In this study, a novel accurate fault location algorithm is presented for two-terminal transmission lines. In contrast to conventional methods, the proposed algorithm not only utilises asynchronous samples recorded during the fault but also needs no line parameters and identification of fault type. In the presented fault-locating method, distributed parameter line model in the time domain and asynchronous data of the terminals collected during fault are applied. Fault locating as an optimisation problem has been solved by the heuristic algorithm of teaching–learning-based optimisation, and the decision variables of fault location, synchronisation time and line parameters are estimated simultaneously. The performance of the presented method was tested with different fault incidence angles, a variety of fault types, and under several system and fault conditions using the MATLAB/Simulink. These tests demonstrate the high accuracy of the presented method. Also, the proposed method did not show any dependence on the impedance of the Thevenin sources of the two sides of the line, the fault impedance and the fault incidence angle. Furthermore, it was not affected by the high resistance of the fault and the network structure.