22-10-2014, 09:33 AM
Abstracts: As we find in most of the investment casting industries, after completion of whole process they find various defects in final product and try to decrease them by the selection of approximate change in process parameters which is not a convenient way to achieve defect free product. So we are going to develop an artificial intelligence neural network system. Using that we can reach more closely to the specific parameters which are responsible for particular defect.Our IDP is to develop an artificial intelligence neural network system to achieve the defect free investment casting product. The application of artificial intelligence in investment casting process can increase the accuracy of predicting the process parameter such as injection time, injection pressure ,melt temperature and die temperature and other physical factors like gate location complexity of the die cast parts and geometry of the parts that are not taken in to consideration. To improve the system capabilities, these effects have to be studied and then incorporated in training the system. A rule based experts system can be incorporated into the existing neural network system to develop an optimization system for the investment casting. The main objective of any industry is to increase the productivity in our case by using artificial intelligence the desirable parameters like quality of products,surface finishing, strength, dimension accuracy etc. will increase and undesirable parameters such as various defects, deformation, rejection will decrease. We have taken the several readings for the peocess of investment casting to find defects considering various parameters then we alwaysthe effects of different parameters on defect and plot the graph of then against % defects using matlab software. Future scope:- To get more concentrated optimum output of this problem the equation including of the variable can be use to achieve the optimum point in neural network system.