基于粒子滤波的耐火材料声发射源信号恢复方法研究

发布时间：2018-04-23 13:57

本文选题：耐火材料 + 声发射源信号　；参考：《武汉科技大学》2016年硕士论文

【摘要】：耐火材料在各种热工设备和高温容器中作为抵抗高温作用的结构材料和内衬,对保障高温设备安全运行具有重要意义。受实际工况和环境的影响,耐火材料损伤严重危害设备正常运行,甚至炉内产品质量。因此对耐火材料损伤识别具有重要意义。基于声发射信号分析的耐火材料损伤识别已成当前研究热点,但受传递通道影响,失真声发射源信息直接影响损伤识别的准确性。为此本文对声发射源信号的恢复展开了研究,主要研究内容如下:(1)针对声发射系统同时包含非线性和线性情况的特点,本文提出基于改进Rao-Blackwellized粒子滤波(RBPF)的源信号恢复方法,通过结合状态空间模型来对系统状态即源信号进行估计,并从理论和实验两个角度验证了改进RBPF在耐火材料声发射信号上的恢复可行性及优势。(2)为了建立合适的声发射系统模型,本文针对声发射信号和地震波信号具有相似性的特点,通过引入地震波模型并修改参数建立了声发射系统状态空间模型,然后结合标准PF、RBPF和改进RBPF对实验采集的耐火材料损伤声发射信号进行源信号恢复,根据仿真验证对三种恢复信号分配了权值,求得了最接近真实值的信号,将求得的信号与三种算法的恢复结果进行对比之后,得出改进RBPF算法最优的结论。(3)为了进一步说明源信号恢复效果,对采集信号和源信号进行了功率谱分析,两种情况下基体开裂的主频成分分别为6~8kHz和5~9kHz,基质裂纹扩展的主频成分分别为40~60kHz和10~20kHz,通过之前分析发现改进RBPF的恢复效果更具有代表性,使用该算法后基体开裂和基质裂纹扩展所对应的主频成分应分别为5~9kHz和10~20kHz。本文提出的基于改进RBPF算法的信号恢复方法不仅对耐火材料声发射源信号进行了有效的恢复,而且对其损伤模式的识别提供了更为准确的依据,为材料损伤研究提供了新的思路。
[Abstract]:Refractories are used as structural materials and liners to resist high temperature in various thermal equipment and high temperature vessels, which is of great significance to ensure the safe operation of high temperature equipment. Under the influence of actual working conditions and environment, the damage of refractories seriously endangers the normal operation of equipment and even the quality of products in the furnace. Therefore, it is of great significance to identify the damage of refractories. The damage identification of refractories based on acoustic emission signal analysis has become a hot topic at present, but the distortion of acoustic emission source information directly affects the accuracy of damage identification due to the influence of transmission channels. In this paper, the recovery of acoustic emission source signal is studied. The main research contents are as follows: (1) aiming at the nonlinear and linear characteristics of acoustic emission system, a method of source signal recovery based on improved Rao-Blackwellized particle filter is proposed in this paper. By combining the state space model to estimate the state of the system, that is, the source signal, In order to establish a suitable acoustic emission system model, this paper aims at the similarity between acoustic emission signal and seismic wave signal, and verifies the feasibility and advantage of improving RBPF recovery on refractory acoustic emission signal from both theoretical and experimental points of view in order to establish a suitable acoustic emission system model, this paper aims at the characteristics of similarity between acoustic emission signal and seismic wave signal. The state space model of acoustic emission system is established by introducing seismic wave model and modifying parameters, and then the source signal of experimental refractory damage acoustic emission signal is recovered by combining standard PFN RBPF and improved RBPF. According to the simulation results, the weights are assigned to the three kinds of recovery signals, and the signals closest to the true values are obtained. After comparing the obtained signals with the recovery results of the three algorithms, In order to further explain the recovery effect of the source signal, the power spectrum analysis of the collected signal and the source signal is carried out. In both cases, the main frequency components of matrix cracking are 6~8kHz and 5kHz, respectively, and the main frequency components of matrix crack propagation are 40~60kHz and 100.20kHz, respectively. It is found that the recovery effect of improved RBPF is more representative by previous analysis. The main frequency components of matrix cracking and matrix crack propagation should be 5~9kHz and 1020kHz, respectively. The proposed signal recovery method based on improved RBPF algorithm not only effectively recovers the acoustic emission signals of refractories, but also provides a more accurate basis for the identification of damage patterns. It provides a new idea for the study of material damage.
【学位授予单位】：武汉科技大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TQ175.1;TN713

【参考文献】