Journal of Multidisciplinary Engineering Science and Technology (JMEST) ISSN: 3159-0040 Vol. 2 Issue 3, March - 2015 www.jmest.org JMESTN42350521 376 In Process Density of HDPE Pipe Material Prediction Using Artificial Neural Network in a Polymer Extruder Bekir Cirak* Siirt University, Engineering and Architectural Faculty, Department of Mechanical Engineering, Kezer Campus, Siirt – Turkey Tel: +9004842234008 Fax: +9004842234009 GSM:+9005072053443 bekircirak@mynet.com Abstract—The pipe extruder process from the traditional pipe products. The model is based on three layer neural network with back propagation learning algorithm. The training data are collected by the experimental setup in the laboratory. The predicted density of neural network model, coincide well with the experimental density. A three layer artificial neutral network ANN model was used for the description of extrusion density. The studies employ experimental data obtained from capillary flow experiments using HDPE molten. On comparing the experimental data, the predictions using the ANN model predictions; it is found that the ANN model is capable of predicting the extrusion density well. The neural network model shows how the significant parameters influencing pipe material density can be found. A three layer multi-layer perceptron artificial neural network was used to correlate the response of the densitymeter to the measured molten density. The density of the extruded pipe was varied by varying the processing conditions over a wide practical range. In this study, an artificial neural network approach for exploring the prediction of the pipe extruder process parameters, molten HDPE density of pipe extrusion product is derived. Keywords—Density; HDPE; Pipe; Extrusion; ANN 1. Introduction The objective of this work is to develop a correlation between the measured properties of the extruded polymer melt and the density of the pipe material. A robust in line monitoring technique can provide plastics manufacturers with the ability to respond to material and process variations accurately and fast in order to maintain the quality and cost of the extruded material within specifications. In addition, a real-time monitoring technique can help to automate many of the plastics manufacturing processes; therefore minimizing the need for the scarce highly skilled machine operators. In pipe extrusion processes, product density is considered to be the most significant factor affecting the production cost and profitability of the manufacturing process. Density of HDPE (High Density Poly Ethylen) can be controlled by varying the type and amount of compound additives, processing parameters, or both [1]. Compound additives; such as blowing agent, nucleation agent, and process aid have a significant effect on the foam density and their effect was shown to be interrelated [2]. Processing parameters such as the temperature of the barrel’s heating zones and screw speed have also shown a considerable effect on the foam density [3]. Many publications have presented studies on the design and performance of foam extrusion processes [4–7]. Artificial neural network (ANN) models have been studied in recent years, with an objective of achieving humanlike performance in many fields of knowledge engineering. Neural network applications are growing rapidly as artificial intelligence tools in the area of speech recognition, pattern recognition, and in robotics and communication [8]. 2. Material and experimental procedure The densitymeter is installed at the exit end of the extruder. The connecting of densitymeter and transducers is similar to the capillary rheometer as shown in Fig. 1. The melt density is measured basing on the measurement of the pressure drop and the flow rate through the process line. Single pressure and temperature transducers are installed in the barrel to measure the pressure and temperature.[9] Figure 1. Connecting of densitymeter and transducers In this experimental study used pipe extruder and compartments. This equipment is shown in Fig. 2. Therefore the means of compartments are follows; A- Extruder motors, B- Coupling elements, C- Barrel and three heater zones, D- Densitymeter, E- Hopper, F- Command panels.