Slug is produced when the roll waves on the liquid surface undergo constructive interference and the resulting wave height increases beyond the critical height. Slug flow is associated with large aeration that develops a sudden surge in pressure and associated mechanical impact on pipe wall and bends. Such phenomena lead to flow accelerated corrosion (FAC) which results in degradation of pipe thickness with time resulting in cracks from where the gas and liquid leaks. Hence, to reduce the chances of failure in industrial pipes that are mostly opaque (flow pattern cannot be visualized) and associated hazard in nuclear power plant industries, it is important to develop mechanisms for detecting the occurrence of slug flow. The present research proposes a methodology for recognizing the onset of slug in opaque industrial pipes using the pressure signals measured from the pipeline. Recurrence analysis of pressure signal establishes different kinds of recurrence plots for stratified-wavy flow and slug flow regime. Also, the transition can be identified using the recurrence quantification parameters. The quantified parameters like recurrence rate, determinism, and entropy show sharp jump in value at the onset of slug from wavy regime.

当液体表面的滚动波受到相长干涉并且由此产生的波高超过临界高度时，就会产生段塞。团状流与大量曝气有关，曝气会导致压力突然激增，并对管壁和弯头产生相关的机械冲击。这种现象会导致流动加速腐蚀 (FAC)，这会导致管道厚度随时间退化，从而导致气体和液体泄漏处出现裂纹。因此，为了减少在核电厂工业中大部分不透明（无法可视化流动模式）的工业管道的故障机会和相关危险，开发用于检测团状流发生的机制非常重要。本研究提出了一种使用从管道测得的压力信号识别不透明工业管道中段塞开始的方法。压力信号的递归分析为分层波浪流和段塞流状态建立了不同类型的递归图。此外，可以使用重复量化参数来识别转变。量化参数（如复发率、确定性和熵）在波浪状态的段塞开始时显示出值的急剧跳跃。