Project/Area Number |
04650090
|
Research Category |
Grant-in-Aid for General Scientific Research (C)
|
Allocation Type | Single-year Grants |
Research Field |
材料力学
|
Research Institution | Kyushu Institute of Technology |
Principal Investigator |
CHEN Dai-heng Kyushu Institute of Technology, Faculty of Computer Science and Systems Engineering, Associate professor, 情報工学部, 助教授 (90217266)
|
Project Period (FY) |
1992 – 1993
|
Project Status |
Completed (Fiscal Year 1993)
|
Budget Amount *help |
¥2,100,000 (Direct Cost: ¥2,100,000)
Fiscal Year 1993: ¥600,000 (Direct Cost: ¥600,000)
Fiscal Year 1992: ¥1,500,000 (Direct Cost: ¥1,500,000)
|
Keywords | Inverse problem / Non-destructive examination / Detection of crack / Body force method / Gradient search / 逆間題 / 匂配探索法 |
Research Abstract |
In this research, the location and the size of the two-and three- dimensional surface crack and the loading sress were detected by the data of the strains measured around the region of the crack. Based on the basic concept of body force method that the strain field due to a crack can be represented by the embedded body force dublets, the problem is reduced to determination of the magnitude and location of the force doublets so as to produce the same strain field as measured around the crack. The problem is solved in two main steps : (1) select the starting point of gradient search, and (2) search for the final solution by using the method of gradient seach from the selected point. To select the starting point, the strain field around the crack is represented by the force doublets concentrated at one point. This approximation enables us to search for the starting point in a subspace with smaller dimensions. To search for the final solution, the strain field around the crack is represented by the force doublets concentrated at several points. This approximation enables us to calculate the strain field around the crack in a short time. The results show that the crack location, the crack size and the loading stress are determined efficiently with good accuracy and short CPU time. The calculations were made using the workstation SUN SPARK station 1, and the CPU time needed to solve the two- and three-dimensional inverse problem was within one and three minutes, respectively.
|