油液中磨粒在线监测系统的设计和研究

发布时间：2018-02-20 16:35

本文关键词： 润滑监测故障预测信号处理磨粒识别故障诊断　出处：《燕山大学》2015年硕士论文　论文类型：学位论文

【摘要】：本论文设计完成了一套润滑油磨粒在线监测系统。系统利用电磁感应的原理,实时监测润滑油液中铁磁性磨粒的尺寸和浓度信息,预测设备的运行趋势,实现故障预警。传感器对于大于7 5 u m的铁磁性磨粒具有良好的检测效果。系统设计主要包括硬件部分和软件部分:硬件部分包括高频激励源电路、模拟调理电路和数字处理电路。激励源部分通过设计、搭建硬件电路实现用于初级线圈工作的高频交流信号的稳定输出。模拟电路部分包括前置放大电路、带通滤波电路、真有效值转换电路和调零放大电路,通过模拟调理电路,使模拟信号输出范围在0-3 V内,以获取高信噪比的数字信号。数字电路部分的微控制器为S T M 3 2 F 1 0 3 V E T 6,包括采样频率选择、实时存储等功能模块。最后绘制、加工了P C B板,并完成了整个监测系统硬件平台的搭建。软件部分设计、实现了模拟信号的采集和预处理,数字信号处理、液晶屏显示以及储存等功能。数字信号处理主要是针对润滑油液中磨粒信号的局部加权回归散点平滑算法(L O W E S S)的设计实现和极值转换两个部分的内容。通过L O W E S S算法抑制、消除了信号中的噪声,提高了检测的性能;通过极值转换得到了信号的峰峰值,获取了反映磨粒尺寸和浓度的信息,以便基于油液中磨粒的分布预测设备的运行状态。论文最后通过两组对比试验,对研制的在线监测系统的性能进行了验证,实际润滑油液磨粒监测实验证明系统的监测精度可满足风电齿轮箱润滑油中磨粒检测的工程要求。
[Abstract]:An on-line monitoring system for lubricating oil abrasive particles is designed and completed in this paper. The system uses the principle of electromagnetic induction to monitor the size and concentration information of ferromagnetic abrasive particles in lubricating oil in real time, and to predict the running trend of the equipment. The sensor has a good effect on the detection of ferromagnetic abrasive particles larger than 7.5 u m. The system design mainly includes the hardware part and the software part: the hardware part includes the high-frequency exciting source circuit. Analog conditioning circuit and digital processing circuit. The excitation source part is designed to build the hardware circuit to realize the stable output of the high frequency AC signal used for the primary coil work. The analog circuit includes preamplifier circuit, bandpass filter circuit, and so on. Through analog conditioning circuit, the output range of analog signal is within 0-3 V. In order to obtain the digital signal with high signal-to-noise ratio (SNR), the microcontroller of the digital circuit is S T M 3 2 F 10 3 V E T 6, which includes sampling frequency selection, real time storage and so on. Finally, the PCB board is drawn and fabricated. The hardware platform of the whole monitoring system is built, the software part is designed, the analog signal is collected and preprocessed, and the digital signal processing is realized. The digital signal processing is mainly aimed at the design and implementation of local weighted regression scatter point smoothing algorithm for abrasive signal in lubricating oil liquid and extreme value conversion. The suppression is carried out by means of the L O W E S algorithm. The noise in the signal is eliminated and the detection performance is improved. The peak and peak value of the signal is obtained by extreme value conversion, and the information reflecting the size and concentration of the abrasive particle is obtained. In order to predict the running state of the equipment based on the distribution of abrasive particles in oil, the performance of the on-line monitoring system is verified by two groups of comparative tests. It is proved that the monitoring precision of the system can meet the engineering requirements of abrasive particle detection in the lubricating oil of wind power gear box.
【学位授予单位】：燕山大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TE96;TP274

【参考文献】