If you do not get good linearization results, you can use the Simulink® Control Design™ troubleshooting tools to diagnose and fix linearization issues. For more information on linearization troubleshooting, see Linearization Troubleshooting Overview.
Some Simulink blocks, can produce poor linearization results. Typically, you must specify custom linearizations for such blocks. For more information, see When to Specify Individual Block Linearization.
To find the cause of poor linearization results, use the Simulink Control Design troubleshooting tools.
To diagnose whether you used the correct operating point for linearization, simulate the model at the operating point you used for linearization.
Incorrect placement of linearization I/O points can result in inappropriately excluded blocks from linearization.
Incorrect loop opening placement causes unwanted feedback signals in the linearized model.
Padé approximation of time delays in your model may cause insufficient phase lag.
View the linearization results for blocks with diagnostic messages indicating configuration warnings, unsupported blocks, and blocks that automatically linearize using numerical perturbation.
Troubleshooting the linearization of large models is easier using a divide-and-conquer strategy.
Incorrect sample time and rate conversion methods can cause poor linearization results in multirate models.
Large Simulink models and blocks with complex initialization functions can cause slow linearization.
Some Simulink blocks, including those with sharp discontinuities, can produce poor linearization results. Typically, you must specify custom linearizations for such blocks.
Specify the linearization of any block, subsystem, or model reference without having to replace this block in your Simulink model.
You can specify a substitute linearization for a block or subsystem in your Simulink model using a custom function on the MATLAB® path.
Augment the linearization of a block with additional time delay dynamics, using a block linearization specification function.
Blocks that do not have preprogrammed analytic Jacobians linearize using numerical perturbation. Changing the perturbation level changes the linearization results.
Subsystems that contain PWM signals do not linearize well due to discontinuities in the signal.
You can use System Identification Toolbox™ software to identify a linear system for a model component that does not linearize well, and use the identified system to specify its linearization.
You can linearize models using exact representations or Padé approximations of continuous-time delays.
Since linearization occurs at a specific moment in time, the trigger event for an externally-scheduled subsystem never happens.
Blocks that have nondouble precision inputs or outputs and have no preprogrammed exact linearization automatically linearize to zero. Linearizing such blocks requires converting all signals to double precision.