In situ health monitoring of multiscale structures and its instantaneous verification using mechanoluminescence and dual machine learning.

Extensive changes in the legal, commercial and technical requirements in engineering fields have necessitated automated real-time structural health monitoring (SHM) and instantaneous verification. An integrated system with mechanoluminescence (ML) and dual artificial intelligence (AI) modules with subsidiary finite element method (FEM) simulation is designed for in situ SHM and instantaneous verification. The ML module detects the exact position of a crack tip and evaluates the significance of existing cracks with a plastic stress-intensity factor (PSIF; K P ). ML fields and their corresponding K p M L values are referenced and verified using the FEM simulation and bidirectional generative adversarial network (GAN). Well-trained forward and backward GANs create fake FEM and ML images that appear authentic to observers; a convolutional neural network is used to postulate precise PSIFs from fake images. Finally, the reliability of the proposed system to satisfy existing commercial requirements is validated in terms of tension, compact tension, AI, and instrumentation.

Improving the Sensitivity of the Mechanoluminescence Composite through Functionalization for Structural Health Monitoring.

Over the past few years, considerable effort has been directed toward the development and improvement of mechanoluminescence (ML)-based stress sensing as an efficient nondestructive inspection technique. One of the challenges in ML stress sensing is the limited luminescent intensity and sensitivity of the ML-epoxy composite film to the local stress field. Herein, we present a novel approach for increasing the sensitivity of ML composites made of an epoxy resin matrix and SrAl2O4:Eu2+, Dy3+ particles functionalized with (3-aminopropyl)triethoxysilane. We performed a tensile test on an epoxy/ML composite specimen to investigate the effect of surface modification of ML particles on the luminescent sensitivity. A series of characterization analyses were performed on the modified surface to investigate the interfacial bonding. In addition, we applied the modified ML/epoxy paint to one side of the tensile specimen with an artificial invisible notch on the other side to visualize the stress field via light intensity (LI) distribution and then compared the results through a finite-element analysis (FEA). Surface modification of ML particles increased the sensitivity and introduced new chemical bonds, corresponding to a larger stress transfer through interfacial bonding rather than mere mechanical locking. In addition, the applied ML film on the notched specimen could visualize the specific pattern of LI reflecting the presence of a crack, which was confirmed by the FEA simulation. This implies that the proposed method of enhancing the ML film is promising for nondestructively predicting the presence, shape, and residual life of a crack in a specimen.

Digital Image Correlation Compatible Mechanoluminescent Skin for Structural Health Monitoring.

Monitoring structural health using mechanoluminescent (ML) effects is widely considered as a potential full-field and direct visualizing optical method with high spatial and temporal resolution and simple setup in a noncontact manner. The challenges and uncertainties in the mapping of ML field to effective strain field, however, tend to limit significant commercial ML applications for structural health monitoring systems. Here, however, quantification problems are resolved using the digital image correlation (DIC) method. Specifically, an image containing mechanically induced photon information is processed using a DIC algorithm to measure the strain field components, which enables the establishment of a calibration curve when the ML field is mapped onto the effective strain field using pixel level information. The results show a linear relationship between effective strain and ML intensity despite the plastic flow in ML skin. Furthermore, the calibration curve allows for easy conversion of ML field to effective-strain field at the crack-tip plastic zone of the alloy structure, retaining its spatial resolution. The compatibility of ML skin with the DIC algorithm not only enables the quantification of the ML effects of several organic/inorganic ML materials, but may also be useful in elucidating the fundamentals of the trap-controlled mechanism.

Determining the Fracture Process Zone Length and Mode I Stress Intensity Factor in Concrete Structures via Mechanoluminescent Technology.

The mechanoluminescent (ML) technology that is being developed as a new and substitutive technology for structural health monitoring systems (SHMS) comprises stress/strain sensing micro-/nanoparticles embedded in a suitable binder, digital imaging system, and digital image processing techniques. The potential of ML technology to reveal the fracture process zone (FPZ) that is commonly found in structural materials like concrete and to calculate the stress intensity factor (SIF) of concrete, which are crucial for SHMS, has never been done before. Therefore, the potential of ML technology to measure the length of the FPZ and to calculate the SIF has been demonstrated in this work by considering a single-edge notched bend (SENB) test of the concrete structures. The image segmentation approach based on the histogram of an ML image as well the skeletonization of an ML image have been introduced in this work to facilitate the measurement of the length of ML pattern, crack, and FPZ. The results show ML technology has the potential to determine fracture toughness, to visualize FPZ and cracks, and to measure their lengths in structural material like concrete, which makes it applicable to structural health monitoring systems (SHMS) to characterize the structural integrity of structures.

An Extremely Inexpensive, Simple, and Flexible Carbon Fiber Electrode for Tunable Elastomeric Piezo-Resistive Sensors and Devices Realized by LSTM RNN.

Here, we describe the utility of a carbon fiber (CF) electrode that is inexpensive, simple, and flexible and can be embedded with elastomeric nanocomposite piezo-resistive sensors fabricated from silicone rubber (Ecoflex) blended with carbon nanotubes (CNTs) and various wt % of silicone thinner to tune the sensitivity and softness range. The performance of the CF electrode was evaluated on the basis of piezo-resistive responses from the sensors subjected to dynamic sinusoidal compressive strains at different levels and frequencies. The responses were positive-pressure effects with rate-dependent asymmetric nonlinear hysteresis characteristics. Developing a mathematical model to describe the rate-dependent asymmetric nonlinear hysteresis behavior is technically impossible; therefore, we employed artificial intelligence-based hysteresis modeling, long short-term memory recurrent neural network, to describe the hysteresis nonlinearity. The debonding strength of the CF electrode was determined in the pull-off testing and was found to be much higher than that of a copper wire electrode. The debonding mechanism was further elucidated via an in situ resistance profile. The importance of a robust conductive interface between a CF electrode and a nanocomposite was experimentally demonstrated. It was found that the inherent piezo-resistance of the CF was negligible compared with the piezo-resistance of the sensor; therefore, the signals from the sensor were free of interference. We believe CF-embedded tunable piezo-resistive sensors could be used in biomedical devices, artificial e-skins, robotic touch applications, and flexible keyboards where the required stretchability of the electrode can be introduced via an appropriate geometrical design.

Deep-Learning Technique To Convert a Crude Piezoresistive Carbon Nanotube-Ecoflex Composite Sheet into a Smart, Portable, Disposable, and Extremely Flexible Keypad.

An extremely simple bulk sheet made of a piezoresistive carbon nanotube (CNT)-Ecoflex composite can act as a smart keypad that is portable, disposable, and flexible enough to be carried crushed inside the pocket of a pair of trousers. Both a rigid-button-imbedded, rollable (or foldable) pad and a patterned flexible pad have been introduced for use as portable keyboards. Herein, we suggest a bare, bulk, macroscale piezoresistive sheet as a replacement for these complex devices that are achievable only through high-cost fabrication processes such as patterning-based coating, printing, deposition, and mounting. A deep-learning technique based on deep neural networks (DNN) enables this extremely simple bulk sheet to play the role of a smart keypad without the use of complicated fabrication processes. To develop this keypad, instantaneous electrical resistance change was recorded at several locations on the edge of the sheet along with the exact information on the touch position and pressure for a huge number of random touches. The recorded data were used for training a DNN model that could eventually act as a brain for a simple sheet-type keypad. This simple sheet-type keypad worked perfectly and outperformed all of the existing portable keypads in terms of functionality, flexibility, disposability, and cost.

An extremely simple macroscale electronic skin realized by deep machine learning.

Complicated structures consisting of multi-layers with a multi-modal array of device components, i.e., so-called patterned multi-layers, and their corresponding circuit designs for signal readout and addressing are used to achieve a macroscale electronic skin (e-skin). In contrast to this common approach, we realized an extremely simple macroscale e-skin only by employing a single-layered piezoresistive MWCNT-PDMS composite film with neither nano-, micro-, nor macro-patterns. It is the deep machine learning that made it possible to let such a simple bulky material play the role of a smart sensory device. A deep neural network (DNN) enabled us to process electrical resistance change induced by applied pressure and thereby to instantaneously evaluate the pressure level and the exact position under pressure. The great potential of this revolutionary concept for the attainment of pressure-distribution sensing on a macroscale area could expand its use to not only e-skin applications but to other high-end applications such as touch panels, portable flexible keyboard, sign language interpreting globes, safety diagnosis of social infrastructures, and the diagnosis of motility and peristalsis disorders in the gastrointestinal tract.

A Mechanoluminescent ZnS:Cu/Rhodamine/SiO2/PDMS and Piezoresistive CNT/PDMS Hybrid Sensor: Red-Light Emission and a Standardized Strain Quantification.

We developed a hybrid strain sensor by combining mechanoluminescent ZnS:Cu/rhodamine/SiO2/PDMS composites and piezoresistive CNT/PDMS for qualitative and quantitative analysis of onsite strain. The former guarantees a qualitative onsite measure of strain with red-light emission via mechanoluminescence (ML) and the latter takes part in accurate quantification of strain through the change in electrical resistance. The PDMS matrix enabled a strain sensing in a wider range of strain, spanning up to several hundred percent in comparison to the conventional rigid matrix composites and ceramic-based ML sensors. Red-light emission would be much more effective for the visualization of strain (or stress) when ML is used as a warning sign in actual applications such as social infrastructure safety diagnosis, emergency guide lighting, and more importantly, in biomedical applications such as in the diagnosis of motility and peristalsis disorders in the gastrointestinal tract. Despite the realization of an efficient red-light-emitting ML in a ZnS:Cu/rhodamine/SiO2/PDMS composite, the quantification and standardization of strain throughout the ML has been far from complete. In this regard, the piezoresistive CNT/PDMS compensated for this demerit of mechanoluminescent ZnS:Cu/rhodamine/SiO2/PDMS composites.

Mechanoluminescence of SrAl2O4:Eu(+), Dy(3+) under cyclic loading.

The mechanoluminescence (ML) of SrAl2O4:Eu(+), Dy(3+) (SAO) has been of particular interest based on the possibility that these materials could be used as nondestructive, reproducible stress (or load) sensors. However, there has been no in-depth study of ML under a cyclic load. It was found that a cyclic load generated harmonics in the ML response. The second harmonic term exhibiting a doubled frequency was significant, but the others could be ignored. In addition, hysteresis behavior was observed in the ML and was examined by comparison with the hysteresis that is typical in piezoelectricity.