KEYWORDS: Switching, Device simulation, Control systems, Switches, Signal attenuation, Control systems design, Performance modeling, Systems modeling, Smart structures, Computing systems
Many forced systems are prone to undesirable levels of oscillation if lightly damped modes are present with their frequency range of operation. Rotary systems, for example, can experience these problems during the speed up or speed down stage of operation. Resonant motion can damage or effect the accuracy of operation of such systems and is, therefore, highly undesirable. Many closed-loop controllers avoid this by suppressing the mode itself such that at resonance the modal vibration amplitude is small (i.e. highly damped). The current work presents an alternative novel switching controller, which suppresses the system not by applying a high amount of damping but rather by moving the resonant mode such that it is never excited. From the basis of an accurate plant model, two pole-placement controllers are designed and implemented both in simulation and experiment on a cantilever smart structure. These controllers are shown to successfully change the natural frequency value, while retaining the same damping ratio value. A novel switching system is employed that calculates the optimal switching position by running a simulation of the desired systems in parallel to the controlled open-loop system. Moreover, the system minimises transients occurred by switching back and forth between controllers, thus increasing the efficiency of the system. By comparing the experimental results to a conventional high damping pole-placement controller that applies a similar amount of control effort, a lower overall level of amplitude suppression can be seen.
KEYWORDS: Systems modeling, Device simulation, Actuators, Simulink, Data modeling, Process modeling, Sensors, Finite element methods, Dynamical systems, Control systems
The ability to model, investigate and control the behavior of dynamic systems in a simulation environment is highly desirable due to time and cost benefits. A new technique has been developed allowing finite element models to be integrated with Simulink for dynamic simulation and control. The technique is presented by the modeling of a fixed-free cantilever beam with bonded piezoelectric patches. A description of the modeling technique is presented detailing the process of model creation, including input and output variable determination, and exportation to Simulink as a state-space model. A comparison of simulated and experimental open-loop behavior is provided. Furthermore the free and forced system behavior both observed, and simulated with velocity feedback controllers (VFB) is presented. Conclusions are drawn regarding the capabilities and restrictions of the developed technique in comparison to modeling using a system identification technique. The author's views on the techniques application to non-linear system modeling and potential for optimizing sensor and actuator locations are presented.
Modern design briefs increasingly stipulate the need to reduce the weight of high technology structures due to the large costs involved in transport. As a result, these light structures are prone to unwanted vibrations. The uncertain surrounding environment necessitates the assumption of no fixed base to operate sensing/actuation devices with reference to. These conditions have made piezoceramic devices ideal candidates to operate as distributed actuators to apply vibration control to smart structures. The limited actuation of these devices, however, makes it essential to consider the problems associated with actuator saturation on the resulting controllers. This paper compares Pole-placement and Generalised Minimum Variance controllers for application to a vibrating cantilever beam. Certain desired closed-loop pole-positions are found to lead to limit cycling behaviour with the Pole-placement controller in both experiment and simulation. Subsequent application of saturation compensation is shown to return stability to the closed-loop system. Due to the control signal penalising property of the Generalised Minimum Variance controller this method is shown not to require compensation. Utilising the Nelder-Mead simplex algorithm, efficient controllers have been obtained for both controller types. Experimental results have shown that the controllers are able to suppress both the free and forced disturbances of the experimental system in the presence of control signal saturation and plant model mismatch. The results show that a developed reduced order model results in a controller performance that is comparable to that using a larger order Auto Regressive with eXogenous inputs model.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.