Search for: All records

Cracks are the defects formed by cyclic loading, fatigue, shrinkage, creep, and so on. In addition, they represent the deterioration of the structures over some time. Therefore, it is essential to detect and classify them according to the condition grade at the early stages to prevent the collapse of structures. Deep learning-based semantic segmentation convolutional neural network (CNN) has millions of learnable parameters. However, depending on the complexity of the CNN, it takes hours to days to train the network fully. In this study, an encoder network DenseNet and modified LinkNet with five upsampling blocks were used as a decoder network. The proposed network is referred to as the “CrackDenseLinkNet” in this work. CrackDenseLinkNet has 19.15 million trainable parameters, although the input image size is 512 × 512 and has a deeper encoder. CrackDenseLinkNet and four other state-of-the-art (SOTA) methods were evaluated on three public and one private datasets. The proposed CNN, CrackDenseLinkNet, outperformed the best SOTA method, CrackSegNet, by 2.2% of F1-score on average across the four datasets. Lastly, a crack profile analysis demonstrated that the CrackDenseLinkNet has lesser variance in relative errors for the crack width, length, and area categories against the ground-truth data. The code and datasets can be downloaded at https://github.com/preethamam/CrackDenseLinkNet-DeepLearning-CrackSegmentation . 

Abstract Emerging transformable lattice structures provide promising paradigms to reversibly switch lattice configurations, thereby enabling their properties to be tuned on demand. The existing transformation mechanisms are limited to nonfracture deformation, such as origami, instability, shape memory, and liquid crystallinity. In this study, we present a class of transformable lattice structures enabled by fracture and shape-memory-assisted healing. The lattice structures are additively manufactured with a molecularly designed photopolymer capable of both fracture healing and shape memory. We show that 3D-architected lattice structures with various volume fractions can heal fractures and fully restore stiffness and strength over two to ten healing cycles. In addition, coupled with the shape-memory effect, the lattice structures can recover fracture-associated distortion and then heal fracture interfaces, thereby enabling healing of lattice wing damages, mode-I fractures, dent-induced crashes, and foreign-object impacts. Moreover, by harnessing the coupling of fracture and shape-memory-assisted healing, we demonstrate reversible configuration transformations of lattice structures to enable switching among property states of different stiffnesses, vibration transmittances, and acoustic absorptions. These healable, memorizable, and transformable lattice structures may find broad applications in next-generation aircraft panels, automobile frames, body armor, impact mitigators, vibration dampers, and acoustic modulators. 

Abstract Topological field‐effect transistor is a revolutionary concept that physical fields are used to switch on and off quantum topological states of the condensed matter. Although this emerging concept has been explored in electronics, how to realize it in the acoustic realm remains elusive. In this work, a class of magnetoactive acoustic topological transistors capable of on‐demand switching on and off topological states and reconfiguring topological edges with external magnetic fields is presented. The key mechanism is to harness magnetic fields to tune air‐cavity volumes within acoustic chambers, thus breaking or preserving the inversion symmetry to manifest or conceal the quantum valley Hall effect. To switch the topological transport beyond the in‐plane routes, a magneto‐tuned non‐topological band gap to allow or forbid the wave transport out‐of‐plane is harnessed. With the reversible magnetic control, on‐demand switching of topological routes to realize topological field‐effect waveguides and wave regulators is demonstrated. Analogous to the impact of semiconductor transistors on modern electronics, this work may expand the scope of topological acoustics by achieving unprecedented functions in acoustic modulation.