Artificial intelligence (AI) assisted fatigue fracture recognition based on morphing and Fully Convolutional Networks

Fatigue fracture is one of the most common metallic component failure cases in manufacturing industries. The observation on fractography can provide the direct evidence for failure analysis. In this study, an image segmentation method based on Fully Convolutional Networks (FCNs) was proposed to figure out the boundary between fatigue crack propagation and fast fracture regions from optical microscope (OM) fractography images. Furthermore, novel morphing-based data augmentation method was adopted to enable few-shot learning of sample images. The proposed framework can successfully segment two categories, namely the crack propagation and fast fracture regions, thus differentiating the boundary of two regions in one image. The artificial intelligence (AI) assisted fatigue analysis architecture can complete the failure analysis procedure in 0.5 second and prove the feasibility of fatigue failure analysis. The segmentation accuracy of developed network achieves 95.4% for the fatigue crack propagation region, as well as 97.2% for the fast fracture region, which possesses comparable accuracy to the segmentation results using Mask R-CNN Regional Convolutional Neural Network (Mask R-CNN), one state-of-the-art deep learning networks.

Failure Analysis of Bolt of Rear Mounting Trunion of an Aero-Engine: A Case of Bending Induced Chevron Pattern as well as Fatigue Failure on the Same Fracture Surface

Present work describes the failure investigation of failed bolt of starboard rear mounting trunion of an aero-engine. Multiple fracture initiation points are noticed. This is a classic case of a single bending type of load initiating reversed bending fatigue as well as chevron pattern on the same fracture surface. Visually observed bending phenomenon supports the each type of failure mode. More interestingly, point of initiation of fast as well as the reversed bending fatigue failure is the same, although those two events have been found to be separate phenomena. It has been established that two different fatigue crack fronts, typical of reversed bending fatigue phenomenon propagated towards each other to make half of the cross-section fractured, while the another half failed by chevron patterned fast fracture. In this, the fast fracture of one half has preceded the reversed bending fatigue fracture of the other half, although the former is not responsible for happening of the later. Modes of fracture and factors influencing have been established in this article with emphasis on circumstantial evidence involving background information and visual examination, supported often by the open literature. Presence of cadmium (Cd) and its possible source, residence time and relative presence on differently fractured surfaces have offered important clues on establishing the sequence and relative inter-dependence of the said two fracture types. Presence of cadmium on the fracture surfaces, multiple crack initiation sites and numerous well-revealed secondary cracks on Branson cleaned fracture surface indicate that the cracks pre-existed on the material even before the cadmium plating and manifestation of chevron pattern is its extreme revelation. This pre-existing chevron pattern primarily aggravated the present failure through bending fatigue phenomenon in the later stages. Low alloy steel (ASTM grade 16) with presently used hardness (340 HV) level does not seem to suit the present application, as it is clear from its extreme brittleness as manifested by pre-existing cracks.

Fast Fracture Evaluation of Steam Generator Channel Heads Considering High Carbon Macro-Segregation

The Fracture evaluation is important in the structural integrity analysis of nuclear equipment which is subjected to the effects of neutron irradiation. The increment of ductile and brittle transition temperature is mainly due to the neutron irradiation, thermal ageing and strain ageing. In addition to above these factors, the high carbon macro-segregation of low-alloy steels also increases the risk of fast fracture failure as the carbon positive macro-segregation will lead to the increasing transition temperature of low-alloy steels. In this work, a relationship between the carbon content and the increment of transition temperature is developed and is used to the fast fracture failure analysis of the highest carbon content region in steam generator channel head. Results show that ratio between the calculated stress intensity factor considering safety coefficient suggested by ASME design code and the critical stress intensity factor is less than one, which indicates a safe design for the highest carbon content region in steam generator channel head.

Properties of Cladding With Respect to RPV Integrity

Inner surface of most of primary circuit components in PWR/BWR/WWER type reactors is covered by austenitic cladding that serves primarily as anticorrosion protection. This is also supported by the requirements for stress analysis of the vessel by most of the codes — austenitic cladding is not taken into account in the calculation of vessel wall thickness and on allowance of stress intensities for operating conditions. Its effect is taken, in some codes like for WWER components, in fatigue calculation and also for evaluation of vessel resistance against fast fracture during pressurized thermal shock (PTS) events.