Development of near future forecasting method for slope disasters using data assimilation technique

2018 Fiscal Year Final Research Report

• Project/Area Number: 16K01328
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Natural disaster / Disaster prevention science
• Research Institution: Osaka Sangyo University (2018); Osaka University (2016-2017)
• Principal Investigator: Oda Kazuhiro 大阪産業大学, 工学部, 教授 (00185597)
• Co-Investigator(Kenkyū-buntansha): 小泉 圭吾 大阪大学, 工学研究科, 助教 (10362667)
• Research Collaborator: Ito Shin-ichi ; Sakuradani Keiji
• Project Period (FY): 2016-04-01 – 2019-03-31
• Keywords: 斜面災害 / 体積含水率 / データ同化 / 粒子フィルタ / 不飽和浸透 / 数値解析 / 現地計測 / 水分特性曲線

Outline of Final Research Achievements

In this study, the authors attempted to reveal the moisture characteristics of decomposed granite soils by sequential data assimilation in which field measurements of both volumetric water content and rainfall were used. The particle filter method was applied for sequential data assimilation. In the analysis, a saturated-unsaturated seepage finite element analysis was adopted as a simulation model. Six soil moisture parameters could be identified by the particle filter method from the field measurements of volumetric water content. Moreover, the numerical simulation, in which the soil moisture parameters determined from the results of identifying by the particle filter were applied, was able to predict the soil moisture behavior in any state.

Free Research Field

地盤工学

Academic Significance and Societal Importance of the Research Achievements

国土の7割以上が山地で構成される日本では，毎年全国各地で豪雨による斜面災害が発生している．斜面災害は，土中の水分状態に依存しており，現況のそれをモニタリングする現地計測も普及している．一方，数値解析でも土中水分量の予測する不飽和浸透流解析技術は確立されている．但し，現地計測には予測ができない，数値解析には適切なパラメータや境界条件を決定するのが難しい．本研究では，数理情報技術の一つであるデータ同化手法によって両者を結びつけ，精度の高い土中の水分量の予測手法を開発した．また，それを使って，土砂災害の近未来予測手法の基礎的な研究を行った．