| 摘要 |
PURPOSE:To easily judge the presence or absence of any abnormality in plural stages of isotope separating cascades by fitting the analytical value of a product gas to an autoregression model with an abnormality diagnosis device at the time of concentrating the specific components in a raw material gas with said cacades. CONSTITUTION:This is a device of precisely diagnosing the inclusion of foreign components such as air at the time of passing a raw material gas contg. U235 through multiple cascades thereby separating isotope and concentrating U235. The product gas concd. by the cascades is analyzed with an analyzer and the analyzed value 21 thereof is put into an abnormality diagnosis device 12 and is stored in a time series storage device 121. By using the stored value 211 thereof, a covariance calculator 122 calculates a covariance value 212, and from this value, an autoregression model coefft. calculator 123 calculates an autoregression model coefft. 213. Whether this value is within the threshold value of the autoregression model coefft. or off this value is decided by a comparator 124, and is outputted as an abnormality display signal 22, whereby the abnormality of the gas treatment in the cascades is decided precisely. |
