18 Developing a Correlation Criterion (SpaceMAC) for Repeated and Pseudo-repeated Modes 199 Fig. 18.13 Analytical modeshape corresponding to experimental mode at 1328.05 Hz Fig. 18.14 Analytical modeshape corresponding to experimental mode at 1328.77 Hz Table 18.3 spaceMAC3 values Number Frequency1 (Hz) Frequency2 (Hz) Frequency3 (Hz) spaceMAC 1 362.337 363.648 761.104 0.624 2 362.337 363.648 764.163 0.615 3 362.337 363.648 1223.011 0.050 4 362.337 761.104 764.163 0.049 5 362.337 761.104 2322.186 0.046 6 362.337 363.648 2322.186 0.647 7 362.337 363.648 2322.555 0.631 to eliminate those cases as being possible repeated roots cases based upon frequency alone. It has been observed that MAC values between two different experimental results belonging to the same experimental setup were close to unity even for repeated modes. This observation was contrary to expectation. It was the same for different force functions used, signal processing methods and number of reference points. When compared to another method for correlation of closely spaced modes, the results obtained with spaceMAC were clearer as long as an increasing order of multiplicity was followed. This was largely because spaceMAC works with correlation of two modal vector spaces whereas the other method compares a space with a vector. The effect of variation of MAC threshold was also explored. It has been observed that it did not directly affect the spaceMAC calculation for a particular set of modes.
RkJQdWJsaXNoZXIy MTMzNzEzMQ==