Browsing by Subject "robust estimation"
Now showing 1 - 3 of 3
- Results Per Page
- Sort Options
Item type:Article, Access status: Open Access , A review of robust estimation methods applied in surveying(2012) Banaś, MarekThis paper provides a review of popular outlier-robust methods used in surveying. The presented methods have been divided into two groups i.e. active and passive methods. The first group comprises the following methods: the Huber Method, the Hampel Method, the Danish Method, the Gaździcki Method, the Least Absolute Deviation and a Choice Rule of Alternative. The second group of methods is represented by the so called iterative data snooping (IDS) and i-test. There are M-estimation rules presented in the introduction to the review of active methods.Item type:Thesis, Access status: Restricted , Teoria L-estymatorów i M-estymatorów(Data obrony: 2015-09-30) Wardzała, Paulina
Wydział Matematyki StosowanejItem type:Article, Access status: Open Access , Zastosowanie metod estymacji odpornej w obliczeniach nośności granicznej pali(2008) Muszyński, Zbigniew; Rybak, JarosławThis work presents the results of the calculations of foundation pile load failure on the basis of the so-called 80%-method, formulated by Brinch Hansen. The approximation procedure has been analysed from the point of view of: the number of measurements of settlements, taken into account in the calculations, the weights of the settlement measurements and the change in the weights of particular settlement measurements (the elimination of errors by the resistant estimation). With the use of Mathcad environment, an algorithm was developed, which serves the calculating of load failure by the following methods: least squares method, when the same weights are assumed for all the analysed points; least squares method, assuming differentiated weights for particular points; Huber method, assuming the initial differentiation of weights; Danish method, with the assumption, that the weights for particular points are initially differentiated. In the computations, one made use of the last six observations in the aspect of elastic-plastic work of piles (the last six observations before the ending of the load test). The conclusions from the calculations are intended to constitute the starting point for the further development of the method of failure load computation on the basis of the results from the static load test, for the needs of calibrations - introduced, with growing frequency, into dynamic load tests.
