Nonlinear Fault Separation for Redundancy Process Variables Based on FNN in MKFDA Subspace

Table 4

The accuracy with different feature sets with testing data.

Feature set	Combination of variables	Accuracy
Set1	Vab.21	83.521%
Set2	Vab.21, Vab.13	85.731%
Set3	Vab.21, Vab.13, Vab.9	89.652%
Set4	Vab.21, Vab.13, Vab.9, Vab.16	92.123%
Set5	Vab.21, Vab.13, Vab.9, Vab.16, Vab.7	94.547%
Set6	Vab.21, Vab.13, Vab.9, Vab.16, Vab.7, Vab.20	92.532%
⋮	⋮	⋮
Set53	Vab.21, Vab.13, Vab.9, Vab.16, Vab.7, Vab.20, … Vab.42	72.119%