Font Size: a A A

Research On The Key Methods Of Evaluation, Processing And Application Of Important Nuclear Data For Reactor And Code Development

Posted on:2016-09-29Degree:DoctorType:Dissertation
Country:ChinaCandidate:J K YuFull Text:PDF
GTID:1222330503456169Subject:Nuclear Science and Technology
Abstract/Summary:PDF Full Text Request
The independent reactor core physical d esign and simulation for advanced nuclear energy system relies on the foundation of nuclear data with high accuracy. It is important and necessary for some processing procedures to be utilized in the conversion from raw nuclear data in experimental nuclear data base into the ACE-formatted point-wise nuclear data libraries, which can be applied directly in Monte Carlo based neutronics transport calculations by RMC, MCNP and some other similar codes. The domestic researches on key algorithms are limited to the copyright and licenses of these codes developed by foreign institutes, including resonance parameters evaluations, cross sections reconstruction in resolved resonance range, Doppler broadening cross sections, and the unresolved probability tables’ generation and so on.This paper is focus on the key methodologies in nuclear data evaluation, processing and application, aiming at developing the corresponding codes system with independent intellectual property rights, which content in this study is listed as follows:1) Study the nonlinear least squares fitting method, and the latest R-Matrix approximation formula, a code named RRPE is developed to evaluate the resonance parameters in resolved resonance range.2) In view of the resonance reconstruction module-CSP in self-developing reactor nuclear data processing code-RXSP, the defeciency of the convergence criterion in CSP module is handled to expande the R-Matrix Limited based resolved resonance cross section reconstruction.3) Research the approaches of unresolved probability tables’ generation, and develop a PURC module in the framework of RXSP code, which solves the generation efficiency in the traditional but classic algorithm with the assistant of sorting techniques optimization and Open MP-based parallel acceleration in it.4) The study on inner-coupling for Fast Doppler broadening cross secions with temperature dependence and RMC code is carried out, and thus RMC code has the capability of in-line Doppler broadening cross secion during the Monte Carlo based neutronics transport, and thermal cross section interpolation is included as well.Some conclusions are drawn from these above researches, which are listed as follows:In basis of Levenberg-Marquardt method used in data fitting procedure, RRPE code is developed to have the capability of resonance paramters evaluation as accurate as SAMMY code; The improved CSP module can process R-Matrix Limited-formatted resonance parameters from evaluated nuclear data library to produce the point-wise cross sections at zero temperature with the high accuracy as much as NJOY(2012) code; After the optimized sorting algorithm and parallel acceleration is applied, PURC module can generate probability tables by using “Ladder Sampling” approache with the accuracy as much as NJOY code, but much higher efficiency of it;RMC code is improved by inner coupling of Fast Doppler broadening and thermal cross sections interpolations with it, which now has the capbability of rapidly processing cross sections at the problem-depedent temperatures, and has a very good prospective application in neutronics-thermal hydraulic coupling calculations with many nuclides and many temperatures feedback.
Keywords/Search Tags:resonance evaluation, resonance reconstruction, probability tables, in-line Doppler broadening
PDF Full Text Request
Related items