Skip to main content

Table 10 Summary of feature selection techniques

From: A systematic review of real-time detection and classification of power quality disturbances

Set of techniques

Advantages

Disadvantages

Application

Multiple PQDs

Sags

Notches

Refs

%

Refs

%

Refs

%

Handcrafted

Physical interpretation of contextual features

Ease understanding

Time-consuming

Detailed knowledge of features and phenomena is usually required

[33,34,35,36,37,38,39,40,41, 43,44,45,46,47,48,49,50,51,52,53,54, 56,57,58,59,60,61,62, 64, 66,67,68,69,70,71,72,73,74, 77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94, 96,97,98,99, 101, 102, 104, 106,107,108, 110, 112, 113, 115, 117, 118, 120, 122, 126, 128]

77

[132,133,134,135,136,137,138,139,140, 142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157, 160, 161, 163, 166, 169, 172,173,174, 176, 178, 181, 182, 185, 186]

71

[187,188,189,190,191,192,193,194,195,196,197,198,199,200,201,202,203,204,205,206,207,208,209,210]

96

Optimization methods

Yield optimal sets of features

Sound theoretical background

Diverse methods

Not necessarily a physical interpretation of features

High computational burden

They rely on the proper a priori selection of relevant features

[55, 95, 102, 109, 114, 123, 127]

7

[168, 177, 185]

5

-

0

Dimensionality reduction

Physical interpretation of indices

Diverse methods and indices

May lead to some data loss

Handcrafted rules are required in most cases

[42, 62, 75, 76, 98, 100, 102, 110, 112]

9

[141, 162, 170, 177]

7

-

0