Jiří Šejnoha

Homogenization of coupled heat and moisture transport in masonry structures including interfaces

Homogenization of a simultaneous heat and moisture flow in a masonry wall is presented in this paper. The principle objective is to examine an impact of the assumed imperfect hydraulic contact on the resulting homogenized properties. Such a contact is characterized by a certain mismatching resistance allowing us to represent a discontinuous evolution of temperature and moisture fields across the interface, which is in general attributed to discontinuous capillary pressures caused by different pore size distributions of the adjacent porous materials. In achieving this, two particular laboratory experiments were performed to provide distributions of temperature and relative humidity in a sample of the masonry wall, which in turn served to extract the corresponding jumps and subsequently to obtain the required interface transition parameters by matching numerical predictions and experimental results. The results suggest a low importance of accounting for imperfect hydraulic contact for the derivation of macroscopic homogenized properties. On the other hand, they strongly support the need for a fully coupled multi-scale analysis due to significant dependence of the homogenized properties on actual moisture gradients and corresponding values of both macroscopic temperature and relative humidity.

Analysis of coupled heat and moisture transfer in masonry structures

Evaluation of effective or macroscopic coefficients of thermal conductivity under coupled heat and moisture transfer is presented. The paper first gives a detailed summary on the solution of a simple steady state heat conduction problem with an emphasis on various types of boundary conditions applied to the representative volume element—a periodic unit cell. Since the results essentially suggest no superiority of any type of boundary conditions, the paper proceeds with the coupled nonlinear heat and moisture problem subjecting the selected representative volume element to the prescribed macroscopically uniform heat flux. This allows for a direct use of the academic or commercially available codes. Here, the presented results are derived with the help of the SIFEL (More information available at http://mech.fsv.cvut.cz/web/?page=software) (SImple Finite Elements) system.

Probabilistic models for tunnel construction risk assessment

The paper introduces different probabilistic models for prediction of tunnel construction risk. First, a simple probabilistic model for the estimation of the damage due to tunnel construction failures (e.g. cave-in collapses) is proposed. It can be used in conjunction with a deterministic estimate of the construction time/costs as a support for decision-making in tunnel construction projects. The occurrence of failures is modelled as an inhomogeneous Poisson process. The model takes into account the heterogeneity of the environment along the tunnel (changing geological conditions, changing damage potential) and it includes the influence of common factors such as human and organisational aspects. The damages caused by the failures are modelled as uncertain and they are thus represented by full probability distributions in the model. Second, the decision-making under uncertainty in construction projects is discussed. The use of the concept of utility for considering the attitude of the stakeholder to risk is demonstrated. The simple probabilistic model and the decision-making concept are applied to a case study of construction of a 480-m-long tunnel. Third, stochastic models for specific problems of tunnel construction, such as impacts of excavation on surface structures or probabilistic prediction of thickness of rock overburden, are introduced. The use of the models is illustrated on an example from Blanka tunnel in Prague.

A BEM formulation for homogenization of composites with randomly distributed fibers

Abstract In recent papers by the authors, deterministic models of distribution of fibers in composite structures have been studied. For example, problems related to optimization, homogenization, localization, etc., have been solved. The extended Hashin–Shtrikman (H–S) variational principles served as a starting point (eigenparameters were involved in the formulations), and the comparative medium was introduced. The BEM formulations were then admissible and efficient. The formulations of the above-mentioned problems require the restriction of geometry of the fibers to certain ‘locally reasonable’ structures, e.g. to periodic or pseudo-periodic cells. Since the condition of regular distribution of fibers is violated in applications, and a random distribution is more probable, another extension of the H–S principles is needed. In this paper, the problem is extended to the case of statistically distributed fibers. H–S variational principles are formulated in terms of statistical characteristics in the domain and the eigenparameters are also involved, affected by the statistical values. Following the H–S principles, an integral formulation is stated (again, thanks to the use of the comparative medium such a formulation is admissible) in a representative volume, which contains no longer regular geometry of the fibers. The boundary element method has then a special form, which is advantageous particularly for two-phase media. The above-mentioned formulation of H–S variational principles with randomly distributed fields of fibers can be extended to non-linear problems (plasticity, debonding) by introducing transformation fields (eigenstresses or eigenstrains, which are involved in the formulations for completeness). The results form the research presented in this paper basically apply to homogenization of diaphysal implants. But, there is a wide range of applications of the theory introduced in this paper. Due to results from tests on the bearing composite frame of a bicycle, which has a similar structure for certain types of composites of the diaphysal implants, a typical cross-section of the bearing frame of a bicycle is studied as an example. The frame is built of a graphite-epoxy composite.

Analytical model of composite floors with steel fibre reinforced concrete slab subjected to fire

AbstractUnder fire, membrane action plays an important role in the performance of slabs subjected to large deflections. In this paper, a new model is proposed based on a proper approximation of horizontal displacements for a simply supported composite slab. The novelty of the proposed approach consists in a special treatment of the system of shape functions for the “in-plane” displacements. Moreover, a load applied to the slab is divided into two components, so that one component is balanced by the membrane forces, while the second one is transmitted by the bending forces (including transfer of shear and moment). The deflection due to thermal elongations is replaced by the identical deflection caused by a fictitious load. Unknown parameters are calculated using the principle of virtual displacements. The effectiveness of the model is validated by the results obtained from experiments.

Probabilistic approach to damage of tunnel lining due to fire

In this paper, risk is perceived as the probable damage caused by a fire in the tunnel lining. In its first part the traffic flow is described as a Markov chain of joint states consisting of a combination of trucks/buses (TB) and personal cars (PC) from adjoining lanes. The heat release rate is then taken for a measure of the fire power. The intensity λf reflecting the frequency of fires was assessed based on extensive studies carried out in Austria [1] and Italy [2, 3]. The traffic density AADT, the length of the tunnel L, the percentage of TBs, and the number of lanes are the remaining parameters characterizing the traffic flow. In the second part, a special combination of models originally proposed by Bažant and Thonguthai [4], and Kunzel & Kiessl [5] for the description of transport processes in concrete at very high temperatures creates a basis for the prediction of the thickness of the spalling zone and the volume of concrete degraded by temperatures that exceed a certain temperature level. The model was validated against a macroscopic test on concrete samples placed into the furnace.

Hierarchical stochastic model of terrain subsidence during tunnel excavation

In this contribution the Bayesian statistical method is applied to assess the expected probability distribution of the terrain subsidence in the course of tunnel excavation. The approach utilizes a number of simplifying assumptions regarding the system kinematics to arrive at a very simple model with just a few degrees of freedom. This deterministic model together with the intrinsic uncertainties of its parameters and measurement inaccuracies are used to formulate the stochastic model which defines a distribution of the predicted values of terrain subsidence. Assuming the measured data to be fixed, the stochastic model thus defines the likelihood function of the model parameters which is directly used for updating their prior distribution. This way the model parameters can be incrementally updated with each excavation step and the prediction of the model refined.