摘要
传统风电机组故障诊断预警算法准确度较低,预警速度较慢,为此,提出基于工况辨识的风电机组故障诊断预警算法。首先进行故障诊断数据预处理,基于工况辨识,选择合适的风电机组运行参数,采用温度预测的方法,利用观测温度与预期温度的残差来反映设备的运行状态,使用MSET建立风机齿轮箱温度模型,确定预警指标,设计基于工况辨识的风电机组故障诊断预警算法。实验结果表明,设计算法能有效减少诊断预警时间,且其准确度更高。
The traditional fault diagnosis and early warning algorithm of wind turbine has low accuracy and slow early warning speed.Therefore,a fault diagnosis and early warning algorithm of wind turbine based on condition identification is proposed.Firstly,preprocess the fault diagnosis data,select the appropriate operating parameters of the wind turbine based on the condition identification,use the temperature prediction method to reflect the operating state of the equipment through the residual between the observed temperature and the expected temperature,use MSET to establish the temperature model of the wind turbine gearbox,determine the early warning index,and design the fault diagnosis and early warning algorithm of the wind turbine based on the condition identification.Experimental results show that the designed algorithm can effectively reduce the diagnosis and early warning time,and its accuracy is higher.
作者
苏俊
方超
孙猛
SU Jun;FANG Chao;SUN Meng(State Power Investment Group Jiangsu New Energy Co.,Ltd.,Yancheng 224399,China;Shanghai Power Equipment Research Institute Co.,Ltd.,Shanghai 200240,China)
出处
《现代信息科技》
2021年第12期33-35,40,共4页
Modern Information Technology
基金
国家电力投资集团有限公司统筹研发经费支持项目(TC2020FD05)。
关键词
工况辨识
风电机组
故障诊断
预警算法
condition identification
wind turbine
fault diagnosis
early warning algorithm
