Diagnosis of internal cracks in concrete using vibro-acoustic modulation and machine learning

This paper investigates the utility of physics-informed machine learning models for vibro-acoustic modulation (VAM)–based damage localization in concrete structures. Vibro-acoustic modulation is a nonlinear dynamics-based non-destructive testing method, which was initially developed to perform damage detection and later extended to accomplish damage localization. The VAM-based damage (hidden crack) diagnosis is performed by analyzing the damage index pattern on the surface of the component to arrive at the size and location of the hidden damage. Past investigations have employed heuristically selected damage index thresholds as well as computationally expensive Bayesian estimation methods for VAM-based damage localization in two (surface) dimensions. Compared to these studies, the proposed methodology automates the threshold selection (algorithmic instead of heuristic), increases the speed of the probabilistic damage diagnosis process, and enables the estimation of damage depth. We generate training data (damage index) for the machine learning models using the pertinent nonlinear dynamics (finite element) models using different combinations of test parameters. The (supervised) machine learning models are thus informed by computational physics models. These include two types of artificial neural network (ANN) models: classification models that identify whether a sensor location is damaged or not and regression models that enable Bayesian estimation to obtain the posterior probability distribution of damage location and size. The accuracy of machine learning-based diagnosis is evaluated using both numerical and laboratory experiments. The proposed physics-informed machine learning models for VAM-based damage diagnosis are able to achieve an accuracy of about 60–64% in the validation experiments, indicating the potential of these methods for internal crack detection. The results show that for complex (nonlinear dynamics-driven) diagnostic methods, damage index patterns learned from physics models could be successfully used for damage detection as well as localization.

Cracking Diagnosis in Fiber-Reinforced Concrete with Synthetic Fibers Using Piezoelectric Transducers

The addition of short fibers in concrete mass offers a composite material with advanced properties, and fiber-reinforced concrete (FRC) is a promising alternative in civil engineering applications. Recently, structural health monitoring (SHM) and damage diagnosis of FRC has received increasing attention. In this work, the effectiveness of a wireless SHM system to detect damage due to cracking is addressed in FRC with synthetic fibers under compressive repeated load. In FRC structural members, cracking propagates in small and thin cracks due to the presence of the dispersed fibers and, therefore, the challenge of damage detection is increasing. An experimental investigation on standard 150 mm cubes made of FRC is applied at specific and loading levels where the cracks probably developed in the inner part of the specimens, whereas no visible cracks appeared on their surface. A network of small PZT patches, mounted to the surface of the FRC specimen, provides dual-sensing function. The remotely controlled monitoring system vibrates the PZT patches, acting as actuators by an amplified harmonic excitation voltage. Simultaneously, it monitors the signal of the same PZTs acting as sensors and, after processing the voltage frequency response of the PZTs, it transmits them wirelessly and in real time. FRC cracking due to repeated loading ad various compressive stress levels induces change in the mechanical impedance, causing a corresponding change on the signal of each PZT. The influence of the added synthetic fibers on the compressive behavior and the damage-detection procedure is examined and discussed. In addition, the effectiveness of the proposed damage-diagnosis approach for the prognosis of final cracking performance and failure is investigated. The objectives of the study also include the development of a reliable quantitative assessment of damage using the statistical index values at various points of PZT measurements.

Diagnosis of interior damage with a convolutional neural network using simulation and measurement data

Current deep learning applications in structural health monitoring (SHM) are mostly related to surface damage such as cracks and rust. Methods using traditional image processing techniques (such as filtering and edge detection) usually face difficulties in diagnosing internal damage in thicker specimens of heterogeneous materials. In this paper, we propose a damage diagnosis framework using a deep convolutional neural network (CNN) and transfer learning, focusing on internal damage such as voids and cracks. We use thermography to study the heat transfer characteristics and infer the presence of damage in the structure. It is challenging to obtain sufficient data samples for training deep neural networks, especially in the field of SHM. Therefore we use finite element (FE) computer simulations to generate a large volume of training data for the deep neural network, considering multiple damage shapes and locations. These computer-simulated data are used along with pre-trained convolutional cores of a sophisticated computer vision-based deep convolutional network to facilitate effective transfer learning. The CNN automatically generates features for damage diagnosis as opposed to manual feature generation in traditional image processing. Systematic parameter selection study is carried out to investigate accuracy versus computational expense in generating the training data. The methodology is demonstrated with an example of damage diagnosis in concrete, a heterogeneous material, using both computer simulations and laboratory experiments. The combination of FE simulation, transfer learning and experimental data is found to achieve high accuracy in damage localization with affordable effort.

Optimization of Hypoxic Brain Injuries Diagnostics in Full-Term Newborns

The problem of early diagnosis of the central nervous system damage in newborn before the onset of clinical symptoms remains relevant at the present time.The aim of the study was to optimize the hypoxic brain damage diagnosis in full-term newborns by analyzing the concentration of cytokines in the umbilical cord blood.Materials and methods. During the first stage of the study, a prospective analysis of concentrations of interleukins (IL-1β, IL-4, IL-6, IL-8, IL-10), TNF-α and neuron-specific enolase (NSE) in the umbilical cord blood serum of full-term newborns was performed. The second stage of the study included the retrospective analysis of clinical data and instrumental research methods. The main method for diagnosing in the development of hypoxic brain damage in newborns was neurosonography.Results. The development of hypoxic brain damage is evidenced by the concentration of IL-1β over 30.3 pg/ml, IL-4 – over 1.7 pg/ml, IL-6 – over 79.4 pg/ml, IL-8 – over 107.7 pg/ml, NSE – more than 10.3 ng/ml and TNF-α – more than 1.6 pg/ml in umbilical cord blood.Conclusion. The results of the study confirmed that the comprehensive assessment of the cytokines concentration in the umbilical cord blood improves the hypoxic brain damage diagnosis in newborns. Analysis of the level of these markers immediately after the birth will optimize the management tactics of newborns who have undergone hypoxic exposure in antenatal and intranatal period.