Identification and robust control of smart structures using artificial neural networks

This paper describes an integrated approach to design and implement robust controllers for smart structures. To demonstrate this procedure, we have designed and fabricated a structural test article incorporating shape memory alloy (SMA) actuators, strain gauge sensors, signal-processing circuits and digital controllers with flexible structures. A neural-network-based structural identification method to determine a state space model of the system from its experimental input/output data is presented. To reduce the learning time required to train a neural network significantly, we have developed an accelerated adaptive learning-rate algorithm. The mathematical model derived using neural networks is compared with models obtained by more conventional and well known methods. Using this model, a modified linear quadratic Gaussian with loop transfer recovery (LQG/LTR) controller is designed for vibration suppression purposes. This robust controller accommodates the limited control effort produced by SMA actuators. A multilayered feedforward neural network is then trained to mimic this controller. These designs are all then realized as digital controllers and their closed-loop performances have been compared. In particular, the robustness properties of the controller have been verified for variations in the mass of the test article and the sampling time of the controller.