Siemens Corporation, Corporate Technology, 755 College Rd. E., Princeton, NJ 08540, USA.
Ultrasonics. 2014 Feb;54(2):516-25. doi: 10.1016/j.ultras.2013.07.019. Epub 2013 Aug 8.
Damage diagnosis for turbine rotors plays an essential role in power plant management. Ultrasonic non-destructive examinations (NDEs) have increasingly been utilized as an effective tool to provide comprehensive information for damage diagnosis. This study presents a general methodology of damage diagnosis for turbine rotors using three-dimensional adaptive ultrasonic NDE data reconstruction techniques. Volume reconstruction algorithms and data fusion schemes are proposed to map raw ultrasonic NDE data back to the structural model of the object being examined. The reconstructed volume is used for automatic damage identification and quantification using region-growing algorithms and the method of distance-gain-size. Key reconstruction parameters are discussed and suggested based on industrial experiences. A software tool called AutoNDE Rotor is developed to automate the overall analysis workflow. Effectiveness of the proposed methods and AutoNDE Rotor are explored using realistic ultrasonic NDE data.
涡轮转子的损伤诊断在电厂管理中起着至关重要的作用。超声无损检测（NDE）越来越多地被用作提供损伤诊断的综合信息的有效工具。本研究提出了一种利用三维自适应超声 NDE 数据重建技术进行涡轮转子损伤诊断的一般方法。提出了体积重建算法和数据融合方案，以便将原始超声 NDE 数据映射回被检测对象的结构模型。使用区域生长算法和距离-增益-尺寸法，利用重建体积自动进行损伤识别和量化。根据工业经验讨论并提出了关键的重建参数。开发了一个名为 AutoNDE Rotor 的软件工具，用于自动化整个分析工作流程。利用真实的超声 NDE 数据，探讨了所提出的方法和 AutoNDE Rotor 的有效性。
