2022年9月27日
Review of data assimilation using prompt neutron decay constant
5th International Workshop on Nuclear Data Covariances (CW2022)
- ,
- 開催年月日
- 2022年9月26日 - 2022年9月30日
- 記述言語
- 英語
- 会議種別
- 口頭発表(招待・特別)
- 主催者
- Institute of Innovative Research, Tokyo Institute of Technology
- 開催地
- Online
- 国・地域
- 日本
This presentation will review our recent research about the applicability of the prompt neutron decay constant α to the data assimilation in order to reduce the nuclear data-induced uncertainty of the neutron multiplication factor keff for a target system.
In the field of reactor physics, the data assimilation has been effectively utilized to improve the accuracy and precision of numerically predicted core characteristics parameters and to update a priori nuclear data that are used as input parameters in the core analysis. For the data assimilation using the integral experiments, critical experimental results collected in databases such as the International Criticality Safety Benchmark Evaluation Project (ICSBEP) are often utilized. For the following reasons, we focus on the utilization of other integral experiments for the data assimilation process: (1) Construction of a new facility for critical experiments is not easy; (2) Because a typical experimental system (e.g., geometry and combination of fuel and other materials) for the critical experiment tends to be complicated, a simpler experimental system is desirable to effectively update the a posteriori target data based on the data assimilation technique.
One of the alternative integral experiments is the prompt neutron decay constant α, which corresponds to a time constant for the fundamental mode of exponential decay of prompt neutron flux. Even not only for the subcritical system with fissile nuclides but also for the non-neutron multiplication system without any fissile nuclide, the α value can be directly measured using the pulsed neutron source method or reactor noise analysis method (e.g., Rossi-α or Feynman-α methods).
As the improvement of the measurement technique for the fundamental mode of α value, we clarified that the robust estimation of the α value can be achieved by applying dynamic mode decomposition to the time series data (e.g., time histograms of the pulsed neutron and Rossi-α methods) measured by the multiple neutron detectors.
Furthermore, in order to accomplish the data assimilation using the α value, we developed the efficient sensitivity analysis of α based on the first-order perturbation theory. In this presentation, we will present application examples of the data assimilation of α using the sensitivity coefficients and covariance matrix of nuclear data.
In the field of reactor physics, the data assimilation has been effectively utilized to improve the accuracy and precision of numerically predicted core characteristics parameters and to update a priori nuclear data that are used as input parameters in the core analysis. For the data assimilation using the integral experiments, critical experimental results collected in databases such as the International Criticality Safety Benchmark Evaluation Project (ICSBEP) are often utilized. For the following reasons, we focus on the utilization of other integral experiments for the data assimilation process: (1) Construction of a new facility for critical experiments is not easy; (2) Because a typical experimental system (e.g., geometry and combination of fuel and other materials) for the critical experiment tends to be complicated, a simpler experimental system is desirable to effectively update the a posteriori target data based on the data assimilation technique.
One of the alternative integral experiments is the prompt neutron decay constant α, which corresponds to a time constant for the fundamental mode of exponential decay of prompt neutron flux. Even not only for the subcritical system with fissile nuclides but also for the non-neutron multiplication system without any fissile nuclide, the α value can be directly measured using the pulsed neutron source method or reactor noise analysis method (e.g., Rossi-α or Feynman-α methods).
As the improvement of the measurement technique for the fundamental mode of α value, we clarified that the robust estimation of the α value can be achieved by applying dynamic mode decomposition to the time series data (e.g., time histograms of the pulsed neutron and Rossi-α methods) measured by the multiple neutron detectors.
Furthermore, in order to accomplish the data assimilation using the α value, we developed the efficient sensitivity analysis of α based on the first-order perturbation theory. In this presentation, we will present application examples of the data assimilation of α using the sensitivity coefficients and covariance matrix of nuclear data.
- リンク情報
-
- 共同研究・競争的資金等の研究課題
- 革新炉の解析精度向上:有効部分空間法を用いた高精度かつロバストな断面積調整
- 共同研究・競争的資金等の研究課題
- 詳細不明な体系における未臨界測定技術の開発
- 共同研究・競争的資金等の研究課題
- 臨界実験に頼らない臨界安全性の予測精度向上:未臨界実験を活用したデータ同化手法
- URL
- http://www.zc.iir.titech.ac.jp/~chiba/CW2022/index.html