Intelligent Structural Health Monitoring Research Articles

Damage Monitoring of Braided Composites Using CNT Yarn Sensor Based on Artificial Fish Swarm Algorithm.

This study aims to enable intelligent structural health monitoring of internal damage in aerospace structural components, providing a crucial means of assuring safety and reliability in the aerospace field. To address the limitations and assumptions of traditional monitoring methods, carbon nanotube (CNT) yarn sensors are used as key elements. These sensors are woven with carbon fiber yarns using a three-dimensional six-way braiding process and cured with resin composites. To optimize the sensor configuration, an artificial fish swarm algorithm (AFSA) is introduced, simulating the foraging behavior of fish to determine the best position and number of CNT yarn sensors. Experimental simulations are conducted on 3D braided composites of varying sizes, including penetration hole damage, line damage, and folded wire-mounted damage, to analyze the changes in the resistance data of carbon nanosensors within the damaged material. The results demonstrate that the optimized configuration of CNT yarn sensors based on AFSA is suitable for damage monitoring in 3D woven composites. The experimental positioning errors range from 0.224 to 0.510 mm, with all error values being less than 1 mm, thus achieving minimum sensor coverage for a maximum area. This result not only effectively reduces the cost of the monitoring system, but also improves the accuracy and reliability of the monitoring process.

Intelligent structural health monitoring of composite structures using machine learning, deep learning, and transfer learning: a review

Structural health monitoring (SHM) methods are essential to guarantee the safety and integrity of composite structures, which are extensively utilized in aerospace, automobile, marine, and infrastructure industry. The deterioration of composite structures is primarily caused by operational and environmental variability. To address this issue, artificial intelligence (AI) techniques are being integrated into the SHM systems to enhance the performance of composite structures via digital transformation and big data analysis. Therefore, the present article aims to provide a critical review of AI models, including machine learning, deep learning, and transfer learning, to preserve and sustain composite structures throughout their life. The article covers the complete SHM process for composite structures, including sensing technologies, data-preprocessing, feature extraction, and decision-making process. Thus, the health monitoring of composites is presented in consideration of modern AI techniques, accompanied by the identification of current challenges and potential future research directions.

A multi-sensor mapping Bi-LSTM model of bridge monitoring data based on spatial-temporal attention mechanism

The premise of intelligent structural health monitoring is to build a digital benchmark model representing the multi-sensor mapping relationship. Previously, researchers have used machine learning methods such as Long-and-Short-Term-Memory (LSTM) networks for modeling multi-sensor mapping relationships. However, the typical LSTM networks treat multi-dimensional input data equally, ignoring the possible correlations of different periods and sensor placements. Consequently, they can hardly obtain an accurate model for long-time series and multidimensional datasets. To address the issues, by introducing a two-layer attention mechanism in the temporal and spatial dimensions, an improved Bi-direction LSTM neural network model embedding the attention mechanism is proposed. The model focuses on grasping the non-stationary response process and the spatial correlation of multi-sensors. The accuracy and efficiency of the proposed approach are verified through two case studies. In addition, based on the visualization of attention weight assignments, the spatial–temporal attention distribution is studied to give an engineering interpretation.

Compression after multiple impact strength of composite laminates prediction method based on machine learning approach

The intelligent structural health monitoring system that can evaluate the structural safety online is the future development trend, in which the strength online prediction is the key step. This study developed a machine learning (ML) method to predict the compression-after-impact (CAI) strength of carbon/glass hybrid laminates subjected to multiple impacts at different impact positions online, which can help to find and replace damaged materials quickly to prevent irreversible disasters caused by accidental impact. Firstly, a finite element model verified by experiments was established to obtain the data of training ML model. Secondly, the eXtreme Gradient Boosting (XGBoost) model was utilized to predict the CAI strength of the composites subjected to multiple impacts at different distances between impact positions (DBIP). In addition, the feature importance of impact parameters based on the SHapley Additive exPlanations (SHAP) method was also studied. The results showed that the prediction accuracy and efficiency of ML-based method were better than that of FEM. Impact energy was the most significant factor affecting CAI strength, and DBIP cannot be ignored. The proposed method has great potential in online structural integrity monitoring systems of high-performance composite structures.

Study on Health Monitoring and Fatigue Life Prediction of Aircraft Structures

In this paper, some progress and achievements in aircraft integrity requirements, structural health monitoring, load spectrum measurement and life assessment research were presented. Several concepts of structural health monitoring were analyzed and compared, and the basic flow chart for health monitoring and life prediction of an aircraft structure was given. The selection of control points, construction of load/strain equations and stress calculation of control points were also described. Reliable IAT (Individual Aircraft Tracking) and life monitoring methods and software for IAT were developed for a certain type of aircraft, and fatigue life prediction of an aging aircraft was conducted based on actual measurement of load spectrum. The main features such as damage calculation, life evaluation and result output were discussed, and the future research focuses relating to intelligent structural health monitoring were finally explored.

New Sensing and Radar Absorbing Laminate Combining Structural Damage Detection and Electromagnetic Wave Absorption Properties

Within the paradigm of smart mobility, the development of innovative materials aimed at improving resilience against structural failure in lightweight vehicles and electromagnetic interferences (EMI) due to wireless communications in guidance systems is of crucial relevance to improve safety, sustainability, and reliability in both aeronautical and automotive applications. In particular, the integration of intelligent structural health monitoring and electromagnetic (EM) shielding systems with radio frequency absorbing properties into a polymer composite laminate is still a challenge. In this paper, we present an innovative system consisting of a multi-layered thin panel which integrates nanostructured coatings to combine EM disturbance suppression and low-energy impact monitoring ability. Specifically, it is composed of a stack of dielectric and conductive layers constituting the sensing and EM-absorbing laminate (SEAL). The conductive layers are made of a polyurethane paint filled with graphene nanoplatelets (GNPs) at different concentrations to tailor the effective electrical conductivity and the functionality of the material. Basically, the panel includes a piezoresistive grid, obtained by selectively spraying onto mylar a low-conductive paint with 4.5 wt.% of GNPs and an EM-absorbing lossy sheet made of the same polyurethane paint but properly modified with a higher weight fraction (8 wt.%) of graphene. The responses of the grid's strain sensors were analyzed through quasi-static mechanical bending tests, whereas the absorbing properties were evaluated through free-space and waveguide-based measurement techniques in the X, Ku, K, and Ka bands. The experimental results were also validated by numerical simulations.

Response Prediction Based on Temporal and Spatial Deep Learning Model for Intelligent Structural Health Monitoring

Machine learning models have recently demonstrated the ability to mine structural health monitoring data. While existing machine learning models can provide better performance than the previous univariate time-series model, it is still an open challenge of fully mining the spatial and temporal characteristics of the structural response to increase their accuracy. In addition, the heterogeneous correlation is always missed in traditional models. This article proposes a heterogeneous structural response prediction (HSRP) framework based on the deep learning model to improve the performance. The HSRP framework cannot only make full use of spatial and temporal correlations but also mine the correlation between heterogeneous responses. Motivated by recent studies in machine learning, an attention module is introduced to learn the correlation between different responses and initial the weights of sensors and past response. The convolutional neural network is also implemented to extract the spatial features and the long short-term memory network is used to extract the weekly, daily, and hourly patterns of structural response. A real-world data set collected from a bridge is used to evaluate the performance of the proposed model on single-step prediction and multistep prediction. The experimental results show that the proposed model outperforms several widely used benchmark models. Furthermore, additional experiments and evaluations are implemented to investigate the sensitivity and robustness of the proposed model.

Displacement Data Imputation in Urban Internet of Things System Based on Tucker Decomposition With L2 Regularization

Missing data are critical deficiency in the investigation of displacement measurement in urban Internet of Things system. In the insight of recovering missing displacement data, this article presents a data-driven and high-dimensional gap-imputation method, Tucker decomposition with L2 regularization. Results on the global navigation satellite system (GNSS) time series collected from an intelligent structural health monitoring system show that the recovery accuracy is improved compared with some popular benchmark methods. When the missing rate is 50%, compared with singular spectrum analysis, singular value decomposition, CP optimization algorithm, <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$k$ </tex-math></inline-formula> -nearest neighbors, and Tucker decomposition via alternating least squares, Tucker decomposition with L2 regularization can improve the average mean absolute error by about 4.74, 4.95, 5.82, 2.29, and 5.67 mm for all locations. It can be concluded that the consideration of multiple temporal correlations is necessary for missing data imputation. Compared with matrix decomposition, tensor decomposition can improve the ability for high-dimensional correlations in the GNSS time series.

Seismic Damage-Level Prediction Model of Cable-Stayed Bridge Using Nonlinear Time History Analysis for Intelligent Structural Health Monitoring under Earthquake Events

The current condition of a bridge should be given the highest priority to ensure its safety and its serviceability to the bridge users. Therefore, this paper explained the development of an intelligent decision-support tool in Bridge Health Monitoring system using a Neural Network model as a prediction of seismic damage-level for cable-stayed bridge. A total of eight earthquake loads scaled to various Peak Ground Acceleration (PGA) values. The input and output data which were fed into Feed Forward Artificial Neural Network (ANN) for damage level prediction model were developed based on acceleration responses from the Non-Linear Time History (NLTH) analysis of the cable-stayed bridge. A total of 16620 data were used as the input data. The damage-level categorization is based on FEMA 356. Data used for the ANN training are 70% for training, 15% for validation, and 15% for testing. The damage level prediction can greatly help bridge authority in order to maintain their bridge structure integrity by identifying and predicting the probability of damage occurring under earthquake loads.

Fiber Bragg Grating Smart Material and Structural Health Monitoring System Based on Digital Twin Drive

The damage self‐diagnosis function puts forward higher requirements for the research and development of intelligent structural health monitoring, and in most cases, it is necessary to monitor the load first, especially the monitoring of the impact load. The development of smart materials and structures is based on advanced sensing systems. In order to achieve this purpose, a high‐speed demodulation system based on fiber grating with double long period grating is studied, and then, a damage self‐diagnosis system based on fiber grating is constructed. The system can realize the strain distribution and impact load monitoring of the structure. After quantitative analysis of the signal, an advanced information identification method is used to realize the impact load location. This paper focuses on the data preprocessing process of bridge health monitoring. In view of the characteristics of high data complexity, large amount of data, and many noise components in the process of structural monitoring, this paper adopts basic data cleaning for the original data set, including data dimensionless and missing value processing. Based on the digital twin technology, the composition of the digital twin KNN model of bridge swivel construction monitoring and management is analyzed, and the digital twin system architecture of bridge swivel construction monitoring and management is built. The function display of the monitoring platform, including setting a variety of permission login modes, displaying BIM model, geographic information, and weather environment; monitoring data entry and addition, deletion, and modification; data chart analysis and export; and email warning, to verify the feasibility of the application of digital twin technology in bridge monitoring, and the advantages of the intelligent monitoring system are obtained. The strain error is found to be less than 15.48 μℇ in the research, which is within the range of the fiber grating. This method can effectively monitor and forecast these distributed, nonlinear, strongly coupled, multivariable, and time‐varying complex structures. By monitoring bridges, the original monitoring data of bridges can be obtained, and scientific research data and analysis services can be provided. In particular, the damage caused by shock and vibration is monitored, so that the accumulation of damage can be detected before it threatens the safety of the structure, so that the damaged structure can be repaired in time to ensure the safe operation of the structure.

Research on Intelligent Structural Health Monitoring System

The basic concepts of structural health monitoring are explained, the research status and its application in engineering are outlined, and the key activities of the health monitoring system are introduced. Finally, the application of statistical methods and machine learning methods in the process of damage identification is summarized.

Mortar-diatom composites for smart sensors and buildings

To address the challenge of appropriate cementitious materials that can support future optical fibre sensors with reduced strains for smart city infrastructure, reducing structural heterogeneity by integrating diatomous earth is proposed. A series of samples made up of commercial grade instant mortar and an increased percentage of diatomous earth are fabricated and characterised. Mindful of the growing importance of “in-the-field” measurements and diagnostics to wider construction technology evolution, hardness properties are measured using Mohs scratch test and portable field durometers; the latter was shown to be effective in post-plasticity assessment of these cement composites. Within experimental error low added concentrations improve cement hardness whilst making the cement more plastic. Structural colour is introduced by the diatoms and novel characterisation using a smartphone and polymer lens proposed and demonstrated. The implications for CO2 trapping and intelligent structural health monitoring (ISHM) for internet-of-things (IoT) in smart city infrastructure are discussed.

Structural health monitoring of an adhesively bonded CFRP aircraft fuselage by ultrasonic Lamb Waves

The aviation industry faces the challenge of offering aircraft that are lighter, more economical, and safer. One of the solutions is to increase the use of composites. For these materials, adhesive bonding has proven to be the appropriate joining technology. To check these adhesive joints, costly and time-consuming maintenance measures are carried out. An intelligent Structural Health Monitoring (SHM) system can extend these intervals and allow the use of a predictive maintenance system. This paper describes the method of Ultrasonic Lamb Waves for monitoring a adhesively bonded Carbon Fiber Reinforced Polymer (CFRP) aircraft fuselage. Prior to this, the production of a segment of a fuselage and the characterization of the materials (CFRP and adhesive) is shown. Afterwards the method of Ultrasonic Lamb Waves with the use of piezoelectric transducers and signal processing based on the Reconstruction Algorithm for Probabilistic Inspection of Damage (RAPID) algorithm are explained. At the end, the experimental evaluation of an undamaged and a damaged fuselage structure is done. The results have shown the possibility of RAPID algorithm for damage detection on adhesive. An outlook on future work is given.

Transparent graphene films with a tunable piezoresistive response

We demonstrate a graphene-coated polymer film as a transparent piezoresistive strain sensor. Strain sensitivity was tuned by controlling the thickness and density of the graphene layer as well as the adhesion force between graphene and the substrate. Chemically reducing the graphene oxide, thereby maximizing its adhesion to the substrate, while minimizing the coating density and thickness resulted in excellent transparency and high gauge factors. The fabrication method presented allows design optimization of transparent, high-performance, and sensitivity-tunable piezoresistive sensors, which can make inroads into such applications as wide-area strain sensors, self-sensing flexible electronics, and intelligent structural health monitoring.

用于低频振动测量的弯曲伸张型光纤布拉格光栅加速度传感器

The intelligent structural health monitoring method, which uses a fiber Bragg grating (FBG) sensor, is a new approach in the field of civil engineering. However, it lacks a reliable FBG-based accelerometer for taking structural low frequency vibration measurements. In this letter, a flextensional FBG-based accelerometer is proposed and demonstrated. The experimental results indicate that the natural frequency of the developed accelerometer is 16.7 Hz, with a high sensitivity of 410.7 pm/g. In addition, it has a broad and flat response over low frequencies ranging from 1 to 10 Hz. The natural frequency and sensitivity of the accelerometer can be tuned by adding mass to tailor the sensor performance to specific applications. Experimental results are presented to demonstrate the good performance of the proposed FBG-based accelerometer. These results show that the proposed accelerometer is satisfactory for low frequency vibration measurements.

Autonomous Impact Damage Monitoring in a Stiffened Composite Panel

This study is concerned with the detection and characterization of impact damage in a stiffened woven composite structure using high frequency Lamb waves and low frequency modal vibrations. The geometric and material complexities of the structure present practical difficulties in the direct analysis of both wave propagation and modal vibration data using theoretical constructs. Improved ultrasonic and vibration test setups consisting of distributed, high fidelity, and surface mounted sensor arrays are used here to determine changes in the dynamical properties of the composite structural components in the presence of damage. The sensors are assumed to provide both the low frequency global response (i.e., modal frequencies and mode shapes) of the structure to external loads and the (local) high frequency signals due to wave propagation effects in either passive or active mode of the ultrasonic array. A damage index, comparing the measured dynamical response of two successive states of the structure is introduced as a determinant of structural damage. The method relies on the fact that the dynamical properties of a structure change with the initiation or growth of damage. A diagnostic imaging tool is used for the interpretation and graphical representation of the indices to enable automated monitoring of the changes in the indices at a given instant of time. The value of the index at a given sensor increases with the proximity of the damage to the sensor. A sensitivity analysis is carried out in an effort to determine a threshold value of the index below which no reliable information about the state of health of the structure can be estimated. It is shown that the automated procedure is able to identify a defect right from its appearance, with some degree of confidence. The feasibility of developing a practical intelligent structural health monitoring system (ISHMS), based on the concept of `a structure requesting service when needed,' is discussed.

Damage identification for structural health monitoring using fuzzy pattern recognition

Uncertainty abounds with in situ structural performance assessment and damage detection in Structural Health Monitoring (SHM). Most research in SHM focuses on statistical analysis, data acquisition, feature extraction and data reduction. We introduce a method to improve pattern recognition and damage detection by supplementing Intelligent Structural Health Monitoring (ISHM) with fuzzy sets. Intuitively we know that damage does not occur as a Boolean relation (one of two values, true or false) but progressively. Bayesian updating is used to demarcate levels of damage into fuzzy sets accommodating the uncertainty associated with the ambiguous damage states. The new techniques are examined to provide damage identification using data simulated from finite element analysis of a prestressed concrete bridge without a priori known levels of damage.

Introduction to the use of wavelet multiresolution analysis for intelligent structural health monitoring

This paper suggests the use of wavelet multiresolution analysis (WMRA) as a reliable tool for digital signal processing in structural health monitoring (SHM) systems. A damage occurrence detection algorithm using WMRA augmented with artificial neural networks (ANN) is described. The suggested algorithm allows intelligent monitoring of structures in real time. The probability of damage occurrence is determined by evaluating the wavelet norm index (WNI) representing the energy of a signal describing the change in the system dynamics due to damage. An example application of the proposed algorithm is presented using a finite element simulated structural dynamics of a prestressed concrete bridge. The new algorithm showed very promising results.Key words: structural health monitoring, neural networks, wavelet analysis, signal processing, damage detection.