Uncertainties in the interpretation of computer results on the behaviour of LMFBR fuel pins

Since about ten years a large number of computer programs has been developed for the simulation of the behaviour of fuel pins in a LMFBR. The applications for these programs are manifold. They can be used to explain the results of the post-irradiation examination of test pins and improve the understanding of the irradiation behaviour of these pins. They give valuable assistance in the preparation and optimal design of irradiation experiments. They are indispensable in the detailed analysis of the influence of individual material phenomena. Moreover, they can be used to analyse the influence of individual design parameters on the fuel pin behaviour with the aim to optimize the pin performance and establish appropriate specifications. Further on, they can be used to predict the behaviour of fuel pins in nuclear power reactors under the scheduled operation and analyse the consequences of off-normal operation with the aim to find out critical conditions, maybe for reasons of safety or the optimization of operation. The computer program can only serve its purpose if several conditions are fulfilled. Sufficient knowledge of all the data necessary for its application is one of these conditions. The experience gained with many computer runs resulted in some critical considerations. It follows that the knowledge of those data is still insufficient in many cases and large uncertainties have to be accepted in the interpretation of computed results. Some sources of uncertainties are demonstrated. Their influence on the computed results is discussed. It is clear that, due to the complexity of the problem, only some selected examples can be treated. The computer calculation results presented have been provided by the computer programs SATURN and TEXZ respectively