Defuzzification

Defuzzification is the process of performing output renormalization that converts a set of controller output values to single point values and maps the point values of the controller output to the physical domain (Eatherley et al., 1994). Conceptually, the job of the defuzzifier is to assign a weighted point to W with the FS C' obtained as a result of the inference engine. Then, inverse fuzzification can be specified as a mapping of the set C' to the well-spoken deportation space z* e W in the fuzzy deportation space W c R in the world of speech output.

There are many choices to determine this well-performed controller output z* but when choosing a diffraction scheme, you need to consider the four criteria you seek:

• • Validity: The point z* should represent C' from an intuitive point of view. For example, it could roughly lie in the middle of the support of C' or have a higher tier of members at C'.
• • Calculation simplicity: This criterion is particularly important for fuzzy tenancy given that the fuzzy controller operates in real time.
• • Continuity: A small increase in C' should not have a large increase in z*.
• • Clarity: This is how the deepening method produces a unique value for z* without change.

Centroid Defuzzifier

This method is known as part of gravity or a region. This technique was matured by Sugeno. This is the most surprisingly used technique. The only downside to this method is that it is difficult to compute for the classified member functions.

Bisecting Defuzzifier

A bisector is a vertical line that divides a region into two subregions of the same region. Sometimes it coincides with the centerline, but not invariably.

Weighted Average Defuzzifier

This method is usually limited to a symmetric output membership function, and the weighted average method is formed by weighting each maximum membership value to each membership function in the output.

Midpoint of Maximum Defuzzifier

In this defuzzy method, only the clever rules with the highest fulfillment tier are considered.

Largest of Maximum Defuzzifier

The largest of the maximum values is the sharpest value tapped with ZLOM, taking the largest of all z bested in [z„ zj.

Smallest Maximum Defuzzifier

Select the smallest output with maximum membership function with good value ZSOM. That is, in the smallest of maximum, the smallest of all z given to [zj, z₂] is selected.