中文摘要：

风电机组状态监测对于风电场特别是海上风电场降低维护成本,提高运行水平具有重要的实用价值。采用温度趋势分析的方法对风电机组齿轮箱的运行状态进行监测。利用非线性状态估计（nonlinear state estimate technology,NSET）方法建立齿轮箱正常工作状态下的温度模型并用其进行温度预测。通过合理构造过程记忆矩阵,使模型覆盖齿轮箱的正常工作空间。当齿轮箱工作异常时,其动态特性偏离正常工作空间,NSET温度模型预测残差的分布特性发生改变。采用滑动窗口方法实时计算残差的统计分布特性,当残差的均值或标准差的置信区间超出预先设定的阈值时,发出报警信息,提示运行人员检查设备状态。为模拟齿轮箱的故障情况,在机组数据采集与监视控制系统（supervisory control anddata acquisition,SCADA）数据中加入人为温度偏移。通过对该模拟故障的分析,新的状态监测方法能够及时发现齿轮箱的异常状态,达到实时在线状态监测的目的。

英文摘要：

Condition monitoring（CM） of wind turbine can greatly reduce the maintenance cost for wind farm especially offshore wind farm.A new condition monitoring method using temperature trend analysis for a wind turbine gearbox was proposed.Nonlinear state estimate technique（NSET） was used to construct the normal behavior model of the gearbox temperature.With a proper construction of memory matrix,the NSET model can cover the normal working space for the gearbox.When the gearbox has an incipient failure,the residuals between NSET model estimates and the measurement temperature will become significant.A moving window statistical method was used to detect the changes of the residual mean value and standard deviation in a timely manner.When one of these parameters exceeds predefined thresholds,an incipient failure was flagged.In order to simulate the gearbox fault,manual temperature drift was added to the initial SCADA data.Analysis of the simulating gearbox fault shows that the new condition monitoring is effective.