Learn to estimate linear statespace models using the prediction-error method (PEM) in this 24-minute tutorial video from the System Identification with Julia series. Master parameter estimation for linear ODEs through hands-on demonstrations covering linear ODE models with correction, experimental data analysis, and non-parametric transfer-function estimates. Explore the implementation of PEM, validate results, and compare findings with true models while discovering advanced PEM options. Work alongside using the provided notebook file and leverage tools like JuliaSim and ControlSystemIdentification. Follow structured segments including experimental data examination at 4:03, PEM implementation at 8:27, and validation techniques at 10:32. Access comprehensive resources including JuliaSim's enterprise platform for modeling and simulation, and dive deeper into classical system identification using the ControlSystemIdentification documentation.
System Identification with Julia: Prediction-Error Method for Linear Models - Part 4
Overview
Syllabus
Linear ODE model with correction
Experimental data
Non-parametric transfer-function estimate
PEM
Validation
Compare with the true model
PEM advanced options
Taught by
JuliaHub
Related Courses
-
Linear ARX Models in System Identification with Julia - Part 2
-
Parameter Calibration for Nonlinear ODEs in System Identification - Part 9
-
Time-varying Parameter Estimation in System Identification with Julia - Part 10
-
System Identification with Julia: Experiments and Excitation - Part 6
-
System Identification with Julia: Prefiltering Data for Model Estimation - Part 5
-
System Identification with Julia: Model Validation - Part 7
