Computer calculations concerning a nuclear waste repository will require a high accuracy in the predictions of the deformations and the stability of the repository over very long times and therefore a hight level of sophistication in geomechanical modeling. For the Waste Isolation Pilot Plant (WIPP) project near Carlsbad, New Mexico a benchmark II problem was stated in order to cornpare various computer codes with respect to their capability for calculating the thermal mechanical response of a hypothetical drift configuration for nuclear waste experiments and storage demonstration. Structural computations were made with ANSALT, a new FEM-code, especially developed for the analysis of thermomechanical processes related to the storage of heat producing radioactive wastes in rock salt. The computed results are illustrated and discussed.
Stabilitatsberechungen fuer ein Endlagerbergwerk fur nukleare Abfalle erfordern geomechanische Modelle und Rechenverfahren, die es erlauben, zuverlassige Vorhersagen ueber thermomechanische Auswirkungen der Einlagerung auch ueber sehr lange Zeitraume zu machen. Fuer das "Waste Isaolation Pilot Plant (WIPP)" project in New Mexico, U.S.A., wurden Vergleichsberechnungen mit verschiedenen Rechenprogrammen durchgefuehrt, um die Möglichkeiten, die Genauigkeit und Leistungsfahigkeit der verschiedenen Rechenprogramme zu untersuchen. Im Rahmen dieser Studie wurden Berechnungen auch mit dem von der BGR in Zusammenarbeit mit CD nach neuesten Erkenntnissen der Salzmechanik und der Rechentechnik entwickelten FEM-Programm AN SALT durchgefuehrt. Die Berechnungsergebnisse fuer eine hypothetische Lagerstrecke fuer warmeentwickelnden radioaktiven Abfall unter Zugrundelegung einer differenzierten geologischen Situation werden dargestellt und erlautert.
Les calculs de stabilite pour des installations souterraines d''enfouissement requièrent des modèles geomecaniques aussi que des procedes de calculs permettant de faire, meme sur de très lonques periodes, des pronostics sûrs en ce qui concerne les consequences thermomechaniques d''enfouissement. Pour Le "Waste IsoIation Pilot Plant (WIPP) ", Projet a New Mexico, U. S.A., ont ete effectues des calculs de comparaison, avec differents programmes de calcul afin d''etudier les possibilites, l''exactitude et le rentement des differents programmes de calculs. C''est dans le cadre de cette tude que les calculs ont ete realises avec egalement le FEMProgramm ANSALT, mis au point par le BGR en collaboration avec le CD d''apres les theoris les plus recentes de la mecanique du sel et de la technique de calcul. On y decrit et y expliquer les resultats de calculs concernant une qalerie pour des dechets radioactifs thermogenes en prenant pour base une situation geologique differenciee.
The design of a nuclear waste repository is based on a multiple barrier storage concept taking into account a system of sequential or interconnected natural and technical barriers (Albrecht et al 1980) The natural barrier provided by a rock salt formation could act as an encapsulated system due to the very low permeability and the high ductility of rock salt contrary to other host rock types. However, the assessment of the integrity of the rock salt mass over very long times among other things requires a thorough understanding of the long-term thermomechanical process taking place due to the decay heat of high level waste. In the framework of the US!FRG bilateral agreement on cooperation in research and development for nuclear waste management the BGR, Hannover, in cooperation with CD, Hamburg, had the chance to participate in the second benchmark problem for.the Waste Isolation Pilot Plant(WIPP) initiated by Sandia National Laboratories, Albuquerque NM in order to compare different thermal-structural computational codes. The purpose of the WIPP project is to investigate the feasibility of storing defence transuranic waste in the bedded salt formation of Southeastern New Mexico and to perform experiments with defence high level waste (Hunter 1979). Structural calculations are used for reliable and detailed predictions for the WIPP experiments and storage demonstration as well as for extrapolating the longterm behavior of the repository to hundreds or thousands of years needed for nuclear waste isolation.
It is obvious that geomechanical modeling of an underground structure can only reach a certain level of accuracy because the rock mass itself will always remain unknown up to a certain extend. Beyond that all input data for a computer calculation like the material properties, the loading and the boundary conditions may vary within the investigated system, whereby the exact variability of those values will not be known. The engineer''s approach to overcome this general difficulties is a continuous improvement to the model appropriate to the improved knowledge of the input data within an integrated cycle. The features of this cycle are the establishment of a constitutive modei for the material behavior, the quantification of site relevant input data, the prediction and monitoring of the structural response and from that the necessary feedback to check the validity of the constitutive law, input data and modeling process in an iterative manner.
