants. The paper may be useful to specialists and researchers in the field of control of various technical devices.