L-shape triple defects in a phononic crystal for broadband piezoelectric energy harvesting.

This study proposes a phononic crystal (PnC) with triple defects in an L-shape arrangement for broadband piezoelectric energy harvesting (PEH). The incorporation of defects in PnCs has attracted significant attention in PEH fields owing to properties such as energy localization and amplification near the defect. Several studies have been conducted to enhance output electric power of PnC-based PEH systems with single defects. However, it is susceptible to the limitations of narrow bandwidth. Recently, double-defect-incorporated systems have been proposed to widen the PEH bandwidth via defect-band splitting. Nevertheless, the PEH performance rapidly decreases in the frequency range between the split defect bands. The limitations of single- and double-defect-incorporated systems can be resolved by the incorporation of the proposed design concept, called the L-shape triple defects in a PnC. The isolated single defect at the top vertex of the letter 'L' compensates for the limitations of double-defect-incorporated systems, whereas the double defects at the bottom vertices compensate for the limitations of the single-defect-incorporated systems. Hence, the proposed design can effectively confine and harvest elastic-wave energy over broadband frequencies while enhancing the application of single and double defects. The effectiveness of the proposed design concept is numerically validated using the finite element method. In the case of a circular hole-type PnC, it is verified that the PnC with L-shape triple defects broadens the bandwidth, and improves the output voltage and electric power compared with those of single- and double-defect-incorporated systems. This study expands the design space of defect-incorporated PnCs and might shed light on other engineering applications of the frequency detector and elastic wave power transfer.

A Domain Adaptation with Semantic Clustering (DASC) method for fault diagnosis of rotating machinery.

Recently, substantial research has explored the development of deep-learning-based methods to diagnose faults in rotating machinery. For these diagnosis methods, it is difficult to obtain high target diagnosis accuracy when the amount of labeled data obtained pertaining to the rotating machinery under study is insufficient or in cases involving a discrepancy in the distribution types found in the training and test data sets. To deal with this research need, the paper outlines a new method, a domain adaptation with semantic clustering (DASC), capable of diagnosing faults in rotating machinery. The method outlined in this research learns both domain-invariant and discriminative features. The method reduces the domain discrepancy by minimizing the domain-related loss. In addition, by defining an additional loss, which is called semantic clustering loss, and reducing it at multiple feature levels, the DASC method learns features that make samples better semantically clustered, according to their health conditions. Consequently, fault diagnosis performance for target rotating machinery can be enhanced through the use of the DASC approach. The effectiveness of the DASC approach is confirmed by examining various fault diagnosis situations with domain discrepancies across the source and target domains, using experimental data from three bearing systems. Also, various analyses are explored to better understand the advantages of the DASC method.

A deep transferable motion-adaptive fault detection method for industrial robots using a residual-convolutional neural network.

Recently, in various industrial fields, including automated manufacturing processes, industrial robots are becoming indispensable equipment; these robots perform repetitive tasks and increase the productivity of the production line with consistent precision and accuracy. Thus, fault diagnostics of industrial robots is an essential strategy to prevent the significant economic losses that can be caused by a sudden stop of a production line due to an industrial robot fault However, previous data-driven industrial robot fault diagnostics are limited because a pre-trained model built for a specific motion may not accurately or consistently detect faults in other motions, due to motion discrepancies. To overcome this difficulty, in this paper, we propose a deep transferable motion-adaptive fault detection method that uses torque ripples for fault detection of industrial robot gearboxes. The proposed method is composed of two stages: (1) a residual-convolutional neural network is used to enhance the performance of feature extraction for simple motions, after first refining raw torque signals by filtering out the motion-dependent signals (2) a binary-supervised domain adaptation is performed to detect faults adaptively on multi-axial motions through adversarial contrastive learning. The efficacy of the proposed method was validated using experimental data from unit-axis and multi-axial welding motions collected from a real industrial robot testbed. The proposed method showed superior fault detection accuracy for the motion adaptation task, as compared to existing methods.

Double defects-induced elastic wave coupling and energy localization in a phononic crystal.

This study aims to investigate elastic wave localization that leverages defect band splitting in a phononic crystal with double defects through in-depth analysis of comparison of numerical and experimental results. When more than one defect is created inside a phononic crystal, these defects can interact with each other, resulting in a distinctive physical phenomenon from a single defect case: defect band splitting. For a phononic crystal consisting of circular-hole type unit cells in a thin aluminum plate, under A0 (the lowest antisymmetric) Lamb waves, both numerical simulations and experiments successfully confirm the defect band splitting phenomenon via frequency response functions for the out-of-plane displacement calculated/measured at the double defects within a finite distance. Furthermore, experimental visualization of in-phase and out-of-phase defect mode shapes at each frequency of the split defect bands is achieved and found to be in excellent agreement with the simulated results. Different inter-distance combinations of the double defects reveal that the degree of the defect band splitting decreases with  the increasing distance due to weaker coupling between the defects. This work may shed light on engineering applications of a multiple-defect-introduced phononic crystal, including broadband energy harvesting, frequency detectors, and elastic wireless power transfer.

Intelligent Steam Power Plant Boiler Waterwall Tube Leakage Detection via Machine Learning-Based Optimal Sensor Selection.

Boiler waterwall tube leakage is the most probable cause of failure in steam power plants (SPPs). The development of an intelligent tube leak detection system can increase the efficiency and reliability of modern power plants. The idea of e-maintenance based on multivariate algorithms was recently introduced for intelligent fault detection and diagnosis in SPPs. However, these multivariate algorithms are highly dependent on the number of input process variables (sensors). Therefore, this work proposes a machine learning-based model integrated with an optimal sensor selection scheme to analyze boiler waterwall tube leakage. Finally, a real SPP test case is employed to validate the proposed model's effectiveness. The results indicate that the proposed model can successfully detect waterwall tube leakage with improved accuracy vs. other comparable models.

Autonomous health management for PMSM rail vehicles through demagnetization monitoring and prognosis control.

Autonomous vehicles are playing an increasingly importance in support of a wide variety of critical events. This paper presents a novel autonomous health management scheme on rail vehicles driven by permanent magnet synchronous motors (PMSMs). Firstly, the PMSMs are modeled based on first principle to deduce the initial profile of pneumatic braking (p-braking) force, then which is utilized for real-time demagnetization monitoring and degradation prognosis through similarity-based theory and generate prognosis-enhanced p-braking force strategy for final optimal control. A case study is conducted to demonstrate the feasibility and benefit of using the real-time prognostics and health management (PHM) information in vehicle 'drive-brake' control automatically. The results show that accurate demagnetization monitoring, degradation prognosis, and real-time capability for control optimization can be obtained, which can effectively relieve brake shoe wear.

A new piezoelectric energy harvesting design concept: multimodal energy harvesting skin.

This paper presents an advanced design concept for a piezoelectric energy harvesting (EH), referred to as multimodal EH skin. This EH design facilitates the use of multimodal vibration and enhances power harvesting efficiency. The multimodal EH skin is an extension of our previous work, EH skin, which was an innovative design paradigm for a piezoelectric energy harvester: a vibrating skin structure and an additional thin piezoelectric layer in one device. A computational (finite element) model of the multilayered assembly - the vibrating skin structure and piezoelectric layer - is constructed and the optimal topology and/or shape of the piezoelectric layer is found for maximum power generation from multiple vibration modes. A design rationale for the multimodal EH skin was proposed: designing a piezoelectric material distribution and external resistors. In the material design step, the piezoelectric material is segmented by inflection lines from multiple vibration modes of interests to minimize voltage cancellation. The inflection lines are detected using the voltage phase. In the external resistor design step, the resistor values are found for each segment to maximize power output. The presented design concept, which can be applied to any engineering system with multimodal harmonic-vibrating skins, was applied to two case studies: an aircraft skin and a power transformer panel. The excellent performance of multimodal EH skin was demonstrated, showing larger power generation than EH skin without segmentation or unimodal EH skin.