贝叶斯估计动态汽车衡分时段数据融合研究

Multi-period data fusion research of dynamic vehicle weighbridge based on Bayesian estimation

数据融合是一个多级、多层面的数据处理过程,主要完成对来自多个信息源的数据进行自动检测、关联、相关、估计及组合等处理。该文通过分析动态汽车衡的称重原理,对两路称重传感器不同时段的输出数据进行分析,提出基于贝叶斯估计的数据融合方法。实验检定表明:采用这种融合方法的动态汽车衡称重误差小,克服了动态汽车衡由于车辆振动、路面不平和传感器灵敏度分散性、传感器线性度误差等因素对称量结果的影响,准确度高。

Data fusion is a multi-level and multi-layered data process, whose main function is to detect, correlate, relate, estimate and assemble the data of multiple sources automatically. Based on the principle of dynamic vehicle weighbridge, the paper analyzed the output data of two road weighing sensors at different times, and proposed a theory of data fusion method based on Bayesian estimation. The experimental verification shows that the dynamic vehicle weighbridge error is small by using this method. It overcomes the dynamic truck scale caused by vehicle vibration, road surface roughness, the sensitivity of the sensor, the sensor dispersion linearity error and other factors on the weighing results with high accuracy.