基于BP网络算法的大地测量误差检测技术  

Geodetic survey error detection technology based on BP network algorithm

作　　者：王怀宝[1] WANG Huai-bao(School of Surveying and Prospecting Engineering, Jilin Jianzhu University, Changchun 130118, China)

机构地区：[1]吉林建筑大学测绘与勘查工程学院,长春130118

出　　处：《沈阳工业大学学报》2020年第3期292-297,共6页Journal of Shenyang University of Technology

基　　金：吉林省教育科学“十二五”规划重点项目(ZD15078).

摘　　要：针对大地测量检测时间长、检测过程成本较高,且检测结果准确度较低的问题,提出一种基于BP神经网络算法的大地测量误差检测方法.对大地测量的基本原理进行分析,通过对测量所得数据的综合计算得到待测量目标相对位移及旋转角度相关测量结果,构建基于BP网络的测量误差预测模型;将测量结果输入模型,得到的输出值即为预测误差,利用动态贝叶斯检验算法判断测量结果是否准确.结果表明,所提测量误差检测方法的检测结果准确率在90%以上,且检测过程所需时间与成本消耗低于实验对比方法,证实了所提方法的检测准确率及检测效率.Aiming at the long detection time,high cost of detection process and low accuracy of detection results of geodetic survey,a geodetic survey error detection method based on BP neural network algorithm was proposed.The basic principle of geodetic survey was analyzed.Through the comprehensive calculation of measured data,the related measurement results of relative displacement and rotation angle of measured targets were obtained.In addition,the measurement error prediction model based on BP network was constructed.The measurement result was input into the model,and the obtained output value was the prediction error.The dynamic Bayesian test algorithm was used to judge whether the measurement result was accurate.The results show that the detection accuracy of as-proposed error detection method is above 90%,and the time and cost needed in the detection process are lower than those of the experimental comparison method,confirming the detection accuracy and efficiency of as-proposed method.

关 键 词：BP网络 大地测量 误差检测 误差检测概率 动态贝叶斯算法 映射函数 相对位移 旋转角度 

分 类 号：TP212[自动化与计算机技术—检测技术与自动化装置]