Journal of Civil Structural Health Monitoring

Understanding the behavior of historic structures that have undergone structural changes, restorations, and damage over time is still a significant challenge for structural engineers, particularly in those countries subject to high seismic risk, such as Italy. The study of built heritage for its prevention and conservation is an active research topic, due to the numerous uncertainties present in historic structures. Finite element modelling has become the most common and accessible method to study the behavior of complex masonry structures, however, the gap between numerical and experimental analysis may lead to erroneous results. Model updating techniques can reduce the discrepancy between the behavior of the numerical models and the testing results. The goal of this work is to illustrate a methodology to integrate the information derived from local, global, and geotechnical investigations into the finite element model of the masonry historical church of San Giovanni in Macerata, considering the Douglas–Reid model updating method. The PRiSMa laboratory of Roma Tre University carried out local investigations such as sonic tomography, video endoscopy and double flat jack tests, along with five ambient vibration tests that were processed through the operational modal analysis to extrapolate the dynamic properties of the building (modal frequency, modal shape vector and modal damping). The combined use of global, local and geotechnical information implemented in the methodology effectively reduced the uncertainties of the model and led the refinement and validation of the most relevant structural parameters.

Accurate finite element models are needed in many applications of Civil Engineering. Non-structural elements (NSEs) often interfere with the main structure, altering its stiffness and modal signature. Neglecting such interaction, although a common practice in design, may lead to unreliable predictions of the structural dynamic response and to biased interpretations of experimental vibrational tests. In the literature, the role of NSEs in vibration-based finite element model calibration is well documented for NSEs working “in parallel” with the main structure (e.g. masonry infills in buildings, pavements or railings in bridges) but remains essentially unexplored for NSEs working “in series” with the main structure (e.g. non-structural appendages like suspended ceilings, piping systems, storage tanks and antennas, but also partitions and façades in their out-of-plane modes). Through the analysis of numerical and experimental case studies, this paper shows that “in series” NSEs accidentally resonating with some structural mode may deeply contaminate the overall modal behaviour and severely invalidate model calibration, unless their role is properly addressed while performing experimental modal analysis and structural modelling. Two alternative calibration strategies are finally discussed which prove effective in the presence of resonant NSEs, based on, respectively, excluding/including the NSEs from/into the model structure.

For brick-concrete buildings which are easily damaged by mining activities, this study conducted measurements using a terrestrial laser scanning and adopted the wall length as a deformation monitoring index. First, the point cloud data of the individual walls are extracted by segmenting the building-point cloud data and redistributing the wall-point cloud data. Second, after individual walls were rotated, the boundary points of all the wall-point clouds were estimated. Third, based on boundary points, the top boundary lines are fitted using the weighted iterative least squares method by applying the constraint that two adjacent (top) boundary lines intersect at a common point. Fourth, multiple regions of interest were constructed by performing downward translations on each wall, and the length of the wall was calculated at these regions. Finally, for a brick-concrete building in the coal mining area, the difference between the wall lengths of two consecutive measurements at the same interest-region was estimated to obtain the deformation information about the building. The results show that the deformation estimated using the proposed method is consistent with the actual deformation of the building as reflected by the wall fractures. The method exhibits millimeter-level accuracy in determining the wall length, with an absolute error in the range – 6 to 6 mm.

This paper presents the development of a new software code for the fully autonomous management of permanent integrated Structural Health Monitoring (SHM) systems installed in highway bridge structures. The code was developed within the framework of a research project funded by Anas S.p.A., the largest public infrastructure manager in Italy. The software program includes all the necessary steps to conduct SHM within the statistical pattern recognition paradigm, including automated dynamic identification, modal tracking, filtering of environmental effects, and damage detection through novelty analysis. Additionally, the software suite includes specific modules for processing and analysis of seismic events and structural reliability analysis of bridges, as well as specific functionalities for span-wise identification of long multi-span bridges. Moreover, a novel automated density-based tracking algorithm is developed. The potential of P3P is illustrated through two real application case studies: (i) a long multi-span bridge, the Trigno V Bridge in Italy; and (ii) the Z-24 Bridge benchmark. This work demonstrates the effectiveness of the developed code for handling large monitoring databases within the framework of SHM as a statistical pattern recognition, and currently P3P is in phase of being applied by Anas S.p.A for the management of a large number of bridges of the Italian roadway system.

Có hàng ngàn cầu dầm gỗ phân bố khắp khu vực miền nông thôn Australia, chủ yếu được giám sát thông qua việc kiểm tra trực quan. Kinh nghiệm từ những thất bại trong quá khứ đã dẫn đến sự nhận thức rõ ràng rằng kiểm tra trực quan với khoảng thời gian nhiều tháng hoặc năm là không đủ để phát hiện những hỏng hóc tiềm ẩn do quá tải và sự suy thoái sinh học. Một cây cầu bị quá tải hôm nay có thể bị hỏng vào ngày mai và cần phải triển khai hệ thống giám sát sức khỏe kết cấu (SHM) để có thể phát hiện sớm tình trạng quá tải. Nhu cầu này đang trở nên ngày càng cấp bách với mong đợi gia tăng để đáp ứng nhu cầu về tải trọng cao hơn trong các giai đoạn vận chuyển sản phẩm theo mùa. Việc đo biến dạng giữa nhịp của các dầm có thể được sử dụng để xác định các chỉ số an toàn phục vụ cho việc đánh giá độ an toàn của kết cấu. Việc phát hiện hư hỏng theo thời gian thực trong các cầu dầm gỗ bằng cách sử dụng camera tốc độ cao và các phương pháp dựa trên laser mang lại những lợi thế độc đáo và có thể dẫn đến các kỹ thuật đo lường tiết kiệm chi phí. Công trình này báo cáo về việc sử dụng các phương pháp giám sát liên tục để xác định độ tin cậy kết cấu của các dầm cầu gỗ. Một số ứng dụng của hệ thống cảm biến biến dạng dựa trên laser được thảo luận liên quan đến việc giám sát độ tin cậy kết cấu của hai cầu dầm gỗ cũ hơn ở miền nông thôn New South Wales, Australia. Các phương pháp thực nghiệm và phân tích được trình bày và sử dụng để chứng minh rằng xác suất thất bại có thể dễ dàng được xác định liên tục thông qua hệ thống SHM.

In this study, a multi-bandwidth wavelet transform of signals with mixed frequencies from multiple delamination mechanisms is proposed to achieve the time and frequency resolution required in delamination detection. It was applied to detect six embedded defects, simulating shallow and deep delamination, in a 60 in. × 36 in. × 7.25 in. reinforced concrete slab from impact echo signals recorded at 40 points using a portable seismic property analyzer. Test results indicated that the proposed transform successfully detected all the defects and was twice as accurate as Fourier transform for the identification of resonant frequencies. Out of the 40 measurement points, the false detection points were reduced from 22 with Fourier transform to 10 with the proposed transform.

Ambient vibration is an unknown excitation source that may produce stationary or non-stationary signals. Under such circumstances, traditional feature extraction techniques may not yield relevant features to damage and not provide reliable results of damage identification. The main objective of this article is to propose a data-driven method based on the concept of statistical pattern recognition for locating damage under ambient vibration and unpredictable signal nature in terms of simultaneously stationary and non-stationary behavior. This method is generally comprised of a three-level hybrid algorithm for feature extraction and new statistical distance metrics for feature analysis. The proposed feature extraction method aims at providing new damage-sensitive features in three levels including (1) analyzing the nature of measured vibration signals in terms of stationarity or non-stationarity, and normalizing non-stationary signals by detrending and differencing techniques, (2) modeling each vibration signal by an Autoregressive Moving Average (ARMA) model along with extracting the model residuals, and (3) estimating the power spectral density of residual samples as a new spectral-based feature. To identify the location of damage via spectral-based features, this article proposes two new spectral-based measures called Jeffery’s and Smith’s distances. The major contributions of this study include proposing a new feature extraction method for dealing with the problem of unpredictable vibration nature and introducing two new distance metrics for damage identification. Experimental vibration measurements of a well-known laboratory structure are utilized to verify the proposed methods. Results demonstrate that these approaches succeed in accurately extracting relevant features and locating damage under ambient vibration and unpredictable signal nature.

Corrosion-induced wall thickness loss (CIWTL) can reduce the strength and integrity of a pipe, threatening its normal operation. Effective detection of CIWTL in pipes helps ensure their safe operation. This paper presents a novel methodology based on the reflected L(0,1) guided wave to quantitatively detect CIWTL in a continuous pipe. Investigating the effects of CIWTL and propagation length on time-of-flight (TOF) variation of the L(0,1) guided wave showed that increasing the accumulated propagation length of the L(0,1) mode improved its sensitivity to CIWTL. The reflected L(0,1) guided wave, which had a longer accumulated propagation length in a certain range, was generated by making discontinuities on both sides of a localized section within a continuous pipe. Then, the TOF variation of the reflected wave was proposed as a CIWTL-sensitive feature, and a quantitative relationship between the TOF variation of the reflected wave and CIWTL was theoretically established for quantifying the CIWTL of the pipe section. High-resolution measurement of CIWTL could be achieved through increased accumulated propagation length. Additionally, this methodology could be applied to measure CIWTL in the next pipe section and extended to realize the distributed detection of CIWTL in a continuous pipe. The effectiveness of this methodology was validated experimentally. The experimental results indicated that the L(0,1) mode was clearly reflected from artificial discontinuities, CIWTL in the pipe was sensitively identified and accurately quantified using the proposed method, and the values of the CIWTL measured by the proposed method were consistent with those measured by ultrasonic testing (UT). This methodology has higher estimation performance for CIWTL than current guided wave-based (GWB) methods.