Dynamites of Naval Craft - System Identtftcatton 
Figure 9. The good agreement between the observed motion responses 
and the predicted trajectories is exhibited in this graph, and the final 
estimated parameters are listed below together with the actual (true) 
value used in the computer simulation that generated the observed 
data. 
Table 3 
Hydrofoil Estimated Parameters From Noisy Data 
A NLTCALSS PLSBeLe pee? on FOUN 
m= mC, 300-350 375 
m.. = mC, 900 892 
C, 50 491 
C, “ik6 . 167 
C, 2.2 2.053 
Ce OF ~521 
CL ZulS 1, 829 
Cy, 55 457 
Cy ce xs) 0 
Cio ee 0 
In this particular case the values of mC, and mC? 
(where m is the craft mass) did not change at all, throughout the 
total time of computation from the original initial guess value. The 
good estimates of the added masses, together with the probable 
small influence of the differences in values of added mass on the mo- 
tions, resulted in satisfactory trajectory tracking with these values. 
Similarly the nonlinear lift terms, represented by Cgand Cj, did not 
appear to have much influence either since the motions (and angles) 
were small, so that the data itself did not allow an adequate estimate 
of these values, just as in the case with the iteration method discussed 
previously. While there were some small differences in the estimated 
values of C, r Ce 5 CL and C. as compared to the actual values, 
1661 
