Authors & Affiliations
Golub Ye.V.
A.I. Leypunsky Institute for Physics and Power Engineering, Obninsk, Russia
Golub Ye.V. – Scientific Researcher, A.I. Leypunsky Institute for Physics and Power Engineering. Contacts: 1, pl. Bondarenko, Obninsk, Kaluga reg, Russia, 249033. Tel.: (484) 399-49-75, e-mail:
Abstract
Accuracy of calculation results is restricted by dispersion of input parameters. Usually algorithms design calculations are not linear functions of the inputs and results of the calculation presented as a nominal value with a symmetrical field of dispersion, or a one-way field dissipation towards a possible dangerous overheating of the element (in stock), are useful in particular cases and in the early stages of design. In the general case result scattering is not symmetrical because of the complex nonlinear relationships between input and output parameters.
To evaluate the results of validation analysis (with sufficient measurement accuracy), or dispersion of more precise calculation result is necessary an valuation of insufficiency of calculation accuracy by scattering of input parameters in the fields of tolerance.
The aim of the work was to evaluate the accuracy of the calculation and statistical characteristics of result dispersion caused only a random scattering of input parameters.
A revised algorithm for calculation of asymmetry, and displacement of the nominal value was received for a multiparameter generalized function of random arguments.
The method was applied to estimate the influence of manufacturing tolerances of the core elements on uncertainty and the possible displacement of the mean value of heating in a standard triangular heat exchange channel (relative to the nominal calculation result) of the reactor – stand core.
The resulting set of statistical characteristics of the deviation of the results of calculation of the nominal values due to the technological uncertainties of the input parameters is tentatively indicates a possible limit of accuracy of the calculation and can serve as a criterion of matching the thermal-hydraulic calculation and experiment. In the case of significant discrepancy we can assume significant methodological error or significant influence of lack of precision in the measurement or influence of significant operating errors.
Keywords
core elements, process tolerance, the statistical characteristics of dispersion, heating, cross-section, coolant, validation
Article Text (PDF, in Russian)
UDC 621.039.51