基于Kalman-BP协同融合模型的含沙量测量

发布时间：2018-03-06 08:36

  本文选题：黄河含沙量　切入点：卡尔曼滤波　出处：《应用基础与工程科学学报》2016年05期 　论文类型：期刊论文

【摘要】：针对黄河含沙量测量易受环境因素影响而导致测量结果不准确的问题,提出基于卡尔曼和BP神经网络(Kalman-BP)的协同融合模型,将含沙量、水温和流速等传感器输出值经过卡尔曼滤波器进行滤波处理;然后经BP神经网络模型对含沙量信息和环境量信息进行多传感器数据融合;最后建立了含沙量测量的反演模型.为了比较Kalman-BP神经网络的协同处理方法的融合效果,在相同环境下还进行了一元线性回归模型和多元线性回归模型的含沙量数据处理,并进行了误差分析比较.实验结果表明,Kalman-BP神经网络协同融合模型的测量误差较小,提高了含沙量测量系统的精度.
[Abstract]:In order to solve the problem that the measurement of sediment content in the Yellow River is easily affected by environmental factors, a cooperative fusion model based on Kalman and BP neural network Kalman-BP) is proposed. The output values of the sensor such as water temperature and velocity are filtered by Kalman filter, and then the information of sediment content and environment are fused by BP neural network model. Finally, the inversion model of sediment content measurement is established. In order to compare the fusion effect of the cooperative processing method of Kalman-BP neural network, the data processing of single linear regression model and multivariate linear regression model is carried out in the same environment. The experimental results show that the measurement error of Kalman-BP neural network cooperative fusion model is small and the precision of sediment content measurement system is improved.
【作者单位】： 华北水利水电大学信息工程学院;郑州大学水利与环境学院;
【基金】：国家科技重大专项(2014ZX03005001) 水利部黄河泥沙重点实验室开放课题基金项目(2012005) 河南省高校科技创新团队支持计划(13IRTSTHN023) 河南省高等学校重点科研项目计划(14B170012,15A510003)
【分类号】：TV149.1
，

本文编号：1574092