In the present work, a probabilistic framework for fatigue life prediction and reliability assessment of an engine high pressure turbine disc is proposed to incorporate the effects of load variatio...

Probabilistic fatigue life prediction and reliability assessment of a high pressure turbine disc considering load variations

A neural-network technique can exploit the power of machine learning to mine the exponentially large data sets characterizing the state space of condensed-matter systems. Topological transitions and many-body localization are first on the list. Classifying phases of matter is key to our understanding of many problems in physics. For quantum-mechanical systems in particular, the task can be daunting due to the exponentially large Hilbert space. With modern computing power and access to ever-larger data sets, classification problems are now routinely solved using machine-learning techniques1. Here, we propose a neural-network approach to finding phase transitions, based on the performance of a neural network after it is trained with data that are deliberately labelled incorrectly. We demonstrate the success of this method on the topological phase transition in the Kitaev chain2, the thermal phase transition in the classical Ising model3, and the many-body-localization transition in a disordered quantum spin chain4. Our method does not depend on order parameters, knowledge of the topological content of the phases, or any other specifics of the transition at hand. It therefore paves the way to the development of a generic tool for identifying unexplored phase transitions.

/pdf/learning-phase-transitions-by-confusion-3tgj8gh1jx.pdf

Learning phase transitions by confusion

Aerospace is a key market driver for the advancement of additive manufacturing (AM) due to the huge demands in high-mix low-volume production of high-value parts, integrated complex part geometries and simplified fabrication workflow. Rapid and significant progress has been made in the laser additive manufacturing (LAM) of aeroengine materials, including the advanced high-strength steels, nickel-based superalloys and titanium-based alloys. Despite the extensive investigation of these three families of materials by the research community, there is a lack of comprehensive review on LAM of high strength steels, and existing gaps in published reviews on Ti-based alloys and Ni-based superalloys. Furthermore, although emerging materials such as high/medium entropy alloys and heterostructured materials exhibit promising mechanical performance, rigorous characterization, testing, qualification, and certification are still required before actual application in engine parts. Thus, it is still important and relevant to have a deep understanding on the relationship between process parameters – microstructures – mechanical properties in these widely used aeroengine materials, to drive the development of superior high-value alloys. This review aims to provide a critical and in-depth evaluation of laser powder bed fusion (LPBF) and laser directed energy deposition (LDED) technologies of the mentioned aeroengine materials. The review will summarize the material properties, performance envelops and outlines the research gaps of these aeroengine materials. Furthermore, perspectives on research opportunities, materials development, and new R&D approaches of LAM for the aeroengine materials are also highlighted.

Progress and perspectives in laser additive manufacturing of key aeroengine materials

Fracture and fatigue in additively manufactured metals

A machine-learning fatigue life prediction approach of additively manufactured metals

Prediction of fatigue–crack growth with neural network-based increment learning scheme

The present work aims to investigate the mechanism of crack initiation induced by internal pores, which are inevitable in additive manufacturing (AM), and the influence of internal pores on the fatigue performance of directed energy deposited (DED) Ti-6.5Al-2Zr-Mo-V. After fatigue test under constant amplitude alternating stress at three stress levels, thirty-one pieces of DED Ti-6.5Al-2Zr-Mo-V specimens were found that cracks initiating from internal pores. Scanning electron microscope (SEM) and its accessories, such as energy dispersive spectrometry (EDS) and electron backscattered diffraction (EBSD), were used to analyze the characteristics of pore defects and clarify the mechanism of crack initiation. The results show that the specificity of the microstructure affected by the DED process and pore defects, such as segregation of Al and the existence of incomplete grain boundaries, are the main causes of crack initiation. Then, the crack initiation modes were divided into three types, and a classification model was established that can make the effect of pore defects on fatigue life clearer and more intuitive.

Internal pores in DED Ti-6.5Al-2Zr-Mo-V alloy and their influence on crack initiation and fatigue life in the mid-life regime

Fatigue crack growth in diffusion-bonded Ti-6Al-4V laminate with unbonded zones

A Fatigue Life Prediction Approach for Laser-Directed Energy Deposition Titanium Alloys by Using Support Vector Regression Based on Pore-Induced Failures

Xiaofan He

Papers

Prediction of fatigue–crack growth with neural network-based increment learning scheme

Internal pores in DED Ti-6.5Al-2Zr-Mo-V alloy and their influence on crack initiation and fatigue life in the mid-life regime

Fatigue crack growth in diffusion-bonded Ti-6Al-4V laminate with unbonded zones

A Fatigue Life Prediction Approach for Laser-Directed Energy Deposition Titanium Alloys by Using Support Vector Regression Based on Pore-Induced Failures

Developing an accelerated flight load spectrum based on the nz-N curves of a fleet