Estimation of structure boundary conditions for virtual machine tools by data assimilation of machining vibration

2022 Fiscal Year Final Research Report

• Project/Area Number: 21K20400
• Research Category: Grant-in-Aid for Research Activity Start-up
• Allocation Type: Multi-year Fund
• Review Section: 0301:Mechanics of materials, production engineering, design engineering, fluid engineering, thermal engineering, mechanical dynamics, robotics, aerospace engineering, marine and maritime engineering, and related fields
• Research Institution: Kyoto University
• Principal Investigator: Yamato Shuntaro 京都大学, 工学研究科, 特定助教 (00908486)
• Project Period (FY): 2021-08-30 – 2023-03-31
• Keywords: 逐次データ同化 / 粒子フィルタ / 境界条件推定 / 工作機械 / デジタルツイン / 時間領域シミュレーション / 低次元化モデル

Outline of Final Research Achievements

In order to automatically calibrate the boundary conditions (contact stiffness and damping between element parts, etc.) of a machine tool structural model, a study was conducted to estimate the boundary conditions using real-time data of excitation force (cutting force) and vibration. A digital twin system that sequentially updates the probability distribution of the boundary parameters each time data is input is proposed by assimilating data from time-domain reduced-order simulations of structural vibration and the time response of a limited number of accelerometers installed at arbitrary points using a particle filter. The validity of the estimated boundary parameters was demonstrated experimentally. It was also suggested that the uncertainty of the estimated model and important boundary parameters could be inferred from the spread and change of the updated probability distribution.

Free Research Field

機械工学

Academic Significance and Societal Importance of the Research Achievements

測定した加工振動データを用いてバーチャル工作機械モデル上の境界条件を推定し，シミュレータを逐次更新していくための基盤技術を構築した．これは，収集した実稼働データとシミュレータが有機的に連動し，機械の使用状態に応じて適応・成長する真のデジタルツインを実現するうえで重要な技術である．また，推定された確率分布の挙動から，稼働状況に応じた重要なモデルパラメータの示唆や信頼性区間の評価，さらには機械状態異常の監視などに繋げられる可能性があり，今後更なる発展が期待できる．