Creep feed grinding is widely used in manufacturing supperalloy materials. These materials are usually used in aircrafts, gas turbines, rocket engines, petrochemical equipments and other high temperature applications. The objective of this paper is to model and predict the grinding forces of the creep feed grinding of these materials using the neural network. This model is then used to select the working conditions (such as depth of cut, the wheel speed and workpiece speeds) to prevent the surface burning and to maximize the material removal rate. The results show that the combined neural network and an optimization system are capable of generating optimal process parameters. The outcomes of the paper are now used to apply the optimal working conditions for grinding the turbine blades.

No abstract

No abstract

No abstract

Flanging is a common sheet metal forming operation to produce structural sheet metal components. Flanging is used to give a component smooth rounded edge, and to provide jointing and assembling of components. In this paper, the simulation and experimental results of two types of z-flanging are investigated. First, a z-flange forming which incorporates mating drawbeads on the main and backup punches is studied. Drawbeads are used in commercial stretch flange operations to control or limit the rate of cut-out expansion. A z-flange forming is then performed on a blank with pre-punch created holes. The non-uniform deformation of the hole and undulating are two main important defects. Stretch z-flange forming is simulated using the explicit finite element (FE) LS-DYNA software. In three dimensions FE simulation, four different types of shell elements have been used to determine the one showing the best agreement with the experimental results.