Phong B. Dao

Cointegration approach for temperature effect compensation in Lamb-wave-based damage detection

Lamb waves are often used in smart structures with integrated, low-profile piezoceramic transducers for damage detection. However, it is well known that the method is prone to contamination from a variety of interference sources including environmental and operational conditions. The paper demonstrates how to remove the undesired temperature effect from Lamb wave data. The method is based on the concept of cointegration that is partially built on the analysis of the non-stationary behaviour of time series. Instead of directly using Lamb wave responses for damage detection, two approaches are proposed: (i) analysis of cointegrating residuals obtained from the cointegration process of Lamb wave responses, (ii) analysis of stationary characteristics of Lamb wave responses before and after cointegration. The method is tested on undamaged and damaged aluminium plates exposed to temperature variations. The experimental results show that the method can: isolate damage-sensitive features from temperature variations, detect the existence of damage and classify its severity.

Data normalisation for Lamb wave–based damage detection using cointegration: A case study with single- and multiple-temperature trends

This article presents an application of the cointegration technique for temperature effect removal (i.e. data normalisation) in Lamb wave–based damage detection. The method is based on the concept of stationarity of time series. Analysis of cointegrating residuals and stationary statistical characteristics – before and after the cointegration process – are used for damage detection. The method is validated using Lamb wave data from undamaged and damaged aluminium plates instrumented with low-profile, surface-bonded piezoceramic transducers. Two temperature variation cases (single- and multiple-temperature trends) are investigated. The experimental results show that the cointegration process can remove undesired temperature effects and accurately detect damage.

Nonlinear Cointegration Approach for Condition Monitoring of Wind Turbines

Monitoring of trends and removal of undesired trends from operational/process parameters in wind turbines is important for their condition monitoring. This paper presents the homoscedastic nonlinear cointegration for the solution to this problem. The cointegration approach used leads to stable variances in cointegration residuals. The adapted Breusch-Pagan test procedure is developed to test for the presence of heteroscedasticity in cointegration residuals obtained from the nonlinear cointegration analysis. Examples using three different time series data sets—that is, one with a nonlinear quadratic deterministic trend, another with a nonlinear exponential deterministic trend, and experimental data from a wind turbine drivetrain—are used to illustrate the method and demonstrate possible practical applications. The results show that the proposed approach can be used for effective removal of nonlinear trends form various types of data, allowing for possible condition monitoring applications.

Stationarity-Based Approach for the Selection of Lag Length in Cointegration Analysis Used for Structural Damage Detection

Cointegration has been recently brought to structural health monitoring SHM as a new methodology for dealing with the problem of environmental and/or operational variability in monitored structures. However, it is well known that the choice of lag length in cointegration analysis has a strong influence on damage detection results. The article presents a new approach for optimal lag length selection in cointegration analysis used for structural damage detection. This new method is based on stationarity analysis of data representing undamaged condition. The proposed method is validated using Lamb wave data under the effects of temperature variations and vibroacoustic data obtained from nonlinear vibroacoustic modulation experiments with different low-frequency vibration or modal excitations. The results demonstrate the effectiveness of the method for structural damage detection based on SHM data heavily affected by environmental or operational conditions.

Operational Condition Monitoring of Wind Turbines Using Cointegration Method

This paper presents a cointegration-based method for condition monitoring and fault detection of wind turbines. The proposed method is based on the residual-based control chart approach. The main idea is that cointegration is a property of some sets of nonstationary time series where a linear combination of the nonstationary series can produce a stationary residual. Then the stationarity of cointegration residuals can be used in a control chart as a potentially effective damage feature. The method is validated using the experimental data acquired from a wind turbine drivetrain with a nominal power of 2 MW under varying environmental and operational conditions. Two known abnormal problems of the wind turbine are used to illustrate the fault detection ability of the method. A cointegration-based procedure is performed on six process parameters of the wind turbine where data trends have nonlinear characteristics. Analysis of cointegration residuals—obtained from cointegration process of wind turbine data—is used for operational condition monitoring and fault/abnormal detection. The results show that the proposed method can effectively monitor the wind turbine and reliably detect abnormal problems.

Cointegration and wavelet-analysis-based approach for Lamb-wave-based structural damage detection

The paper demonstrates how to remove the undesired temperature effect from Lamb wave data in order to detect structural damage more precisely and reliably. The method used is based on the cointegration technique and wavelet analysis. The former is built on the analysis of non-stationary behaviour whereas the latter brings the concept of multiresolution decomposition of time series. Instead of directly using Lamb wave data for damage detection, three approaches are used: (1) analysis of the variance of wavelet coefficients of Lamb wave responses before cointegration, (2) analysis of the cointegrating residuals obtained from the cointegration process of Lamb wave responses, and (3) analysis of the variance of wavelet coefficients of Lamb wave responses after cointegration. These approaches are tested on undamaged and damaged aluminium plates that have been exposed to temperature variations. The experimental results show that the first approach still exhibits temperature variability and damage cannot be detected. In contrast the second and third approaches can isolate damage-sensitive features from temperature variations, detect the existence of damage and classify its severity.

Damage Detection Using Cointegration Technique and Wavelet Analysis of the Post-Cointegrated Lamb Waves

This paper presents an application of Lamb-wave-based damage detection under varying temperature conditions. The method used is based on the cointegration technique and wavelet analysis that are partially built on the analysis of non-stationary behaviour and multi-resolution decomposition of time series, respectively. Instead of directly using Lamb wave data for damage detection, two approaches are used: (1) analysis of cointegrating residuals obtained from the cointegration process of Lamb wave responses and (2) analysis of stationary characteristics of the multi-level wavelet decomposed cointegrating residuals. These two approaches are tested on undamaged and damaged aluminium plates exposed to temperature variations. The experimental results show that the method can isolate damage-sensitive features from the temperature effect and reliably detect damage.