基于贝叶斯运行模态分析法的砂轮架动态特性分析

Dynamic Characteristics Analysis of Wheelhead Based on Bayesian OMA Approach

为了研究砂轮架的固有模态和在工作环境中的谐波模态,运用贝叶斯理论和运行模态分析相结合的模态参数识别技术——分步测试的贝叶斯运行模态分析法对环境状态下的砂轮架进行模态识别,获得了结构的动态特性,包括固有频率、振型以及阻尼特性等,并对外激励和预测误差水平以及信噪比进行了评估;将该模态识别结果与传统试验模态分析结果进行对比,验证了贝叶斯运行模态分析法的应用可行性;进一步在工作环境下对砂轮架进行了振动测试,对比开机前的模态识别结果和开机后砂轮架的功率谱密度图,成功地区分出砂轮架的固有模态和实际工作环境中由结构周期激励引起的谐波模态。

In order to study the wheelhead’s intrinsic modes from natural environment and harmonic modes under working conditions ,a modal parameter identification technology based on Bayesian theory and OMA - Bayesian OMA approach was adopted .It was applied to identify the wheelhead’s modal property by ambient vibration tests using different setups .The dynamic characteristics obtained included the natural frequencies ,mode shapes and damping ratios ,which were compared with traditional experimental modal analysis results then to verify the feasible practical applications of Bayesian OMA approach .The estimation for the level of excitation ,prediction error and signal-to-noise ratio also added to the approach’s uniqueness .Furthermore ,vibration tests were conducted for the wheelhead under operational situations .The contrast between identified results from natural environment and power spectral density（PSD） in actual working conditions helps to distinguish the wheelhead’s in‐trinsic modes and harmonic modes caused by periodic excitation successfully .