Setsuo Arita

Cable Identification Method for Power Plants

Before cables can be replaced in power plants, they must be properly identified. We have developed a new cable identification method that does not need contact with the conductor of the cable before replacement. The method uses two techniques. The first is based on inductive coupling by a high-frequency search signal that is injected into the cable. The second is based on the change in the cable signal propagation characteristic, which occurs by applying or by removing a magnetic core on the cable. We carried out an experiment in an environment like that of actual cable laying conditions to confirm the effectiveness of our developed method and found that the cables could be identified accurately.

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...

Development of Inherently Safe Technologies for Large Scale BWRs: (5) Operation Support System for Plant Accidents

When accident events are caused by a large-scale natural disaster, conditions beyond those at the plant site may affect the accident. As well, quick diagnosis and recognition of damaged equipment are necessary. We have been developing inherently safe technologies for boiling water reactor (BWR) plants in response to these. An operation support system for plant accident events is one of these technologies. Our operation support system identifies accident events and predicts the progression of plant behavior.The system consists of three main functions: sensor integrity diagnosis, accident event identification, and plant simulation functions.The sensor integrity diagnosis function diagnoses whether sensor signals have maintained their integrity by correlating redundant sensors with the plant design information.The accident event identification function extracts a few of candidate accident events using alarm and normal sensor signals received by the sensor integrity diagnosis function. The scale and position of the accident event are determined by comparing plant simulation results with normal sensor signals.The plant simulation function uses a detailed three-dimensional model of the nuclear reactor and plant. This simulation can predict future plant behavior on the basis of identified accident events.This proposed operation support system provides available results of accident event identification and plant condition prediction to plant operators. This system will reduce the occurrence of false identifications of accident events and human errors of operators.Copyright

Plant Simulation System for Developing ABWR Automatic Power Regulator System

We developed a plant simulation code system to design and verify an automatic power regulator (APR) system for an advanced boiling water reactor (ABWR). The point kinetic plant simulator AUTSIMU is an off-line and real-time online plant simulation code that can simulate the start-up process from a subcritical condition of cold water to a 100% rated power condition as well as the reactor shutdown process. The three-dimensional effect of neutron detector signals in the core was simulated by combining predetermined three-dimensional analysis and point kinetic analysis results. The AUTSIMU code was tested using ABWR start-up experimental data. The calculated results were in good agreement with the measured ones. Less than one minute was needed in the off-line mode to simulate a ten-hour transient with an ordinary personal computer. We confirmed that simple AUTSIMU models could effectively simulate ABWR start-up behaviors. The APR hardware was effectively verified in the online mode through factory tests before being shipped to the nuclear site.

Development of Instrumentation System for Severe Accidents

The accident at Tokyo Electric Power Company’s Fukushima Dai-ichi Nuclear Power Plant (TF-1 accident) caused severe situations and resulted in a difficulty to measure important parameters to monitor plant conditions. Therefore, we have studied to select the important parameters (SA parameters) that should be monitored at the severe accident and are developing the Severe Accident Instrumentations and Monitoring Systems (SA-keisou) that could measure the parameters in severe accident conditions. (SA-keisou in Japanese means “severe accident instrumentation and monitoring systems). In this study, the boiled water reactor (BWR) is mainly explained.This study is a part of the results of the collaborative project by domestic electric power companies and plant manufacturers that is carried out as the Safety Enhancement for LWRs program by Agency for Natural Resources and Energy.Copyright

Reliability Enhancement of Reactor Protection System for BW R Plants.

The realization method is presented for a logic configuration of the reactor protection system (RPS) for BWRs which enhances their reliability.Higher reliability requires a majority voting logic even when a single logic component has been bypassed. Furthermore the load driver of the RPS on the output side must actuate two separate operation objects. To fulfill the above conditions, a general realization of the 2-out-of-4 logic with a minimum hardware component is proposed. Then, the structure of the microprocessor-configured RPS employing this logic is presented.The reliability of the above RPS is evaluated by computations. The results show that the average frequency of improper operation is less than that with the conventional 1-out-of-2-twice logic under the conditions that the repair rate is 0.1 and the failure detection rate is above 92%. The average probability of failure to operate is reduced by two orders of magnitude under the above conditions.

Method for improving control accuracy of nuclear plant sub-loop controller.

Development of the sub-loop control system is an important step leading to full-automatic control of nuclear power plants. In this study, characteristics of sub-loop controller were investigated. Firstly, a program which simulates electrical and mechanical behavior of the valve controller was developed to evaluate control accuracy. Then, using this simulation program, error sensitivity analysis was done to get the major error sources for theo pen-loop control method.The analytical results showed that these sources were voltage change of the power supply and load friction torque change of the mechanism. Lastly, an improved control method was proposed and applied to the sub-loop controller to reduce the control error. Theoretical and experimental results confirmed that the control error for the new method was less than 0.2% for Iong periods such as plant life time.