Annals of Nuclear Science and Engineering最新文献

This paper is concerned with the radiological protection criteria used for setting limits to public radiation risks. It describes the formal basis of the criteria in current use with extensive reference to the recommendations used to formulate statutory limits. Attention is drawn to the limitations of the present concepts for use in estimating risks, and brief comments are made on likely future developments.

It is shown that the current concepts used for defining acceptable industrial augmentations to the radiation environment of the general public are derived from a formalized set of dose limits. These dose limits do not provide an index of risk, furthermore the practical criteria used to supplement them are intended only to ensure that doses received do not exceed the limits. The practical criteria do not provide a means of adequately estimating doses received or risks incurred.

Properly applied, with experience and professional judgement, the present criteria are useful in setting norms of good practice in design and operation. Because of inconsistencies in the dose limits, and the assumptions and approximations in estimating potential doses, any practical rules are necessarily imprecise, and a factor of safety is applied when it is necessary to be certain of working within statutory limits.

The numerical interpretation of concepts based on dose limits will undergo continuous development mainly by advances in knowledge of the pathways to man, human metabolism, and in the mathematical procedures used for deriving organ doses from this data. However, the quantification of risks to populations involves eliminating ambiguities in dose-risk relationships which are orders of magnitude greater than the error in dose estimations.

The integral over time of some local density is found. The method is based on solving stationary regular and adjoint equations. It is applied to a problem from time dependent reactor theory and tested by a numerical example.

The effective diffusion coefficient in a cylindrical cell is calculated from the Benoist formula by applying the first-flight collision probability method in one-velocity. Following his formula, the diffusion coefficient is determined by a component of the tilted flux. The anisotropy and the space variation of the tilted flux is taken into account by using the Legendre expansion. The expansion coefficients are determined by the generalized first-flight collision probability. At the cell boundary, a new reflection condition is derived for the generalized collision probability instead of the usual isotropic reflection boundary condition. From numerical examples, it is seen that values of the radial diffusion coefficient calculated from the present method agree with those from the exact method in a square cell. Further, when the first term of the Legendre expansion is included, rapid convergence of the radial diffusion coefficient can be seen.

The results obtained from measurements of thermocouple noise signals from a graphite moderated gas cooled reactor are presented.

An overall survey demonstrated that in the frequency range covered (up to 0·15 Hz), the reactor behaved as a closely coupled system with no significant phase differences between different parts of the core. Furthermore, the temperature noise power spectra could be divided into two distinct frequency regions in this range. In the lower part of the range the observed fluctuations were found to be due to overall reactor power fluctuations, and in the upper part to heat transfer coefficient variations.

A simple model of the noise processes in the reactor has been developed and verified by consideration of the transfer function between coolant and fuel temperatures.

The channel gas outlet thermocouple response, which was required for reactor fault studies, was measured by conventional rod oscillation techniques and noise analysis using both intrinsic fluctuations and refuelling transients. A comparison of the results showed that noise analysis, especially using refuelling transients, was capable of replacing the rod oscillation technique. This gave the advantages of not interfering with the normal operation of the reactor, reducing the experimental time and extending the frequency range of the measurements.