摘要
状态监测和早期故障诊断对大型发电机的安全、经济运行具有非常重要的作用。定子温度类故障是发电机故障中的一大类。为了给发电机定子温度类故障的早期诊断提供判据 ,该文对水 氢 氢冷却的汽轮发电机定子绕组温度在不同运行工况下标准值的计算方法进行了研究。先从理论上建立了各线圈出口水温度和线棒槽内检温计温度与发电机结构和运行参数之间的统一模型 ,然后以正常运行时不同稳定工况下的大量实际数据为基础 ,分别用人工神经网络(ANN)和最小二乘法对模型进行了辨识。通过与实测数据的对比 ,用人工神经网络计算的最大误差小于 1℃ ,而最小二乘法的误差也不超过 2℃。
Condition monitoring and fault diagnosis plays a very important role for the safe and economic operation of large turbogenerators. Fault concerning stator temperature is an important type of turbogenerator fault. To provide criterion for early detection of fault concerning stator temperature, this paper studies computation method of standard temperature of stator windings for turbogenerators cooled by water and hydrogen. First, general models are set up for stator bar coolant outlet water temperature and temperature detected by detectors embedded in individual slot theoretically, taking account of their relationships with turbogenerator structure and operating parameters. Then, on the basis of large quantity of data in different steady and healthy operating points, artificial neural network(ANN)and the least squares method are used to identify these models. Compared with measured values, the maximal error calculated by artificial neural network is less than 1℃,while that by the least squares method is no more than 2℃. These models have been put into use in an on line turbogenerator fault diagnosis expert system of some power plant successfully.
出处
《中国电机工程学报》
EI
CSCD
北大核心
2001年第8期79-83,共5页
Proceedings of the CSEE
关键词
汽轮发电机
故障诊断
定子绕组
温度标准值
人工神经网络
fault diagnosis
standard temperature
artificial neural network
the least squares method
