Moving Principal Component Analysis Research Articles

Moving-Principal-Component-Analysis-Based Structural Damage Detection for Highway Bridges in Operational Environments.

With the deterioration of bridge performance and ever-increasing amounts of traffic, bridge safety is becoming a concern for the engineering community. A method that can assess a bridge's condition in real time is urgently needed. The main factors that hinder bridge condition assessment are the uncertain operational environments. A new moving principal component analysis (MPCA)-based method is developed for structural damage detection in bridges in operational environments in this paper. Two main operational environmental factors, the environmental temperature and traffic loads, are studied in the assessment process to verify the robustness and practicality of the proposed method. The numerical and experimental results show that the proposed method is effective and accurate in assessing the bridge's condition in operational environments.

Machine-learning-based damage identification methods with features derived from moving principal component analysis

This paper aims to propose machine-learning-based damage identification methods with features derived from moving principal component analysis (MPCA) to improve the damage identification performance for engineering structures. Previously, machine learning algorithms have usually used structural responses as inputs directly. These methods show low damage identification capabilities and are susceptible to noise. In this paper, the eigenvectors of structural responses derived from MPCA are employed as inputs instead. Several traditional machine learning algorithms are applied for verification. The results demonstrate that as compared to strains and frequencies, their eigenvectors as inputs for machine learning algorithms render better performances for damage identification.

A temperature-driven MPCA method for structural anomaly detection

An important issue in structural health monitoring (SHM) is to develop appropriate algorithms that can explicitly extract meaningful changes in measurements due to structural anomalies, especially damage. However, the effects due to environmental factors, especially temperature variations may produce significant misinterpretations. Consequently, developing solutions to identify the structural anomaly, accounting for temperature influence, from measurements, is crucial and highly anticipated. This paper presents a Temperature-driven Moving Principal Component Analysis method, designated as Td-MPCA, for anomaly detection. The Td-MPCA introduces the idea of blind source separation (BSS) for thermal identification with intent to enhance the performance of Moving Principal Component Analysis (MPCA) for anomaly detection. To achieve this target, temperature-induced strain variations are first investigated and revealed by employing Independent Component Analysis based on maximization non-Gaussianity, also known as Fast ICA. Afterwards, the MPCA is adopted for anomaly detection on the separated temperature-related response. Three case studies are provided in this paper to evaluate the proposed method. The first one is a numerical truss bridge with a simulated 5% stiffness reduction. The results confirm that Td-MPCA is more sensitive than MPCA in detecting anomalies, where the simulated stiffness loss fails to be detected by MPCA. The second case study is on an experimental truss bridge where two damage scenarios are introduced and interpreted. The detection results show that Td-MPCA outperforms MPCA since the damage is identified at the expected time by Td-MPCA but not by MPCA. The third case study is an in-situ curved viaduct in Switzerland. Data acquired during both construction period and normal service period has been used for interpretation. Results demonstrate that Td-MPCA is able to identify the date of change in construction process without any delay when compared with the application of MPCA only.

Thermal strain extraction methodologies for bridge structural condition assessment

This paper presents a feature extraction method to uncover the temperature effects on bridge responses, which combines mode decomposition, data reduction, and blind separation. For mode decomposition, empirical mode decomposition (EMD) and ensemble empirical mode decomposition (EEMD) have been executed, followed by principal component analysis (PCA) for data size compression. Independent component analysis (ICA) is then employed for blind separation. The unique feature of the proposed method is the blind separation, which enables temperature-induced response to be extracted from the mixed structural responses without any prior information of the loading conditions and structural physical models. This study further evaluates the effects of extracting a temperature-induced response on damage detectability when using Moving Principal Component Analysis (MPCA). Numerical analysis for a truss bridge model is first conducted to evaluate the proposed method for thermal feature extraction, followed by a real truss bridge test in the structural laboratory at the University of Warwick. Results from the numerical study show that the method enables the separation of a temperature-induced response; and furthermore, the EEMD, utilized in mode decomposition, has a positive influence on the blind separation compared with EMD when combined with PCA and ICA. Finally, the real truss bridge test demonstrates that the feature extraction method can enhance the probability of MPCA to uncover the damage as the MPCA fails to discern damage without the proposed method.

Performance-based structural health monitoring through an innovative hybrid data interpretation framework

The utilization of Structural Health Monitoring (SHM) for performance-based evaluation of structural systems requires the integration of sensing with appropriate data interpretation algorithms to establish an expected performance related to damage or structural change. In this study, a hybrid data interpretation framework is proposed for the long-term performance assessment of structures by integrating two data analysis approaches: parametric (model-based, physics-based) and non-parametric (data-driven, model-free) approaches. The proposed framework employs a network of sensors through which the performance of the structure is evaluated and the corresponding maintenance action can be efficiently taken almost in real-time. The hybrid algorithm proposed can be categorized as a supervised classification algorithm. In the training phase of the algorithm, a Monte-Carlo simulation technique along with Moving Principal Component Analysis (MPCA) and hypothesis testing are employed for simulation, signal processing, and learning the underlying distribution, respectively. The proposed approach is demonstrated and its performance is evaluated through both analytical and experimental studies. The experimental study is performed using a laboratory structure (UCF 4-Span Bridge) instrumented with a Fiber Brag Grating (FBG) system developed in-house for collecting data under common bridge damage scenarios. The proposed hybrid approach holds potential to significantly enhance sensor network design, as well as continuous evaluation of the structural performance.

An integrated approach for structural health monitoring using an in-house built fiber optic system and non-parametric data analysis

Multivariate statistics based damage detection algorithms employed in conjunction with novel sensing technologies are attracting more attention for long term Structural Health Monitoring of civil infrastructure. In this study, two practical data driven methods are investigated utilizing strain data captured from a 4-span bridge model by Fiber Bragg Grating (FBG) sensors as part of a bridge health monitoring study. The most common and critical bridge damage scenarios were simulated on the representative bridge model equipped with FBG sensors. A high speed FBG interrogator system is developed by the authors to collect the strain responses under moving vehicle loads using FBG sensors. Two data driven methods, Moving Principal Component Analysis (MPCA) and Moving Cross Correlation Analysis (MCCA), are coded and implemented to handle and process the large amount of data. The efficiency of the SHM system with FBG sensors, MPCA and MCCA methods for detecting and localizing damage is explored with several experiments. Based on the findings presented in this paper, the MPCA and MCCA coupled with FBG sensors can be deemed to deliver promising results to detect both local and global damage implemented on the bridge structure.

Damage detection using data-driven methods applied to moving-load responses

Developed economies depend on complex and extensive systems of infrastructure to maintain economic prosperity and quality of life. In recent years, the implementation of Structural health monitoring (SHM) systems on full-scale bridges has increased. The goal of these systems is to inform owners of the condition of structures, thereby supporting surveillance, maintenance and other management tasks.Data-driven methods, that involve tracking changes in signals only, are well-suited for analyzing measurements during continuous monitoring of structures. Also, information provided by the response of structures under moving loads is useful for condition assessment.This paper discusses the application of data-driven methods on moving-load responses in order to detect the occurrence and the location of damage. First, an approach for using moving-load responses as time series data is proposed. The work focuses on two data-driven methods – Moving principal component analysis (MPCA) and Robust regression analysis (RRA) – that have already been successful for damage detection during continuous monitoring. The performance of each method is assessed using data obtained by simulating the crossing of a point-load on a simple frame.

Combined Model-Free Data-Interpretation Methodologies for Damage Detection during Continuous Monitoring of Structures

Despite the recent advances in sensor technologies and data-acquisition systems, interpreting measurement data for structural monitoring remains a challenge. Furthermore, because of the complexity of the structures, materials used, and uncertain environments, behavioral models are difficult to build accurately. This paper presents novel model-free data-interpretation methodologies that combine moving principal component analysis (MPCA) with each of four regression-analysis methods—robust regression analysis (RRA), multiple linear analysis (MLR), support vector regression (SVR), and random forest (RF)—for damage detection during continuous monitoring of structures. The principal goal is to exploit the advantages of both MPCA and regression-analysis methods. The applicability of these combined methods is evaluated and compared with individual applications of MPCA, RRA, MLR, SVR, and RF through four case studies. Result showed that the combined methods outperformed noncombined methods in terms of damage detectability and time to detection.

Evaluating two model-free data interpretation methods for measurements that are influenced by temperature

Interpreting measurement data to extract meaningful information for damage detection is a challenge for continuous monitoring of structures. This paper presents an evaluation of two model-free data interpretation methods that have previously been identified to be attractive for applications in structural engineering: moving principal component analysis (MPCA) and robust regression analysis (RRA). The effect of three factors are evaluated: (a) sensor-damage location, (b) traffic loading intensity and (c) damage level, using two criteria: damage detectability and the time to damage detection. In addition, the effects of these three factors are studied for the first time in situations with and without removing seasonal variations through use of a moving average filter and an ideal low-pass filter. For this purpose, a parametric study is performed using a numerical model of a railway truss bridge. Results show that MPCA has higher damage detectability than RRA. On the other hand, RRA detects damages faster than MPCA. Seasonal variation removal reduces the time to damage detection of MPCA in some cases while the benefits are consistently modest for RRA.

Near infrared spectroscopy of urine proves useful for estimating ovulation in giant panda (Ailuropoda melanoleuca)

The usefulness of near infrared spectroscopy (NIRS) to monitor urine estrogen concentrations was studied in order to determine optimal timing for breeding captive female giant pandas. NIR spectra of daily urine samples from a female giant panda (Ailuropoda melanoleuca) were acquired in the period between March 1st and 25th, 2007 (n = 53). Estrone-3-glucuronide (E1G) concentrations in the samples were also measured by enzyme immunoassay (EIA). Transmittance spectra of all urine samples were obtained in the wavelength range from 1100 to 2432 nm (excluding the range from 1884 to 2012 nm) with sample thickness of 1 mm. Partial least square regression was applied to the spectra and good correlation was obtained between E1G concentration measured by EIA and predicted values by NIR (R2 = 0.94, SECV = 10.04 ng ml−1). The results of both soft-independent modeling of class analogy (SIMCA) and moving principal component analysis (MPCA) could detect the time changes in E1G concentration as measured by EIA (the Pearson's correlation coefficients between E1G concentration and the interclass distances of SIMCA or the index of MPCA were r = 0.64 and r = 0.81 respectively, P < 0.01). As for MPCA index, the index sharply dropped on March 24th corresponding to the decrease of the E1G concentration indicating ovulation. Finally, artificial insemination was performed for 3 consecutive days including the peak day, March 24th, and the female became pregnant. These results indicated that NIRS and the following MPCA analysis of the respective urine spectral data could detect the changes of urinary hormones during estrous cycle at a nanogram level. The NIRS can find the optimal timing for breeding quicker and easier than EIA, so this technique can be useful for captive breeding of this threatened species.

Comparison of multivariate statistical process monitoring methods with applications to the Eastman challenge problem

To improve the performance of multivariate statistical process control (MSPC), two advanced methods, moving principal component analysis (MPCA) and DISSIM, have been proposed. In MPCA and DISSIM, an abnormal operation can be detected by monitoring the directions of principal components and the dissimilarity index, respectively. Another important extension of MSPC was made with Multiscale PCA (MS-PCA). The present work investigates the characteristics of several statistical monitoring methods. The monitoring performance is compared with applications to simulated data obtained from a 2×2 process and the Tennessee Eastman process. The superiority of MPCA and DISSIM over the conventional methods comes from the fact that those methods focus on changes in the distribution of process data. Furthermore, the advantages of MPCA or DISSIM over the conventional MSPC and that of MS-PCA are combined, and new methods, termed MS-MPCA and MS-DISSIM, are proposed.

Comparison of statistical process monitoring methods: application to the Eastman challenge problem

Multivariate statistical process control (MSPC) has been successfully applied to chemical procesess. In order to improve the performance of fault detection, two kinds of advanced methods, known as moving principal component analysis (MPCA) and DISSIM, have been proposed. In MPCA and DISSIM, an abnormal operation can be detected by monitoring the directions of principal components (PCs) and the degree of dissimilarity between data sets, respectively. Another important extension of MSPC was made by using multiscale PCA (MS-PCA). In the present work, the characteristics of several monitoring methods are investigated. The monitoring performances are compared with using simulated data obtained from the Tennessee Eastman process. The results show that the advanced methods can outperform the conventional method. Furthermore, the advantage of MPCA and DISSIM over conventional MSPC (cMSPC) and that of the multiscale method are combined, and the new methods known as MS-MPCA and MS-DISSIM are proposed.