Space systems utilize mathematical models to predict the vibrational effects of loads due to launch, stage separation, maneuvering, docking, robotic motions, and reentry. After the mathematical model has been produced, it must be compared to experimental data and modified (or correlated) to match this data. The predictions of the model can only be trusted for design, safety checks, qualification of hardware, and control of the vehicle after this model correlation process has been accomplished. Very similar technologies can be used for monitoring the health of the structure (or identifying the damage accumulating in the structure). In this damage identification process, recently acquired experimental data is compared to previous data or a previously correlated model. Determinations are then made as to the existence, location, and extent of the damage as well as the effect of the damage on the safety and lifetime of the structure.

There are currently two major NASA systems with needs for model correlation and damage identification: the space shuttle orbiter vehicles and the International Space Station. The orbiter vehicles are regularly tested with the Shuttle Modal Inspection System (SMIS) to perform damage identification which has proven to be useful to monitor structural changes in subsurface components of the shuttle orbiter vehicles. Procedures have been developed to extract comparable data sets from the vehicles separated by a period of years, to assure the quality of the data, to understand the dynamics and vibration response of the orbiters, and rank order the results to reflect confidence in the data. However, new technologies are available which may enhance the results and increase the efficiency of the SMIS system. Hence, this project would be used to understand and transfer SMIS experience to the university and external engineers, while providing damage identification enhancements to the current SMIS approach.

There are three activities which will comprise this project's work on Orbiter-related technology. The first activity includes analysis of an existing data set from four composite plates of construction similar to orbiter control surfaces with three different types of engineered flaws. These plates are interesting in that they have a high density of measurements, require plate modeling, exercise a non-contact sensor (a scanning laser vibrometer), and simulate known damage in composites. Model correlation and damage ID of this data set has been one early focus of this ISSO project. This work has driven the development of techniques and algorithms for model correlation to process larger data sets.

The second orbiter-related activity includes analysis of a Shuttle Modal Inspection System (SMIS) data set from the space shuttle Endeavor. This data is currently being analyzed using recently developed modal identification packages to determine the utility of these tools to SMIS. Also novel damage identification tools will be applied to this data sot. Comparisons can be made between right and left wings as well as previous data sets from the same vehicle. This project is also intended to provide insight and experience into the SMIS procedures. Figure 1 shows the differences as a function of frequency in structural dynamics between the left and right wing of the space shuttle Endeavor. These differences are currently being characterized to determine if one wing has undergone a structural change.

The primary activity for the orbiter-related work will center around testing and analysis of a test article from the actual orbiter vehicle. This article includes the upper half of the orbiter tail as well as the upper set of rudders. Hence there will be three subsystems to model, test, correlate, and mate. Each substructure will be tested individually, followed by a test of the entire structure. Model correlation and damage detection algorithms for substructured systems will be evaluated and modified in this phase of the research. Currently, transportation of this test article from California is being arranged.

The space station is currently entering final production and assembly phases. Therefore, model correlation and damage ID technology needs to flow into NASA for application on this project. The space station will be one of the most difficult applications of model correlation and damage identification technology to date. The structure will be assembled in stages on-orbit over a period of several years and at each phase of the assembly a new model (which is an assembly of previous models) needs to be correlated. Several complicating factors also exist: all measurements must be performed on-orbit; the available sensors are extremely limited; the inputs are not optimized for this application; and very little time is available for testing. Several critical model correlation technologies are being developed as part of this ISSO project. Also, damage identification technology is under development to enhance, or in some cases, offset on-orbit inspection of the International Space Station. This is desirable as astronaut EVA time will be much better spent on activities other than inspection or maintenance. Hence, there is a need for a more space-station oriented test articles to deal with issues specific to that application. Some of these issues include: ultra-low frequency modes, changing mass properties, ambient environment testing, mixed sensor sets, multiple configurations, and control system interaction. A Generic Space Station test-bed is currently being developed to support this phase of the research.

Another technology which is necessary for both model correlation and damage identification is data acquisition and sensor technology. Both correlation and damage ID require a large amount of data to be acquired from the structure. Also the specific algorithms to perform correlation and damage ID must be tailored to take advantage of the type of data which is available (usually this is strain, displacement, velocity, or acceleration). Since NASA management has seen the need to acquire new sensors to enable these technologies, this project will integrate these unique sensors into the test program to the maximum extent possible. This not only assists NASA in selecting the proper sensors, but assures that the model correlation and damage ID procedures are compatible with the most advanced sensors available. An initial effort to define and demonstrate sensor concepts for the space station has taken place. This has produced two equipment demonstrations which have given the ISSO project an immediate impact on NASA planning for ground and on-orbit testing.

Current plans for the experimental activities related to the Orbiter Vertical Tail Structure and the Generic Space Station test article include both traditional and non-contact sensors such as laser Doppler velocimeter, laser ranging radar, video-based triangulation, and combinations of these. This will allow comparisons to be performed between novel and traditional sensors. Also, tests specific to a space station application of such sensors will be performed such as measurements at distance, temperature and vacuum effects, simulated reboost inputs, and signal processing issues. An initial comparative data set using the X-35 vehicle (which is similar to the orbiter application) has been acquired as part of an equipment demonstration. The demonstration proved that completely remote data acquisition operations from a long distance (approximately 80 feet) can be performed with laser technology. This part of the ISSO effort has the potential to initiate a near term space-based experiment.