基于MSET方法的风电机组齿轮箱预警仿真研究

Wind Turbine Gearbox Prognostic Simulation Research Based on MSET Method

风电机组常年在复杂工况下连续运行,机组对设备的可靠性等方面有很高的要求,齿轮箱重要部件的损伤和失效都会直接或间接导致机组停机,从而导致计划外的更换和维护成本。利用MSET(Multivariate State Estimation Technique)方法对风电机组的齿轮箱轴温进行状态估计,得出齿轮箱轴温的估计值。然后利用滑动窗口残差统计方法对齿轮箱轴温的估计值和实际值的残差进行分析,残差均值能显示出齿轮箱轴温的实时运行状态,当残差均值曲线超出设定的阈值区间时,系统开始预警,表明设备运行异常。仿真结果表明基于MSET方法的风电机组齿轮箱预警系统能够实时监测机组齿轮箱的运行状态。

Wind turbine has been continuous operation under complex condition, and all aspects of the unit equipment have to require high performances such as reliability. Any damage or invalidation of the gearbox important components would make fans stop working directly or indirectly, and it could produce unplanned replacement and maintenance. MEST （Multivariate State Estimation Technique） was used to carry out the state Estimation for the unit gear box bearing temperature to get the gear box bearing temperature estimated values. The sliding window residual statistical method was then used to analyze the residual between the estimated values and the actual values, the residual mean curve was then obtained which expressed the gear box bearing temperature real-time operation state. When residual mean curve went beyond set threshold value interval, the system began prognostic and indicate the equipment has fault. The simulation result shows that the gearbox prognostic system based on MSET method can monitor unit gear box operation real-time.