Structural Health Monitoring of Research-Scale Wind Turbine Blades

Taylor, Stuart G.,Farinholt, Kevin M.,Park, Gyu Hae,Farrar, Charles R.,Todd, Michael D.,Lee, Jung Ryul Trans Tech Publications, Ltd. 2013 Key Engineering Materials Vol.558 No.-

<P>This paper presents ongoing work by the authors to implement real-time structural health monitoring (SHM) systems for operational research-scale wind turbine blades. The authors have been investigating and assessing the performance of several techniques for SHM of wind turbine blades using piezoelectric active sensors. Following a series of laboratory vibration and fatigue tests, these techniques are being implemented using embedded systems developed by the authors. These embedded systems are being deployed on operating wind turbine platforms, including a 20-meter rotor diameter turbine, located in Bushland, TX, and a 4.5-meter rotor diameter turbine, located in Los Alamos, NM. The SHM approach includes measurements over multiple frequency ranges, in which diffuse ultrasonic waves are excited and recorded using an active sensing system, and the blades global ambient vibration response is recorded using a passive sensing system. These dual measurement types provide a means of correlating the effect of potential damage to changes in the global structural behavior of the blade. In order to provide a backdrop for the sensors and systems currently installed in the field, recent damage detection results for laboratory-based wind turbine blade experiments are reviewed. Our recent and ongoing experimental platforms for field tests are described, and experimental results from these field tests are presented. <I>LA-UR-12-24691</I>.</P>

Multi-scale wireless sensor node for health monitoring of civil infrastructure and mechanical systems

Taylor, Stuart G.,Farinholt, Kevin M.,Park, Gyuhae,Todd, Michael D.,Farrar, Charles R. Techno-Press 2010 Smart Structures and Systems, An International Jou Vol.6 No.5

This paper presents recent developments in an extremely compact, wireless impedance sensor node (the WID3, $\underline{W}$ireless $\underline{I}$mpedance $\underline{D}$evice) for use in high-frequency impedance-based structural health monitoring (SHM), sensor diagnostics and validation, and low-frequency (< ~1 kHz) vibration data acquisition. The WID3 is equipped with an impedance chip that can resolve measurements up to 100 kHz, a frequency range ideal for many SHM applications. An integrated set of multiplexers allows the end user to monitor seven piezoelectric sensors from a single sensor node. The WID3 combines on-board processing using a microcontroller, data storage using flash memory, wireless communications capabilities, and a series of internal and external triggering options into a single package to realize a truly comprehensive, self-contained wireless active-sensor node for SHM applications. Furthermore, we recently extended the capability of this device by implementing low-frequency analog-to-digital and digital-to-analog converters so that the same device can measure structural vibration data. The compact sensor node collects relatively low-frequency acceleration measurements to estimate natural frequencies and operational deflection shapes, as well as relatively high-frequency impedance measurements to detect structural damage. Experimental results with application to SHM, sensor diagnostics and low-frequency vibration data acquisition are presented.

Multi-scale wireless sensor node for health monitoring of civil infrastructure and mechanical systems

Stuart G. Taylor,Kevin M. Farinholt,Gyuhae Park,Michael D. Todd,Charles R. Farrar 국제구조공학회 2010 Smart Structures and Systems, An International Jou Vol.6 No.6

This paper presents recent developments in an extremely compact, wireless impedance sensor node (the WID3, Wireless Impedance Device) for use in high-frequency impedance-based structural health monitoring (SHM), sensor diagnostics and validation, and low-frequency (< ~1 kHz) vibration data acquisition. The WID3 is equipped with an impedance chip that can resolve measurements up to 100 kHz, a frequency range ideal for many SHM applications. An integrated set of multiplexers allows the end user to monitor seven piezoelectric sensors from a single sensor node. The WID3 combines on-board processing using a microcontroller, data storage using flash memory, wireless communications capabilities, and a series of internal and external triggering options into a single package to realize a truly comprehensive, self-contained wireless active-sensor node for SHM applications. Furthermore, we recently extended the capability of this device by implementing low-frequency analog-to-digital and digital-to-analog converters so that the same device can measure structural vibration data. The compact sensor node collects relatively low-frequency acceleration measurements to estimate natural frequencies and operational deflection shapes, as well as relatively high-frequency impedance measurements to detect structural damage. Experimental results with application to SHM, sensor diagnostics and low-frequency vibration data acquisition are presented.

Active-sensing platform for structural health monitoring: Development and deployment

Taylor, Stuart G,Raby, Eric Y,Farinholt, Kevin M,Park, Gyuhae,Todd, Michael D SAGE Publications 2016 Structural health monitoring Vol.15 No.4

<P>Embedded sensing for structural health monitoring is a rapidly expanding field, propelled by algorithmic advances in structural health monitoring and the ever-shrinking size and cost of electronic hardware necessary for its implementation. Although commercial systems are available to perform the relevant tasks, they are usually bulky and/or expensive because of their high degree of general utility to a wider range of applications. As a result, multiple separate devices may be required in order to obtain the same results that could be obtained with a structural health monitoring-specific device. This work presents the development and deployment of a versatile, Wireless Active-Sensing Platform, designed for the particular needs of embedded sensing for multi-scale structural health monitoring. The Wireless Active-Sensing Platform combines a conventional data acquisition ability to record voltage output (e.g. from strain or acceleration transducers) with ultrasonic guided wave-based active-sensing, and a seamlessly integrated impedance measurement mode, enabling impedance-based structural health monitoring and piezoelectric sensor diagnostics to reduce the potential for false positives in damage identification. The motivation, capabilities, and hardware design for the Wireless Active-Sensing Platform are reviewed, and three deployment examples are presented, each demonstrating an important aspect of embedded sensing for structural health monitoring.</P>

Fatigue crack detection performance comparison in a composite wind turbine rotor blade

Taylor, Stuart G,Park, Gyuhae,Farinholt, Kevin M,Todd, Michael D SAGE Publications 2013 Structural health monitoring Vol.12 No.3

<P>This article presents the detection performance results for multiple detectors or test statistics, using different active-sensing hardware systems in identifying the presence and location of a through-thickness fatigue crack in a 9-m composite wind turbine rotor blade. The rotor blade underwent ~8.5 million cycles of fatigue loading until failure, when a 30-cm-long crack surfaced on the leading edge portion of the blade’s transitional root area. The rotor blade was cantilevered on a 7-ton test stand and excited using a hydraulically actuated resonant excitation system, which drove the rotor blade at its first natural frequency. Through the course of the test, data were collected using two distinct types of acquisition hardware: one designed for ultrasonic-guided wave interrogation and the other for diffuse wave field interrogation. This article presents the fatigue crack detection performance results for several hardware and test statistic combinations.</P>

Analysis of Time Domain Active Sensing Data from CX-100 Wind Turbine Blade Fatigue Tests for Damage Assessment

Mijin Choi,Hwee Kwon Jung,Stuart G. Taylor,Kevin M. Farinholt,Jung-Ryul Lee,Gyuhae Park 한국비파괴검사학회 2016 한국비파괴검사학회지 Vol.36 No.2

This paper presents the results obtained using time-series-based methods for structural damage assessment. The methods are applied to a wind turbine blade structure subjected to fatigue loads. A 9 m CX-100 (carbon experimental 100 kW) blade is harmonically excited at its first natural frequency to introduce a failure mode. Consequently, a through-thickness fatigue crack is visually identified at 8.5 million cycles. The time domain data from the piezoelectric active-sensing techniques are measured during the fatigue loadings and used to detect incipient damage. The damage-sensitive features, such as the first four moments and a normality indicator, are extracted from the time domain data. Time series autoregressive models with exogenous inputs are also implemented. These features could efficiently detect a fatigue crack and are less sensitive to operational variations than the other methods.

Analysis of Time Domain Active Sensing Data from CX-100 Wind Turbine Blade Fatigue Tests for Damage Assessment

Choi, Mijin,Jung, Hwee Kwon,Taylor, Stuart G.,Farinholt, Kevin M.,Lee, Jung-Ryul,Park, Gyuhae The Korean Society for Nondestructive Testing 2016 한국비파괴검사학회지 Vol.36 No.2

This paper presents the results obtained using time-series-based methods for structural damage assessment. The methods are applied to a wind turbine blade structure subjected to fatigue loads. A 9 m CX-100 (carbon experimental 100 kW) blade is harmonically excited at its first natural frequency to introduce a failure mode. Consequently, a through-thickness fatigue crack is visually identified at 8.5 million cycles. The time domain data from the piezoelectric active-sensing techniques are measured during the fatigue loadings and used to detect incipient damage. The damage-sensitive features, such as the first four moments and a normality indicator, are extracted from the time domain data. Time series autoregressive models with exogenous inputs are also implemented. These features could efficiently detect a fatigue crack and are less sensitive to operational variations than the other methods.