1984 Vol. 9, No. 1
Die Untersuchung mehr oder minder komplizierter Bewegungsmuster der Erdkruste oder anderer Objekte,die mit Hilfe geodätische Netzeüberwacht werden,ist eine wissenschaftliche Aufgabe,die nicht schematisch,sondern nur individuell ausgefuhrt werden kann.Ebenso wie in anderen Bereichen der Wissenschaft muβ das Instrumentarium des Analytikers vielfältig und flexibel sein,um die Vielfalt der möglichen Erscheinungen erfassen zu können.In der vorliegenden Arbeit wird es unternommen,allgemeine Methoden zur geometrischen und statistischen Untersuchung geodätischer Netze auf konformität und Affinität darzustellen.Die gezeigten Methoden sind gut geeignet für interaktive Arbeit an graphischen Bildschirmen.Neben der Einzelpunktanalyse können finite Elemente,Teilnetze,Übergangszonen oder ganze Netze untersucht werden.Multivariates Design der Vergleichsnetze ist kein Hindernis.Sowohl homogene Verformungen als auch Unstetigkeitsstellen können erfaβt und quantitativ und qualitativ beschrieben werden.Besonderer Wert wird auf geometrische Strenge gelegt.
This paper introduces the definition of the differential coefficient from matrix to matrix,and applies this more generalized definition to prove the minimality of the variance of the adjusted results.This method of proving is considered to be more rigorous than the usually adopted ones.
The Wuhan district Doppler positioning observations have been adjusted with a software package prepared by author for the purpose of investigating the prospects of its application in geodetic surveying and various engineering projects,The potential accuracy of Doppler positioning has been thus further developed by proper processing method.The experimental result has been very satisfactory as is shown in the f first part of this paper.Then,a series of analysis and comparison have been made and finally a summary is given with some recommendations for further processing or designing of Doppler positioning networks.
The iteration method with variable weights would be an effective method for the location of gross errors in the adjustment.After a general view of several different and available weight functions of this method a new weight function and a corresponding test quantity are developed based on the a posteriori variance estimation of the least squares.The advantages of the suggested method are justified through a comparative investigation of the different weight functions.The introduction of this weight function into danish method can also produce a good result.
In this paper the methods and experiments for applying the digital terrain model(DTM) to improve the classification accuracy are introduced.The forest area of Jiu-gong mountains in Hubei province of China was selected as a test area.The author's experimental results indicated that forest types were seriously confused in classified classes when the classification processing had been based on original MSS images only.As soon as the auxiliary data,which refletcs the terrain relief influences to reflecative spectrum of ground feature,were introduced into processing,the classification accuracy could be improved.The average classification accuracy in the test area had been increased to 70.4% in case the combination processing of original MSS images with solar incident angle data was taken.
Aimed at a reiterative free network,this paper puts forward the conception of the displacement weight R and the principle of the solution to the rank-defect normal equations according to the fact that the weighted norms of unknown vectors are the least.In this paper are included:the deductions of the equations,analysis of their methods and examples,and a few examples in computer calculations.Although this paper deals only with the plane control network,the method that the paper puts forward may also be used to adjust other types of free network.
The paper first analyzes the effect of refraction on triangle closure in extensive tri angulation chain.Then in considering the effects of random error,refraction and error of centring correction,the formula for estimating the correlation coefficient ρ△ between two adjacent triangle closures is derived,and the relation between the ratio Kf of the random m.s.e.to mf,which is the angle m.s.e.gd:\PDF\.pdfiven by Ferrero criterion,and the estimation of ρ△,and the formula to estimate the m.s.e.of observation of any certain angle in a chain are further deduced.The latter two formulas are Kf2=A-ρ△/1/3+A,Kf=mn/mf,A=s02-4.33Ds/3(s02+Ds) and the approximate formula for estimating the variance of any given angle α mα2=mf2+B-2B/1+Ds/s02COSα,B=mf2(1-Kf)2,where mn is the random m.s.e of the observed angle awithout refraction effects,and s0,Ds the mean value taken over all sides in the chain and the variance to the mean.Applying the above formulas to the Chinese first order triangulation chain,the result is that Kf2≈0.61=0.61,i.e.,the effect of random errors is 61% in mf2,and that of nonuniform refraction fields is about 39% in the chain.In addition,it may be considered that ρ△=-0.13 and the tendency for large angles to be less accurate in the chain both are caused by the effect of non-uniform refraction fields.
This paper deals with the probability and practicality of the Laser Ranging to active fault.Applying the mathematical model based on the Kapoller parameters,an analog computation for district Songpan,in Szechuan Province has been carried out.Some problems are discussed in this paper,such as,orbital parameters have to be slacked off and two computational methods could be used for this purpose;it will not be convergent if only one circle data is used;the zeropoint error for a clock can't be solved and so forth.