Roundness Error Of The Rotational Parts Based On Wavelet Transform Isolation And Assessment Methods

Roundness error is an important indicator for high-precision rotary parts,in the industrial scene, the roundness error is detected similarly, for example using dial- gauge, vernier caliper or micrometer as tool for two-point or three-point measurement, the reliability is poor, efficiency is low. However, effective instruments such as roundness instrument and three dimensional measuring machine is difficultly used for production widely,because it is expensive and high demand for environment.With the rapid development of science and technology, high-precision product is needed. Accuracy increased and the larger size part present a new study for roundness error detection.Combining with theoretical analysis, computer simulation and experimental verification, in application of laser non-contact detection for rotational parts's roundness error ,in order to improve accuracy, roundness error's separation and algorithms based on wavelet theory is proposed in this study. The feasibility of using wavelet transform to fiter signal is analyzed theoretically; it is studied to separate the actual integrated error signal for workpiece's surface profile applying with wavelet multi-resolution analysis, according to the theoretical analysis, carried out the computer simulating for surface simulation signals and the experimental verification.The main research work in this study:(1) Discusses the basic theory and characteristics of wavelet transform, wavelet transform have the merits of multi-resolution analysis, it can distinguish mutations and noise in the signal effectively, it is very suitable for non-stationary signal's processing; duing to the characteristics of the roundness error's the original sampling signal that it is constituted with the multi-frequency signals , it is proposed to use wavelet theory for sampled signals filtering.(2) It is described to use wavelet multi-resolution analysis to decomposite surface geometry error signal for workpiece, which is showed that owing to application of the wavelet transform technique, on the one hand external noise's interference on the original sampled signal is excluded, and on the other hand, shape error, surface waviness and surface roughness is obtained at the same time from the workpiece surface integrated topography error signal.(3) The center and radius's equation of the least-squares circle is established which is used to evaluated roundness error based on wavelet transform. In accordance with the characteristics of the surface of circular contour, the error signal is separated under computer simulation, roundness error curve is obtained, which is substituted into the equation ,and the location and size of the least square circle is knowed, then the value of roundness error is get. (4) By experiment, the detect accuracy of roundness error is compared and verified which is based on wavelet algorithm, it is proved furtherly that using wavelet algorithm to separate and roundness error can improve the detect accuracy, and it is reliable.In conclusion, this study will provide an available approaches for the research on roundness error at theory and the detection technology , and there will have certain value of theoretical study and practical application.