Search

Acquiring instrumentation signals generated from nuclear power plants (NPPs) is essential to maintain nuclear reactor integrity or to mitigate an abnormal state under normal operating conditions or severe accident circumstances. However, various safety-critical instrumentation signals from NPPs cannot be accurately measured on account of instrument degradation or failure under severe accident circumstances. Reactor vessel (RV) water level, which is an accident monitoring variable directly related to reactor cooling and prevention of core exposure, was predicted by applying a few signals to deep neural networks (DNNs) during severe accidents in NPPs. Signal data were obtained by simulating the postulated loss-of-coolant accidents at hot- and cold-legs, and steam generator tube rupture using modular accident analysis program code as actual NPP accidents rarely happen. To optimize the DNN model for RV water level prediction, a genetic algorithm was used to select the numbers of hidden layers and nodes. The proposed DNN model had a small root mean square error for RV water level prediction, and performed better than the cascaded fuzzy neural network model of the previous study. Consequently, the DNN model is considered to perform well enough to provide supporting information on the RV water level to operators. Keywords: Deep neural networks, Genetic algorithm, Reactor vessel water level, Safety-critical instrumentation signals, Severe accident

A serious threat to the integrity of the reactor core, reactor coolant system, or containment is incurred if proper and essential actions to mitigate accidents cannot be taken owing to insufficient information about the internal states of the nuclear power plant (NPP). Therefore, this study was carried out to develop a model capable of mitigating the risk of severe accidents by accurately predicting the internal states of an NPP containment. A deep fuzzy neural network (DFNN) is a method in which syllogistic fuzzy reasoning is relatively efficient and inference capability is enhanced. In this study, the internal states of an NPP containment, hydrogen concentration and pressure, are predicted using a rule-dropout DFNN, as little NPP information is available under the circumstances of severe accidents. In addition, the performance of the proposed rule-dropout DFNN model is compared with that of other fuzzy neural network variations to verify the enhancement in the accuracy of the DFNN. The developed rule-dropout DFNN model is expected to be capable of providing accident monitoring information in advance for accident mitigation, as its prediction error for the hydrogen concentration and pressure in the containment is low.

Fukushima accident was worse by instrument inability. Eventually, the accident was not mitigated and keeping the integrity of reactor was failed since the operators was not able to quickly understand the situation and take necessary actions. Therefore, in this study, reactor vessel (RV) water level considered as one of the parameters to keep the integrity of reactor is predicted in loss of coolant accident (LOCA) situation using the deep neural network (DNN) method. This is in an effort to provide supporting information under the severe circumstance. The simulation data obtained from modular accident analysis program (MAAP) are applied to the DNN method to check the prediction performance of the RV water level. The prediction performance of RV water level using the proposed DNN model is presented as root mean square error (RMSE). Although the data of several circumstances among a variety of LOCAs are applied, good prediction performance is shown using the proposed DNN method.

Providing information about the leak flow rate caused by a loss-of-coolant accident (LOCA) to nuclear power plant (NPP) operation personnel is a key to the management and mitigation of severe post-LOCA circumstances at NPPs where active safety injection systems do not actuate. The leak flow rate is a function of break size, differential pressure (i.e., difference between internal and external reactor vessel pressure), temperature, and so on. In this study, the break position and size were first identified and predicted, and then, the leak flow rate was predicted using a fuzzy neural network (FNN). The FNN was developed using training data and validated using independent test data. The data were generated from simulations of the optimized power reactor 1000 (OPR1000) using MAAP4 code. The data for training the FNN model were selected among the acquired data using the subtractive clustering method, and FNN performance was improved. The developed FNN model was sufficiently accurate to be used for predicting leak flow rate, which is useful information for managing severe post-LOCA situations.

The reactor protection system (RPS) is designed to minimize a propagation of the abnormal or accident conditions of nuclear power plants. We have developed a KINCS-RPS prototype by using POSAFE PLC. The main design feature of the KNICS RPS is to satisfy the APR 1400 design requirements and to improve the availability and maintainability. The application software and hardware of the RPS are being developed according to the safety critical software life cycle and the safety H/W criteria. Especially, the software requirement specification of the RPS application software is written in a formal language for a correct software verification and validation. Also, the KNICS RPS has undergone an equipment qualification (EQ) test. The RPS prototype satisfied the nuclear EQ criteria except for some minor points. It is expected that the digital reactor protection system being developed through the 2nd phase of the KNICS project will be able to be applied to the existing NPPs for a replacement or to a new nuclear power plant.

In this paper, the design concept of the smart type motor operator valve for nuclear power plant was described. The development objective of the smart valve is to achieve superior accuracy, long-term reliability, and ease of use. In this reasons, developed smart valve has fieldbus communication such as deviceNet and Profibus-DP, auto-tuning PID controller, self-diagnostics, and on-line calibration capabilities. And also, to achieve pressure, temperature, and flow control with internal PID controller, the pressure sensor and transmitter were included in this valve. And, temperature and flow signal acquisition port was prepared. The developed smart valve will be performed equipment qualification test such as environment, EMI/EMC, and vibration in Korea Test Lab. And, the valve performance is tested in a test loop which is located in Seoul National University Lab. To apply nuclear power plant, the software is being developed according to software life cycle. The developed software is verified by independent software V and V team. It is expected that the smart valve can be applied to an existing NPPs for replacing or to a new nuclear power plants. The design and fabrication of smart valve is now being processed.

Valves integrated with a smart positioner have been developed at many sites. We have been developing a smart valve with a sensor for a nuclear power plant, which is performed a process sensing function and a process control function automatically. The smart valve has the functions as follows; self-sensing, auto-tuning, multi-communication, self-diagnosing, etc. The smart valve should satisfy the nuclear criteria for a nuclear application. In this paper, we developed the requirements and criteria for the valve, which contain the requirements and criteria for the valve body, the sensing part, the control part, and the communication part. We produced a prototype of the smart valve which could control the pressure on the basis of the requirements. Additionally, we established the test criteria such as the environment test, the seismic test, and the EMI test. On the basis of the requirements and criteria, we are developing a prototype of the gate type-smart valve which contains the valve body, the control circuit, the communication circuit, and the application software. The smart valve can reduce the installation costs and the maintenance costs of nuclear power plants.