In order to improve the fault location accuracy of transmission lines , a detection method based on the ensemble empirical mode decomposition algorithm(EEMD) and the difference-entropy(DE) for transmission lines fault locating is proposed. Firstly, the improved differential entropy signal processing method is used to calculate the entropy of the transmission line current. It can be judged that whether the current has mutated by comparing the entropy with the normal value. Secondly, the falut current is decomposed by EEMD, which can effectively eliminate the noise in the fault signal and avoid the aliasing of the mode. In this process, the multiple intrinsic mode functions(IMF) components that accurately reflects the fault information is obtained. And then, the magnitude of the current variation will be quantified by the differential analysis of the differential entropy method. Besides, the corresponding relation between the degree of change of each signal point and the time will also be established, so that the position of the wave head and the initial time of the fault initial wave arriving at each monitoring point can be determined. Finally, the fault location of transmission lines will be determined by using the double terminal method of traveling wave. The simulation results show that the method can be used to solve the problem of current signal noise and mode aliasing in fault location of transmission lines, which can effectively improve the accuracy of fault location.