Application of Multiattention Mechanism in Power System Branch Parameter Identification

Table 2

Experimental results of grid dataset on index branch conductance .

Branch conductance	None	50 dB	30 dB	dropData	dropEdge	50 dB + DD + DE
LinearRegression	0.0052	0.0134	0.1019	0.2138	0.2922	0.3481
SVR	0.0985	0.1097	0.1583	0.0885	0.3196	0.3110
RF	0.0710	0.0714	0.1258	0.0769	0.3232	0.3192
XGboost	0.0754	0.0983	0.1642	0.1013	0.1257	0.1343
lightGBM	0.1043	0.1342	0.1871	0.1222	0.1465	0.1543
KNN	0.0951	0.1011	0.1498	0.1228	0.1028	0.1435
Bagging	0.0772	0.0992	0.1345	0.1173	0.1274	0.1371
FCN	0.0706	0.0777	0.2494	0.0723	0.0751	0.0751
GTN	0.0672	0.0682	0.0680	0.0678	0.0680	0.0681

The order of RMSE index used by branch conductance is . Here, we choose RMSE as the comparative index. The highest values for the different metrics are highlighted in bold.