Faculty and Student Teams Program

Project Descriptions

Los Alamos National Laboratory
Damage Detection in High Performance Ship Structures

Requesting applications from science and engineering faculty members at institutions serving students underrepresented in science, engineering, mathematics and technology

Project Description

The process of implementing a damage detection strategy for aerospace, civil and mechanical engineering infrastructure is referred to as structural health monitoring (SHM). Here damage is defined as changes to the material and/or geometric properties of these systems, including changes to the boundary conditions and system connectivity, which adversely affect the system’s performance.  The SHM process involves the observation of a system over time using periodically sampled dynamic response measurements from an array of sensors, the extraction of damage-sensitive features from these measurements, and the statistical analysis of these features to determine the current state of system health.  For long term SHM, the output of this process is periodically updated information regarding the ability of the structure to perform its intended function in light of the inevitable aging and degradation resulting from operational environments.  After extreme events, such as earthquakes or blast loading, SHM is used for rapid condition screening and aims to provide, in near real time, reliable information regarding the integrity of the structure.
 
Our approach is to address the SHM problem in the context of a statistical pattern recognition paradigm [1]. In this paradigm, the process can be broken down into four parts: (1) Operational Evaluation, (2) Data Acquisition and Cleansing, (3) Feature Extraction and Data Compression, and (4) Statistical Model Development for Feature Discrimination.  When one attempts to apply this paradigm to data from real world structures, it quickly becomes apparent that the ability to cleanse, compress, normalize and fuse data to account for operational and environmental variability is a key implementation issue when addressing Parts 2-4 of this paradigm.  These processes can be implemented through hardware or software and, in general, some combination of these two approaches will be used.

This project will apply statistical analysis and machine learning tools to strain and acceleration data acquired during sea trials of a high performance prototype navy ship with the intention of identifying novelties in these sensor readings that might be indicative of damage.  This application is particularly challenging because of the physical size of these structures, the widely varying and often extreme operational and environmental conditions associated with these sea trials, lack of data from known damage conditions, limited sensing that was not designed specifically for SHM, and the management of the vast amounts of data that can be collected during a mission even with this limited sensing.  The scope of this project involves algorithm development and subsequent verification and validation of these algorithms.  Tasks include:

1. Time series modeling.  Develop linear and nonlinear time series models of measured data.  Analyze changes in the residual errors of these predictive models for indications of change in the sensor readings.  First validate these models on data from a laboratory test structure that LANL has developed for SHM studies.  Then apply them to sensor data from the sea trials.
2. Statistical classification procedures.  Develop statistical classification procedures to identify sensor reading outliers based on the time series models developed in Task 1. Validate these procedures on data from a laboratory test structure that LANL has developed for SHM studies.  Then apply these classification procedures to sensor data from the sea trials.
3. Develop data management strategies.  The sea trials generate tremendous amounts of data over relatively short time intervals (60+ sensors sampling at 100Hz for several hours).  The team will be asked to develop strategies for compressing and fusing these data in a manner that still allows for robust damage detection.

The Engineering Institute at Los Alamos National Laboratory is seeking students and faculty to assist in the analysis of these data and development of outlier detections algorithms that can eventually be deployed on ships for near-real-time monitoring during sea trials.

Qualifications of the Ideal Candidates

Faculty: Ph.D. in Computer Science, Mathematics and Statistics, Aerospace, Mechanical, Civil or Electrical Engineering with experience in machine learning, statistical analysis and signal processing.  Works well both in collaborative environment with researchers and students.  Experience teaching and mentoring students.  Currently teaches and collaborates with students in his/her field. Possess good written and verbal communication skills.  Willing to work at LANL for extended period during the summer.  

Students: Working toward s a BS/BA in engineering, computer science or mathematics and statistics degree with interest in working on a project involving  machine learning, statistical analysis and/or signal processing as applied to mechanical response measurements.  Works well both in collaborative environment with researchers, faculty and other students.  Possess good written and verbal communication skills.  Willing to work at LANL for extended period during the summer.  Most importantly, is interested in learning new technology that is beyond what they have had in their formal classroom environment.

For More Information Contact:
Scott Robbins
Education and Post-doc Office
P.O. Box 1663, MS M709
Los Alamos National Lab
Los Alamos NM 87544
srobbins@lanl.gov
505 667 3639 (O)
505 665 6871 (F)

[1] C. R. Farrar, S. W. Doebling and D. A. Nix “Vibration-Based Structural Damage Identification” Philosophical Transactions of the Royal Society: Mathematical, Physical & Engineering Sciences, 2001, 359(1778) pp. 131 – 149
[2] C. R. Farrar and K. Worden “An Introduction to Structural Health Monitoring,” Philosophical Transactions of the Royal Society A, 365, February 2007, pp. 303-315.

Support and Financial Commitments

See Financial Information.