This article presents a method to experimentally estimate the direct and cross-coupled dynamic coefficients of tilting-pad journal bearings of vertical hydro-generators and other similar rotating machinery for damage detection purposes. Based on a simplified second-order model of a journal bearing in the state-space, the method employs only the usually monitored vibrations, the shaft radial relative, and the bearing radial absolute vibrations originated by the hydro-generator residual unbalance or by hydraulic excitations in the turbine rotor. This article shows that the method was successfully tested using the shaft and bearing vibration signals synthesized by a mathematical model of a 700 MW hydro-generator, even when these signals are contaminated with random noise. This article also shows the method’s performance when applied to real vibration signals acquired from the modeled hydro-generator. Finally, it discusses the possible measures to improve the method’s efficiency.

本文提出了一种通过实验估算立式水轮发电机和其他类似旋转机械的可倾瓦滑动轴承的直接和交叉耦合动态系数的方法，以进行损坏检测。该方法基于状态空间中轴颈轴承的简化二阶模型，仅采用通常监测的振动、轴径向相对振动和由水轮发电机残余不平衡或液压激励引起的轴承径向绝对振动在涡轮转子中。本文表明，使用由 700 MW 水力发电机的数学模型合成的轴和轴承振动信号成功地测试了该方法，即使这些信号受到随机噪声的污染。本文还展示了该方法应用于从建模水轮发电机获取的真实振动信号时的性能。最后讨论了提高该方法效率的可能措施。