Soft sensing of melt temperature in polymer extrusion

Precise monitoring techniques are invaluable to any process for diagnosing its operational health, safety concerns and also for achieving good process control. In polymer extrusion, it is quite difficult to visually observe the melt inside barrel during the process operation and hence the level of control of the process operational quality is highly dependent upon the process monitoring techniques. Currently, a number of physical sensing devices are widely available in industry for monitoring of parameters such as melt temperature, melt pressure, screw speed and so forth. However, there are some limitations to use physical sensors in process measurements due to several constraints such as their access requirements, disruptive effects on the melt flow, fragility, complexity, etc. Thus, the application of soft sensing techniques should be highly useful for improved process monitoring and hence for advanced process control. In this work, a general discussion is made on the soft sensors and soft sensing applications in polymer extrusion. Then, a soft sensor concept is proposed for the die melt temperature profile prediction in polymer extrusion. The simulation results showed that the proposed technique can predict the temperature profile across the melt flow in real-time with good accuracy. Eventually, the importance of developing of such soft sensing techniques is discussed while providing some of the possible directions for future research.