Structural health monitoring (SHM) of high-rate, mechanical systems in dynamically harsh environments presents many challenges over traditional SHM applications. Damage in these systems must be detected and quantified in tens to hundreds of microseconds in order to have sufficient time to react and mitigate damage. The computation speeds and robustness of sliding mode observers (SMOs) for state, parameter, and disturbance estimation for linear and nonlinear systems make them an attractive approach for real-time SHM of high-rate systems. This paper investigates a novel SMO combined with a recursive least squares parameter estimator to detect and track changing system parameters. The observer is simulated on a one degree-of-freedom system with time-varying model parameters to mimic damage. This paper focuses on practical considerations for SMOs for high-rate systems, such as the effects of measurement noise and sampling rates on the estimator’s accuracy and convergence speeds.

This content is only available via PDF.
You do not currently have access to this content.