Shear Stress Prediction Using FEA-ANN Hybrid Modeling Of Eicher 11.10 Chassis Frame

Abstract

The chassis serves as a backbone for supporting the body and different parts of the automobile. It should be rigid enough to withstand the shock, twist, vibration and other stresses. Along with strength (Shear Stress), an important consideration in chassis design is to have adequate bending stiffness (Deflection). The main objective of the research is to develop an ANN model for shear stress prediction. The chassis frame is made of two side members joined with a series of cross members. The number of cross members, their locations, cross-section and the sizes of the side and the cross members becomes the design variables. The chassis frame model is to be developed in Solid works and analyzed using Ansys. Since the no. of parameters and levels are more, the probable models are too many. The weight reduction of the sidebar is achieved by changing the Parameters using the orthogonal array. Then FEA is performed on those models. ANN model is prepared using the results of FEA. For the ANN modeling, the standard back-propagation algorithm is found to be the best choice for training the model. A multi- layer perception network is used for non-linear mapping between the input and output parameters. This model can save material used, production cost and time.