Showing papers by "Charles R. Farrar published in 2002"

A review of structural health monitoring literature 1996-2001

Abstract: Staff members at Los Alamos National Laboratory (LANL) produced a summary of the structural health monitoring literature in 1995. This presentation will summarize the outcome of an updated review covering the years 1996 2001. The updated review follows the LANL statistical pattern recognition paradigm for SHM, which addresses four topics: 1. Operational Evaluation; 2. Data Acquisition and Cleansing; 3. Feature Extraction; and 4. Statistical Modeling for Feature Discrimination. The literature has been reviewed based on how a particular study addresses these four topics. A significant observation from this review is that although there are many more SHM studies being reported, the investigators, in general, have not yet fully embraced the well-developed tools from statistical pattern recognition. As such, the discrimination procedures employed are often lacking the appropriate rigor necessary for this technology to evolve beyond demonstration problems carried out in laboratory setting.

A Review of Structural Health Review of Structural Health Monitoring Literature 1996-2001.

Abstract: Staff members at Los Alamos National Laboratory (LANL) produced a summary of the structural health monitoring literature in 1995. This presentation will summarize the outcome of an updated review covering the years 1996 - 2001. The updated review follows the LANL statistical pattern recognition paradigm for SHM, which addresses four topics: (1) Operational Evaluation; (2) Data Acquisition and Cleansing; (3) Feature Extraction; and (4) Statistical Modeling for Feature Discrimination. The literature has been reviewed based on how a particular study addresses these four topics. A significant observation from this review is that although there are many more SHM studies being reported, the investigators, in general, have not yet fully embraced the well-developed tools from statistical pattern recognition. As such, the discrimination procedures employed are often lacking the appropriate rigor necessary for this technology to evolve beyond demonstration problems carried out in laboratory setting.

Statistical Damage Classification Under Changing Environmental and Operational Conditions

Abstract: Stated in its most basic form, the objective of damage diagnosis is to ascertain simply if damage is present or not based on measured dynamic characteristics of a system to be monitored. In reality...

Classifying Linear and Nonlinear Structural Damage Using Frequency Domain ARX Models

Abstract: Structural health monitoring can be viewed as a problem in statistical pattern recognition involving operational evaluation, data cleansing, damage identification, and life prediction. In damage id...

Statistical damage classification using sequential probability ratio tests.

Abstract: The primary objective of damage detection is to ascertain with confidence if damage is present or not within a structure of interest. In this study, a damage classification problem is cast in the context of the statistical pattern recognition paradigm. First, a time prediction model, called an autoregressive and autoregressive with exogenous inputs (AR-ARX) model, is fit to a vibration signal measured during a normal operating condition of the structure. When a new time signal is recorded from an unknown state of the system, the prediction errors are computed for the new data set using the time prediction model. When the structure undergoes structural degradation, it is expected that the prediction errors will increase for the damage case. Based on this premise, a damage classifier is constructed using a sequential hypothesis testing technique called the sequential probability ratio test (SPRT). The SPRT is one form of parametric statistical inference tests, and the adoption of the SPRT to damage detection problems can improve the early identification of conditions that could lead to performance degradation and safety concerns. The sequential test assumes a probability distribution of the sample data sets, and a Gaussian distribution of the sample data sets is often used. This assumption, however, might impose potentially misleading behavior on the extreme values of the data, i.e. those points in the tails of the distribution. As the problem of damage detection specifically focuses attention on the tails, the assumption of normality is likely to lead the analysis astray. To overcome this difficulty, the performance of the SPRT is improved by integrating extreme values statistics, which specifically models behavior in the tails of the distribution of interest into the SPRT.

Structural health monitoring using wireless sensing systems with embedded processing

Abstract: A damage detection system was developed with commercially available wireless sensors. Statistical process control methods were used to monitor the correlation of vibration data from two accelerometers mounted across a joint. Changes in correlation were used to detect damage to the joint. All data processing was done remotely on a microprocessor integrated with the wireless sensors to allow for the transmission of a simple damaged or undamaged status for each monitored joint. Additionally, a portable demonstration structure was developed to showcase the capabilities of the damage detection system to monitor joint failure in real time.

Damage detection in mechanical structures using extreme value statistics

Abstract: The first and most important objective of any damage identification algorithms is to ascertain with confidence if damage is present or not. Many methods have been proposed for damage detection based on ideas of novelty detection founded in pattern recognition and multivariate statistics. The philosophy of novelty detection is simple. Features are first extracted from a baseline system to be monitored, and subsequent data are then compared to see if the new features are outliers, which significantly depart from the rest of population. In damage diagnosis problems, the assumption is that outliers are generated from a damaged condition of the monitored system. This damage classification necessitates the establishment of a decision boundary. Choosing this threshold value is often based on the assumption that the parent distribution of data is Gaussian in nature. While the problem of novelty detection focuses attention on the outlier or extreme values of the data i.e. those points in the tails of the distribution, the threshold selection using the normality assumption weighs the central population of data. Therefore, this normality assumption might impose potentially misleading behavior on damage classification, and is likely to lead the damage diagnosis astray. In this paper, extreme value statistics is integrated with the novelty detection to specifically model the tails of the distribution of interest. Finally, the proposed technique is demonstrated on simulated numerical data and time series data measured from an eight degree-of-freedom spring-mass system.

Consideration of environmental and operational variability for damage diagnosis

Abstract: Damage diagnosis is a problem that can be addressed at many levels. Stated in its most basic form, the objective is to ascertain simply if damage is present or not. In a statistical pattern recognition paradigm of this problem, the philosophy is to collect baseline signatures from a system to be monitored and to compare subsequent data to see if the new 'pattern' deviates significantly from the baseline data. Unfortunately, matters are seldom as simple as this. In reality, structures will be subjected to changing environmental and operational conditions that will affect measured signals. In this case, there may be a wide range of normal conditions, and it is clearly undesirable to signal damage simply because of a change in the environment. In this paper, a unique combination of time series analysis, neural networks, and statistical inference techniques is developed for damage classification explicitly taking into account these natural variations of the system in order to minimize false positive indication of true system changes.

Application of frequency domain ARX features for linear and nonlinear structural damage identification

Abstract: Many different vibration-based dynamic input-output and output only data features have been used to identify structural damage and assess structural integrity. Since structural damage introduces linear or nonlinear variations into all of these features, all of them might give positive indications of damage but may not distinguish between linear or nonlinear types of damage. This information can sometimes be used to more reliably diagnose damage by first, helping to distinguish between damage, which is inherently nonlinear, and healthy nonlinearities in a baseline structure; and second, serving as an absolute damage prognosis indicator which, together with prior information about the structural mechanics, determined the degree to which a structure is damaged. A set of potential features that distinguish between linear and nonlinear damage are discussed here. These features are auto-regressive exogenous dynamic transmissiblity model coefficients in the frequency domain. The auto-regressive coefficients are used to characterize the nonlinear nature of damage states and the exogenous coefficients are used to characterize the linear nature of such states. After reviewing the theoretical development of this data model, experimental measurements from a three-story test structure are analyzed using these model coefficients and statistical features are extracted from the coefficients. By using two complementary features, a better indication of the severity of damage is obtained.© (2002) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

Experimental modal analysis and damage detection in a simulated three story building

Abstract: This paper is a continuation of a study entitled Damage Detection in Building Joints by Statistical Analysis [1] in which accelerometer data were acquired from a simulated three-story building driven by an electrodynamic shaker attached to the base of the structure. Joint damage and environmental conditions were simulated and data were collected systematically for comparison. Operational variability was introduced by changing the shaker input amplitudes and frequency range. An AutoRegressive model with Exogenous Inputs (ARX) was fit to the collected data and the standard deviations of the residual errors between ARX predictions and the measured data were used as the damage sensitive features. A Sequential Probability Ratio Test (SPRT) was used to make damage detection decisions. The test produced promising results, but was shown to be sensitive to the operational and environmental variability. This investigation was conducted as part of a conceptual study to demonstrate the feasibility of detecting damage in structural joints caused by seismic excitation.

Utilizing the sequential probability ratio test for building joint monitoring

Abstract: In this application of the statistical pattern recognition paradigm, a prediction model of a chosen feature is developed from the time domain response of a baseline structure. After the model is developed, subsequent feature sets are tested against the model to determine if a change in the feature has occurred. In the proposed statistical inference for damage identification there are two basic hypotheses; (1) the model can predict the feature, in which case the structure is undamaged or (2) the model can not accurately predict the feature, suggesting that the structure is damaged. The Sequential Probability Ratio Test (SPRT) develops a statistical method that quickly arrives at a decision between these two hypotheses and is applicable to continuous monitoring. In the original formulation of the SPRT algorithm, the feature is assumed to be Gaussian and thresholds are set accordingly. It is likely, however, that the feature used for damage identification is sensitive to the tails of the distribution and that the tails may not necessarily be governed by Gaussian characteristics. By modeling the tails using the technique of Extreme Value Statistics, the hypothesis decision thresholds for the SPRT algorithm may be set avoiding the normality assumption. The SPRT algorithm is utilized to decide if the test structure is undamaged or damaged and which joint is exhibiting the change.

Laboratory and Field Validation of a Wireless Sensing Unit Design for Structural Monitoring

Abstract: There exists a clear need to monitor the performance of civil structures over their operational lives. Current commercial monitoring systems suffer from various technological and economic limitations that prevent widespread adoption. The wires used to route measurements from system sensors to the centralized data server represent one of the greatest limitations since they are physically vulnerable and expensive from an installation and maintenance standpoint. In lieu of cables, the introduction of low-cost wireless communications is proposed. The result is the design of a prototype wireless sensing unit that can serve as the fundamental building block of wireless modular monitoring systems (WiMMS). The prototype unit is validated with a series of tests conducted in the laboratory and the field. In particular, the Alamosa Canyon Bridge is employed to serve as a full-scale benchmark structure to validate the performance of the wireless sensing unit in the field.

Damage detection using frequency domain ARX models and extreme value statistics

Abstract: Structural health monitoring (SHM) is fast becoming a field of great importance as engineers seek for new ways to ensure the safety of structures throughout their designed lifetime. Current methods for analyzing the dynamic response of structures often use standard frequency response functions to model linear system input/output relationships. However, these functions do not account for the nonlinear response of a system, which damage often introduces. In this study, an auto-regressive model with exogenous inputs (ARX) in the frequency domain is used to extract damage sensitive features, explicitly considering the nonlinear effect in the frequency domain. Furthermore, because of the non-Gaussian nature of the extracted features, extreme value statistics (EVS) is employed to develop a robust damage classifier. The applicability of the ARX model combined with EVS to nonlinear damage detection is demonstrated using vibration data obtained from a laboratory experiment of a three-story building model.

Structural health monitoring system design using finite element analysis

Abstract: The project described in this report was performed to couple experimental and analytical techniques in the field of structural health monitoring and damage identification. To do this, a finite element model was constructed of a simulated three-story building used for damage identification experiments. The model was used in conjunction with data from the physical structure to research damage identification algorithms. Of particular interest was modeling slip in joints as a function of bolt torque and predicting the smallest change of torque that could be detected experimentally. After being validated with results from the physical structure, the model was used to produce data to test the capabilities of damage identification algorithms. This report describes the finite element model constructed, the results obtained, and proposed future use of the model.

Experimental demonstration of local attractor variance as a damage indication feature

Abstract: In vibration-based structural damage assessment, the proper selection and extraction of appropriate features remains an important component of the overall process. These features ideally must be both sensitive to a particular damage scenario and insensitive to ambient influences within some statistical confidence. A novel approach has been introduced in previous works which described the local attractor variance ratio as a state-space-based feature (in combination with appropriate chaotic excitation input), which was shown in numerical experiments to be superior to modal-based features in both detecting and quantifying damage under several scenarios. In this work, we apply this new technique to a simple five-degree-of-freedom experimental system, where damage is induced through a spring stiffness change. In addition, we present a variation of this feature, both of which seek to quantify states space distortion with damage.

Singularity detection using holder exponent.

Abstract: A signal processing technique called Holder exponent is presented to detect the presence of a discontinuity and when the discontinuity occurs in a dynamic signal. Wavelet transforms are incorporated with the Holder exponent to capture the time varying nature of discontinuities, and a classification procedure is developed to quantify when changes in the Holder exponent are significant, The proposed Holder exponent analysis is applied to acceleration response of a mechanical system with a rattling internal part. The experimental results demonstrate the effectiveness of the Holder exponent for identifying certain types of events that introduce discontinuities into the measured dynamic response data.

Los Alamos dynamics summer school

Abstract: A unique summer educational program focusing on engineering structural dynamics has been developed and implemented at Los Alamos National Laboratory (LANL). The purpose of this summer school is to expose a select group of students to the broad field of engineering dynamics with the hopes that they will be motivated to pursue this area of research in their graduate studies. The summer school activities included: 1) lectures on various engineering topics such as computational structural dynamics, experimental modal analysis, random vibrations, signal processing, etc.; 2) a distinguished lecturer series in which prominent guest lecturers give talks about cutting edge research in structural dynamics; 3) field trips; and 4) an eight week project having both an analytical and an experimental component. LANL is motivated to pursue such an education program because it represents a proactive approach to recruiting new hires that allows time to better assess candidates' technical abilities and their abilities to perform in a team environment. This article will discuss the details of the program and its implementation.