MATERIAL CHARACTERISTIC VALUE ESTIMATION SYSTEM AND METAL PLATE MANUFACTURING METHOD

• Main Authors: FUNAKAWA YOSHIMASA, KOJIMA MAYUMI
• Format: Patent
• Language: English; Japanese
• Published: 25.03.2022
• Subjects: CONTROL OR REGULATING SYSTEMS IN GENERAL; CONTROLLING; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MECHANICAL METAL-WORKING WITHOUT ESSENTIALLY REMOVINGMATERIAL; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS ORELEMENTS; PERFORMING OPERATIONS; PHYSICS; PUNCHING METAL; REGULATING; ROLLING OF METAL; TRANSPORTING

To provide a material characteristic value estimation system which predicts a material characteristic value with high accuracy and also to provide a metal plate manufacturing method by which a yield of products is improved through a proper change of manufacturing conditions of following processes based on the material characteristic value that the material characteristic value prediction system predicts.SOLUTION: A material characteristic value prediction system 100 comprises a material characteristic value prediction part that acquires input data including facility output factors of metal plate manufacturing facilities, disturbance factors and component values of a metal plate being manufactured and predicts a material characteristic value of a manufactured metal plate using a prediction model into which input data are inputted. The prediction model comprises a machine learning model that is generated through a machine learning into which input data are inputted and from which manufacturing condition factors are outputted and a metallurgy model into which manufacturing condition factors are inputted and from which a material characteristic value is outputted.SELECTED DRAWING: Figure 1 【課題】材料特性値を高精度に予測可能な材料特性値予測システムを提供する。また、その材料特性値予測システムが予測した材料特性値に基づいて後の工程の製造条件を適切に変更することによって、製品の歩留まりを向上させることが可能な金属板の製造方法を提供する。【解決手段】材料特性値予測システム１００は、金属板を製造する設備における設備出力因子、外乱因子及び製造中の金属板の成分値を含む入力データを取得し、入力データを入力する予測モデルを用いて、製造される金属板の材料特性値を予測する、材料特性値予測部を備え、予測モデルは、入力データを入力して製造条件因子を出力する、機械学習によって生成された機械学習モデルと、製造条件因子を入力して材料特性値を出力する金属学モデルと、を含む。【選択図】図１