Contents of Journal of Mechanical Engineering 53, 5 (2002) 


 

J. MURIN, V. KUTIS: Beam element with variation of cross section 
with effect of large axial forces                                        261

Y. LI, X. LIU, Z. PENG: A fuzzy method approach for identification 
of joint parameters of mechanical structures                             280

G. H. WOUBE, O. STRYCEK: Research results of an axial-centripetal 
diffuser (in Slovak)                                                     289

H. KLASTERKA: Dynamic properties of a wire probe (in Czech)              309

Abstracts



Beam element with variation of cross section with effect of large axial forces

JUSTIN MURIN, VLADIMIR KUTIS

This paper deals with the new beam element with arbitrary continuous variation of the cross section, allowing also for the effect of large axial forces. This element is not sensitive to fineness of the FEM mesh and fulfils the equilibrium conditions both locally and globally for statically determined and indetermined tasks. The convergence and accuracy of solving of critical loads by our new element have been compared with results which were obtained by the ANSYS computer code for the same problems.

A fuzzy method approach for identification of joint parameters of mechanical structures

YANGMIN LI, XIAOPING LIU, ZHAOYANG PENG

In order to build an accurate dynamic model and to realize optimal control for mechanical structures, the joint surface parameters are the key parameters to be found at first. A method using fuzzy optimization to identify the joint parameters of the mechanical structure is presented in this paper. It is based on a combination of experimental test results of the whole structure with finite element models of the substructures. Theoretical formulae and numerical analysis are shown. In the proposed method the influence of different modes onto the corresponding joint parameters are taken into consideration via weighted factors. At the same time, in this method a fuzzy equation is used, based on the consideration that there exist some uncertain errors in both the experimental values and finite element models. The numerical results demonstrate that the new method can overcome the disadvantages of the least square method that uses a definite equation involving the experimental values and finite element models. By introducing weighted factors and a fuzzy unequal equation, we can obtain more accurate results.

Research results of an axial-centripetal diffuser

GETU HAILU WOUBE, OLDRICH STRYCEK

This paper deals with the interaction between an impeller and a newly developed type of diffuser. Two diffusers of this type were investigated. The work is divided into two main parts: theoretical and experimental. The theoretical part analyzes the computational method now often used and deals with its application to the axial-centripetal diffuser. The experiment investigates the velocity and pressure field distribution and the energy transformation using a ball velocity probe. The experimental results are compared with the theoretical ones.

Dynamic properties of a wire probe

HYNEK KLASTERKA

The paper deals with using hot-wire anemometry for turbulence characteristic measurement. The described method of measurement is based on the fact that the wire probe of an anemometer is sensitive to the fluctuations of all three velocity components and to the fluctuations of fluid temperature and fluid pressure. The aim of this article is showing some effects which influence the dynamic behaviour of the wire probe, e.g. the influence of heat loss to wire supports, the influence of heat inertia of the wire probe and the influence of the frequency of fluid parameters fluctuations. The equation of heat balance of the wire including the cooling effects of the wire supports is derived in the first part of the article. The response of the wire probe at various types of fluctuations is introduced in the second part. Mathematical expressions for dynamic sensitivities of wire probe are presented. It is shown, that dynamic sensitivities are generally dependent on time constant and on the frequency of measured fluctuations.