| Project/Area Number |
03555086
|
|---|
| Research Category |
Grant-in-Aid for Developmental Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
計測・制御工学
|
|---|
| Research Institution | Oita University |
|---|
Principal Investigator |
ENOKIZONO Masato Oita Univ., Faculty of Engineering, Associate Prof., 工学部, 助教授 (40136784)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
NAGATA Shoichiro Miyazaki Univ., Faculty of Engineering, Research Associate, 工学部, 助手 (20218001)
TODAKA Takashi Oita Univ., Faculty of Engineering, Research Associate, 工学部, 助手 (50163994)
AKITA Masanori Oita Univ., Faculty of Engineering, Associate Prof., 工学部, 助教授 (40192900)
|
|---|
| Project Period (FY) |
1991 – 1992
|
| Project Status |
Completed (Fiscal Year 1992)
|
| Budget Amount *help |
¥7,600,000 (Direct Cost: ¥7,600,000)
Fiscal Year 1992: ¥2,000,000 (Direct Cost: ¥2,000,000)
Fiscal Year 1991: ¥5,600,000 (Direct Cost: ¥5,600,000)
|
|---|
| Keywords | Non-Destructive Testing / Magnetic Senser / Rotational Magnetic Flux / Neural Network / Lissajous's Figure / Barkhausen Noise / Acoustic Analysis / Fatigue / 2次元磁気センサ- / 形状推定 / ニュ-ロ |
|---|
| Research Abstract |
Studies on practical application of the magnetic sensor using rotational magnetic flux and non-destructive testing (NDT) system have been curried out. The summary of the results is shown below. (1) Nondestructive testing with magnetic sensor using rotational magnetic flux.
In conventional eddy current NDT's, the magnetic sensor has been driven by dc or alternative current. Therefore, the generated magnetic field in magnetic materials is arranged in one direction. However, in order to estimate shape or position of an unknown defect, a two-dimensional alternating magnetic flux density vector is necessary. We have proposed the NDT system using rotational magnetic flux. In this system, the magnitude and phase value are obtained as the signals of defect. (2) Rotational magnetic sensor with neural network for non-detractive testing.
In application of the sensor for NDT system, we have developed a new NDT method with neural network. In this system, the magnitude and phase value are measured and
… More
used to obtain information about defect. These values include the information about the shapes or positions of an unknown defect. We applied a neural network technique to estimation of a defect shape. We have reported that the estimated defects show good agreement with the given one in a simple example. (3) Acoustic analysis of the Barkhausen noise in stressed magnetic material.
Because increase of fatigue in magnetic materials is known as a cause of breakdown of material, developing techniques in detecting fatigue before breakdown are very important. For the detection of flaw in the magnetic materials with NDT, we notice the Barkhausen noise occurred in magnetizing proses of ferromagnetic materials. We have investigated acoustic analyses of the noise on sample materials with thermal fatigue. As acoustic analyses, obtaining power of the waveform, zero crossing rate analysis and frequency analysis are performed. These results showed that the fatigue of the materials and spectral feature of the noise were closely concerned. We have reported that The Barkhausen noise gets similar to the white noise with increasing the fatigue of the specimen. Less
|
