Research On Image Detection Of Wear Debris By Capacitance Array In Lubricating Oil Of Aviation Engine

Fatigue wear,which occur on bearings and other mechanical parts of aircraft engine,is a major factor of aviation accidents.It is related to material,design,working environment and load,but results are debris that generated by the abrasive surface and fall into the lubrication system.Aviation engine wear debris monitoring mainly use the oil sample analysis methods,which is can't meet the requirements of real-time monitoring.On-line detection of debris in lubrication system of aviation engine,can give effective early warning of emergency engine failure to avoid accidents,which has theoretical significance and engineering application value.In view of the requirements,techniques of on-line capacitance array imaging and wear feature extraction is studied.A new method of measurement with curved plates array in ring channel sensor is put forward.This has been verified on finite element model and designed demonstration platform.Research-work is mainly done in the following aspects:Put forward a new method of capacitance array imaging measure.For aviation engine debris on-line detection,analysis the piecewise constant sparse mixture features of debris.Present math model of capacitor array debris detection,with multi-objects micro-scale,piecewise constant sparse mixed medium conditions,parameters of capacitor array sensor and excitation strategy were researched also.A 12-plates array sensor is designed;Detecting system and metal granular measure test system are built.It provides a platform for detecting and verifying the wear debris detection.Key technical characteristics are verified use this system.Self-adaptive cellular genetic ungradient algorithm(SA-CGA)aiming at solve non-convex nonlinear PDE inversion problem is proposed.Inversion imaging effect of multiple metal-debris is verified by capacitance array data.LFLCS algorithm,based on math morphology image processing method,is put forward for feature recognition and measurement the size and number of multi-particles.Feasibility of multi-particles detection and wear feature recognition measurement are verified.