1988 Fiscal Year Final Research Report Summary
Development of On-line Criticality Surve:Ilance system for Nuclear Fuel Re-processing Facility
| Project/Area Number |
62580183
|
|---|
| Research Category |
Grant-in-Aid for General Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
Nuclear engineering
|
|---|
| Research Institution | Osaka University |
|---|
Principal Investigator |
YAMADA Sumasu Associate Professor, Dept. of Nuclear Engi. Osaka University, 工学部, 助教授 (00029133)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
IIDA Toshiyuki Assistant Prof. Dept. of Nucl. Eng. Osaka University, 工学部, 助手 (60115988)
SUMITA Kenji Professor, Dept. of Nucl. Engi. Osaka University, 工学部, 教授 (80028977)
|
|---|
| Project Period (FY) |
1987 – 1988
|
| Keywords | ARMA process / adaptive filter / RELS method / on-line parameter estimation / subcriticality / criticality surveillance / break frequency / 残差の分散行列 |
|---|
| Research Abstract |
This project has studied the applicability of the adaptive filter for an inexpensive online criticality surveillance system. First, it has been shown that the time series data of the fluctuation of the output signal from a neutron detector is a Auto-Regressive Moving Average(ARMA) process under the assumption that the subcritical system can be described in the framework of a point reactor kinetics approximation and hence the subcriticality can be obtained from the break frequency(Fc) of the ARMA model identified with an adaptive filter applied for the time series data. Then, computer simulations and analyses of the data from a reactor (UTR-KINKI) have been carried out in order to find the most appropriate conditions for signal reprocessing and the most appropriate algorithm for parameter estimation. The followings are the results: 1. The sampling frequency for A/D conversion should be 20Fc-200Fc. 2. A low pass filter of the order more than 4 with the break frequency at Fs/2 should be u
… More
sed for avoiding the aliasing. 3. A high pass filter of the order more than 6 with the break frequency between 0.1Fc-0.01Fc for reduction of the DC and lower frequency component. 4. The Recursive Extended Least Squares Method has been best for estimating the ARMA(1,1) model parameters of the subcritical system if appropriate initial conditions aree set. 5. For quick response for parameter change, a new approach has been proposed and it has been shown that a step change in reactivity can be detected using about 1000 data which correspond to 0.05sec if Fs=20KHz. These results indicate that the proposed approach is effective for detection of a sudden subcriticality change in the system. Since this approach need not to prepare a special neutron detector or neutron source, the criticality surveillance system based on this approach is expected to be simple and inexpensive. However, furhter studies are needed on space dependency and the problem of the neutron source strength for practical application. Less
|
