水轮发电机推力轴承在线检测方法研究

Study of on-line monitoring for thrust bearings of hydro-electric generator

分析了水轮发电机推力轴承传统在线检测方法的性能特点,建立了一种新型的反射式光纤位移传感器的数学模型和一种基于前向神经网络的光纤传感器静态误差修正模型,并将其用于推力轴承油膜厚度的测试系统中。这种可补偿型光纤位移传感器的输出与传感器光源的特性、环境介质以及被测目标表面性质无关。通过将多点测量值求和后取平均值,消除了由于几路接收光纤特性非均质引起的信号波动。此外,给出了针对推力轴承不同瓦面材料的在线检测方法,在试验台的试验中检测系统显示了良好的性能。

The characteristics of the traditional on-line monitoring method of hydro-generator thrust bearing is analyzed. An oil film thickness measuring system is applied to set up a new kind of optical fiber reflective sensor model, the outputs of the sensor are independent of the characteristics of the light source, the losses in the ambient medium and the target reflectivity. Additionally, the distance information is averaged over several points on the target surface, which reduces signal fluctuations due to inhomogeneities. An optical fiber sensor model of correcting static error based on BP artificial neural network (ANN) is set up also. According to the actual operation in thrust bearing test board, the designed system shows better performance.