摘要
为了能更准确、容易地在线诊断出同步发电机转子绕组匝间短路故障,提出了一种基于支持向量回归机的励磁电流预测方法.利用同步发电机正常运行时不同工况下的机端电压、有功功率、无功功率和励磁电流来建立发电机励磁电流的支持向量回归机预测方法.利用该方法预测正常运行时所需励磁电流,并与在线实测的励磁电流进行比较,误差(相对误差)超过阈值就诊断为发生匝间短路故障.通过微型同步发电机动态模拟实验表明,该方法的精度优于BP神经网络法和遗传规划法.
For more accurate and easy online diagnosis of synchronous generator rotor winding interturn short-circuit fault,this paper puts forward a novel field current prediction method based on support vector regression( SVR) machine. We use terminal parameters( voltage,active power,reactive power) and rotor field current under different fault-free operating conditions of synchronous generator to create this SVR prediction method. If online measured terminal parameters are input to this method,a predicted field current will be obtained. Then,by comparing the predicted field current with the corresponding online measured field current,if error( relative error) exceeds a specific threshold,a synchronous generator rotor winding inter-turn short-circuit fault is diagnosed to occur. The micro-synchronous generator dynamic simulation results show that this method has better accuracy than BP neural network method and genetic programming( GP) method.
引文
[1]Auckland D W,Pickup I E D,Wu Y T,et al.Novel approach to alternator field winding interturn fault detection[J].IEEE Proceedings-Generation,Transmission Distribution,1995,142(2):97-102.
[2]马宏忠.电机状态监测与故障诊断技术[M].北京:机械工业出版社,2008.
[3]马宏忠,李之昆,彭钱.GP法在发电机转子绕组匝间短路诊断中的应用[J].继电器,2004,32(22):16-19.
[4]丁坚勇,李劲波,陈允平.基于人工神经网络的同步电机转子匝间故障在线诊断[J].电力系统及其自动化学报,1999,11(4):44-48.
[5]蔡艳宁,胡昌华.基于支持向量回归机的陀螺漂移预测模型[J].中国惯性技术学报,2007,15(5):593-597.
[6]鲍可进,高飞.基于SVR的动平衡加工误差补偿方法[J].传感器与微系统,2010,29(10):14-17.
[7]田有文,唐晓明.基于支持向量机的微机保护装置状态评估的研究[J].电力系统保护与控制,2009,37(4):66-69.
[8]邓乃扬,田英杰.数据挖掘中的新方法—支持向量机[M].北京:科学出版社,2004.
[9]吴鹏,宋文龙.基于支持向量机理论的植物根系图像边缘检测方法[J].农机化研究,2012,34(7):66-69.