No abstract available

[en] A new method for multivariate autoregressive modeling of a vector time series is described, which has an advantage in guaranteeing a less-biased estimation in a least-squares sense by ensuring orthogonality between the components of a residual vector. An expression for a multivariate autoregressive model with a zero-lag coefficient matrix is derived theoretically and utilized for a model based on a weighted least-squares procedure. Model parameters such as regression matrices and residual matrix are determined recursively for each order by using the Light Water Reactor (LWR) algorithm. It is shown that the present model differs from the ordinary one in estimating open-loop properties, though these models are equivalent for closed-loop estimates. (author)

No abstract available

[en] Several methods developed in different fields have been applied to reactor noise analysis; the probability distribution method in the probability theory, the Langevin technique in the statistical physics, the time series analysis in the statistics, and the Kalman filter in the control theory. For reactor diagnosis or reactor safety, it is important to understand relations among them and their merits. To make full use of signals included in reactor noise, the above mentioned methods and techniques are discussed systematically from the physical viewpoint of the contraction of information and coarse-graining in time and space. Properties of noise sources and spatial dependence of model are also discussed. (author)

[en] A generalization of the signal transmission path method for the ARMA model has been performed with the aim of evaluating reactor noise diagnostic measurements. (author)

[en] This paper is concerned with a study of the fine structure of reactor noise. The power spectrum of reactor noise is represented in the frequency-time space using procedures of Bayesian AR-modelling. Results from the application to a number of power reactor noise data retrieve information on tomographical structures of peculiar reactor noise as well as apparently stationary reactor noise. Some comments are referred to the introduction of this idea to the STP analysis. (author)

[en] The present states and development prospects of reactor noise analysis and earlier period failure diagnoses technique of power reactor based on the 5th specialists meeting on reactor noise are briefly introduced

No abstract available

[en] A minicomputer-based on-line diagnosis algorithm was developed and its verification test was performed. This algorithm identifies the plant status by applying a statistical pattern recognition technique to various multidimensional noise pattern vectors. The test was performed on-line with a simple analog simulator which was modeled on the experimental fast reactor JOYO. Seven pattern vectors were used to diagnose three categories of abnormality including abnormal control rod vibration. Reference patterns were learned under only one degree of severeness for each abnormality category. Test results showed the effectiveness of the proposed algorithm. This indicates that the number of reference patterns to be prepared can be significantly decreased. (author)

[en] In this paper a new recursive algorithm for estimating the parameters of the Autoregressive Moving Average (ARMA) model from measured data is presented. The Yule-Walker equations for the case of the ARMA model are derived from the ARMA equation with innovations. The recursive algorithm is based on choosing an appropriate form of the operator functions and suitable representation of the (n + 1)-th order operator functions according to those with lower order. Two cases, when the order of the AR part is equal to that of the MA part, and the general case, were considered. (Author)