Using feature extraction algorithm reduces the speed of intelligent fault location for HV power transmission line. To improve the speed, a novel high voltage power transmission line intelligent fault location scheme is proposed, combining with voltage amplitude and Support Vector Regression (SVR). Firstly, a 220 kV/300 km transmission line power system is set up by Matlab, all various fault impedances, various types of faults at different locations and various fault inception angles are simulated on the system. The system uses single-end measurements and only collects voltage fault signals. The collected fault voltage signals are inputted into low-pass filter to eliminate the noises. Then, voltage amplitude from 1/2 cycle of post fault signals are extracted and applied as fault feature signals. A SVR is trained and verified with the fault feature signals. Consequently, the precise location of fault on the transmission line is realized. The simulation results show that the system not only improves the speed of fault location to a large extent, but also has high accuracy of fault location. This work is supported by National Natural Science Foundation of China (No. 61403395).