摘要
可靠性是测量网优化设计和数据质量控制要考虑的重要因素之一。在独立观测情况，测量网的可靠性可用多余度（或称多余观测分量）来表示。多余度越小，该观测位置的可靠性越差。但在相关观测情况，多余度的值域超出［０，１］区间，已不能反映可靠性特征。本文建议采用一种新的可靠性度量指标：γｉ＝ （ Ｐ Ｒ）ｉｉ／ｐ ｉｉ，其中 Ｐ 是观测权阵， Ｒ 是平差因子阵。γｉ 的值域在［０，１］区间，γｉ 越小，该观测位置的可靠性越差。本文认为，通常包含在内部和外部可靠性指标中的非中心参数 λ０，由主观选择的弃真概率 α０，检验功效β０ 以及模型误差个数ｂ０ 决定，且一经选定，λ０ 是常数，因此可靠性指标可不考虑 λ０ 因子。系统的可靠性完全由 Ｒ 和 Ｐ 决定。本文论证，新的指标 γｉ 是一种统一的指标，既适于独立观测，也适于相关观测；还可将内部和外部可靠性统一起来。最后对三个算例的有关指标进行了分析比较。
The reliability is one of the important factors considered in the optimal design of surveying networks and the control of data quality. For the independent observations, the reliability of a surveying network may be indicated by the redundancy measure (or called as component of the number of redundant observations). But for correlated observations, the redundancy measures is over [0,1], at times even some of them are negative, so that the redundancy measures may not reflect the characteristics of the reliability. A new measurable indication of reliability is suggested in this paper, i.e. γ i=(PR) ii /p ii , here P is the weight matrix of observations, R the adjustment factor matrix, the subscript i represents the place of the observation i.γ i is in [0,1]. The smaller γ i is, the poorer the reliability corresponding to the observation i is. The new indication γ i is united one. It is suitable not only for independent observations, but also correlated. In addition, it may unite the internal and external reliability. The opinion of the author is that the reliability of surveying network is completely determined by the matrices by R and P, so that the noncentral parameter λ 0 is not needed to be included in the indication of reliability.
出处
《测绘学报》
EI
CSCD
北大核心
1999年第3期189-194,共6页
Acta Geodaetica et Cartographica Sinica
基金
国家自然科学基金
中科院资源与生态环境研究重点项目
国家攀登项目子课题
中科院动力大地测量学开放研究实验室项目
关键词
测量
相关观测
测量网
可靠性
数据质量
correlated observation, internal reliability, external reliability, component of the number of redundant observation (or redundant measure),Marginally Detectable Bias (MDB)
