观测误差

The text analyses GPS observation model, the characteristic of kinds of errors fountain and the adopting measures in the observation, and processes the thorough research to Bernese GPS Software Ms-Windows Version 4.2. It makes the most stable satellites as the main error. The differential way adopts non-differential way and double differential way. The positioning precision of the single point is decimeter level. RMS is within 0.01~0.02; If the ambiguity right and without week skip, the observation datum of a EPOCH-BY-EPOCH can make the short base line（S0km）positioning precision reach centimeter level.

本文分析了GPS的观测模型、观测量中各种误差源的特点及应采取的措施，并对Bernese GPS Software Ms-Windows Version 4.2 软件进行了深入的研究，它是以最稳定的卫星作为主差，采用非差和双差两种差分方法，单点定位精度在分米级，RMS能在0.01~0.02以内；在模糊度已求解正确条件下，若没有周跳，一个历元的观测数据就可以使短基线（S0km）定位精度达到厘米级。

Observationalgross error, such as cycle slip, and deleting nonlinear high-order term above quadraticcan make the observational equation emerge model error.

非线性二次以上高次项舍去，以及周跳等观测粗差等原因，使观测方程产生模型误差。

The maximum permissible error is the unique basis to judge the qualification of tilt instrument.

在地震行业标准《地震观测仪器进网技术要求地壳形变观测仪》的编制研究中，第一次提出以"最大允许误差"作为地形变观测仪器的主要技术指标，也是判别倾斜仪是否合格的惟一依据。

To improve the pointing accuracy of the telescope mount and the capability of daylight tracking, it is necessary to do more study on the error of telescope axes. The present thesis analyses many kinds of factors that cause the pointing error of telescope mount, and discusses the method for correcting the pointing error by setting up telescope mount model in detail.

本文详细分析了望远镜机架指向产生偏差的各种因素，讨论了通过建立望远镜机架指向误差模型的方法来修正望远镜机架指向误差的方法，通过使用CCD观测恒星来求解望远镜机架指向误差模型参数。

But sometimes we find the data that is combined by true explained variable and measurement error when we constructed back to regression analysis. We call this kind of regression model as measurement error model.

但是在建构回归分析时，有时会发现所得到的资料是由真实解释变数和测量误差所结合而成的观测值，像这种具测量误差的回归模式，我们称为测量误差模式〈measurement error model〉。

In order to calibrate the model, data of six typhoons such as the flood peak and the peak time are collected to compare with the result of the simulations. The result shows that the difference of the inflow hydrograph between them is very close. The average error of the flood peak is 5.21%. Furthermore, the error of the peak time is less than one hour but Wenny Typhoon is an exception by more than two hours. Therefore, the model could be applied to simulate the rainfall event at Shih-men watershed.

为验设模式模拟精确性，本研究以具代表性之六场台风降雨模拟，针对石门水库入流历线之观测流量与模拟值比较，结果显示模拟之水库入流历线与观测值差异不大，其尖峰流量误差绝对个平均为5.21%；另在预测洪峰到达时间方面，除温妮台风迟迟2小时外，其余均在1小时内，显示本分析模式可应用於石门水库上游集水区暴雨事件模拟。

The paper also discusses how to properly determine the tolerance of the longitudinal and lateral closure for the CPⅢ network by analog calculation to realize effective inspection in detecting the gross errors of field measurement in the CPⅢ network with the method of closure, then eliminates and resurveys field measurement with gross errors to improve the precision and strengthen reliability of the CPⅢ network.

通过模拟计算，确定了CPⅢ网纵横向闭合差极限误差的建议值，并据此进行CPⅢ网外业观测数据中是否含有粗差的检核。剔除并重测含有粗差的外业观测数据，可以有效提高CPⅢ网的可靠性和精度。

Presently, the reading work of the dial indicator in our country still adopts the manual method. This method is restricted by such subjective factors as observing angel and distance as well as fatigue strength and so on, which results in not only large error and low reliability but also low efficiency and great labor costs.

目前国内对这类仪表的判读工作主要是采用人工观测仪表表盘指针来读数的方法，这种方法受人的主观因素如人的观测角度、观测距离及疲劳强度等的影响，误差较大，可靠性不高，且工作效率低，劳动强度大。

The measurement comprehensive precision evaluating formula considering the weight of baseline observation is put forward in the paper.By checking calculation and analysis,the views on those problems have been proved to be reasonable and representative,and it can be used to revise and to complete the corresponding standards.

其中的( 1 )和( 2 )类误差在仪器检测时表现为加、乘常数及其检测误差，它是文献［1，2，3］规定的必检项目，也是影响测距精度的主要误差源，因此必须对新的和使用中的EDM 的常数进行定期检测，并用检测值对距离观测值施加有效的改正，以保证和提高距离测量的精度。

Throughout comparing the above estimating methods, we have the following results: the estimators of the moments of the errors does not depend on the random effects, and that of the random effects does not depend on the errors, and then the corresponding asymptotic variances are very simple and optimal; when the random effects are multivariate, we can not construct different estimating equations for the random effects and errors respectively, which results that the asymptotic covariances of estimation are very complex and then the estimating efficiency is bad.

比较上述两种估计法，我们发现：当随机效应是一维的时侯，误差的各阶矩的估计不依赖不可观测的随机效应，随机效应的估计也不依赖误差，因此，估计的渐近方差结构特别简单也是最优的；而当随机效应是多维的，因为随机效应的协变量的影响，我们没有办法针对随机效应和误差的各阶矩分别建立估计方程，这导致所得的估计的渐近方差或者协方差矩阵特别复杂，从而估计的效果不是很好。

Plunder melds and run with this jewel!

掠夺melds和运行与此宝石！

My dream is to be a crazy growing tree and extend at the edge between the city and the forest.

此刻，也许正是在通往天国的路上，我体验着这白色的晕旋。