基于正交多项式拟合的称重传感器非线性校正.pdf

第44卷第4期 中国测试 Vol.44 No.4 2018年4月 CHINA MEASUREMENT TEST April袁2018 doi院10.11857/j.issn.1674-5124.2018.04.017 基于正交多项式拟合的称重传感器非线性校正 黄永刚 渊中国铁道科学研究院标准计量研究所袁北京 100015冤 摘 要院因受自身材质尧工艺的限制以及外界环境影响袁机车车辆称重传感器实际的输入输出关系存在非线性特性袁 造成测量结果的非线性误差袁制约其测量精度遥对此袁该文采用曲线拟合法尧以正交多项式做最小二乘拟合建立非线 性校正环节袁对机车车辆称重传感器进行非线性校正袁来减小非线性特性对测量结果造成的影响袁提高测量精度袁同 时避免正规方程系数矩阵的病态问题遥实验结果表明袁校正后系统的非线性得到明显改善袁测量精度得到显著提高袁 同时该方法操作简单尧校正效率高尧结果稳定可靠袁具有较好的工程应用价值遥 关键词院机车车辆称重传感器曰非线性校正曰正交多项式曰最小二乘法曰曲线拟合 文献标志码院A 文章编号院1674-5124渊2018冤04-0091-05 The non-linear correction of the weighing sensor based on orthogonal polynomial fitting method HUANG Yonggang 渊Standards Metrology Research Institute袁China Academy of Railway Sciences袁Beijing 100015袁China冤 Abstract: Because of the material袁 process restrictions and external environmental impact袁 the actual input-output relationship of rolling stock weighing sensor presents non-linear characteristics袁 which causes non-linear error of measurement results and limits the measurement accuracy袁 thus non-linear correction is needed. The curve fitting method and least squares method based on orthogonal polynomial are proposed to establish non-linear correction links袁 so as to achieve non- linear correction for rolling stock weighing sensor and thereby reduce the influence of non-linear characteristics on the measurement results and improve the measurement accuracy袁 and avoid the ill-posed problem of the matrix of formal equations at the same time. The test results show that the system nonlinearity of corrected system is obviously improved袁 and the m