With the development of intelligent transportation systems and advanced driver assistance systems, early prediction of driver’s intentions is required. Among that, lane change is a basic maneuver driver conducted during driving, improper lane changing could cause severe traffic accident. Thus, it is urgent to develop lane change prediction systems, especially which could work in the initial phase of lane change. Hence, in the present study, a six degree of freedom motion-based driving simulator with 18 drivers were involved in a lane change experiment, data of steering wheel angle and steering wheel velocity during both lane change and lane keep phase were collected and compared. The results showed that, in the phase up to 0.55s before the onset of lane change, means and variances of the steering wheel velocity showed a significant difference with that in lane keep phase, which suggested a potential indicator for lane change detection before the normal onset of lane change. Therefore, Delta -Lognormal theory, which had been developed for rapid human movements, was used to construct a model for steering wheel velocity, during the intention phase and open-loop phase of lane change. Results showed that the Delta –Lognormal model fitted the intention phase and open-loop phase of lane change very well. This model could be used to predict lane change in the future, thus used in vehicle active safety and intelligent transportation system.