Research Of Roughness Of Surface Polished By Abrasive Waterjet

In this dissertation, single factor experiment and Taguchi experiment were carried out. The dimensional analysis (DA) and quadratic polynomial regression were applied to construct the molds. And the relationships among the roughness of polished surface (ROPS) and the process parameters of abrasive waterjet (AWJ) polishing were explored qualitatively and quantitatively. The influential trend of each process parameter upon the ROPS was analyzed and reported. And the models to describe these trends were created. At the end of this dissertation, a model to depict the mathematic relationship among the ROPS, the initial surface roughness of workpiece, and all process parameters was also founded. And it can serve as a theory to plan the polishing process and predict the ROPS qualitatively and quantitatively. The main researches are the following:1. In the Taguchi experiment to polish the40CrMnMo7model steel and BK7glass, the significance order of process parameters upon ROPS and optimized combination of process parameters were reported. In the variation ranges of process parameters in this experiment, the optimized combination of process parameter to polish these two materials is:green SiC, mass ratio of abrasive to water of0.08, nozzle diameter of0.4064mm, pump pressure of4.5MPa. standoff distance of10mm, jet angle of75°, trasverse speed of16mm/s. And the DA was used to create a Ra model with the mean of percent error of4.27%, the variance of percent error of7.18%, and the coefficient of determination of0.950, and a R2model with the mean of percent error of9.82%, the variance of the percent error of13.98%, and the coefficient of determination of0.882. This fact demonstrates that the DA can be used to create a useful model to predict the ROPS.2. A series of single factor experiments were designed to polish the40CrMnMo7model steel. The influential trend of each process parameter upon the ROPS of40CrMnMo7model steel was investigated. And the corresponding models to depict these trends were created. The process parameter ranges in which to obtain the optimized ROPS were also recognized. By observing the topography of polished surface, the plastic shear and the plastic extrusion are discovered as two main effects which affect the roughness of the surface polished by AWJ. The severer the plastic shear or plastic extrusion is, the greater the ROPS changes. What’s more, the more continuous and more even the veins of polished surface are, the lower the ROPS is.3. The Taguchi experiment to polish the40CrMnMo7model steel was designed. The significance order of the polishing process parameters upon the ROPS and the optimized combination of process parameter were reported. In the variation ranges of the polishing parameters of this experiment, the optimized combination of process parameters to polish the40CrMnMo7model steel was:nozzle diameter of0.4064mm, jet angle of90°, standoff distance of25mm, pump pressure of3.33MPa, polishing pitch of0.4mm, polishing times of2, and traverse speed of12mm/s. By adding some results of above single factor experiments, DA. non-elimination quadratic polynomial regression (NEQPR), and back-elimination quadratic polynomial regression (BEQPR) were applied to constitute3ROPS models. After comparing these3models, it was found that:the predictive ability of the DA model is better than BEQPR model, but weaker than NEQPR one. and its creation needs the least data, therefore, DA is a very useful modeling method; for NEQPR model, its structure is the most complex, and its constitution needs the most data, but its predictive ability is the best; like the NEQPR model, the constitution of BEQPR needs the most data, although its predictive ability is the weakest, it demonstrates the interactions of machining parameters.4. It was discovered that the cross section shape of the jet-machined zone is subtriangle. On the basis of this discovery, a few assumptions, and experimental data of Chapter3and Chapter5, a theoretic model was constructed to describe the mathematic relationship of ROPS, initial surface roughness, and process parameters after one time polishing. The test results show that the mean and variance of percent error between predictive values and the actual ones are2.12%and2.09%, respectively. Thereby, the accuracy of this model is pretty well. And in the creating process of ROPS model, the mathematic relationships of the process parameters with the jet velocity, machining depth, and material erosion rate were also constructed respectively.