CMS- 0624985 Steven D. Glaser, University of California Brekeley)
This project will use acoustic emission methods to develop tools for inspection and monitoring, which allows direct detection of structural deterioration prior to failure. New interpretation algorithms and transducer designs will be developed to account for current difficulties, such as working in a heterogeneous media, using few sensors, and placing the sensors far from the damage source.
The project focuses on three problems for which no sensor currently exists. They are: i - development of a micro but hardened embeddible high-fidelity AE sensor sensitive to particle displacements; ii - development of a micro three-component high-fidelity AE sensor; iii - development of a high fidelity AE sensor specially for low acoustic impedance materials. The project will leverage close ties between the P.I. and Prof. Christian Grosse of Stuttgart University, Germany. We will use the large scale test beds and actual field monitoring that will be made for his EU project, and gain a wireless platform with onboard processing power, built specifically for AE.
In carrying out our project we have shed light on three important engineering and seismological problems, mostly based on our development of a novel acoustic emission sensor. Acoustic emissions (AE) are very small seismic events that travel through a material due to an internal cracking or frictional sliding – earthquakes on a small scale. We have increased our understanding of fundamental behaviors of earthquake faults through what we call nano-seismology; located growing cracking in structures like bridge decks; and furthered computer modeling of the effects of friction on blocks sliding down slopes during an earthquake. We have worked with researchers in Germany, Israel and Japan during this project, broadening our dissemination and extending our understanding of problems and solutions. 1. We perfected an absolutely calibrated high-fidelity nano-seismic sensor now commercially available to researchers and practitioners. This Glaser-type conical displacement sensor, previously crafted by hand, is now commercially available (KRN Services). The sensors are unique in that they extend the limits of what vibrations can be measured far beyond previous laboratory devices, and measure surface displacements as small as 1 picometer (1 trillionth of a meter) over the 20 kHz to 2 MHz frequency. We also developed a novel procedure to absolutely calibrate all AE and ultrasonic sensors. 2. These unique sensors allowed us to analyze laboratory earthquakes in a detail not previously possible. By studying earthquakes in a controlled laboratory environment we can identify and understand earthquake source mechanisms lost in the noise of field temblors. We call this new interpretation of acoustic emission nano-seismology. Using our novel measurement methods we were able to isolate the physical origins of laboratory earthquakes by determining fault force-time functions, similar to how low frequency earthquakes are found buried within tremor. Our laboratory earthquakes duplicate the tremor-like signals so interesting to seismologists as well as locally repeating earthquakes, depending on fault material properties modeled. 3. We demonstrated that for laboratory earthquakes there are distinct spectral changes in the recorded seismicity with increasing healing time, in addition to a modest increase in fault strength and stress drop. This finding is important for understanding repeating earthquake sequences and the effect of variations in earthquake recurrence time on the relationship between fault healing and earthquake generation. These laboratory results are seen in observations of repeating earthquake sequences on the Calaveras and San Andreas faults, which show similar spectral changes when recurrence time is perturbed by a nearby large earthquake. 4. We have derived a simplified method to locate active damage on plate-like structures. A critically important "plate-like" structure is a concrete bridge deck, and our work can point out where and when a crack occurs on the deck. We do this using a seismic method called beam-forming. The techniques avoids the to-date insurmountable problems of sensitivity, accurate waveform arrival determination, and the large number of sensors and wiring required for using traditional AE techniques. By arranging the sensors into a few small clusters the need for precise time synchronization between spatially distance sensors is also eliminated. These characteristics make beam-forming AE well suited for use with wireless sensor networks. 5. We also examining the role friction plays in preventing a rock from sliding rock down a slope during an earthquake. This result leads to much better methods to estimate seismic slope safety. We demonstrated that velocity-dependent friction should be integrated into dynamic rock slope analysis to obtain realistic results when strong ground motions are considered. The role of interface friction was studied by slow direct shear tests and rapid shaking table experiments. We presented an analytical solution for dynamic, single and double face sliding as well as a computer modeling solution using 3D-DDA.
