| Project/Area Number |
18K13672
|
|---|
| Research Category |
Grant-in-Aid for Early-Career Scientists
|
| Allocation Type | Multi-year Fund |
|---|
| Review Section |
Basic Section 18020:Manufacturing and production engineering-related
|
|---|
| Research Institution | Okayama University |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2018-04-01 – 2021-03-31
|
| Project Status |
Completed (Fiscal Year 2020)
|
| Budget Amount *help |
¥4,160,000 (Direct Cost: ¥3,200,000、Indirect Cost: ¥960,000)
Fiscal Year 2020: ¥390,000 (Direct Cost: ¥300,000、Indirect Cost: ¥90,000)
Fiscal Year 2019: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2018: ¥2,730,000 (Direct Cost: ¥2,100,000、Indirect Cost: ¥630,000)
|
|---|
| Keywords | 研削砥石 / データマイニング / ランダムフォレスト / 平面研削 / 決定木 / ドレッシング / 意思決定支援 |
|---|
| Outline of Final Research Achievements |
In this study, we used random forest, a data mining method, to construct a system that can determine the abrasive grain, grain size, and bonding strength from various combinations of material property values. In addition, the usefulness of the system was verified. The verification was carried out by grinding experiments on general materials and difficult-to-cut materials using the recommended grinding wheels by the system. As a result of a grinding experiment of Inconel 718, which is a difficult-to-cut material that does not exist in the learning database, the amount of wear of the recommended grinding wheels (PA) was reduced by 12%. From this result, we were able to verify the usefulness of this system constructed by random forest method.
|
| Academic Significance and Societal Importance of the Research Achievements |
本研究によりデータマイニング手法の有用性を示すことで，非熟練技能者育成を支援できるだけでなく，隠された暗黙知の体系化の根幹技術となれることが予測される．研削加工における意思決定が体系化されることにより，日本の生産を支える根幹となる中小企業や町工場の技能者が，研削砥石を決定する際のコストや時間の低減につながる．製造現場の抱える，従業者の高齢化，後継者不足，販売価格の低下やそれに起因する産業構造の弱体化など，数々の課題を解決できると考えられる．
|
