Font Size: a A A

Research On Impact Load Monitoring System Of Aviation Structure Based On ARM

Posted on:2018-06-14Degree:MasterType:Thesis
Country:ChinaCandidate:Y GaoFull Text:PDF
GTID:2322330536487355Subject:Measuring and Testing Technology and Instruments
Abstract/Summary:PDF Full Text Request
With the improvement of aircraft performance,the structure of aircrafts is becoming more and more complicated.These in-service aircraft structures are inevitably subject to various forms of impact load which can cause varying degrees of damage easily and can be a potential threat to the safety of structures.Therefore,the research on the real-time monitoring of impact loads is of great practical significance.With the urgent engineering need,a set of online real-time monitoring system based on ARM has been developed and the main contents are as follows:Firstly,the theories of acoustic emission technology and the principle of the use of piezoelectric sensors were summarized.With the introduction of the propagation characteristics of acoustic emission signals,the wavelet transform was chosen to be the processing technique of signals.Then the procedure to obtain the time of arrival based on wavelet transform was emphatically analyzed and a fast algorithm for non-orthogonal wavelet transform was proposed.Secondly,the localization algorithms of cross correlation function and back-ground neural network were studied.According to the needs of the actual monitoring system,two algorithms were improved.Thirdly,the hardware and the software designs of the on-line impact monitoring system were described in detail.The system was mainly controlled by STM32F407 which has the characteristics of miniaturization(179.5mm×140mm),portable,low power consumption and low cost.Finally,impact location experiments were conducted on the model of trailing edge to validate the system and two location methods.The results were achieved with satisfactory accuracy(the positioning error of cross correlation method was less than 20 mm and the error of back-ground neural network was less than 10 mm except the boundary),requiring little computational time(nearly 5s)and displaying clear positioning images in the monitoring area of 600×600 mm.
Keywords/Search Tags:impact monitoring, ARM, cross correlation, back-ground neural network, acoustic emission technology
PDF Full Text Request
Related items