In this paper, an optimization method is proposed for the switching of the main and auxiliary branches of a Novel Controllable Commutation Converter (CLCC). Based on the theoretical model of commutation failure, the AC-side ground fault is decomposed into a DC component, harmonic component and overvoltage component, and a multidimensional fault dataset is constructed by combinatorial analysis. Using the Radviz visual multidimensional data analysis method, the correspondence between the control-and-protection thresholds of CLCC main- and auxiliary-branch switching under multi-component conditions is established, and the optimal switching thresholds under different short-circuit conditions of the AC system are determined. Finally, the validity of the method is verified by a field artificial short-circuit test, which provides theoretical basis and technical support for the reasonable determination of field setting.