基于齿轮箱物理结构的风机状态监测数据预处理方法

A method of pre-processing wind turbine condition monitoring data based on the physical structure of gearbox

风机状态监测能实时了解风机的健康状态,降低风力发电场的维护成本。风机状态监测应用产生的原始数据体量大速率高、并发性强,对风场局域网数据传输的压力大,对风场控制中心数据分析的计算和存储能力要求高。文章提出了一种基于齿轮箱物理结构的数据预处理方法,在数据源端结合齿轮箱结构参数和轴转速信息,将振动数据根据轴同步转速脉冲进行周期域处理,可将数据容量减小至原始文件的2%以下。该方法在最大化保留风机传动链各部件健康信息的前提下缩小了数据量,降低了网络传输带宽占用,减少了数据存储空间,提高了风机状态监测系统的数据分析效率,为风机状态监测和故障诊断提供了有效的数据源。

Wind turbine condition monitoring can monitor the healthy state of the wind turbine in real time and reduce the maintenance cost of wind farm.The raw data generated by this application has the characteristics of large amount of data,high speed and strong concurrency,which puts great pressure on network transmission and requires high caleulation and storage capacity of local data analysis.In this paper,a data preprocessing method based on the physical structure of the gearbox is proposed.Combined with the structural parameters of the gearbox and the information of shaft speed,the vibration data that is time domain is converted to cycle domain data based on shaft synchronous speed pulses and procesed at the data source,and the data capacity can be reduced to 2% of the original file.After applying the proposed method,on the premise of keeping the health information of each part of the wind turbine drive chain maximally,the amount of data is reduced,so as to reduce the bandwidth occupied by the network transmission,reduce the data storage space,improve the efficiency of data analysis and reduce the system request response time.