基于模糊距离的核电厂瞬态分段识别方法

Transient Identification in Nuclear Power Plants Based on Transient Division and Fuzzy Euclidean Distance

近年来发展的核电厂瞬态识别技术,可为操纵员提供处于发展阶段的故障信息,有助于了解核电厂状态并及时采取相应的操作动作,保证核电厂的安全运行。将瞬态过程曲线分为两段,前段利用聚类方法用于快速识别,后段利用提取的瞬态过程的特征进行更准确的识别。利用待识别瞬态与参考瞬态间的模糊距离描述二者的相近程度,可以消除噪声等扰动的影响,并得到更符合认知习惯的结果。利用高温气冷堆核电厂仿真机的故障数据验证瞬态识别方法的有效性。

The transient identification techniques were recently developed to alert the operators about the faults in their early stages, therefore corrective actions can be taken in time to keep the safety of nuclear power plants. In this paper, the transients were split into two parts： the first part is identified by clustering method, while the other is identified by comparing three features of signals. The similarity between the on-line and the reference transients was described by Fuzzy Euclidean distance, which conforms to human understanding habits. The method was verified by simulator data of Pebble Bed Modular High Temperature Gas-cooled Reactor （HTR-PM） with Tsinghua University. It is shown that the transients can be correctly and quickly identified.