轧辊故障诊断方法研究

发布时间：2018-05-09 12:34

本文选题：混沌 + 轧辊　；参考：《燕山大学》2012年硕士论文

【摘要】：随着非线性动力学研究的不断进步，混沌理论的提出，这使得对非线性学科的研究出现了重大突破。混沌是指确定的系统在一定条件下所呈现的不确定的或不可预测的貌似随机现象。裂纹轧辊在轧制时，非线性现象很强，，因此，本文将混沌理论引入轧辊裂纹故障诊断。首先介绍了混沌的定义及其特点，为有效的将有用的信号提取出来，进行了小波去噪的研究，编制了小波降噪的Matlab程序，对信号进行降噪，取得了良好的效果。然后对相空间重构理论进行了介绍，并利C-C方法求得嵌入维数m和延迟时间τ。通过分析正常与裂纹状态下轧辊轧制运动特性，求解混沌振动信号的最大Lyapunov指数、关联维数、Kolmogorov熵特征参数，并对正常与裂纹及裂纹程度不同的状态下的特征参数进行分析，从另一个角度探索了轧故障诊断的新方法，能过混沌特征量的变化对轧辊出现故障进行诊断。时间序列预测是用被预测事物的过去和现在的观测数据，构造依时间变化的序列模型，并借助一定的规则推测未来，是预测机械运动状态的前提。混沌应用于时间序列预测把许多以前被看作随机信号的现象，现在都可以用混沌理论去解释。利用相空间重构理论，基于回声状态网络进行时间序列预测，通过对时间序列预测的研究，为轧辊运行状态的预测奠定基础，使我们对轧辊预测所处的状态是否出现故障及故障的严重程度做出一个判断，这具有很强的实际应用空间，可以提前做好预防措施提高轧机设备的安全性，提高经济效益有着重要的实际应用价值。
[Abstract]:With the development of nonlinear dynamics and the development of chaos theory, there has been a great breakthrough in the study of nonlinear dynamics. Chaos is an uncertain or unpredictable seemingly random phenomenon under certain conditions. The nonlinear phenomenon of cracked roll is very strong, so the chaos theory is introduced into the fault diagnosis of roll crack in this paper. Firstly, the definition and characteristics of chaos are introduced. In order to extract useful signals effectively, wavelet denoising is studied, and the Matlab program of wavelet de-noising is compiled, and good results are obtained. Then the theory of phase space reconstruction is introduced, and the embedding dimension m and delay time 蟿 are obtained by C-C method. The maximum Lyapunov exponent of chaotic vibration signal and the correlation dimension of Kolmogorov entropy characteristic parameter are solved by analyzing the roll rolling motion characteristics under normal and crack state, and the characteristic parameters under the condition of normal crack and crack degree are analyzed. From another point of view, a new method of rolling fault diagnosis is explored, which can diagnose the roll fault through the change of chaotic characteristic quantity. Time series prediction, which is based on the past and present observation data of things being predicted, constructs a time-dependent sequence model and predicts the future with certain rules, which is the premise of predicting the state of mechanical motion. Chaos is applied to the prediction of time series. Many phenomena that used to be regarded as random signals can now be explained by chaos theory. The theory of phase space reconstruction is used to predict time series based on echo state network. Through the research of time series prediction, it lays a foundation for the prediction of roll running state. It makes us make a judgment on whether the roll is in the state of failure and the severity of the malfunction. It has a strong practical application space and can improve the safety of rolling mill equipment by taking preventive measures in advance. It has important practical application value to improve economic benefit.
【学位授予单位】：燕山大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TG333;TH165.3

【参考文献】