Multiple Input Multiple Output (MIMO) testing requires the use of multiple shakers to excite the structure with uncorrelated signals supplied to each of the shakers. This requires that all the shakers be simultaneously attached to the structures for the modal test, as well as separate independent uncorrelated output sources from the data acquisition system.
While it seems possible to test the structure with one shaker and then move that shaker to all the different shaker locations, conducting the test in this manner typically results in an inconsistent set of frequency response functions. This can be due to a variety of reasons such as an inconsistent mass load distribution from roving transducers or environmental changes altering the structures mass and stiffness properties. When the different sets of data are combined, the resulting frequency responses are not as consistent as when all the data is collected simultaneously. The best measurement results have been achieved when all of the test data is acquired in a single snapshot, eliminating any issues related to time invariance or structure stationarity.
To illustrate this point, reciprocal frequency response measurements were taken with a single shaker moved between two different locations using a SISO approach. Each measurement was taken twice, first using a random and then a burst random signal to illustrate the differences. The results are shown below. Notice that the random signal has more variance and suffers from leakage even though a Hanning window was used. Clearly there is a difference in the two measurements shown; these two measurements should be exactly the same.
This measurement was repeated with a MIMO approach. Again a random excitation with a Hanning window and a burst random excitation were used. The variance using the random excitation can still be seen in the measurement even using the MIMO approach. Notice that the burst random MIMO approach provides the best measurement overall with a good frequency response where reciprocity is observed in the measurement.
图15 – SISO互易性FRF，随机激励（左侧） – 猝发随机（右侧）
图16 – MIMO互易性FRF，随机（左侧） – 猝发随机（右侧）