一种改进的小波分析消噪方法及其在健康监测中的应用

结构物的变形可以根据压电智能材料的正压电效应来监测,但是由于压电材料是电学材料,易受外界环境的干扰,预将压电材料用于实际工程的健康监测,好的滤波方法是必备的条件之一。采用小波分析对信号进行降噪和故障诊断是其有效的应用之一,本文在小波多分辨分析阈值法的基础上,对阈值进行了修正,仿真结果表明其效果优于传统阈值法;将该方法应用于利用压电陶瓷传感器所监测到的一框架剪力墙模型的振动信号降噪中,其结果表明该法可以较有效去除噪声。

NOISE CONTROL FOR HEALTH MONITORING SIGNAL OF A CIVIL STRUCTURE BASED ON PZT SENSORS

Two piezoelectric ceramic patches are stuck on the root of a reinforced concrete frame-shear wall which undergoes simulated earthquake in order to monitor its response. Signal denoising and damage detection are the most useful applications of wavelet which can be used to analyse signal in time and frequency domains in the mean time. An improved threshold method based on wavelet multi-resolution analysis is adopted to extract the true in-situ monitoring signal. Simulation result indicates that this method is better than the traditional threshold. A good result is obtained from the monitoring signal acquired by using PZT sensors.