Integrator Limited

Integrate signal

• 

Libraries:
Simulink / Continuous

Description

The Integrator Limited block is identical to the Integrator block with the exception that the output of the block is limited based on the upper and lower saturation limits. See Limiting the Integral for details.

Simulink^® treats the Integrator block as a dynamic system with one state. The block dynamics are given by:

$$\{\begin{array}{c}\dot{x}(t)=u(t)\\ y(t)=x(t)\end{array}x(t_0)=x_0$$

where:

• u is the block input.

• y is the block output.

• x is the block state.

• x₀ is the initial condition of x.

While these equations define an exact relationship in continuous time, Simulink uses numerical approximation methods to evaluate them with finite precision. Simulink can use several different numerical integration methods to compute the output of the block, each with advantages in particular applications. Use the Solver pane of the Configuration Parameters dialog box (see Solver Pane) to select the technique best suited to your application.

The selected solver computes the output of the Integrator block at the current time step, using the current input value and the value of the state at the previous time step. To support this computational model, the Integrator block saves its output at the current time step for use by the solver to compute its output at the next time step. The block also provides the solver with an initial condition for use in computing the block's initial state at the beginning of a simulation. The default value of the initial condition is 0. Use the block parameter dialog box to specify another value for the initial condition or create an initial value input port on the block.

Use the parameter dialog box to:

• Define upper and lower limits on the integral

• Create an input that resets the block's output (state) to its initial value, depending on how the input changes

• Create an optional state output so that the value of the block's output can trigger a block reset

Use the Discrete-Time Integrator block to create a purely discrete system.

Defining Initial Conditions

You can define the initial conditions as a parameter on the block dialog box or input them from an external signal:

• To define the initial conditions as a block parameter, specify the Initial condition source parameter as internal and enter the value in the Initial condition field.

• To provide the initial conditions from an external source, set the Initial condition source parameter to external. An additional input port appears on the block.

  

  Note

  If you select the Limit output parameter, the initial condition must be within the saturation limits. If the initial condition is not within the block saturation limits, the block displays an error message.

Wrapping Cyclic States

Several physical phenomena, such as oscillators and machinery that exhibits rotational movement, are cyclic, periodic, or rotary in nature. Modeling these phenomena in a Simulink block diagram involves integrating the rate of change for the periodic or cyclic signals to obtain the state of the movement. Over long simulation time spans, this approach can result in the states that represent periodic or cyclic signals integrating to large values. Computing trigonometric values, such as sine or cosine, for these signals takes longer as the values become larger due to angle reduction. As the signal values become larger, solver performance and accuracy degrades.

One approach for overcoming this drawback is to reset the angular state to 0 when it reaches 2π (or to –π when it reaches π, for numerical symmetry). This approach improves the accuracy of sine and cosine computations and reduces angle reduction time. But it also requires zero-crossing detection and introduces solver resets, which slow down the simulation for variable step solvers, particularly in large models.

To eliminate solver resets at wrap points, the Integrator block supports wrapped states that you can enable by checking Wrap state on the block parameter dialog box. When you enable Wrap state, the block icon changes to indicate that the block has wrapping states.

The Integrator block supports wrapping states that are bounded by upper and lower values parameters of the wrapped state. The algorithm for determining wrapping states is given by:

$$y=\{\begin{array}{cc}x& x\in [x_l,x_u)\\ x-(x_u-x_l)\lfloor \frac{x-x_l}{x_u-x_l}\rfloor & \text{otherwise}\end{array}$$

where:

• x_l is the lower value of the wrapped state.

• x_u is the upper value of the wrapped state.

• y is the output.

The support for wrapping states provides these advantages.

• It eliminates simulation instability when your model approaches large angles and large state values.

• It reduces the number of solver resets during simulation and eliminates the need for zero-crossing detection, improving simulation time.

• It eliminates large angle values, speeding up computation of trigonometric functions on angular states.

• It improves solver accuracy and performance and enables unlimited simulation time.

Limiting the Integral

To limit the output signal to a specified range of values, select Limit output and specify the saturation limits. When the output reaches one of the limits, the integral action is disabled to prevent integral wind up. During a simulation, you can change the limits but not whether the output is limited. The block determines output signal value using these criteria:

• When the integral value is less than or equal to the lower saturation limit, the output signal value is the lower saturation limit.

• When the integral value is between the lower saturation limit and the upper saturation limit, the output is the integral value.

• When the integral value is greater than or equal to the upper saturation limit, the output signal value is the upper saturation limit.

To generate a signal that indicates when the state is limited by the saturation limits, select Show saturation port. A second output port appears on the block.

The saturation signal has one of three values:

• 1 — State limited by upper saturation limit

• 0 — State not limited

• –1 — State limited by lower saturation limit

When you limit the Integrator block output, the block has three zero crossing signals: one to detect when the integral value exceeds the upper saturation limit, one to detect when the integral value is less than the lower saturation limit, and one to detect when the integral value changes from being saturated to not being saturated.

Note

By default, the Limit output parameter is enabled for the Integrator Limited block, with the Upper saturation limit parameter value set to 1, and the Lower saturation limit parameter value set to 0.

Resetting the State

The block can reset its state to the specified initial condition based on an external signal. To cause the block to reset its state, select one of the External reset choices. A trigger port appears below the block's input port and indicates the trigger type.

• Select rising to reset the state when the reset signal rises from a negative or zero value to a positive value.

• Select falling to reset the state when the reset signal falls from a positive value to a zero or negative value.

• Select either to reset the state when the reset signal changes from zero to a nonzero value, from a nonzero value to zero, or changes sign.

• Select level to reset the state when the reset signal is nonzero at the current time step or changes from nonzero at the previous time step to zero at the current time step.

• Select level hold to reset the state when the reset signal is nonzero at the current time step.

The reset port has direct feedthrough. If the block output feeds back into this port, either directly or through a series of blocks with direct feedthrough, an algebraic loop results (see Algebraic Loop Concepts). Use the Integrator block's state port to feed back the block's output without creating an algebraic loop.

Note

To be compliant with the Motor Industry Software Reliability Association (MISRA™) software standard, your model must use Boolean signals to drive the external reset ports of Integrator blocks.

About the State Port

Selecting the Show state port check box on the Integrator block's parameter dialog box causes an additional output port, the state port, to appear at the top of the Integrator block.

The output of the state port is the same as the output of the block's standard output port except for the following case. If the block is reset in the current time step, the output of the state port is the value that would have appeared at the block's standard output if the block had not been reset. The state port's output appears earlier in the time step than the output of the Integrator block's output port. Use the state port to avoid creating algebraic loops in these modeling scenarios:

• Systems that use self-resetting integrators

• Passing states from one enabled subsystem to another

  Note

  When updating a model, Simulink checks that the state port applies to one of these two scenarios. If not, an error message appears. Also, you cannot log the output of this port in a referenced model that executes in Accelerator mode. If logging is enabled for the port, Simulink generates a "signal not found" warning during execution of the referenced model.

Specifying the Absolute Tolerance for the Block Outputs

By default Simulink software uses the absolute tolerance value specified in the Configuration Parameters dialog box (see Error Tolerances for Variable-Step Solvers) to compute the output of the Integrator block. If this value does not provide sufficient error control, specify a more appropriate value in the Absolute tolerance field of the Integrator block dialog box. The value that you specify is used to compute all the block outputs.

Examples

Create Self-Resetting Integrator

When you need an Integrator block to reset based on the block output value, you can use the state port to avoid creating an algebraic loop.

Open the model SelfResettingIntegratorAlgLoop. The Integrator block in the model is configured to reset when the reset signal changes from a positive value to a value of zero or a negative value. The reset signal is created by subtracting the current block output value from 1 so that the block resets itself each time the output value reaches 1.

mdl = "SelfResettingIntegratorAlgLoop";
open_system(mdl)

Because the reset port has direct feedthrough, using the output signal to calculate the value of the reset signal creates an algebraic loop. The output signal value depends on the value of the reset signal and vice versa. Because of this mutual dependence, the software cannot calculate either value.

When you try to simulate the model, the software issues diagnostics about the algebraic loop. To see the algebraic loop in the model, use the Simulink.BlockDiagram.getAlgebraicLoops function. The Algebraic Loops viewer opens and shows that the model contains one real algebraic loop.

Simulink.BlockDiagram.getAlgebraicLoops(mdl);

The algebraic loop is highlighted in the model.

The Integrator block calculates the state value before the output value in each time step. Because the state value is available before the output value, you can use the state value to create a self-resetting Integrator block without creating an algebraic loop.

To see this solution, open the model SelfResettingIntegrator. The Integrator block has the state port enabled. Instead of using the output signal to calculate the value of the reset signal, the model uses the state value.

mdl2 = "SelfResettingIntegrator";
open_system(mdl2)

To verify that using the state instead of the output breaks the algebraic loop, call the Simulink.BlockDiagram.getAlgebraicLoops function again.

Simulink.BlockDiagram.getAlgebraicLoops(mdl2);

No algebraic loops were found.

To see the self-resetting Integrator in action, simulate the model.

simOut = sim(mdl2);

To view the output and reset signals in the Scope, double-click the Scope block. The Integrator block output signal oscillates between a value of 0 and 1. The output value resets to 0 each time the reset signal value is 0.

Pass States Between Enabled Subsystems

You can use the state port to prevent creating an algebraic loop when you need an Integrator block in one enabled subsystem to use the state of an Integrator block in another enabled subsystem.

Open the model EnabledSubsystemStatesAlgLoop. The model contains two enabled subsystems, A and B. A Constant block provides the first input to each subsystem. The output from subsystem A is the other input for subsystem B. The output from subsystem B is the other input for subsystem A. A Pulse block provides the enable signal for both subsystems. The Pulse block enables subsystem B directly. The Pulse block output is inverted using a Logical Operator block to create the enable signal for subsystem A. As a result, the execution of the enabled subsystems alternates between subsystem A and subsystem B as the value of the Pulse block output changes.

mdl = "EnabledSubsystemStatesAlgLoop";
open_system(mdl);

The enabled subsystems each contain an Integrator block that uses an input port to provide the initial condition for the block. The Enable block in each subsystem is configured to reset the states each time the subsystem executes. When the state of the Integrator block resets, the initial value comes from the initial condition input port, which is connected to the output of the Integrator block in the other subsystem. Passing the state from one subsystem to another allows for continual integration of the signal while alternating execution between the subsystems.

However, to compute the output value from subsystem B, the solver needs the output from subsystem A and vice versa. Connecting the output of one Integrator block to the input of the other creates an algebraic loop. To see the loop in the model, use the Simulink.BlockDiagram.getAlgebraicLoops function. The Algebraic Loops viewer opens and shows that the model contains one real algebraic loop.

Simulink.BlockDiagram.getAlgebraicLoops(mdl);

The algebraic loop is highlighted in the model.

To avoid creating this algebraic loop, you can use the state port instead of the output port to pass the state of the Integrator block in one subsystem to the initial condition of the Integrator block in the other. The solver computes the block state value at an earlier point in each time step, so the output for subsystem B no longer depends on the output from subsystem A and vice versa.

To see this solution, open the model EnabledSubsystemStates. The input, enable, and output signals are the same, but the output from subsystem A no longer acts as a second input for subsystem B and vice versa.

mdl2 = "EnabledSubsystemStates";
open_system(mdl2)

The software does not support connecting the state port of the Integrator block to an output port on an enabled subsystem. The state port of the Integrator block in subsystem A is connected to the initial condition port for the Integrator block in subsystem B using GoTo and From blocks with global visibility. The state port of the Integrator block in subsystem B connects to the initial condition port for the Integrator block in subsystem A using the same strategy.

To verify that using the state port resolved the algebraic loop, use the Simulink.BlockDiagram.getAlgebraicLoops function again.

Simulink.BlockDiagram.getAlgebraicLoops(mdl2);

No algebraic loops were found.

Simulate the model. To view the results, double-click the Scope block. The Scope window displays the enable signal and the output signals from each subsystem.

out = sim(mdl2);

For another example of a system that passes state values between enabled subsystems, see Building a Clutch Lock-Up Model.

Extended Examples

Automotive Suspension

Model a simplified half-car model that includes an independent front and rear vertical suspension. The model also includes body pitch and bounce degrees of freedom. The example provides a description of the model to show how simulation can be used to investigate ride characteristics. You can use this model in conjunction with a powertrain simulation to investigate longitudinal shuffle resulting from changes in throttle setting.

• Open Model

Model Stick-Slip Friction and Hard Stops in Mass-Spring-Damper System

One way you can incorporate hard stops and friction changes from stick-slip motion into a mass-spring-damper model.

• Open Live Script

Model an Anti-Lock Braking System

Model a simple Anti-Lock Braking System (ABS). The model simulates the dynamic behavior of a vehicle under hard braking conditions. The model represents a single wheel, which may be replicated a number of times to create a model for a multi-wheel vehicle.

• Open Model

Ports

Input

Port_1 — Integrand signal

real scalar or array

Signal that needs to be integrated.

Data Types: double

External Reset — Reset state to initial conditions

real scalar or array

Reset the state to the specified initial conditions based on an external signal. See Resetting the State.

Dependencies

To enable this port, enable the External Reset parameter.

Data Types: Boolean

x₀ — Initial condition

real scalar or array

Set the initial condition of the block's state from an external signal.

Dependencies

To enable this port, set the Initial Conditions parameter to external.

Data Types: double

Output

Port_1 — Output signal

real scalar or array

Output the integrated state.

Data Types: double

Port_2 — Show output saturation

-1 | 0 | 1

Indicate when the state is being limited. The signal has a value of 1 when the integral is limited by the specified Upper saturation limit. When the signal is limited by the Lower saturation limit, the signal value is -1. When the integral is between the saturation limits, the signal value is 0. See Limiting the Integral.

Data Types: double

Port_3 — State

real scalar or array

Output the state of the block. See About the State Port.

Dependencies

Enable this port by enabling the Show state port parameter.

Data Types: double

Parameters

External reset — Reset states to their initial conditions

none (default) | rising | falling | either | level | level hold

Specify the type of trigger to use for the external reset signal.

• Select rising to reset the state when the reset signal rises from a negative or zero value to a positive value, or a negative value to zero value.

• Select falling to reset the state when the reset signal falls from a positive value to a zero or negative value, or from a zero value to negative value.

• Select either to reset the state when the reset signal changes from zero to a nonzero value, from a nonzero value to zero, or changes sign.

• Select level to reset the state when the reset signal is nonzero at the current time step or changes from nonzero at the previous time step to zero at the current time step.

• Select level hold to reset the state when the reset signal is nonzero at the current time step.

Programmatic Use

Block Parameter: ExternalReset

Type: character vector , string

Values: 'none' | 'rising' | 'falling' | 'either' | 'level' | 'level hold'

Default: 'none'

Initial condition source — Option to set initial condition using external signal

internal (default) | external

Select source of initial condition:

• internal — Get the initial conditions of the states from the Initial condition parameter.

• external — Get the initial conditions of the states from an external signal. When you select this option, an input port appears on the block.

Programmatic Use

Block Parameter: InitialConditionSource

Type: character vector, string

Values: 'internal' | 'external'

Default: 'internal'

Initial condition — Initial state

0 (default) | real scalar or array

Set the initial state of the Integrator block.

Tips

Simulink software does not allow the initial condition of this block to be inf or NaN.

Dependencies

Setting Initial condition source to internal enables this parameter.

Setting Initial condition source to external disables this parameter.

Programmatic Use

Block Parameter: InitialCondition

Type: scalar or vector

Default: '0'

Limit output — Limit block output values to specified range

off (default for Integrator) | on (default for Integrator Limited)

Limit the block's output to a value between the Lower saturation limit and Upper saturation limit parameters.

• Selecting this check box limits the block output to a value between the Lower saturation limit and Upper saturation limit parameters.

• Clearing this check box does not limit the block output values.

Dependencies

Selecting this parameter enables the Lower saturation limit and Upper saturation limit parameters.

Programmatic Use

Block Parameter: LimitOutput

Type: character vector , string

Values: 'off' | 'on'

Default: 'off'

Upper saturation limit — Upper limit for the integral

inf (default) | scalar | vector | matrix

Specify the upper limit for the integral as a scalar, vector, or matrix. You must specify a value between the Output minimum and Output maximum parameter values.

Dependencies

To enable this parameter, select the Limit output check box.

Programmatic Use

Block Parameter: UpperSaturationLimit

Type: character vector, string

Values: scalar | vector | matrix

Default: 'inf'

Lower saturation limit — Lower limit for the integral

-inf (default) | scalar | vector | matrix

Specify the lower limit for the integral as a scalar, vector, or matrix. You must specify a value between the Output minimum and Output maximum parameter values.

Dependencies

To enable this parameter, select the Limit output check box.

Programmatic Use

Block Parameter: LowerSaturationLimit

Type: character vector , string

Values: scalar | vector | matrix

Default: '-inf'

Wrap state — Enable wrapping of states

off (default) | on

Enable wrapping of states between the Wrapped state upper value and Wrapped state lower value parameters. Enabling wrap states eliminates the need for zero-crossing detection, reduces solver resets, improves solver performance and accuracy, and increases simulation time span when modeling rotary and cyclic state trajectories.

If you specify Wrapped state upper value as inf and Wrapped state lower value as -inf, wrapping does not occur.

Dependencies

Selecting this parameter enables Wrapped state upper value and Wrapped state lower value parameters.

Programmatic Use

Block Parameter: WrapState

Type: character vector, string

Values: 'off' | 'on'

Default: 'off'

Wrapped state upper value — Specify upper value for wrapped state

pi (default) | scalar | vector

Upper limit of the block output.

Dependencies

Selecting Wrap state enables this parameter.

Programmatic Use

Block Parameter: WrappedStateUpperValue

Type: scalar or vector

Values: '2*pi'

Default: 'pi'

Wrapped state lower value — Specify lower value for wrap state

-pi (default) | scalar | vector

Lower limit of the block output.

Dependencies

Selecting Wrap state enables this parameter.

Programmatic Use

Block Parameter: WrappedStateLowerValue

Type: scalar or vector

Values: '0'

Default: '-pi'

Show saturation port — Enable saturation output port

off (default) | on

Select this check box to add a saturation output port to the block. When you clear this check box, the block does not have a saturation output port.

Dependencies

Selecting this parameter enables a saturation output port.

Programmatic Use

Block Parameter: ShowSaturationPort

Type: character vector , string

Values: 'off' | 'on'

Default: 'off'

Show state port — Enable state output port

off (default) | on

Select this check box to add a state output port to the block. When you clear this check box, the block does not have a state output port.

Dependencies

Selecting this parameter enables a state output port.

Programmatic Use

Block Parameter: ShowStatePort

Type: character vector , string

Values: 'off' | 'on'

Default: 'off'

Absolute tolerance — Absolute tolerance for block states

auto (default) | real scalar or vector

• If you enter auto or –1, then Simulink uses the absolute tolerance value in the Configuration Parameters dialog box (see Solver Pane) to compute block states.

• If you enter a real scalar, then that value overrides the absolute tolerance in the Configuration Parameters dialog box for computing all block states.

• If you enter a real vector, then the dimension of that vector must match the dimension of the continuous states in the block. These values override the absolute tolerance in the Configuration Parameters dialog box.

Programmatic Use

Block Parameter: AbsoluteTolerance

Type: character vector, string, scalar, or vector

Values: 'auto' | '-1' | any positive real scalar or vector

Default: 'auto'

Ignore limit and reset when linearizing — Option to use default reset and limit behavior for block during linearization

off (default) | on

When you select this parameter, the software ignores reset and limit parameters for this block during linearization and uses default values for those parameters.

Tip

Enable this parameter when you want to linearize a model around an operating point that causes the integrator to reset or saturate.

Programmatic Use

Block Parameter: IgnoreLimit

Type: character vector | string

Values: 'off' | 'on'

Default: 'off'

Enable zero-crossing detection — Enable zero-crossing detection

on (default) | off

Select to enable zero-crossing detection. For more information, see Zero-Crossing Detection.

Programmatic Use

Block Parameter: ZeroCross

Type: character vector | string

Values: 'off' | 'on'

Default: 'on'

State Name (e.g., 'position') — Assign unique name to each state

' ' (default) | character vector | string

• To assign a name to a single state, enter the name between quotes, for example, 'velocity'.

• To assign names to multiple states, enter a comma-delimited list surrounded by braces, for example, {'a', 'b', 'c'}. Each name must be unique.

• The state names apply only to the selected block.

• The number of states must divide evenly among the number of state names.

• You can specify fewer names than states, but you cannot specify more names than states.

  For example, you can specify two names in a system with four states. The first name applies to the first two states and the second name to the last two states.

• To assign state names with a variable in the MATLAB^® workspace, enter the variable without quotes. A variable can be a character vector, string, cell array, or structure.

Programmatic Use

Block Parameter: ContinuousStateAttributes

Type: character vector, string

Values: ' ' | user-defined

Default: ' '

Block Characteristics

Data Types	double
Direct Feedthrough	yesa
Multidimensional Signals	no
Variable-Size Signals	no
Zero-Crossing Detection	yes
a Ports of this block have different direct feedthrough characteristics.

Extended Capabilities

C/C++ Code Generation

Generate C and C++ code using Simulink® Coder™.

• Consider discretizing the model

• Not recommended for production code

Version History

Introduced before R2006a