Analog and Labview- Based Control of a Maglev System with Ni-elvis

This paper investigates the control of a low cost verticalaxis maglev system for mechatronics and controls education. The tabletop maglev system consists of an electromagnetic coil that levitates a ferrous object using an infrared sensor to determine the object’s position. Based on the sensor output, the controller adjusts the coil current, thus changing the magnetic field controlling the levitated object’s position. A second electromagnetic coil is used to provide known disturbances. The paper develops the underlying theory for magnetic levitation and presents the results of experiments with classical controllers implemented both as analog circuits and in software-based virtual instruments. Analog controllers, such as PID-type controllers, were implemented as simple circuits on National Instruments’ Educational Laboratory Virtual Instrumentation Suite (NI-ELVIS) prototyping board. NI-ELVIS offers a LabVIEW-based prototyping environment for readily experimenting with different controller circuits. It consists of a multifunction data acquisition device and a custom-designed benchtop workstation with a prototyping board. In addition to analog control circuits, a suite of LabVIEW-based controllers were developed which offer in software a rapid way to change control strategies and gains and explore the effect on the physical system. INTRODUCTION Levitation is the stable equilibrium of an object without contact and can be achieved using electric or magnetic forces. In a magnetic levitation, or maglev, system a ferromagnetic object is suspended in air using electromagnetic forces. These forces cancel the effect of gravity, effectively levitating the object and achieving stable equilibrium. In this paper, comparisons of control strategies implemented in hardware and software are conducted. The controllers are analyzed for stability, robustness, system response, and mechanical impedance. The maglev physical system investigated here is a laboratory prototype often studied in engineering education. It is representative of real-world applications of maglev technology in bearings and high speed transportation systems (Nagurka and Wang, 1997). Tabletop maglev systems are available commercially from manufacturers such as Quanser and ECP (http://www.quanser.com/ and http://www.ecpsystems.com/), and descriptions of simple test stands showcasing the principles of maglev have been described in the literature (e.g., Cicion, 1996; Xie, 2003). In theory, the magnetic force produced by passing current through an electromagnet can exactly counteract the weight of the object. In practice, the electromagnetic force is sensitive to small disturbances that can induce acceleration forces on the levitated object, causing the object to be unbalanced. The main function of the controller is to maintain both static and dynamic equilibrium between the magnetic force and the object’s weight, in the face of disturbances, using the input of the sensor to obtain the position of the object. Though various strategies for non-linear control currently exist, this paper focuses on the application of classical linear controllers to a linearized model of the non-linear maglev system. The controllers provide robust closed-loop performance of maglev systems that can levitate a variety of suspended masses despite disturbances (both mechanical and electrical). Analog PID-type control circuits are implemented using National Instruments Educational Laboratory Virtual Instrumentation Suite (NI ELVIS). In addition, the controllers are implemented digitally (in software) using NI LabVIEW. Fig. 1: Maglev Testbed Developed at Marquette University Shown Levitating AA Battery. (Power Supply for Electromagnet in Background Shows 18.0V at 1.60A.) 1 Copyright © 2005 by ASME A tabletop single-axis maglev system, shown in Fig. 1, was first developed as a Marquette University senior design project (Craig, et al, 1998). The system is approximately 25 cm wide and 40 cm tall and requires power supplies for the electromagnet (18V, 3A) and the control circuitry (15V, 0.5A). The testbed, which had been used to explore different control strategies, was modified by inserting a second electromagnetic coil below the levitated object for providing known disturbances, as shown in Fig. 2.