Natural Excitation Signal Processing Method Of Structural Health Monitoring Studies

Structural health monitoring is an advanced interdisciplinary technology, while signal processing and feature extraction based on the vibration of the structure is one of the key technologies for health monitoring. In this thesis, space launch tower and other large steel structures based on natural environment incentive problems has been built, started from the random decrement technique (RDT), used Hilbert-Huang Transform (HHT) to extract the damage features from signals, calculated the first-order model of the instantaneous frequency of natural variation (Fimf) through intrinsic mode function (IMF), then identify the normal and damage characteristics.The main contents and conclusions are as follows:(1) According to the incentive of the natural environment unpredictable, studied random decrement technique to remove the random component, and to get the freedom response inspired by the initial incentive as a signal pre-processing methods input by signal analysis;through the introduction of the Natural Excitation Technique (NExT) to complete the signal pre-processing.(2) Introduced the HHT method and verified its effectiveness on dealing with stationary signal, this paper analyzed the principles and steps of the damage features exacted by this method, and built rocket launch tower experimental model.(3) Used Hilbert spectrum and marginal spectrum to extract features of damage,. Compared the signal frequency of the intrinsic mode function Fimf and each node in the injury of variation to identify the presence and the location of damage.(4) Compared the method of RDT-HHT, NExT-HHT and TMF-HHT, which the RDT-HHT was the best in the thesis.Research results show that the method of Structural Health Monitoring system was proposed and verified in this thesis, which has great significances for the construction and practical application of the health monitoring system.