【作者】 陈敏；

【导师】 徐志胜； 赵望达；

【作者基本信息】 中南大学 ， 防灾减灾工程及防护工程， 2008， 博士

【摘要】 我国人口众多建筑密集,是建筑火灾多发国家之一。火灾除了造成人类生命和财产的巨大损失之外,对建筑物本身的影响也是非常严重的。因此了解火灾高温对混凝土结构的影响,检测火灾后混凝土损伤是提高火灾后建筑物的再利用率,减少火灾带来财产损失的一个关键问题。所以,对于占建筑物材料绝大多数的混凝土结构在火灾后的缺陷检测就显得尤为重要。超声结构损伤诊断技术是当前结构工程学科十分活跃的研究领域并有广阔的工程应用前景。小波分析是一种新的时频两维分析方法,在信号的消噪和奇异性检测中有着广泛的应用,相关的理论和技术正在不断完善。本文着重对小波分析方法在超声检测信号处理中的应用进行了研究。主要研究内容有:(1)根据损伤力学及弹性波的传播理论,结合火灾后混凝土结构损伤的特点,建立了火灾后损伤混凝土介质模型,推导出了火灾后混凝土构件中的波动方程;获得了超声波在火灾后损伤混凝土中传播的规律。(2)采用小波分解去噪后再重构的方法,可在基本不损害原信号的前提下很好地消除信号中的噪声,易于火灾后混凝土损伤的识别。(3)采用奇异性理论处理接收的超声信号,通过小波变换找出信号的奇异点,奇异点的时间参数对应结构中的位置参数和损伤位置,计算出奇异点的Lipschitz指数,判断火灾后混凝土内部缺陷的位置及参数。(4)开发了便携式火灾后混凝土损伤超声检测仪,采用干耦合换能器可提高检测速度,提高了检测效率和检测精度。(5)对超声回波信号进行小波包变换处理,运用BP神经网络对火灾后混凝土结构损伤进行快速诊断,得到结构的损伤程度。该方法简单易行、精度高,可以在实际检测中推广应用。本文研究成果,对超声检测技术在火灾后混凝土结构损伤检测领域中的应用具有一定的指导意义。经过研究不断完善的混凝土超声检测技术,可以更综合直观地获得火灾后混凝土结构内部质量的准确评估。对普通的混凝土缺陷检测也可推广应用,在有效检测和提高混凝土结构工程质量方面有着良好的推广应用前景。更多还原

【Abstract】 Every year there is a tremendous number of fire hazards in our country for the huge population and crowed buildings, which cause significant loss to our country. Besides the extensive damage of life and property brought by the fire, the effect on the buildings is also extraordinary severe. Therefore, knowing the variation of the build bearing capacity, detecting the flaw in concrete structure after fire is a key to renew the using ability of the buildings after fire and reduce the loss caused by fire. Detecting flaw of the concrete structure after fire are especially important because majority of the structures are made in concrete. Damage diagnosis technique using ultrasonic is an active research domain in current structure engineering and has strong engineering background, the correlative theory and technology are developing continuously. Wavelet analysis is a new time-frequency analysis method which has raised a research upsurge internationally in recent years. It is regarded as a breakthrough after Fourier analysis. It is applied widely to eliminate noise interference efficiently and detection singularity on signal. It is proved that the wavelet analysis detects all kinds of singularity discontinuous point effectively. The method of elimination noise interference efficiently and localize flaws accurately by wavelet transform was introduced in the article. The most research work deal with:(1) According to damage mechanics and the theory of elastic waves propagation; combining the characteristic of damaged concrete, the damaged media model of concrete is presented. The perfect concrete is regard as isotropic medium. The wave equation of ultrasonic propagation in concrete is established. The rule of ultrasonic transmits in concrete after fire is obtained. (2) Using the method of reconstructing signal after wavelet transform and elimination noise, the noise in ultrasonic signal is reduced perfectly. It is more easily to distinguish the damage of concrete.(3) Analyzing the received ultrasonic signal using the theory of singularity, the singular point is found by wavelet transform. The time parameter of singular point is equivalent with the position parameter of damage in concrete. Calculating the Lipschitz exponent is ascertained.(4) The ultrasonic diagnose instrument is developed. The technology of coupling without using liquid can detect concrete rapidly. The efficiency and accuracy is improved.(5) Disassembling the received ultrasonic signal using wavelet packet, BP artificial neural network is used in the rapid diagnosis the damage of structural after fire. The results show that this method can reduce the complexity of diagnosis and improve efficiency and precision.The achievement in this disquisition will make the ultrasonic diagnosis techniques be applied in concrete structure after fire. Author deeply believes that the technique detecting concrete after fire using ultrasonic will have vast prospect in application. It will be more accurately and quickly that assess the quality of concrete after fire.更多还原