Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
PREDICTION METHOD, PREDICTION PROGRAM AND INFORMATION PROCESSING APPARATUS
To prevent prediction precision of a water level of a river, predicted from rainfall data measured actually by a rain gauge when it rains locally and heavily and a learning model, from decreasing.SOLUTION: There is provided a prediction method that causes a computer to: acquire a plurality of pieces of rainfall information measured by a plurality of rain gauges corresponding to an object river for which a learning model is used to predict a water level; determine, based upon the plurality of pieces of rainfall information having been acquired, whether it rained locally and heavily in a region where the riber flows; use, when it is determined that it rained locally and heavily, at least one of the plurality of pieces of rainfall information to correct the plurality of respective pieces of rainfall information to larger values; and use the plurality of pieces of rainfall information having been corrected and the learning model to predict the water level of the river.SELECTED DRAWING: Figure 1
【課題】局地的大雨が発生したときに実際に雨量計が計測した雨量データと学習モデルとから予測される河川の水位の予測精度が、過小方向に低下することを防ぐ。【解決手段】学習モデルを使って水位予測を行う対象の河川に応じた複数の雨量計により計測された複数の降雨量情報を取得し、取得した前記複数の降雨量情報に基づき、前記河川が流れる地域に局地的大雨が発生したかを判定し、前記局地的大雨が発生したと判定した場合、前記複数の降雨量情報のそれぞれを、前記複数の降雨量情報のうちの少なくとも一つの降雨量情報を用いて、より大きい値に補正し、補正した前記複数の降雨量情報および前記学習モデルを用いて、前記河川の前記水位予測を実行する、ことをコンピュータが実行する予測方法が提供される。【選択図】図１
PREDICTION METHOD, PREDICTION PROGRAM AND INFORMATION PROCESSING APPARATUS
To prevent prediction precision of a water level of a river, predicted from rainfall data measured actually by a rain gauge when it rains locally and heavily and a learning model, from decreasing.SOLUTION: There is provided a prediction method that causes a computer to: acquire a plurality of pieces of rainfall information measured by a plurality of rain gauges corresponding to an object river for which a learning model is used to predict a water level; determine, based upon the plurality of pieces of rainfall information having been acquired, whether it rained locally and heavily in a region where the riber flows; use, when it is determined that it rained locally and heavily, at least one of the plurality of pieces of rainfall information to correct the plurality of respective pieces of rainfall information to larger values; and use the plurality of pieces of rainfall information having been corrected and the learning model to predict the water level of the river.SELECTED DRAWING: Figure 1
【課題】局地的大雨が発生したときに実際に雨量計が計測した雨量データと学習モデルとから予測される河川の水位の予測精度が、過小方向に低下することを防ぐ。【解決手段】学習モデルを使って水位予測を行う対象の河川に応じた複数の雨量計により計測された複数の降雨量情報を取得し、取得した前記複数の降雨量情報に基づき、前記河川が流れる地域に局地的大雨が発生したかを判定し、前記局地的大雨が発生したと判定した場合、前記複数の降雨量情報のそれぞれを、前記複数の降雨量情報のうちの少なくとも一つの降雨量情報を用いて、より大きい値に補正し、補正した前記複数の降雨量情報および前記学習モデルを用いて、前記河川の前記水位予測を実行する、ことをコンピュータが実行する予測方法が提供される。【選択図】図１
PREDICTION METHOD, PREDICTION PROGRAM AND INFORMATION PROCESSING APPARATUS
予測方法、予測プログラム及び情報処理装置
WATANABE YUTA (Autor:in) / SUZUKI TAKASHI (Autor:in)
31.08.2020
Patent
Elektronische Ressource
Japanisch
| Europäisches Patentamt | 2020
FLOW PREDICTION PROGRAM, FLOW PREDICTION METHOD AND INFORMATION PROCESSING DEVICE
| Europäisches Patentamt | 2022
INFORMATION PROCESSING DEVICE, PREDICTION MODEL, INFORMATION PROCESSING METHOD, AND PROGRAM
| Europäisches Patentamt | 2024
PREDICTION APPARATUS, PREDICTION SYSTEM, PREDICTION METHOD, AND PROGRAM
| Europäisches Patentamt | 2019
WATER LEVEL PREDICTION PROGRAM, WATER LEVEL PREDICTION METHOD, AND INFORMATION PROCESSING DEVICE
| Europäisches Patentamt | 2021