The tank experiment resulted in the six parameters as enumerated below.

(1) Positions of the floating body model units in a period between the start of measurement and 2.0 seconds and in the period between 100 seconds and 102 seconds.

(2) The relative horizontal displacement, Cx-Dx, between position C of the joint surface of floating body model unit B and position D of the joint surface of floating body model unit A.

(3) The inclinational difference, DF-AC, between the inclination AC of the base floating body model unit B and the inclination DF of the floating body model unit A.

(4) The relative vertical displacement, Cz-Dz, between position C of the joint surface of the floating body model unit B and position D of the joint surface of the floating body model unit A.

(5) Wave height.

(6) Temperature of the joining jig of the SMA.

Positions of the floating body model units are indicated being shifted at an interval of 10 mm for each time in the z (vertical) direction. Experimental results for λ/L=2 as part of the results are shown in Figures 6-12.

　

　

　

　

　

　

　

　

　

(1) Comparing the position in the period from the start of measurement to 2.0 seconds with that of the period from 100 seconds to 102 seconds, it is understood that two floating body model units oscillate separately at the start of the measurement, while at 100 seconds from the start of measurement both floating body model units oscillate almost in unison.

(2) From the results of relative horizontal displacements Cx-Dx, it is understood that both floating body model units are drawn to each other by about 40 mm and finally they touch each other as viewed under any condition. From these facts, it has been found out that two floating body model units can be drawn to each other against wave forces by the restoring forces of the SMAs.

(3) From the experimental result revealing that the inclinational differences DF-AC concentrate around 0, it is understood that both floating body model units sustained nearly the same inclination. Further, in this experiment only the upper sections of floating body model units were joined, while their bottom sections were free to oscillate. It is assumed that by applying similar joining jigs to their bottom sections; the whole of both floating body model units can be integrated almost completely.

(4) The results of the relative vertical displacement Cz-Dz indicate a difference of around ± 5mm. This difference is considered to be caused by the inclinations of the floating body model units since position C is distanced horizontally about 55 mm from position D and also considerable vertical noise exists. Further, since the LED target of the position sensor, which was used in this study, measures the target position optically, reflected waves appearing from the water surface are considered to be another source of noise.

(5) The model is assumed to be 1/80 of an actual apparatus:

Incidentally, the unit of Mega-float (Phase I) is as follows: 100m x 20m x 2m.

　

　

From the above mentioned, we can develop a joining jig using SMAs as a means for an at-sea joining method in the construction of a large marine floating structure.

　

　

We deeply express our gratitude to Dr. Kunihiro Hoshino of the Ocean Engineering Division, the National Maritime Research Institute (NMRI), for his guidance about motion measurement. This study was supported by the Interdisciplinary General Joint Research Grant for Nihon University (Representative: K. Shingu).

　

Industrial Material Center, 1991. Handbook of Production and Processing of New Materials and New Metals. pp.535-552. (Written in Japanese).

　

Shimamune, S., M. Matsuura and S. Nakano, 1998. Afloat Joining Method of Very Large Floating Structures and Response Analysis with a Mathematical Modeling of Joint Mechanism, 14^th Ocean Engineering Symposium, The Society of Naval Architects of Japan, pp. 487-494. (Written in Japanese).

　

Technological Research Association of Mega-Float. 1996, 1997, 1998, 1999. Study Reports of Very Large Floating Type Marine Structures. (Written in Japanese).