Fault Location Method in Nonsolid-Earthed Network Based on Spatial Domain Image Fusion and Convolution Neural Network

Table 5

Results of fault location under different phase-mode transformation matrixes.

Fault types	Fault distance (km)	Clarke (km)	Range error of Clarke (km)	This paper (km)	Range error of this paper (km)
AG	0.5	0.7350	0.2350	0.5554	0.0554
	1.0	1.0652	0.0652	0.8966	-0.1034
	2.2	2.5064	0.3064	2.4352	0.2352
	4.5	491	0.3991	4.4899	-0.0101
	5.5	6.7651	1.2651	5.5770	0.0770
	6.0	6.6974	0.6974	6.0206	0.0206
	7.9	7.4039	-0.4961	8.0608	0.1608
BG	0.5	3.2778	2.7778	0.6003	0.1003
	1.0	3.8437	2.8437	0.8025	-0.1975
	2.2	3.7938	1.5938	2.3667	0.1667
	4.5	5.3339	0.8339	4.5479	0.0479
	5.5	5.4917	-0.0083	5.4859	-0.0141
	6.0	5.5436	-0.4564	6.0423	0.0423
	7.9	5.4927	-2.4073	8.0606	0.1606
CG	0.5	0.5973	0.0973	0.6811	0.1811
	1.0	0.9620	-0.0380	0.8551	-0.1449
	2.2	2.3238	0.1238	2.3071	0.1071
	4.5	4.7555	0.2557	4.5175	0.0175
	5.5	6.7567	1.2567	5.3396	-0.1604
	6.0	6.7053	0.7053	5.9139	-0.0861
	7.9	7.4334	-0.4666	7.8825	-0.0175