When the fusion efficiency is low, the method based on information fusion technology will directly affect the accuracy of the grounding line selection . Respectively using EEMD and FFT to analyze the zero-sequence component of steady-state and transient, this paper sets up faulty measurement function to calculate fault measures of feeders and establishes fault measures of different methods based on information gain,finally, sample fault measures are founded by fusing the two fault measures. Then the depth information fusion line selection is conducted by using the least squares support vector machine (LSSVM) of the simplex optimum-seeking algorithm and the information gain degree with the sample falut measures as characteristic input. Simulation results show that the above method is of high accuracy and sensitivity in fault line selection.