• Yarlagadda, Prasad Kdv

Abstract

Pressure die casting is an important production process. In pressure die casting,the ® rst setting of process parameters is established through guess work. Expertsuse their previous experience and knowledge to develop a solution for a newapplication. Due to rapid expansion in the die casting process to producebetter quality products in a short period of time, there is ever increasingdemand to replace the time-consuming and expert-reliant traditional trial anderror methods of establishing process parameters. A neural network system isdeveloped to generate the process parameters for the pressure die casting process.The system aims to replace the existing high-cost, time-consuming and expertdependenttrial and error approach for determining the process parameters. Thescope of this work includes analysing a physical model of the pressure die casting® lling stage based on governing equations of die cavity ® lling and the collection offeasible casting data for the training of the network. The training data weregenerated by using ZN-DA3 material on a hot chamber die casting machinewith a plunger diameter of 60 mm. The present network was developed usingthe MATLAB application toolbox. In this work, the neural network was developedby comparing three di€ erent training algorithms: i.e. error backpropagationalgorithm; momentum and adaptive learning algorithm; and Levenberg±Marquardt approximation algorithm. It was found that the Levenberg±Marquardt approximation algorithm was the preferred method for this applicationas it reduced the sum-squared error to a small value. The accuracy of thedeveloped network was tested by comparing the data generated fromthe networkwith those of an expert froma local die casting industry. It was established that byusing this network the selection of process parameters becomes much easier, sothat it can be used by a novice user without prior knowledge of the die castingprocess or optimization techniques.1. IntroductionPressure die casting is an important production process that is extensively used toproduce castings for the electrical, electronic and automobile industries. The processhas its origins in type casting machines developed in 1822. The process showed itsproduction potential as early as the mid 1800s when it had reached a high level ofautomation and mechanical efficiency.In 1894, the ® rst die casting machine was developed, in which molten metal wasforced through an inclined port and out of the nozzle into the die by the central ramactuated by a lever. During the past two decades, the pressure die casting process hasbecome an essential casting production process for the engineering industry. Highproduction rate, excellent surface ® nish and good mechanical properties of the ® n-Revision received March 1999.

Topics

• impedance spectroscopy
• surface
• zinc
• zinc alloy
• die casting