Pollution flashover is a serious accident with a wide range of impacts in power grid. Accurate and timely classification of pollution severity is both a key to preventing pollution flashovers and a crucial challenge. This paper proposes an innovative method based on PTR and MCS3VM for classification of insulator pollution severity. The introduction of time-dimension information makes it possible to achieve pixel-level image identification instead of image or region-level image identification. The PTR physical model for pollution severity measurement is established. The relevant features are extracted by using PCA. A semi-supervised classifier is proposed to solve the problem of poor generalization due to insufficient labeled samples in industrial applications. Experimental results verify the satisfactory efficiency and accuracy of the proposed method and an estimation framework for fast, accurate and nondestructive industrial application without tedious work of labeling large amounts of data is concluded.

Localization of fatigue cracks imposes immense significance to ensure the health of the engineering structures and prevent further catastrophic accidents. The nonlinear ultrasonic waves, especially the nonlinear Lamb waves, have been increasingly studied and employed for identifying micro-damages that are usually invisible to traditional linear ultrasonic waves. However, it remains a challenge to locate the fatigue cracks using nonlinear Lamb waves owing to the enormous difficulties in decoding location information from acoustic nonlinearity. Motivated by this, this work presents a data-driven method for precise location of fatigue crack using nonlinear Lamb waves. A 1D-Attention-convolutional neural network is developed to correlate the fatigue crack location with the wavelet coefficients at the second harmonic frequency of Lamb wave signals. The introduction of the Attention layer enables the models to pay more attention to the desired nonlinear features which dominates locating the fatigue crack. In particular, a convenient dataset creation scheme guided by the relative value label is proposed to generate sufficient data commonly required for deep learning approach. In addition, a lightweight single-excite-multiple-receive signal acquisition method is adopted instead of full-matrix capture method used in the traditional research, which highly improves detection efficiency. Numerical simulation and experimental validation manifest that the trained network can be used to establish the complex mapping between the nonlinear ultrasonic signals and the fatigue crack location features, so as to locate barely visible fatigue cracks. Our work provides a promising and practical way to facilitate nonlinear Lamb waves to accurately locate fatigue cracks in large-scale plate-like structures.

The defect classification task is of great benefit to evaluating the safety performance of equipment and providing useful feedback information for discovering production process problems. In this article, we present a semisupervised learning (SSL) framework for transient thermography detection to employ the temporal and spatial information encoded into the three-dimensional transient thermal tensor data and provide pixel-level classification results for defect types. The time- and frequency-domain physical models for the transient thermal evolution of different kinds of defects are established to illustrate the theoretical foundation of defects classification based on transient thermography. The semisupervised multiclass Laplacian support vector machine is proposed to enable involving the abundant unlabeled data for enhancing learning performance in practical industrial applications where labeled samples are insufficient and labeling work is costly and laborious. A case study on silicone insulating materials with various types of artificial simulated internal defects validates the stronger generalized ability of the proposed method. This work, for the first time, proposes an SSL framework in transient thermography-based defect detection studies. It is believed that our proposed method is quite inspired for introducing SSL techniques to transient thermography for preferable performance in practical industrial applications.