76 M. Kreutz and D. J. Rixen Appendix: Results for high correlation of equivalent forces In this section, the results for the presented simulation model in section 3 are given, when only one operational force f1 is applied. It can be seen that this leads to a high correlation of the equivalent forces, see fig. 10. In this case the FRF-estimation (fig. 9) and the decoupling of the joint stiffness (fig. 10) give bad results. 0 0.5 1 1.5 2 2.5 3 10−8 10−7 10−6 f/kHz |Y|/ m N exact estimated coherence 0 0.2 0.4 0.6 0.8 1 Coherence Fig. 9 Magnitude of the assembly FRF|YAJB 2A,2A| from estimation and exact values in the case of only one operational force f1. Exemplary entry for excitation and response on the left side. 0 0.2 0.4 0.6 0.8 1 Correlation 0 0.5 1 1.5 2 2.5 3 105 107 109 ℜ(Z)/ N m 0 0.2 0.4 0.6 0.8 1 Correlation 0 0.5 1 1.5 2 2.5 3 103 106 109 f/kHz ℑ(Z)/ N m (a) Left joint: ZJ 2A,left,2A,left . 0 0.2 0.4 0.6 0.8 1 Correlation 0 0.5 1 1.5 2 2.5 3 106 108 1010 ℜ(Z)/ N m exact estimated fit correlation 0 0.2 0.4 0.6 0.8 1 Correlation 0 0.5 1 1.5 2 2.5 3 104 107 1010 f/kHz ℑ(Z)/ N m (b) Right joint: ZJ 2A,right,2A,right . Fig. 10 Dynamic joint stiffness ZJ 2A,2A in the case of only one operational force f1. This example highlights the importance of checking the correlation of the input signals for the FRF-estimation. In the presented method, these are the equivalent forces f eq 2A , see eq. (12).
RkJQdWJsaXNoZXIy MTMzNzEzMQ==