Font Size: a A A

Research On On-Line Moisture Sensing Technology Of Lubricating Oil Based On Image Distortion

Posted on:2023-02-24Degree:MasterType:Thesis
Country:ChinaCandidate:Y J GaoFull Text:PDF
GTID:2568306851486554Subject:Mechanical design and theory
Abstract/Summary:PDF Full Text Request
Lubricating oil is regarded as the"blood"of mechanical equipment,which is very important to ensure the safe and reliable operation of mechanical equipment.Water pollution is the main inducement of lubricating oil performance deterioration,which directly affects the operation health of mechanical equipment.In order to obtain the moisture content of lubricating in real time,predict the deterioration degree of lubricating oil performance and ensure the safe,efficient and long-life operation state of the equipment,this thesis carries out the research on the on-line moisture sensing technology of lubricating oil based on image distortion.The main work is as follows:(1)Based on the distortion effect of optical imaging caused by random medium,an on-line moisture detection method of lubricating oil based on the characteristic distortion of object image is proposed,and an optical sensing device for on-line moisture detection of lubricating oil based on image distortion is designed and developed,which realizes the effective acquisition of distorted images of target objects caused by lubricating oil with different moisture content.(2)In order to realize the effective quantification and characterization of moisture content of lubricating oil,the construction and extraction method of moisture of lubricating oil quantitative characterization index based on image distortion was studied.The gradient edge detection and marked watershed transform are combined to realize the rapid extraction of the boundary features of the distorted image of the object.Based on this,five quantitative representation indexes of the distortion degree of the target object image,namely radial deviation RD,radial internal deviation RID,area,perimeter and eccentricity,are constructed,and the effective quantification and characterization of different moisture content levels of lubricating oil are realized.(3)Carry out the simulation experiment of on-line moisture detection of lubricating oil.The experimental results show that RD180 and RID0are the most sensitive to the change of moisture content of lubricating oil;The linear relationship between RD180,RID0 and the moisture content of lubricating oil is established through mathematical assignment operation.The linear fitting R2 is 0.98050 and 0.99725 respectively.The fitting formula realizes the accurate quantification of the moisture content of lubricating oil.In addition,compared with WD-1 moisture sensor based on chemical reaction principle,the method proposed in this thesis is very sensitive to moisture change.It is suitable for on-line moisture detection of lubricating oil in the range of 0wt%~1wt%.The measurement accuracy is±0.1%,and the operation is simple,the measurement speed is fast,and there is no need to add any reagent.
Keywords/Search Tags:Lubricating oil analysis, Moisture monitoring, Online sensing, Image distortion, Feature extraction
PDF Full Text Request
Related items