论文部分内容阅读
鉴于监测数据的准确度对反映监测体的性质非常重要,但因受到各种因素的干扰,不可避免地存在粗差或异常值,因此必须对监测数据进行误差处理与分析。进而采用改进的未确知滤波法探测定位粗差,同时引入粒子群算法对改进的未确知滤波法中的参数进行搜索优化,建立了数据预处理模型,将该模型应用于锦屏一级水电站左岸边坡的位移观测值检测中,并与传统的莱因达准则的探测结果进行了对比分析。结果表明,数据预处理模型探测精度较传统方法有明显提高,结果更加科学可靠。
Since the accuracy of the monitoring data is very important for reflecting the nature of the monitoring body, it is inevitable that there is a gross error or anomalous value due to interference from various factors. Therefore, the monitoring data must be processed and analyzed in error. Furthermore, the improved unascertained filter method is used to detect the location gross error. At the same time, particle swarm optimization is introduced to search and optimize the parameters of the improved unascertained filtering method. A data preprocessing model is established. The model is applied to Jinping level In the left bank slope displacement observation of the hydropower station, it is contrasted with the detection results of the traditional Leyden criterion. The results show that the detection accuracy of the data preprocessing model is obviously improved compared with the traditional method, and the result is more scientific and reliable.