Modeling and Identification of a Four-Bar Linkage Mechanism Driven by a Geared DC Motor

Abstract

Four-bar linkage mechanisms are popular systems that are used in a wide range of applications in the industry. However, it is not easy to model and control these mechanisms because parameters such as friction, damping, and stiffness are difficult to measure. Thus, there is a need to develop methods to facilitate the modeling and control of such systems. In this paper, identification methods are investigated for the purpose of estimating and predicting the nonlinear dynamics of these mechanisms. Linear parametric models and neural networks models are studied and applied to identify the dynamics of an experimental four-bar linkage mechanism, driven by geared DC motor, via recording and collecting real-time input/output signals for different operating profiles. This data is then used to optimize the parameters of  both linear models and neural network models in order to identify the system behavior. The identification models are tested in MATLAB simulation and their responses are compared with experimental data.  Even though a high-order linear model was able to capture the plant’s behavior,  the neural network model provided better accuracy in mimicking the nonlinear behavior of the four-bar linkage mechanism. The presented work provides appropriate model for a four-bar linkage mechanism that can be used to design controllers for the described mechanism.
Copyright © 2015 Praise Worthy Prize - All rights reserved.