evalfisOptions

Option set for evalfis function

Description

Use an evalfisOptions object to specify options for the evalfis function.

Creation

Syntax

opt = evalfisOptions

opt = evalfisOptions(PropertyName=Value)

Description

opt = evalfisOptions returns a default option set for evaluating a FIS using evalfis. To modify options, use dot notation.

opt = evalfisOptions(PropertyName=Value) specifies options using one or more name-value arguments. For example, opt = evalfisOptions(NumSamplePoints=51) sets the number of output fuzzy set samples to 51.

Properties

NumSamplePoints — Number of sample points in output fuzzy sets

101 (default) | integer greater than 1

Number of sample points in output fuzzy sets, specified as an integer greater than 1.

To reduce memory usage while evaluating a Mamdani FIS, specify fewer samples. Doing so sacrifices the accuracy of the defuzzified output value.

Reducing the number of samples can make the output area for defuzzification zero. In this case, the defuzzified output value is the midpoint of the output variable range.

Dependencies

evalfis ignores this property when evaluating a Sugeno FIS.

OutOfRangeInputValueMessage — Diagnostic message behavior when an input is out of range

"warning" (default) | "error" | "none"

Diagnostic message behavior when an input is out of range, specified as one of these values:

• "warning" — Report the diagnostic message as a warning.

• "error" — Report the diagnostic message as an error.

• "none" — Do not report the diagnostic message.

When an input value is out of range, corresponding rules in the fuzzy system can have unexpected firing strengths.

NoRuleFiredMessage — Diagnostic message behavior when no rules fire

"warning" (default) | "error" | "none"

Diagnostic message behavior when no rules fire, specified as one of these values:

• "warning" — Report the diagnostic message as a warning.

• "error" — Report the diagnostic message as an error.

• "none" — Do not report the diagnostic message.

When NoRuleFiredMessage is "warning" or "none" and no rules fire for a given output, the defuzzified output value is set to its mean range value.

EmptyOutputFuzzySetMessage — Diagnostic message behavior when an output fuzzy set is empty

"warning" (default) | "error" | "none"

Diagnostic message behavior when an output fuzzy set is empty, specified as one of these values:

• "warning" — Report the diagnostic message as a warning.

• "error" — Report the diagnostic message as an error.

• "none" — Do not report the diagnostic message.

When EmptyOutputFuzzySetMessage is "warning" or "none" and an output fuzzy set is empty, the defuzzified value for the corresponding output is set to its mean range value.

This diagnostic message applies only to Mamdani systems.

Object Functions

evalfis

Evaluate fuzzy inference system

Examples

Create Option Set for Evaluating FIS

Create option set object, specifying the number of sample points for output fuzzy sets.

options = evalfisOptions(NumSamplePoints=51)

options = 
  EvalFISOptions with properties:

                NumSamplePoints: 51
    OutOfRangeInputValueMessage: "warning"
             NoRuleFiredMessage: "warning"
     EmptyOutputFuzzySetMessage: "warning"

Alternatively, create a default option set, and configure properties using dot notation.

options = evalfisOptions;
options.NumSamplePoints = 51;

Extended Capabilities

C/C++ Code Generation

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

Usage notes and limitations:

• When used for code generation, an evalfisOptions object stores its OutOfRangeInputValueMessage, NoRuleFiredMessage, and EmptyOutputFuzzySetMessage properties as character vectors rather than strings.

• When evaluating a fuzzy inference system in Simulink^®, it is recommended to not use evalfis or evalfisOptions within a MATLAB Function block. Instead, evaluate your fuzzy inference system using the Fuzzy Logic Controller block.

Version History

Introduced in R2018a

R2018b: C and C++ code generation support

You can generate code for loading and evaluating a fuzzy inference system (FIS) using MATLAB^® Coder™. For an example, see Generate Code for Fuzzy System Using MATLAB Coder.