Research On Space Debris Damage Pattern Recognition Bassed On Neural Network Technology

With the gradual develop of our long-term, manned space station program, thehypervelocity impact of space debris is becoming increasingly prominent. The in-suitdetect system of space debris is more urgent than ever. The main function of the detectsystem is to locate the impact position and assess the impact damage, while the damagepattern recognition is the main difficult point. At present, most investigators recognizepattern by acoustic emission signal frequency transform, this method can only damagemode for fixed line discrimination and quantitative analysis. The introduction of theconcept of neural networks in the damage pattern recognition using LVQ neuralnetwork method of injury qualitative discrimination, using the BP neural network forquantitative analysis method using neural network impact injury model.This paper studies the injury patterns corresponding to the characteristicparameters, a more accurate characterization of injury patterns Targets damageparameters and acoustic emission waveform characteristic parameters, and separatelyfor each of the characteristic parameters of the parameter extraction methods, includingreal-time Fourier transform, wavelet transform and software filtering. Actually showthat the characteristic parameters obtained by this method to characterize the injurypattern characteristics.The characteristic parameters of the damage mode, the application of numericalsimulation tools provide the neural network input samples, until a series of simulationdata for variable speed and variable distance from the sound source. This data on thenetwork as a sample of the neural network training and validation. The results show thatthe LVQ neural network for the damage model is qualitative discriminant accuracy rateis accurate enough.Research results on space debris in orbit perceptual system developed with thereference value. For the realization of the space broken skin, the spacecraft sufferedimpact damage pattern recognition laid the foundation.