Estimate frequency response, examine frequency-domain
characteristics, validate linearization

A *frequency response* describes the
steady-state response of a system to sinusoidal inputs. Simulink^{®}
Control Design™ software has
both command-line tools and a graphical Linear Analysis Tool for estimating
the frequency response of your system. You can use the estimated response
to validate exact linearization results, analyze linear model dynamics,
or estimate parametric models. For more information about frequency
response estimation, see What Is a Frequency Response Model?.

Frequency response estimation requires an input signal at the linearization input point to excite the model at the frequencies of interest. For more information, see Estimation Input Signals.

Linear Analysis Tool | Linearize Simulink models |

`frestimate` | Frequency response estimation of Simulink models |

`frestimateOptions` | Options for frequency response estimation |

`frest.Sinestream` | Signal containing series of sine waves |

`frest.createFixedTsSinestream` | Sinestream input signal with fixed sample time |

`frest.Chirp` | Swept-frequency cosine signal |

`frest.Random` | Random input signal for simulation |

`frest.createStep` | Step input signal |

`frest.simCompare` | Plot time-domain simulation of nonlinear and linear models |

`frest.simView` | Plot frequency response model in time- and frequency-domain |

`getSimulationTime` | Final time of simulation for frequency response estimation |

`frest.findSources` | Identify time-varying source blocks |

`frest.findDepend` | List of model path dependencies |

**What Is a Frequency Response Model?**

A frequency response describes the steady-state response of a system to sinusoidal inputs.

To estimate a frequency response for your model, first disable blocks that simulate noise and blocks that generate time-varying signals.

**Estimate Frequency Response Using Linear Analysis Tool**

Estimate the frequency response of a Simulink model using a manually constructed sinestream input signal.

**Estimate Frequency Response with Linearization-Based Input Using Linear Analysis Tool**

Estimate the frequency response of a Simulink model using an input signal based on the linearization of the model.

**Estimate Frequency Response at the Command Line**

Estimate the frequency response of a Simulink model
at the MATLAB^{®} command line.

**Analyze Estimated Frequency Response**

Analyze your frequency response estimation results by simulating the estimated model.

**Estimation Requires Input and Output Signals**

To estimate a frequency response model, inject signals at linearization input points and measure the corresponding output signals.

Frequency response estimation requires an input signal at the linearization input point to excite the model at frequencies of interest, such as a chirp or sinestream signal.

**Create Sinestream Input Signals**

Create sinestream input signals using the Linear Analysis Tool or at the command line.

Create chirp input signals using the Linear Analysis Tool or at the command line.

**Modify Estimation Input Signals**

When frequency response estimation produces unexpected results, you can try modifying the input signal properties.

**Disable Noise Sources During Frequency Response Estimation**

Noise sources can interfere with the signals at the linearization output points and produce inaccurate estimation results.

**Estimate Frequency Response Models with Noise Using Signal Processing Toolbox**

You can also estimate a frequency response model using Signal Processing Toolbox™ software, which includes windowing and averaging.

**Estimate Frequency Response Models with Noise Using System Identification Toolbox**

You can also estimate a frequency response model using System Identification Toolbox™ software.

**Effects of Time-Varying Source Blocks on Frequency Response Estimation**

Time-varying source blocks drive the model away from the operating point of the linearized system, which prevents the response from reaching steady state.

**Validate Linearization In Frequency Domain**

You can assess the accuracy of your linearization results by estimating the frequency response of the nonlinear model and comparing the result with the response of the linearized model.

**Validate Linearization In Time Domain**

You can assess the accuracy of your linearization results by comparing the simulated output of the nonlinear model and the linearized model.

**Generate MATLAB Code for Repeated or Batch Frequency Response Estimation**

Generate MATLAB scripts or functions for frequency response estimation using the Linear Analysis Tool.

**Managing Estimation Speed and Memory**

Improve frequency response estimation performance by reducing estimation time and memory requirements.

**Troubleshooting Frequency Response Estimation**

If your estimated frequency response does not match the expected behavior of your system, you can use the time-domain and frequency-domain response plots to help improve the results.

Was this topic helpful?