Research Article
An Improved Power Quality Disturbance Detection Using Deep Learning Approach
Table 7
Evaluation parameters of the three models.
| Actual | CNN + LSTM | CNN + GRU | DCNN | Precision | Recall | F1 score | Precision | Recall | F1 score | Precision | Recall | F1 score |
| C1 | 1.0000 | 0.9902 | 0.9951 | 0.9868 | 0.9901 | 0.9885 | 0.9868 | 0.9901 | 0.9885 | C2 | 0.9938 | 1.0000 | 0.9969 | 0.9938 | 1.0000 | 0.9969 | 0.9938 | 1.0000 | 0.9969 | C3 | 0.9902 | 1.0000 | 0.9951 | 0.9935 | 0.9902 | 0.9918 | 0.9935 | 0.9902 | 0.9918 | C4 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | C5 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | C6 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | C7 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | C8 | 1.0000 | 0.9971 | 0.9986 | 1.0000 | 0.9971 | 0.9986 | 1.0000 | 0.9971 | 0.9986 | C9 | 1.0000 | 1.0000 | 1.0000 | 0.9971 | 0.9971 | 0.9971 | 0.9971 | 0.9971 | 0.9971 | C10 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 0.9967 | 0.9984 | 1.0000 | 0.9967 | 0.9984 | C11 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | C12 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | C13 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 0.9968 | 1.0000 | 0.9984 | C14 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 0.9936 | 0.9968 | 0.9968 | 0.9936 | 0.9952 | C15 | 1.0000 | 1.0000 | 1.0000 | 0.9945 | 1.0000 | 0.9972 | 0.9945 | 0.9917 | 0.9931 | C16 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 0.9939 | 0.9970 | 0.9954 | Average | 0.9990 | 0.9992 | 0.9991 | 0.9979 | 0.9978 | 0.9978 | 0.9971 | 0.9971 | 0.9971 |
|
|