The role of a power transformer protective relay is to rapidly operate the tripping during internal faults and block the tripping during magnetizing inrush. This paper presents a new approach for classifying transient phenomena in power transformer, which may be implemented in digital relaying for transformer differential protection. Discrimination between internal faults, external faults with current transformer saturation and magnetizing inrush currents is achieved by combining wavelet transforms and fuzzy logic. The wavelet transform is applied for the analysis of the power transformer transient phenomena because of its ability to extract information from the transient signals in both time and frequency domain. Fuzzy logic is used because of the uncertainty in the differential current signals and relay settings. MATLAB power system toolbox is used to generate current signals at both sides of a power transformer in a typical system with various conditions. The simulation results obtained show that the new algorithm provides a high operating sensitivity for internal faults and remains stable for external faults and inrush currents.