www.heavensgatel2.fora.pl
Heavens Gate Lineage 2
The time now is Thu 10:28, 26 Dec 2024

franklin et marshall There is noise in the model w

 
Reply to topic    www.heavensgatel2.fora.pl Forum Index -> Forum testowe
View previous topic :: View next topic  

Author Message
logoxgvb




Joined: 30 Aug 2010
Posts: 1580
Read: 0 topics

Warns: 0/5
Location: hdfttv


PostPosted: Sun 2:49, 08 May 2011    Post subject: franklin et marshall There is noise in the model w

There is noise in the model with CARV improve recognition accuracy of modal parameters


Calculation can proceed,[link widoczny dla zalogowanych], authors used a principal component analysis. 2. If the order is high, still without a stable modal parameters of valuations, it might be the following situations: (1) excessive noise; Vibration Engineering 11 (2) the number of observed output is far less than the system degrees of freedom; (3) the true model of the system is not CARV model, or model approximation can not be used CARV CARM model. 3. In the order of a given on the preliminary estimates the coefficient matrix should be followed by observing their relative size,[link widoczny dla zalogowanych], if some of the coefficient matrix of the range is much smaller than the other bulk coefficient matrix norm, then they are treated as zero, re- estimated at this time tend to improve the estimation accuracy. 4. If the concept http displacement signal or speed signal to the signal, then (13) [6 +] zero matrix at this time were exempted from jurisdiction between the delay. The prior knowledge can help improve recognition accuracy. 5_ increased model order, but also must introduce the other modal parameters, the modal parameters of what is real or noise mode mode, the author will be discussed in detail in another article. 6. Model order selection of a variety, such as FPE criteria, AIc criteria. This method is to see whether there is stability in the modal parameter estimation, if the sections, you would rather not use CARV model to identify the modal parameters, because at this time to identify the parameters of the valuation is not credible is worth noting that, there is a kinds of order determination is based on the singularity of the correlation matrix of white to set the stage, the idea is to select the autocorrelation matrix of just the United States the largest non-singular model of the order of elimination is the order of. Mathematically speaking, this method is not error. However, the actual calculations, the impact of the truncation error and the presence of noise, resulting from the correlation matrix of the order of singularity occurs with the actual relationship between the system order is complicated, leading to miscarriage of justice. Simulation results show that, even in the data stack only a few percent of the noise, the use of the order of the information obtained in a considerable number of cases is not credible. 7. The amount of DC Dan CARV model more sensitive to the presence of a DC component may fail so CARV modeling, the use of modal parameters CARV model identification, should be to average the data processing. Also in order to minimize the impact of the truncation error must also be divided according to the standard deviation of effective treatment. Reference to the text spine 1YangQJ, BregantL, HeylenW, . Wuxian Min. when and where sub-sequences and series pits and should be thrown off. Beijing; Machinery Industry Press, 19884 Zhang Jinghui, Hong Zhongyu,[link widoczny dla zalogowanych], Qiu Yang. application of the basic properties of time series vibration system parameters identification (I) a theory , methods, software, Vibration and Shock, 1988; 5 (2): 22 ~ 3l5 Zhang Jinhuai - linear model parameter estimation and improved. long copy: National Defense University through the publication of Du .19926 Wang Li Lin Yang. multidimensional model of the timing control modeling. vibration,[link widoczny dla zalogowanych], testing and diagnosis, 1995Il5 (2): l9-26ImprovementofthePrecisionofModalParame ~ rUsingCARVModelinNoiseConditionLuoWenboWang workers TuLiangyaoXiaS ~ gbo. ZouJingxiang (Dept.ofAstronauticsandMechanics, HarbinInstituteofTechnologyHarbin,[link widoczny dla zalogowanych], 150001) AbstractTheprecisionofcontro] averageregressionvector (CARV) model / orparameteridentificationis

相关的主题文章:


[link widoczny dla zalogowanych]

[link widoczny dla zalogowanych]

[link widoczny dla zalogowanych]


The post has been approved 0 times
Back to top
View user's profile

Display posts from previous:   
Reply to topic    www.heavensgatel2.fora.pl Forum Index -> Forum testowe All times are GMT + 2 Hours
Page 1 of 1

 
Jump to:  
You can post new topics in this forum
You can reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum

fora.pl - załóż własne forum dyskusyjne za darmo
Powered by phpBB © 2001, 2002 phpBB Group.
Theme Designed By ArthurStyle
Regulamin