Linearization is a linear approximation of a nonlinear system that is valid in a small region around the operating point.
For example, suppose that the nonlinear function is $$y={x}^{2}$$. Linearizing this nonlinear function about the operating point x=1, y=1 results in a linear function $$y=2x1$$.
Near the operating point, $$y=2x1$$ is a good approximation to $$y={x}^{2}$$. Away from the operating point, the approximation is poor.
The next figure shows a possible region of good approximation for the linearization of $$y={x}^{2}$$. The actual region of validity depends on the nonlinear model.
Extending the concept of linearization to dynamic systems, you can write continuoustime nonlinear differential equations in this form:
$$\begin{array}{l}\dot{x}(t)=f\left(x(t),u(t),t\right)\\ y(t)=g\left(x(t),u(t),t\right).\end{array}$$
In these equations, x(t) represents the system states, u(t) represents the inputs to the system, and y(t) represents the outputs of the system.
A linearized model of this system is valid in a small region around the operating point t=t_{0}, x(t_{0})=x_{0}, u(t_{0})=u_{0}, and y(t_{0})=g(x_{0},u_{0},t_{0})=y_{0}.
To represent the linearized model, define new variables centered about the operating point:
$$\begin{array}{l}\delta x(t)=x(t){x}_{0}\\ \delta u(t)=u(t){u}_{0}\\ \delta y(t)=y(t){y}_{0}\end{array}$$
The linearized model in terms of δx, δu, and δy is valid when the values of these variables are small:
$$\begin{array}{l}\delta \dot{x}(t)=A\delta x(t)+B\delta u(t)\\ \delta y(t)=C\delta x(t)+D\delta u(t)\end{array}$$
Linearization is useful in model analysis and control design applications.
Exact linearization of the specified nonlinear Simulink^{®} model produces linear statespace, transferfunction, or zeropolegain equations that you can use to:
Plot the Bode response of the Simulink model.
Evaluate loop stability margins by computing openloop response.
Analyze and compare plant response near different operating points.
Design linear controller
Classical control system analysis and design methodologies require linear, timeinvariant models. Simulink Control Design™ automatically linearizes the plant when you tune your compensator. See Choosing a Control Design Approach.
Analyze closedloop stability.
Measure the size of resonances in frequency response by computing closedloop linear model for control system.
Generate controllers with reduced sensitivity to parameter variations and modeling errors (requires Robust Control Toolbox™).
You can use Simulink Control Design to linearize continuoustime, discretetime, or multirate Simulink models. The resulting linear timeinvariant model is in statespace form.
Simulink Control Design uses a blockbyblock approach to linearize models, instead of using fullmodel perturbation. This blockbyblock approach individually linearizes each block in your Simulink model and combines the results to produce the linearization of the specified system.
The blockbyblock linearization approach has several advantages to fullmodel numerical perturbation:
Most Simulink blocks have preprogrammed linearization that provides Simulink Control Design an exact linearization of each block at the operating point.
You can configure blocks to use custom linearizations without affecting your model simulation.
Simulink Control Design automatically removes nonminimal states.
Ability to specify linearization to be uncertain (requires Robust Control Toolbox)
Simulink Control Design lets you perform linear analysis of nonlinear models using a graphical user interface, functions, or blocks.
Linearization Tool  When to Use  

Linear Analysis Tool 
 
linearize 
 
slLinearizer  Batch linearize for varying model parameter values, operating points, and I/O sets.  
Linear Analysis Plots blocks 

In most cases, you should use exact linearization instead of frequency response estimation to obtaining a linear approximation of a Simulink model.
Exact linearization:
Is faster because it does not require simulation of the Simulink model.
Returns a parametric (statespace).
Frequency response estimation returns frequency response data. To create a transfer function or a statespace model from the resulting frequency response data requires an extra step using System Identification Toolbox™ to fit a model.
Use frequency response estimation:
To validate exact linearization accuracy.
When your Simulink model contains discontinuities or nonperiodic eventbased dynamics.
To study the impact of amplitude size on frequency response.
See Describing Function Analysis of Nonlinear Simulink ModelsDescribing Function Analysis of Nonlinear Simulink Models.
How is Simulink linmod
different from Simulink Control Design functionality
for linearizing nonlinear models?
Although both Simulink Control Design and Simulink linmod
perform
blockbyblock linearization, Simulink Control Design functionality
is enhanced by a more flexible user interface and Control System Toolbox™ numerical
algorithms.
Simulink Control Design Linearization  Simulink Linearization  

Graphicaluser interface  Yes See Linearize Simulink Model at Model Operating Point.  No 
Flexibility in defining which portion of the model to linearize  Yes. Lets you specify linearization I/O points at any level
of a Simulink model, either graphically or programmatically without
having to modify your model. See Linearize at Trimmed Operating Point.  No. Only rootlevel linearization I/O points, which is equivalent
to linearizing the entire model. Requires that you add and configure additional Linearization Point blocks. 
Openloop analysis  Yes. Lets you open feedback loops without deleting feedback
signals in the model. See OpenLoop Response of Control System for Stability Margin Analysis.  Yes, but requires that you delete feedback signals in your model to open the loop 
Control linear model state ordering  Yes See Ordering States in Linearized Model.  No 
Control linearization of individual blocks  Yes. Lets you specify custom linearization behavior for both
blocks and subsystems. See Controlling Block Linearization.  No 
Linearization diagnostics  Yes. Identifies problematic blocks and lets you examine the
linearization value of each block. See Linearization Troubleshooting Overview.  No 
Block detection and reduction  Yes. Block reduction detects blocks that do not contribute to the overall linearization yielding a minimal realization.  No 
Control of rate conversion algorithm for multirate models  Yes  No 
Exact linearization supports most Simulink blocks.
However, Simulink blocks with strong discontinuities or eventbased dynamics linearize (correctly) to zero or large (infinite) gain. Sources of eventbased or discontinuous behavior exist in models that have Simulink Control Design requires special handling of models that include:
Blocks from Discontinuities library
Stateflow^{®} charts
Triggered subsystems
Pulse width modulation (PWM) signals
For most applications, the states in your Simulink model should be at steady state. Otherwise, your linear model is only valid over a small time interval.
Choosing the right operating point for linearization is critical for obtaining an accurate linear model. The linear model is an approximation of the nonlinear model that is valid only near the operating point at which you linearize the model.
Although you specify which Simulink blocks to linearize, all blocks in the model affect the operating point.
A nonlinear model can have two very different linear approximations when you linearize about different operating points.
The linearization result for this model is shown next, with the initial condition for the integration x_{0} = 0.
This table summarizes the different linearization results for two different operating points.
Operating Point  Linearization Result 

Initial Condition = 5, State x1 = 5  30/s 
Initial Condition = 0, State x1 = 0  0 