基于曲面拟合的超声波液体浓度高精度测量方法

传感器非线性测量误差及温度偏差是高精度液体浓度实时在线测量误差的主要因素.提出了一种基于浓度-温度-声速(S-T-V)曲面模型拟合算法的超声波液体浓度高精度测量方法.该方法以声速测量为基础,首先对非平稳声速测量数据进行预处理以减小系统非线性测量误差,再通过三次样条插值增加测量数据对后续数据拟合的参与度,然后分别建立S-V、T-V单曲线模型,最后运用最小二乘法完成S-T-V曲面拟合,实现多参数融合.实际测量表明,根据本方法测得氯化钠溶液质量浓度精度较高,受温度影响较小,误差优于±0.1％.

High Precision Measurement Method of Ultrasonic Liquid Concentration Based on Surface Fitting

Abstract: The real-time on-line error of high precision liquid concentration measurement comes from sensor nonlinear measurement error and real-time temperature deviation. In this paper, an ultrasonic liquid concentration measurement is proposed with liquid concentration-temperature-velocity surface model fitting algorithm. Based on sound velocity measurement, the method is operation with several stages. Firstly, non-stationary sound velocity measurement data is preprocessed to reduce systemic nonlinear measurement error, and then cubic spline interpolation is employed to increase the participation of measurement data in subsequent data fitting. Finally, the least square method is used in the S-T-V surface fitting, and to realize multi-parameter fusion. Practical measurement shows that, our method reaches high accuracy of mass concentration of sodium chloride solution with less influence of temperature, and the absolute error of measure is less than ± 0.1%.