基于贝叶斯统计的测量系统误差控制方法仿真

针对传统误差控制方法存在耗时较长、误报率较高以及误差控制精度较低等问题,提出了一种基于贝叶斯统计的测量系统误差控制方法。通过含有区间灰数的模糊子集所处的故障状态,得到测量系统产生误差的条件,利用贝叶斯网络推理算法以及相关的区间运算规则,对规划模型求解,获取测量系统产生误差的概率以及相关部件的重要特征信息。根据求解结果以及相关的空间交汇原理构建待测点坐标、待测距离以及有关于经纬仪观测角的数据模型,通过分析测量系统产生误差的原因,选取满足误差存在条件的数据对网络进行训练,令其逼近非线性函数,完成对测量系统误差的有效控制。实验结果表明,所提方法能够快速、准确地对测量系统产生的误差进行控制。

Simulation of Measurement System Error Control Method Based on Bayesian Statistics

Traditional error control method leads to long time consumption, high false positive rate and low error control precision. Therefore, a method to control the error of measurement system based on Bayesian statistics was presented. the condition for generating error of the measurement system was obtained through the fault state of fuzzy subset with interval gray number. Then, Bayesian network inference algorithm and related interval operation rules were used to solve the planning model, so as to obtain the probability of error in the measurement system and the important feature information of related components. According to the solution result and the related spatial intersection principle, the coordinates of the point to be measured, the distance to be measured and the data model about the observation angle of theodolite were built. By analyzing the reason of error caused by the measurement system, the data satisfying the condition of error was selected to train the network, so as to approximate it to the nonlinear function. Finally, the effective control of error in measurement system was completed. Simulation results show that the proposed method can quickly and accurately control the error generated by the measurement system.