基于工作模态的装运减速器模态参数的识别

发布时间：2018-06-24 09:17

本文选题：装运减速器 + 数值模态分析　；参考：《太原理工大学》2012年硕士论文

【摘要】：随着煤矿资源的整合和安全要求的提高,煤矿综合机械化采煤(综采)装备迎来了发展的黄金期。连续采煤机装运减速器作为连采机的核心部件之一,它的准确设计和可靠运行是现代矿井高产、高效的一个重要保证。由于煤矿极其恶劣的工作环境,减速器故障率往往很高。究其故障来源一是新产品设计、制造、装配过程中的误差；二是工作过程中产品的疲劳、过载、环境污染等。模态分析是研究结构动态特性的重要手段,可以获得结构在内部和外部各种激励作用下的振动响应,从而为结构的优化设计、故障诊断和状态监测提供准确的参考。本文运用有限元分析方法对减速器模型进行数值模态分析,能够得到其在自由状态和约束状态下的模态频率和振型,对减速器的优化设计可提供参考,同时其对应各阶模态频率的振型可以作为模态试验响应测点布置的重要依据。实验模态分析在人工激励下,根据激励和响应的传递函数或脉冲响应函数进行模态参数的识别,其识别结果很好地反映了结构静止状态下的模态特性。但是减速器作为旋转机械,静止和工作状态下结构的振动特性差别很大,工作模态分析更适合分析减速器工作时的动力特性。频域分解技术是工作模态参数识别方法的一种,是对峰值法的延伸,通过在离散频率处对响应信号的功率谱密度矩阵进行奇异值分解来获取模态参数,克服了峰值拾取法的缺点,使识别精度更高,抗噪声能力更强。减速器工作时齿轮啮合会产生谐波激励,使得模态分析结果中既存在结构固有模态还叠加有谐波模态。概率密度分析是信号识别的一种方法,将其运用到固有模态和谐波模态两种信号的分析中,所得两者的概率密度曲线特征存在显著不同。本文将频域分解技术和求解概率密度方法运用于减速器的工作模态参数识别中,使两种模态得以鉴别,从而准确地识别出减速器真实的结构模态参数。将其结果与LMS Test. Lab商用振动分析系统的识别结果进行比较,二者能够较好地吻合,从而证明了频域分解技术是可行的、准确的,其分析结果能够作为减速器优化的参考。考虑到减速器结构和传动的复杂性,本文对减速器的振动信号进行了频谱分析,发现了减速器在空载和满载两种情况下振动特性差别很大,且存在部件频率和传动啮频及其高次谐波频率的共振以及加工和安装误差,这些可为减速器的优化提供准确的参考。
[Abstract]:With the integration of coal resources and the improvement of safety requirements, comprehensive mechanized coal mining (fully mechanized mining) equipment has ushered in the golden period of development. As one of the core components of continuous mining machine, the accurate design and reliable operation of continuous shearer is an important guarantee of high yield and high efficiency in modern mine. Because of the extremely bad working environment in coal mine, the failure rate of reducer is often very high. The source of the fault is the error in the process of new product design, manufacture and assembly, and the fatigue, overload and environmental pollution of the product during the working process. Modal analysis is an important method to study the dynamic characteristics of a structure. It can obtain the vibration response of the structure under various internal and external excitations, thus providing an accurate reference for the optimal design, fault diagnosis and state monitoring of the structure. In this paper, the finite element analysis method is applied to the numerical modal analysis of the reducer model. The modal frequency and mode shape of the reducer model under free and constrained state can be obtained, which can be used as a reference for the optimal design of the reducer. At the same time, the modes corresponding to each order modal frequency can be used as the important basis for the layout of the measuring points in response to the modal test. The modal parameters of experimental modal analysis are identified according to the transfer function or impulse response function of excitation and response under artificial excitation, and the identification results reflect the modal characteristics of the structure in static state. However, as a rotating machine, the vibration characteristics of the structure under static and working conditions are very different, and the working mode analysis is more suitable to analyze the dynamic characteristics of the reducer. Frequency domain decomposition is one of the working modal parameter identification methods, which is an extension of the peak value method. The singular value decomposition of the power spectral density matrix of the response signal at the discrete frequency is used to obtain the modal parameters. It overcomes the shortcoming of peak picking method and makes the recognition accuracy higher and the anti-noise ability stronger. The gear meshing will produce harmonic excitation when the reducer is working, which results in the existence of both the inherent mode and the superposition harmonic mode in the modal analysis results. Probabilistic density analysis is a method of signal recognition. When it is applied to the analysis of natural mode and harmonic mode, the characteristics of probability density curve of the two signals are obviously different. In this paper, the frequency domain decomposition technique and the method of solving probability density are applied to the identification of the working modal parameters of the reducer, so that the two modes can be identified and the real structural modal parameters of the reducer can be accurately identified. The results were compared with that of LMS Test. The identification results of Lab commercial vibration analysis system are compared, and the results are in good agreement with each other, which proves that the frequency domain decomposition technique is feasible and accurate, and the analysis results can be used as a reference for the optimization of reducer. Considering the complexity of the reducer's structure and transmission, the vibration signal of the reducer is analyzed in the frequency spectrum. It is found that the vibration characteristics of the reducer are very different in the case of no-load and full-load. The resonance of component frequency, transmission gnawing frequency and its high harmonic frequency, as well as machining and installation errors, can provide an accurate reference for the optimization of reducer.
【学位授予单位】：太原理工大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TH132.46

【相似文献】