Font Size: a A A

The Detection System For Wear Particles In Oil Based On Digital Image Processing

Posted on:2015-01-12Degree:MasterType:Thesis
Country:ChinaCandidate:Y T DuFull Text:PDF
GTID:2268330425988807Subject:Detection Technology and Automation
Abstract/Summary:PDF Full Text Request
The detection for wear particles in oil is a valid method for wear and tear. According to the analysis of the lubricating oil in mechanical system, obtain the equip-ment lubrication condition and abrasive status information, describe the wear state of the device qualitatively and quantitatively, find out the cause of formation, evaluate the equipment environment condition and predict its failure, determine the fault’s position, reason and type, ensure the safety and efficiency of the equipment. This paper analyzes the characteristics of the detection for wear particles in oil, aiming at the shortage of traditional methods which is based on experience, put forward a set of new detection system for wear particles in oil based on digital image processing.Summarize the development status about the detection for wear particles in oil at home and abroad, learn the knowledge of tribology foundation, digital image processing technology, pattern recognition algorithms and other related information. Describe the work principle, design scheme and implementation methods of this detection system in detail.In this paper, the hardware part of this system is composed of YTF6analysis ferro-graph, BX60two-color microscope and CCD (Charge-coupled Device), which is used to collect the oil wear particles image. While Visual Studio2010is used as the devel-opment platform to build the software framework through using the MFC (Microsoft Foundation Class Library) as the basis for the application interface design, realize the function of wear particles image preprocessing, complex background segmentation, contour extraction, wear particles characteristic parameters and pattern recognition.etc. Support vector machine SVM is applied to identify the wear particles images, especially the case of limited wear particles sample images.The detection system for wear particles in oil based on digital image processing is better than the conventional method relying on expert experience. Integrated the func-tion of image acquisition, processing, calculation and recognition, this system has high speed, accuracy and security.
Keywords/Search Tags:Wear and tear, Wear particle detection, Digital image processing, Pat-tern recognition, Support vector machine
PDF Full Text Request
Related items