| Project/Area Number |
08455381
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
生物・生体工学
|
|---|
| Research Institution | Osaka University |
|---|
Principal Investigator |
SHIOYA Suteaki Osaka University, Faculty of Engineering, Professor, 工学部, 教授 (50026259)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
UCHIYAMA Keiji Osaka University, Faculty of Engineering, Assistant Professor, 工学部, 助手 (60294039)
SHIMIZU Hiroshi Osaka University, Faculty of Engineering, Associate Professor, 工学部, 助教授 (00226250)
中嶋 幹男 大阪大学, 工学部, 助手 (00273590)
|
|---|
| Project Period (FY) |
1996 – 1997
|
| Project Status |
Completed (Fiscal Year 1997)
|
| Budget Amount *help |
¥7,800,000 (Direct Cost: ¥7,800,000)
Fiscal Year 1997: ¥1,900,000 (Direct Cost: ¥1,900,000)
Fiscal Year 1996: ¥5,900,000 (Direct Cost: ¥5,900,000)
|
|---|
| Keywords | temperature controllable gene expression / autoassociative neural network / fault diagnosis / wavelet filter / 状態認識 / ウェーブレット変換 |
|---|
| Research Abstract |
A nonlinear multivariate analysis, artificial autoassociative neural network (AANN), was applied to bioprocess fault detection . In an optimal production process of a recombinant yeast with a temperature controllable expression system, faults in test cases of faulty temperature sensor and plasmid instability of recombinant cells could be detected by the AANN.Since the raw data of measured variables included high frequency noise, a wavelet filter bank (WFB) was applied noise elimination before training of the AANN.The filtering performance of the WFB was compared with those of some classical first order digital filters. The filtered signals at several resolution scales by the WFB were employed as the training data of the AANN.The computing time and summation of square of errors (SSE) in training were compared and appropriate degree of the noise filtering and the density of the training data of the AANN were discussed. High frequency noise in the data could be eliminated by the WFB before the fault diagnosis was performed. The diagnosis system could accurately and immediately detect the faults on-line in the test cases of a faulty temperature sensor and plasmid instability of the recombinant cells. The performance of the feature capturing by the AANN was compared with that by a linear multivariate analysis, principal component analysis (PCA). AJ index defined in this study, using inputs and outputs of the AANN was used for fault detection successfully. The same faults were not detected by linear principal component analysis (PCA). The output of the first unit of the trained AANN functioned effectively for the discrimination of the data in the abnormal cases from the data in the normal cases. By implementing corrective action after fault detection, the final production amount was increased to twice the amount it would have been without diagnosis.
|
