The Research Of Oil Particle Detection Based On Compound EST Sensor

Aircraft engine as a major component of the aircraft,its safety and stability have a high demand.In the event of a flight failure,it will cause a large loss of personal property,so it need to improve the accuracy of aircraft engine fault detection.In the process of aircraft engine work generated a large number of iron and oil composed of solid-liquid two-phase flow system.If we can detect the distribution of two-phase flow,then we can indirect to determine the health of the engine.In this paper,the electrostatic precipitator based on electrostatic induction is used to study the distribution of abrasive grains in oil.The problem of electrostatic tomography is that the amount of information obtained is small and the judgment of the abrasive grains is not accurate.For this reason,this paper is based on the already designed 16-electrode array type electrostatic sensor design a combination of array and needle-type 32-electrode electrostatic sensor to detect the aircraft engine oil.The main work is as follows:1.It uses finite element simulation software COMSOL Multiphysics to design a16-electrode array electrostatic sensor and then simulation.The distribution of sensitivity will be studied.The electrode size of the sensor,the length of the shield and other parameters will be optimize.2.In the 16-electrode array of electrostatic sensors installed in the probe,the array and needle-type combination of static sensors can be designed.The effects of electrode size,thickness of insulation pipe,radius of shield and other parameters on the sensitivity distribution are studied.Its structural parameters will be optimized.3.In the image reconstruction,the image reconstruction algorithm is studied and analyzed.Due to the ill-posedness and less information of the image reconstruction problem in electrostatic tomography,the space resolution of the reconstructed image is extremely low.In order to improve the imaging quality,the study of the electrostatic sensor based on the modified Landweber algorithm was proposed.It is on the basis of linear inversion algorithms and Landweber iterative methods.By observing the resolution of the reconstructed image,the image reconstruction ability of the algorithm is evaluated qualitatively.4.The performance of the electrostatic sensor is evaluated qualitatively.By calculating theregularized mean variance,regularized mean absolute error,image correlation coefficient and algorithm running time and other parameters to determine the advantages and disadvantages of sensor imaging.