This block implements a
numerical fuzzification of an input by associating to it a triangular
membership function.
Input : real
Output : vector of
dimension 4 coding a triangular membership function
Parameters
- If the number of points is equal to 1 and
the parameters "left value" and "right value" are at
0, the triangular membership function is reduced to the singleton
associated to the block input.
- If the number of points is equal to 1 and
the parameters "left value" and "right value" are
different from 0, the triangular membership function has its modal value
defined by the block input and a support defined by the left and the right values.
- If the number of points is higher than 5,
one generates a membership function whose parameters are calculated from
the mean of the points and the standard-deviation. The values of the
parameters "left value" and "right value" are without
effect. The associated computing is based on the approximation of a
probability distribution into a distribution of possibility. The
membership function has the mean as modal value and a support calculated
from 3.2*standard-deviation.