摘要
超高层建筑数量不断增长,对其安全性的研究势在必行。振动变形监测是建筑安全性研究的重要内容。全球卫星导航系统是目前获取振动变形数据的重要途径,由于多路径误差、电离层误差等因素的影响,监测数据通常含有噪声。利用离散型卡尔曼滤波算法,对监测数据进行降噪处理,与经验模态分解算法结果进行对比,选取信噪比及均方根误差对降噪效果进行评价,并利用频谱图及边际谱图进一步论证卡尔曼滤波算法去噪的有效性及其较EMD算法的优越性。
With skyscrapers growing in popularity,it is important to study their safety,and monitoring vibration deformation is an important part of building safety research.The global navigation satellite system(GNSS)is the main technical means of obtaining vibration deformation data,which is affected by multipath error,ionospheric error and other factors,and the collected data includes noise.We use the discrete Kalman filtering algorithm to reduce noise in monitoring data.Compared to the empirical mode decomposition(EMD)results,the noise reduction effect of the two algorithms is evaluated by two indicators:signal-to-noise ratio and squared mean error,which confirms the noise reduction efficiency of the discrete Kalman filtering algorithm and the superiority over EMD algorithm.
作者
王怀宝
苏玉
WANG Huai-bao;SU Yu(School of geomatics and exploration engineering,Jilin Jianzhu university,Changchun 130118,China)
出处
《吉林建筑大学学报》
CAS
2024年第6期68-73,共6页
Journal of Jilin Jianzhu University
基金
吉林省教育厅科研项目(JJKH20190859KJ).
