PHYSICAL PROPERTY PREDICTION METHOD

To provide a physical property prediction method capable of quantitatively visualizing a catalyst state of a plant interior which cannot be directly observed by human eyes while a plant is operating, at each period of time during the operation of the plant for a long time.SOLUTION: A physical proper...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: KITAGAWA KAZUHIRO, HATTORI AYAMI, KO HITOKO, OKUNO YOSHISHIGE, HOSOKI YASUHIRO, UMEHARA KAZUKI, TAKEMOTO SHIMPEI
Format: Patent
Sprache:eng ; jpn
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:To provide a physical property prediction method capable of quantitatively visualizing a catalyst state of a plant interior which cannot be directly observed by human eyes while a plant is operating, at each period of time during the operation of the plant for a long time.SOLUTION: A physical property prediction method includes the steps of: selecting a target property included in a dataset for learning; selecting an input variable included in the dataset for learning according to an engineer knowledge database; imparting a constraint condition to the selected input variable according to the engineer knowledge database; imposing a constraint condition according to the engineer knowledge database and constructing a mathematical model; and predicting a physical property using the constructed mathematical model.SELECTED DRAWING: Figure 3 【課題】長時間のプラント運転の各時刻において、プラント運転中に直接人間の目で観察することができないプラント内部の触媒状態を定量的に可視化できる物性予測方法を提供する。【解決手段】本発明に係る物性予測方法は、学習用データセットに含まれる目的特性を選択する工程と、前記学習用データセットに含まれる入力変数を技術者知見データベースを基に選択する工程と、選択された入力変数に対して技術者知見データベースを基に制約条件を与える工程と、技術者知見データベースに基づいて制約条件を課して数理モデルを構築する工程と、構築された数理モデルを用いて物性を予測する工程と、を含む。【選択図】図3