The quality of plastic parts produced by injection molding process is highly affected by several key process parameters. Examples of important parameters are melt and mold temperature cavity pressure injection velocity. Many studies both theoretical and experimental have confirmed that repeatability of those crucial parameters is essential. This paper presents closed loop control of process parameters in injection molding using predictive control algorithm. Development of system to allow closed loop control to be performed was also performed. This involved the usage of several sensors for temperature cavity pressure as well as screw velocity. A multi input multi output model has been developed in order to conduct barrel heater control. An accurate model incorporated all of the influences due to heat transfer of the surrounding zone was obtained using system identification technique. Therefore good performance of melt temperature control was achieved for both simulations and experimental works. The developed model can also be used for any number of barrel zones on any injection molding machines. Cavity pressure and cavity temperature control was performed by using coolant flow rate as the manipulated variable. Cavity pressure control was conducted during the coolant phase
since it takes almost 80 of the cycle time of the production. Dynamic matrix control which uses the model of cavity pressure during cooling explicitly was employed in this research. The results indicate that by altering the coolant flow rate certain cavity profile can be obtained and maintained throughout the entire production process. Same manipulated variable was also be used to carry out cavity temperature control. A second order model with delay was derived to represent the relationship between the controlled and manipulated variables. Proportional and integral control was employed in this work. The overall control system demonstrated an efficient
and effective cooling strategy. Simplified predictive control SPC was employed to perform the screw velocity control. Velocity control becomes an uneasy task since it has many possible velocity profiles in order to
obtain good products. Velocity profiling is important as it can be used to achieve a constant melt flow front velocity. A position transducer available on the machine was used as a feedback by extracting velocity information of it via high speed analog processing of the position voltage signal. The simulations as well as experimental results indicate that SPC algorithm can effectively be used to control various injection speed profiles.