Rotary kiln is a large key rotating machine forproduction in cement, chemical and metallurgical industries.It is important to measure the straightness deviation andsurface deformation of kiln cylinder as they are importantphysical parameters for evaluating the operation state of thekiln. A new method for extracting the profile features fromthe collected data based on EMD algorithm was proposed. Afiltering criterion using energy percentage of IMF componentsand correlation coefficient method was proposed in order toeliminate the pseudo IMF components. And the physicalmeaning of the effective IMF components was explained. Wecompared the proposed method with the traditional geometrymethod. The experimental results indicated that the proposedmethod is effective in extracting the features of thestraightness deviation and the surface deformation of rotarykiln cylinder. Furthermore, it can not only be used to processthe non-stationary and non-linear signals collected but also beused to eliminate the noise signal. Therefore, it can be taken asa new method for evaluating the operation state of rotary kiln.
展开▼