超声波颗粒监测尺寸标定方法研究

为了解决超声波颗粒监测尺寸标定过程中的非线性问题,提出了一种基于相对精度标定原则的超声波磨粒监测装置中的颗粒尺寸标定方法。首先,将标定颗粒在Malvern2600激光颗粒分析仪上的测试结果作为标定数据,在超声波颗粒监测装置试验中通过混合进化算法进行颗粒与气泡信号的区分。然后,利用Levenberg—Marquardt改进BP算法进行标定曲线拟合,建立超声波颗粒监测装置上的试验数据与标定数据之间的对应关系。最后,本文进行了拟和标定曲线的精度分析。

Test Method of Particle Size Calibration for Ultrasonic Particle Monitoring

Based on the principle of relative calibration, a new method for the size calibration of ultrasonic wear debris monitoring is presented. Firstly, the testing results of the calibrated particles obtained from Malvern Autocounter 2600 are used as the calibration data. In the process of the experiment on the ultrasonic particle monitoring instrument, the combination evolving algorithm is employed to distinguish the particle echo signal from air bubble echo signal. Afterwards, the calibration curve is fitted by using the Levenberg Marquardt algorithm, and the corresponding relationship between the experiment data and the calibration data is set up. Hence, the nonlinear problem during calibration is solved. At the end of the paper, the precision of the calibration curve is analyzed.