||Injection ram velocity to a large degree determines the melt injection rate during the injection phase in an injection molding process, and has strong influences on the molded part quality, such as shrinkage, warpage, and impact strength. An injection molding machine operates under strongly different operating conditions, such as different set-point profiles, barrel temperatures, molds and materials. This causes the ram velocity dynamics to vary significantly, and consequently results in poor control performance for a typical PID controller. This thesis presents a computer control system for the injection ram velocity using a real-time fuzzy logic controller (FLC), together with a fuzzy feedforward controller (FFC). The rule base of the FLC is optimized by analyzing the phase plane characteristics and, the optimal membership functions of FLC are based on the 2k factors design technique. The experimental results reveal that the controller has improved performance over the conventional PID controller, in the response speed, set-point tracking ability, noise rejection, and robustness. In this study, the non-linearity and time-varying characteristics of the injection ram velocity have been investigated by an experimental model as well as a simplified physically-based model. Finally, the possibility of applying an adaptive fuzzy controller (AFC) to the ram velocity control is explored in this study. An AFC is designed and tested for the control of ram velocity during filling. The experimental results show that the controller worked in the cup mold, but failed in the modified flat mold. A possible reason for this may be the strong effect of different molds on the dynamics of ram velocity and the use of cup mold data for the adaptive rate y and the input and output variables fuzzy sets. The application of AFC to injection molding may give a direction for further study.