In cylindrical plunge grinding, a series of workpiece are ground successively without intermediate dressing of the wheel. Through the grinding, the wheel characteristic is continuously changed by the wear, fracture and new exposure of abrasive grains, having a profound effect on the grinding performance. Simulation models based on the behavior of active grains on wheel surfaces in the cylindrical grinding process are developed to make it possible to describe the time dependent result of grinding and to establish the grinding operation standards capable of estimating grinding wheel performance and selecting grinding conditions, in which many parameters in actual grinding operations are considered. This simulation system can display its results in graphical form including the time dependent results and the effects of various parameters as well as optimization capabilities. This simulation has the same effects as many grinding experiments, and capable of selecting the optimum grinding wheels and conditions.
In this thesis, a new method is proposed to simulate the grinding cycle behaviors of cylindrical plunge grinding operations. Generally the plunge grinding cycle consists of three stages: that is, (1) the rough grinding with high velocity ratio, (2) the fine grinding with low velocity ration and (3) the sparkout grinding with low velocity ratio.
At the first stage (rough grinding), the reduction in grinding time is planned by high stock removal rate based on high velocity ratio. On the other hand, the main role at the second stage (fine grinding) and the third stage (the sparkout grinding) is improving the surface quality of the workpiece, especially the decrease of the surface roughness and the damaged surface layer by relatively low velocity ratio, Force generated during grinding cause elastic deformation and deflection of the machine, the grinding wheel and the workpiece. Elastic defection of the grinding system and grinding wheel wear cause the actual stock removal to be less than the controlled infeed input to the machine. Considering wheel wear and deflection, for the moment, the continuity requires that the difference between the controlled and the actual infeed velocities be equal to summation of the time rate of change of radial elastic deflection and wheel wear rate. In order to calculate the actual wheel depth of cut which corresponds to the true radial infeed per workpiece revolution, this continuity condition and equilibrium between normal force and deflection is analysed simultaneously.
So, we can predict the time dependent grinding behaviors from the simulation with the actual wheel depth of cut at any time.
Optimization of machining processes generally requires identification of those operating parameters which will satisfy the criterion or objective. Analytical methods of machining economics are frequently used for optimizing large scale cutting processes, such as turning of milling, according to a criterion of maximum production rate of minimum cost. A simillar approach can also be applied to grinding.
However, with the possible exception of some heavy-duty grinding operations, the application of classic machining economics to grinding optimization has been rather limited. One problem is the lack of a reliable tool-life relationships which can account for the influence of the relevant parameters. But, perhaps more significantly, the indicated optimal conditions are often undesirable or even impossible to attain insofar as they violate production constraints associated with the process, the machine tool, or the part quality. The prediction of the combination of operating parameters at constraint thresholds prior to grinding is not possible, but incipient or actual violations of grinding constraints can be identified during grinding by workpiece inspection after grinding. Practical methods for optimization should include constraint identification as an integral step. Optimization strategies are developed for maximizing removal rate and for minimizing cycle times subjected to production constraints.
