摘要
谷幅变形监测对保障蓄水电站等重大基础设施安全有重要作用。然而,传统谷幅变形监测方法大都选择河谷两岸气象条件较为稳定且一致时观测,很大程度上限制了谷幅变形监测全天时自动化监测的实施。为此,引入测量平差理论,研究谷幅变形全天时自动化监测的气象误差改正方法。首先,采用传统气象改正模型初步改正光电测距的气象误差;其次,采用Fourier级数拟合建立初步气象距离的全天时趋势误差模型,并据此削弱趋势项误差;再次,提出联合稳健卡尔曼滤波和加权最小二乘算法获取全天时谷幅变形观测值;最后,选用了锦屏一级水电站测试上述方法。结果显示,该方法估计的谷幅变形精度约0.9 mm,基本能够满足谷幅变形监测对精度的要求。
Valley deformation monitoring plays an important role in ensuring the safety of major infrastructure such as water storage power station.However,the traditional monitoring methods mostly choose the observation when the meteorological conditions on both sides of the valley are relatively stable and consistent,which greatly restricts the implementation of automatic all-day monitoring of valley deformation.Therefore,this paper introduces the measurement adjustment theory to study the meteorological error correction method of automatic monitoring of valley amplitude deformation in the whole day.Firstly,the meteorological error of photoelectric ranging is corrected by the traditional meteorological preliminary correction model.Secondly,a nonparametric regression adaptive model of all-day trend error of preliminary meteorological correction data is constructed using Fourier series fitting to reduce trend error.Thirdly,a combination of robust Kalman filter and weighted least square is proposed to realize further processing of ranging data,so as to obtain automatic monitoring data of all-day valley amplitude deformation.Finally,the proposed method is used in Jinping I Hydropower Station as a test.The experimental results show that the accuracy of automatic valley deformation estimated by the proposed method can reach 0.9 mm,which can basically meet the accuracy requirements of valley deformation monitoring.
作者
周绿
张凯
刘书明
杨泽发
肖亚子
ZHOU Lü;ZHANG Kai;LIU Shuming;YANG Zefa;XIAO Yazi(PowerChina Zhongnan Engineering Corporation Limited,Changsha 410014,Hunan,China;School of Geosciences and Info-Physics,Central South University,Changsha 410083,Hunan,China)
出处
《水力发电》
CAS
2024年第1期98-102,共5页
Water Power
关键词
谷幅变形
气象改正
非参数误差趋势估计
卡尔曼滤波
加权最小二乘算法
valley deformation
meteorological correction
nonparametric error trend estimation
Kalman filter
weighted least squares
