Takaharu Fukuzaki

Evaluation Tests of Event Identification Method Using Neural Network at Kashiwazaki Kariwa Nuclear Power Station Unit No.4

An event identification method using a neural network has been evaluated in an on-line environment during the plant startup test at Unit No.4 Plant in the Kashiwazaki Kariwa Nuclear Power Station. In the method, the neural network identifies the event from the change pattern of analog data, such as reactor pressure signals, and then the result is confirmed or similar events are discriminated using digital data, such as valve open signals. Before the test the neural network is trained for the events causing a reactor scram by using analysis results. For the test the method is incorporated into a prototype of the alarm handling system which is connected to the plant facilities. Five kinds of analog data are acquired and eight sampled data from each, namely a total of 40 data, are input to the neural network after normalization. The results show that the load rejection, the turbine trip and the main steam isolation valve closure events are correctly identified from 9 kinds of subject events, regardless of th...

Optimization of boiling water reactor control rod patterns using linear search

A computer program for searching the optimal control rod pattern has been developed. The program is able to find a control rod pattern where the resulting power distribution is optimal in the sense that it is the closest to the desired power distribution, and it satisfies all operational constraints. The search procedure consists of iterative uses of two steps: sensitivity analyses of local power and thermal margins using a three-dimensional reactor simulator for a simplified prediction model; linear search for the optimal control rod pattern with the simplified model. The optimal control rod pattern is found along the direction where the performance index gradient is the steepest. This program has been verified to find the optimal control rod pattern through simulations using operational data from the Oyster Creek Reactor.

A Knowledge-Based System for Power Maneuver Planning for Boiling Water Reactors

This paper reports on a prototype knowledge-based system that assists plant site engineers in power maneuver planning for boiling water reactor (BWR) startup and load-following that has been developed. In the conventional method, engineers formulate these plans based on their own expertise and on core simulation programs that require long running times. To more quickly provide a suitable plan, the new system makes an initial plan and modifies it based on both expertise stored in a knowledge base and simplified programs using a one-point BWR core model. The time required for this system is {lt}30 min. The system was tested for 1100-MW (electric) BWRs. The quality of the plans generated by the system is almost the same as that of plans directly generated by experts. The test results show that the system is useful in preparing a suitable plan efficiently and quickly.

Knowledge-based system for core operation management of boiling water reactors

A knowledge-based system, which supports designers in the core operation management of boiling water reactors, has been developed. The system consists of (1) the knowledge base to store heuristics of experienced designers in the forms of rules, functions and frames: (2) the inference program with automatic deletion and generation of those data which have to be modified: and (3) rule refinement and editing support programs to find and correct those rules necessary to obtain a better inference result. Through the performance evaluation of the system, the number of rules for modification of data was reduced by 40% with the inference program, and designers with different experiences reached the same solution. As for the modification of rules in the knowledge base by the system manager, the time for searching and determining the value was decreased by 90%.<<ETX>>

Effective Use of Sensor Readings in On-Line Plant Monitoring and Its Application to Boiling Water Reactor Power Distribution Calculations

Effective use of the measured readings of sensors in on-line plant monitoring has been studied, based on an error theory. Both the measured reading and the calculated one, obtained by an analytical model of the plant, are treated as observed values, and the maximum-likelihood estimator is determined so as to minimize its mean-squared error. The difference between the estimator and the calculated reading is used to adapt the model to the current plant state and to increase accuracy of the calculated reading. The index of systematics, which indicates the mutual independence of the two observed values, has been evaluated to determine the step in the procedure where the above adaptation is to be inserted. The error-theory-based model adaptation procedure has been experimentally applied to boiling water reactor power distribution calculations, and its performance has been verified in simulation calculations at different core states and different numbers of in-core neutron monitors by evaluating the expected error of the calculated readings. Compared to the adaptation, which uses the measured readings instead of the estimator, the error is typically lowered by more than 2% and is less affected by the number of monitors.