[en] Extreme dynamic exterior and interior loads in nuclear structural engineering demand exceptional treatment in analysis and design. Theoretical and experimental results for typical nuclear structural dynamic problems are compared, and conclusions for practical applications are derived. The examples discussed refer to the overall elastic dynamic behaviour of a reactor building as well as to the ultimate resistance of slabs against mass impact and to possible reductions of induced vibrations by special structural design. (Author)

[en] Design and fabrication requeriments of the piping penetrations necessary for maintaninig the containment integrity of the reactor structure require that these penetrations survive a worst possible earthquake occurrence without a leakage failure.Current practice only necessitates that these penetrations be qualified to their first restraint beyond the defined qualification boundaries. In order to protect the qualified portion of a system from the effects of a possible failure in its unqualified portion, this restraint would have to by designed to withstand the load required to rupture the system .Due to the magnitudes of the loads considered this becomes an expensive, inefficient and sometimes impossible task to accomplish. Several design methods which will help cope with this problem while maintaining some control over the run-away costs of a Nuclear installation are here introduced. (Author)

[en] An attempt is made to trace the development of extreme load criteria as it applies to earthquakes, extreme wind, high energy system rupture (LOCA), floods and other manmade and natural external hazards, from 1965 until the present, in the leading nuclear power nations throughout the world. (Author)

[en] The analysis of the structures of nuclear power plants requires the evaluation of the internal forces. This is attained by the solution of a system of equations. This solution takes most of the computing time and memory. One of the ways it can be achieved is based on the Cholesky factor. The structural matrix of the coeficients is transformed into an upper triangular matrix by the Cholesky decomposition. Cholesky factor can be obtained directly from the stiffness matrix of the structural element. The result can thus be obtained in a more precise and quick way. (Author)

[en] A method of creep analysis for heated concrete structures is presented which is direct in that it does not require intermediate solutions at many small intervals of time to retain accuracy of solution. It is approximate to a degree that may be chosen by the analyst and is economic in use. The method is based on an optimization procedure for spatial distributions of stress, which are permitted to vary in time. The analysis provides a means of evaluating the time-dependent weighting parameters to the stress distributions and these in turn define the ceeep solution. The accuracy of solution depends upon the number of basic stress distributions used in the analysis and increases as their number increases. The choice of suitable stress distributions is discussed. Finally, comparative solutions are presented for the analysis of a prestresed concrete containment vessel, for the approximate method described - and using two different degrees of approximation - and from a traditional time-step form of computation. It is concluded that a small number of easily selected basic stress distributions can generate acceptable creep solutions at low cost. (Author)

[en] A short description is given of work in progress on the analysis of prestressed concrete containment vessels for nuclear reactors. Firstly,numerical analysis are described and some results shown. Secondly, a description of a microconcrete model under construction is given: forms, prestressing and loading systems and measuring devices. (Author)

[en] In view of the complexity of the results from studies of high-frequency random vibrations in complex structures, a new technique of analysis (statistical energy analysis or S.E.A.) was developed. Instead of a large amount of results obtained by classical analysis, S.E.A. furnishes global parameters - the vibrational energy of sub-systems. After a revision of the fundamental equations of S.E.A., an example (Acoustic and vibrational interaction between the gas flow and the structure is given. The possibility of using hybrid programmes of S.E.A.-Finite Element Method was investigated. The Newland method of calculating the coupling loss factor is find to be adequate. (Author)

[en] The extreme environmental and abnormal loads are of decisive importance for the statical and dynamical design of the safety related buildings of a nuclear power plant (npp.). The arising forces act for a very short time interval, but as a rule with a load peak, whose intensity has been so far uncustomary in structural engineering practice. The treatment of these loading cases requires dynamic investigations by means of special computer programms. For economical reasons, the non linear behaviour of reinforced concrete (r/c) must be considered in these investigations. For heavily reinforced structural members, whose design is mainly governed by the above mentioned extreme loadings, the utilisation of high tensile steel for example the quality Bst 1080/1320 (reinforcing steel, yield point 1080 N/mm2, ultimate strength 1320 N/mm2) leads to reasonable economical advantages and allows and easier assembly of the reinforcing bars (re-bars). Acc. to German Standards, special high tensile steels can be stressed up to the yield point for loadings with a very low probability. (Author)

[en] Evaluation of punching shear stress in flat plates due to moment load is important in the design of slabs supporting pipe restraints and pipe bumpers. The existing analytical, numerical or empirical solutions are not accurate enough for the analysis of the thick slabs which are used in power plants. The exact analytical solution of this problem is presented, based on the Fourier expansion of the moment load in conjunction with the equilibrium equation of a flat plate. An evaluation of a typical case shows that the widely used empirical solution could be unconservative for slabs with small aspect ratios. In such cases, the use of the exact analytical solution is recommended in order to obtain more accurate and more conservative values of stresses. (Author)

[en] The particular method of Dynamic Relaxation which has been developed at IEA is very simple. The structure is divided into blocks and the displacements are repeatedly calculated from the stresses, and vice-versa, using simple finite difference equations. There are some important advantages when dealing with non-linear structures. Because of the iteractions, non-linear stress-strain relations can be applied very easily and accurately. Also, because there is complete freedom to change the boundaries during the interactions, cracks can be correctly represented, and will grow compatibly with the structure and the loads. The above programs originally written for static loads , can be applied directly to evaluate the effects of dynamic loading. The same interactions which are being used to analyse the dynamic behaviour of the structure, can be used simultaneously to input the changing loads. Some results have already been published for a prestressed concrete cylindrical vessel with an internal explosion. Further work is being carried on with much simpler non-linear structures for which classical analysis can give some approximate solutions, in order to validate the method. (Author)