Model identification, or parameter estimation, is a series of data and signal processing tasks to estimate unknown parameters and coefficients of the dynamic system model.

Before you perform model identification, you need to preprocess the acquired raw data to ensure the data is free from external noise, scaling problems, outliers, or other corruptions. Use the Data Preprocessing VIs to preprocess the data and ensure the data accurately reflects the response of the dynamic system.

In this tutorial, the DC servomotor is a simple dynamic system. The acquired raw data is accurate enough for model identification. Therefore, you can skip preprocessing the raw data.

With the acquired time-domain data, you can perform model identification on any of the following models:

  • Continuous-time transfer function model
  • Continuous-time state-space model
  • Discrete-time transfer function model
  • Discrete-time state-space model

You also can convert the time-domain data to frequency-domain data. With the converted frequency-domain data, you can identify models by using the Frequency-Domain Model Estimation VIs.