An Intelligent Operator Support System for Early State Emergency Operations in Nuclear Power Plants

摘要

Human error is one of the major contributors to the probability of accidents in nuclear power plants. It is important to reduce human error probability in order to secure safety. At present, a computerized procedure with digital MCR has appeared. These procedures have contributed to reducing human error probability. In the early state emergency operation, EIGS can improve the current situation by reducing the burden on the operator. This is because, despite the fact that there is a small amount of work due to ESFAS, all CSFs must be checked. To improve the problem, this paper proposed EIGS. EIGS checks the status of each CSF at the same time and suggests the necessary tasks in accordance with the importance and the diagnostic result to the operator depending on the current state of NPP. In other words, EIGS is an intelligent operating support system that replaces early state emergency operating procedures. For this system, securing the reliability of the instrumentation is important. Hence, it is a prerequisite for this support system to verify the instrumentation. To handle this, validation will be performed through machine learning as a future work.