You can use measured data from your system to identify linear and nonlinear black-box and grey-box models. You can use these models to track parameter changes, forecast system response, and compute the associated response uncertainty. You can then employ these predictions for condition monitoring and fault detection.
You can use online parameter estimation techniques to monitor the condition of your system in real time, diagnose faults, and predict remaining useful life. You can also deploy the online estimation code to an embedded target using MATLAB® Compiler™ or MATLAB Coder™.
This example shows how to create a time series model and use the model for prediction, forecasting, and state estimation.
This example shows how to perform multivariate time series forecasting of data measured from predator and prey populations in a prey crowding scenario.
This example shows how to detect abrupt changes in the behavior of a system using online estimation and automatic data segmentation techniques.
This example shows how to use a data-based modeling approach for fault detection.
You can use an extended Kalman filter for fault detection.
This example shows how to implement an online recursive least squares estimator.