Full annulus URANS simulations are presented to investigate distortion phase shift in a single stage transonic fan. The fan is subject to a 90 degree sector inlet total pressure distortion. Simulation results are presented for choke, design, and near-stall operating conditions. Circumferential profiles of swirl, total pressure, total temperature, power, and phase shift are analyzed at 10%, 30%, 50%, 70%, and 90% span. Phase shift is a measure of the rotational translation of a distortion profile and is valuable for understanding the translational motion of distortion as it passes through the fan. At choke and design operating conditions the phase shift is positive in the direction of rotor rotation. At near-stall the phase shift is negative opposite rotor rotation for the outer 30% span. Local power reaches a maximum value, resulting in the generation of a uniform pressure region. This region causes increased distortion content in the near-stall case over a wide circumferential extent resulting in a negative phase shift.

Characterization of distortion transfer and generation through fans with distorted inlet conditions enables progress towards designs with improved distortion tolerance. The abruptness of transition from undistorted to distorted total pressure regions at the inlet impacts the induced swirl profile and therefore the distortion transfer and generation. These impacts are characterized using URANS simulations of PBS Rotor 4 geometry under a variety of inlet distortion profiles. A 90° and a 135° sector, both of 15% total pressure distortion, are considered. Variants of each sector size, with decreasing levels of distortion transition abruptness, are each applied to the fan. Fourier-based distortion descriptors are used to quantify levels of distortion transfer and generation at axial locations through the fan, principally at the stator inlet. It is shown that a gradual transition in distortion at the inlet results in decreased levels of distortion transfer and generation. The flow physics resulting in this reduction are explored.

Hobby level aircraft design can be a tedious and iterative process. To ease the design process, a function is developed and optimized for hobby aircraft energy usage based on aircraft aspect ratio, coefficient of lift, span, motor throttle, and battery capacity. The function allows for specification of mission range, required payload, motor size, and battery voltage which enables energy calculations for a variety of mission profiles. The function is optimized with the MATLAB gradient-based fmincon constrained optimizer. The optimum configurations for a variety of missions are reported and explored. Results from optimization agree with hobby aircraft design best practices and rules of thumb.

Understanding distortion transfer and generation through fan and compressor blade rows is able to assist in blade design and performance prediction. Using full annulus URANS simulations, the effects of distortion as it passes through the rotor of a transonic fan at five radial locations (10%, 30%, 50%, 70%, and 90% span) are analyzed. The inlet distortion profile is a 90-degree sector with a 15% total pressure deficit. Fourier distortion descriptors are used in this study to quantitatively describe distortion transfer and generation. Results are presented and compared for three operating points (near-stall, design, and choke). These results are used to explain the relationship between inlet total pressure distortion, pressure-induced swirl, total pressure distortion transfer, total temperature distortion generation, and circumferential rotor power variation. It is shown that very large changes in pressure-induced swirl and distortion transfer and generation occur between near-stall and design, but only small changes are seen between design and choke. The greatest changes are shown to be near the tip. Local power variations are shown to correlate with total pressure distortion transfer and total temperature distortion generation.

Although microscopy is often treated as a quasi-static exercise for obtaining a snapshot of events and structure, it is clear that a more dynamic approach, involving real-time decision making for guiding the investigation process, may provide deeper insights, more efficiently. On the other hand, many applications of machine learning involve the interpretation of local circumstances from experience gained over many observations; that is, machine learning potentially provides an ideal solution for more efficient microscopy. This paper explores the potential for informing the microscope's observation strategy while characterising critical events. In particular, the identification of regions likely to experience twin activity (twin interaction with grain boundary) in AZ31 magnesium is attempted, from only local information. EBSD-based observations in the neighbourhoods of twin activity are fed into a machine-learning environment to inform the future search for such events, and the accuracy of the resultant decisions is quantified relative to the number of prior observations. The potential for utilising different types of local information, and their resultant value in the prediction process, is also assessed. After applying an attribute selection filter, and various other machine-learning tools, a decision-tree model is able to classify likely neighbourhoods of twin activity with 85% accuracy. The resultant framework provides the first step towards an intelligent microscopy for efficient observation of stochastic events during in situ microscopy campaigns. Lay descriptionOne role of artificial intelligence is to predict future events after learning from many previous observations. In materials science, various phenomena (such as crack nucleation) are difficult to predict because they have been insufficiently observed. Furthermore, observation is difficult, precisely because their location cannot be predicted, leading to a chicken and egg conundrum. This paper applies machine learning to the search for twin nucleation sites in a magnesium alloy, in an attempt to guide the observation of twin nucleation events in a microscope based on previous observations. As more data is obtained, the accuracy of the location prediction will increase. In the current case, the machine-learning tool achieved 85% accuracy for predicting the location of twin interactions with grain boundaries after several thousand observations. The resultant framework provides the first step towards an intelligent microscopy for efficient observation of stochastic events during in situ microscopy campaigns.

Complex relationships between microstructure and twin formation in AZ31 magnesium are investigated as a function of increasing strain using supervised machine learning. In one approach, strain is incorporated as an implicit attribute in a single predictive model, in a second method, separate decision trees are formed for each strain level. A comparison of the methods shows that the second better uncovers the underlying physics. The correlations revealed are found to exhibit similarities with parameters used in conventional modeling techniques, leading to the conclusion that machine learning has potential to assist in future microstructural modeling.

Magnesium AZ31B sheets of 2 mm thickness were stretch formed using a 101.6 mm diameter punch at room temperature and subsequent increments from 25 to 125 °C. Surface strains were measured using a digital image correlation method in order to ensure that biaxial stretching was achieved. The punch height versus load curve was found to be the same for temperatures of 25 and for 50 °C, while at 75 °C the load for a given punch height was less. This difference seems to indicate a change in deformation mechanism between 50 and 75 °C. Electron Backscatter Diffraction (EBSD) was used to quantify features of the microstructure in the as-received and the strained specimens. Rather than a sudden transition from twinning to slip at low temperatures, it appears that twinning gradually decreases and slip activity increases as temperatures rise across the range from 25 to 125 °C. This confirms recent predictions found in the literature. The twin activity predominantly involves the formation of compression twins which rapidly transform further to create secondary twins for easier strain accommodation.

To explore the driving forces behind deformation twinning in Mg AZ31, a machine learning framework is utilized to mine data obtained from electron backscatter diffraction (EBSD) scans in order to extract correlations in physical characteristics that cause twinning. The results are intended to inform physics-based models of twin nucleation and growth. A decision tree learning environment is selected to capture the relationships between microstructure and twin formation; this type of model effectively highlights the more influential characteristics of the local microstructure. Trees are assembled to analyze both twin nucleation in a given grain, and twin propagation across grain boundaries. Each model reveals a unique combination of crystallographic attributes that affect twinning in the Mg. Twin nucleation is found to be mostly controlled by a combination of grain size, basal Schmid factor, and bulk dislocation density while twin propagation is affected most by grain boundary length, basal Schmid factor, angle from grain boundary plane to the RD plane, and grain boundary misorientation. The machine learning framework can be readily adapted to investigate other relationships between microstructure and material response.