Effect Analysis of Faults in Digital I&C Systems of Nuclear Power Plants

A reliability analysis of digital instrumentation and control (I&C) systems in nuclear power plants has been introduced as one of the important elements of a probabilistic safety assessment because of the unique characteristics of digital I&C systems. Digital I&C systems have various features distinguishable from those of analog I&C systems such as software and fault-tolerant techniques. In this work, the faults in a digital I&C system were analyzed and a model for representing the effects of the faults was developed. First, the effects of the faults in a system were analyzed using fault injection experiments. A software-implemented fault injection technique in which faults can be injected into the memory was used based on the assumption that all faults in a system are reflected in the faults in the memory. In the experiments, the effect of a fault on the system output was observed. In addition, the success or failure in detecting the fault by fault-tolerant functions included in the system was identified. Second, a fault tree model for representing that a fault is propagated to the system output was developed. With the model, it can be identified how a fault is propagated to the output or why a fault is not detected by fault-tolerant techniques. Based on the analysis results of the proposed method, it is possible to not only evaluate the system reliability but also identify weak points of fault-tolerant techniques by identifying undetected faults. The results can be reflected in the designs to improve the capability of fault-tolerant techniques. Keyword: Digital I&C, fault-tolerant technique, fault injection