Various approaches with different point densities spanning a wide bandwidth Optical Accelerometer Laser Rigid Body Low Flexible Modes Higher Flexible Modes Figure 1 – Schematic of Different Transducers for Different Applications In many cases, collection of data points that have common geometric locations is very important especially when the modal test is being conducted for validation of a finite element model of the structure. But many times the collocation of all the measurement points may not be possible. This may result from inadequate communication between all the various test parties conducting the different tests required. But this may also result from the inability to make measurements at all the same locations. This can result from geometric access (or lack thereof) to certain locations on the structure; it may not be possible to locate a larger accelerometer case in the same location where a teardrop accelerometer can be easily placed. This can also result from “line of sight” issues related with lasers where measurement points may not be easily visible from the laser orientation for the test performed. In addition to these geometric problems that may occur, there can often be a mismatch from the total number of measurements that can be obtained from a practical standpoint. For instance, accelerometer measurements will be limited by the number of transducers available and the number of channels available on the data acquisition system. Of course roving patches of accelerometers can be performed (and the total number of measurement points may be several hundred) but the mass loading, time to conduct the test, consideration of time variance, etc. may render this impractical for many cases. In comparison, a laser may be set up to take a large number of measurements (on the order of a thousand or two thousand points) and that test can be performed but the measurements can only be made available on the surface where “line of sight” is not an issue; in addition the time to conduct the test, consideration of time variance, etc. also affect the practicality of the test. Once optical measuring approaches are also considered, the total number of measurement points may reach ten thousand or more. All in all, the range of total measurements possible creates a serious mismatch in the definition of points that describe the structural characteristics over such a wide frequency range with no guarantee that there will be points in common unless extreme care is exercised. But even with such care there may be sets of points that do not have corresponding locations that span the entire frequency ranges that are identified by the different tests that are possible. This paper presents an approach to merge all these different data sets into one complete, cohesive set of mode shapes that have a consistent set of scaled modal vectors that are extracted from multiple, different sets of data extracted from various types of data collection systems. 166
RkJQdWJsaXNoZXIy MTMzNzEzMQ==