基于复杂成分小波分析的混凝土锯切实验及加工状态研究

发布时间：2018-01-19 02:36

本文关键词： 硬软骨料金刚石锯片加工信号监测　出处：《广西大学》2017年硕士论文　论文类型：学位论文

【摘要】：混凝土是一种复合材料,不同骨料力学性质差异较大,使得它的锯切状态相对随机。多数情况下,混凝土的锯切过程难以进行有效监测,锯片在使用途中极易出现偏摆,这样严重影响了加工效率。因此本文采用了金刚石锯片对混凝土内部各组分(硬骨料、软骨料)和混凝土进行了锯切实验研究,为提高混凝土的锯切效率和锯片的利用率提供理论依据和应用指导。主要内容及结论如下:(1)对硬骨料和软骨料的锯切机理进行了研究,分析了硬软骨料在不同锯切用量下锯切力和振动幅值的变化规律以及材料去除机制。实验结果表明软骨料的去除机制存在着轻微的塑性变形和脆性崩碎两种情况,硬骨料则完全是脆性崩碎;x和y方向的锯切力和振动幅值变化趋势较为一致,但锯切用量对y方向的锯切力影响不大;在相同的锯切用量下软骨料的锯切力和振动幅值一般要小于硬骨料;切深对硬骨料的锯切力和振动幅值的影响程度最大,进给速度对软骨料的锯切力和振动幅值的影响程度最大。(2)利用多尺度小波分析方法研究了硬软骨料的加工信号各频率段的特征变化情况。发现硬软骨料的加工信号较为稳定且随机性小,其特定频率段与锯切过程存在着良好的对应关系,合理地利用加工信号可实现对硬软骨料锯切过程的稳定性以及锯片状态的监测。(3)对硬软骨料和混凝土加工信号的各频率段进行了比较。实验结果表明混凝土加工信号中156.25-312.5Hz频率段主要源于锯片与混凝土中硬骨料之间的相互作用,1250-2500Hz频率段主要源于锯片与混凝土中软骨料之间的相互作用。(4)本文的最后以混凝土加工信号中的156.25-312.5Hz和1250-2500 Hz频率段作为锯片状态特征的主要来源,使用蝙蝠算法改进的bp网络预测锯片状态。同时考虑到力和振动传感器采集数据方面的差异,对两种不同信号样本的预测结果进行了融合,并与实际锯片状态值进行了比较,发现可实现对锯片状态的准确判断。
[Abstract]:Concrete is a kind of composite material, and the mechanical properties of different aggregates vary greatly, which makes the sawing state of concrete relatively random. In most cases, the sawing process of concrete is difficult to monitor effectively. It is easy to appear deflection in the use of saw blade, which seriously affects the processing efficiency. Therefore, diamond saw blade is used to study the internal components of concrete (hard aggregate, cartilage) and concrete sawing experiments. In order to improve the sawing efficiency of concrete and the utilization rate of saw blade, the theoretical basis and application guidance are provided. The main contents and conclusions are as follows: 1) the sawing mechanism of hard aggregate and cartilage is studied. The variation of sawing force and vibration amplitude of hard soft aggregate under different sawing contents and the material removal mechanism were analyzed. The experimental results show that the removal mechanism of cartilage has two kinds of cases: slight plastic deformation and brittle breakage. . Hard aggregate is completely brittle crumbling; The variation trend of sawing force and vibration amplitude in the direction of x and y is consistent, but the sawing amount has little effect on the sawing force in the direction of y. The sawing force and vibration amplitude of cartilage under the same sawing dosage are generally smaller than that of hard aggregate. Cutting depth has the greatest influence on sawing force and vibration amplitude of hard aggregate. The influence of feed speed on the sawing force and vibration amplitude of cartilage is the biggest. Multi-scale wavelet analysis was used to study the characteristics of processing signals of hard soft aggregates at different frequencies. It was found that the processing signals of hard soft aggregates were relatively stable and less random. There is a good correspondence between the specific frequency range and the sawing process. The stability of sawing process of hard soft aggregate and the monitoring of saw blade state can be realized by reasonable use of processing signal. The frequency bands of hard soft aggregate and concrete processing signal are compared. The experimental results show that the 156.25-312.5 Hz frequency range of concrete processing signal mainly comes from sawing blade and concrete hard aggregate. The interaction. The frequency range of 1250-2500 Hz is mainly derived from the interaction between saw blade and cartilage in concrete. At the end of this paper, the frequency bands of 156.25-312.5Hz and 1250-2500 Hz in the concrete processing signal are taken as the main source of the state characteristics of the saw blade. The improved BP neural network based on bat algorithm is used to predict the state of saw blade. Considering the difference of data collected by force and vibration sensors, the prediction results of two different signal samples are fused. Compared with the actual saw blade state value, it is found that the accurate judgment of saw blade state can be realized.
【学位授予单位】：广西大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TU528

【参考文献】