METHOD FOR GENERATING PREDICTION MODEL, METHOD FOR PREDICTING CORROSION AMOUNT OF METALLIC MATERIAL, PROGRAM FOR GENERATING PREDICTION MODEL, AND DEVICE OF GENERATING PREDICTION MODEL

• Main Authors: NAKATSUJI KAZUHIRO, OMODA MASATAKA
• Format: Patent
• Language: English; Japanese
• Published: 27.08.2020
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIRCHEMICAL OR PHYSICAL PROPERTIES; MEASURING; PHYSICS; TESTING

To provide a method for generating a production model, a method for predicting the corrosion amount of a metallic material, a program for generating the prediction model and a device for generating the prediction model, capable of easily interpreting a prediction result by unifying the magnitude of impact with high prediction accuracy.SOLUTION: A method for generating a prediction model includes: an influence degree search step of searching the influence degree of an explanatory variable to a prediction variable for each explanatory variable; a similarity calculation step of calculating the similarity between the explanatory variable to be predicted and the explanatory variable of the past performance data by using the influence degree; and a model generation step of generating a regression model using the similarity. In the influence degree search step, the influence degree is searched so that the prediction accuracy by the regression model becomes high.SELECTED DRAWING: Figure 2 【課題】予測精度が高く、影響度の大きさが統一されることにより予測結果の解釈が容易な、予測モデルの生成方法、金属材料の腐食量予測方法、予測モデルの生成プログラムおよび予測モデルの生成装置を提供すること。【解決手段】予測モデルの生成方法は、予測変数に対する説明変数の影響度を、説明変数ごとに探索する影響度探索工程と、予測対象の説明変数と過去の実績データの説明変数との類似度を、影響度を用いて算出する類似度算出工程と、類似度を用いて回帰モデルを生成するモデル生成工程と、を含み、影響度探索工程が、回帰モデルによる予測精度が高くなるように影響度を探索する。【選択図】図２