Detection Of Cabin Particles Based On Particle Impact Noise Detection Method

With the development of weapon equipment system and aerospace industry,higher requirements have been put forward for the reliability of cabin equipment such as missiles and aerospace engines.The presence of loose particles in the cabin equipment causes frequent system failures,causing serious aerospace accidents and economic losses.Therefore,this paper learns from the PIND-based detection method for remainders of sealed electronic equipment,and research on the detection of loose particles in cabin equipment,going to determine the test conditions and the influence rules of various factors,and to achieve different materials and sizes identification of loose particles.Aiming at the factors affecting the detection effect of cabin device particles,a detection method based on sinusoidal vibration excitation is proposed.Using the method of combining ADAMS simulation analysis and experimental research,and the influence of vibration acceleration and frequency,particle material and size parameters are investigated in the particle impact noise detection(PIND).Through the single factor simulation test,the influence law of each factor on the detection effect of loose particles and the best vibration test conditions are studied.The simulation test and vibration test are designed by the orthogonal experiment method,and the results show that the material parameters of the particles is the key factor affecting the detection effect,and the way of particle material and size identification is confirmed.The results provide a theoretical reference for the detection and further identification of loose particles.The material feature extraction of redundant particles and the low recognition efficiency of classification decision-making algorithms,etc.The detection method of remainders in cabin devices is constituted based on Ensemble Empirical Mode Decomposition(EEMD)and Multi-classification Relevance Vector Machine(M-RVM).Referenced the detection and identification method of particles in sealed electronic equipment.A decomposition method based on EEMD is proposed,which effectively avoids the problem of modal aliasing after the particle signal is decomposed by EMD.Combining the Hilbert-Huang transform(HHT)method and Wavelet packet transform method,IMF correlation coefficient sequence,Hilbert spectrum centroid vectors and energy distribution vector are extracted.As the characteristic quantities that describes the feature information of the loose particles.Input the trained M-RVM recognition and classification model to recognize and classify the five types of materials including steel,aluminum,Sn,rubber and plastic.The results show that the cabin remnant material recognition method based on EEMD and M-RVM has a better classification effect,and the overall recognition accuracy rate is 89.2%.In view of the insufficient research on the size identification method of loose particle,and the traditional identification method cannot meet the needs of the particle material detection.Therefore,the method for identifying the size of the cabin loose particles based on chaos theory and cluster analysis is proposed.Analyze the nonlinear and chaotic characteristics of the particle signal,chaotic characteristics such as correlation dimension,Lyapunov exponent and Kolmogrov entropy are calculated as the characteristic quantities to describe the size information of the particles.The clustering algorithm is used to realize the identification and classification of different particle sizes.As a result,the method for identifying based on chaotic characteristics and clustering algorithm can effectively recognition 5 kinds of loose particles with the radius of 1.0mm,1.5mm,2.0mm,2.5mm,and 3.0mm,and the overall recognition accuracy rate is 86%.