Fig 1.
BPNN topology diagram.
Fig 2.
Flow chart of the algorithm structure of this paper.
Table 1.
BFO-PSO-VMD-R-LMS-BPNN detection pseudocodes.
Fig 3.
Freiberg plus human equivalent circuit model.
Table 2.
Experimental setup parameters.
Fig 4.
Iterative plots of k, ɑ, envelope entropy.
Fig 5.
IMFs component.
Fig 6.
(a) Simulated signal (b) Reconstructed signal.
Fig 7.
BPNN detection and prediction.
(a) Actual and detected values (b) Actual and predicted values.
Table 3.
Comparison of the detection results.
Fig 8.
Residual current real measurements for different types.
(a) Grass (b) Poplar (c) Wetland (d) Concrete.
Fig 9.
Iterative plots of k, ɑ, envelope entropy.
Fig 10.
IMFs component.
Fig 11.
Comparison of the original simulated signal and the noise reduction signal.
(a) Original signal (b) Noise reduction signal.
Fig 12.
LMS noise reduction signal (a) Poplar tree (b) Grass (c) Concrete ground.
Fig 13.
BPNN plant electrocution current detection.
(a) Actual and detected values (b) Actual and predicted values.
Fig 14.
Single-phase direct grounding of BPNN line.
(a) Actual and detected values (b) Actual and predicted values.
Table 4.
Comparison of the detection results.
Table 5.
Kolmogorov-Smirnov test data analysis table.