Data Mining for Estimating Fatigue Strength Based on Composition and Process Parameters

2019 ◽  
Author(s):  
Arvind Keprate ◽  
R. M. Chandima Ratnayake
Keyword(s):  
Data Mining ◽  
Fatigue Strength ◽  
Process Parameters ◽  
Fatigue Testing ◽  
Absolute Error ◽  
Processing Parameters ◽  
Coefficient Of Determination ◽  
Low Alloy Steels ◽  
Strength Based ◽  
Alloy Steels

Abstract Accurately estimating the fatigue strength of steels is vital, due to the extremely high cost (and time) of fatigue testing and often fatal consequences of fatigue failures. The main objective of this manuscript is to perform data mining on the fatigue dataset for steel available from the National Institute of Material Science (NIMS) MatNavi. The cross-industry process for data mining (CRISP-DM) approach was followed in the paper, in order to gain meaningful insights from the dataset and to estimate the fatigue strength of carbon and low alloy steels, using composition and processing parameters. Of the six steps of the CRISP-DM approach, special emphasis has been placed on steps 2 to 5 (i.e. data understanding, data preparation, modeling and evaluation). In step 4 (i.e. modeling), a range of machine learning (parametric and non-parametric) is explored to predict the fatigue strength, based on the composition and process parameters. Various algorithms were trained and tested on the dataset and finally evaluated, using metrics such as root mean square error (RMSE), Mean Absolute Error (MAE), Coefficient of Determination (R2) and Explained Variance Score (EVS).

Artificial Intelligence Based Approach for Predicting Fatigue Strength Using Composition and Process Parameters

2020 ◽  
Author(s):  
Arvind Keprate ◽  
R. M. Chandima Ratnayake
Keyword(s):  
Artificial Intelligence ◽  
Fatigue Strength ◽  
Process Parameters ◽  
Fatigue Testing ◽  
Simulated Data ◽  
Machine Learning Algorithms ◽  
Coefficient Of Determination ◽  
Gradient Boosting ◽  
Data Set ◽  
Strength Based

Abstract Accurate prediction of the fatigue strength of steels is vital, due to the extremely high cost (and time) of fatigue testing and the often fatal consequences of fatigue failures. The work presented in this paper is an extension of the previous paper submitted to OMAE 2019. The main objective of this manuscript is to utilize Artificial Intelligence (AI) to predict fatigue strength, based on composition and process parameters, using the fatigue dataset for carbon and low alloy steel available from the National Institute of Material Science (NIMS) database, MatNavi. A deep learning framework Keras is used to build a Neural Network (NN), which is trained and tested on the data set obtained from MatNavi. The fatigue strength values estimated using NN are compared to the values predicted by the gradient boosting algorithm, which was the most accurate model in the OMAE 2019 paper. The comparison is done using metrics such as root mean square error (RMSE), Mean Absolute Error (MAE), Coefficient of Determination (R2) and Explained Variance Score (EVS). Thereafter, the trained NN model is used to make predictions of fatigue strength for the simulated data (1 million samples) of input parameters, which is then used to generate conditional probability tables for the Bayesian Network (BN). The main advantage of using BN over previously used machine learning algorithms is that BN can be used to make both forward and backward propagation during the Bayesian inference. A case study illustrating the applicability of the proposed approach is also presented. Furthermore, a dashboard is developed using PowerBI, which can be used by practicing engineers to estimate fatigue strength based on composition and process parameters.

Proposal of Fatigue Life Equations for Carbon and Low-Alloy Steels and Austenitic Stainless Steels as a Function of Tensile Strength

2013 ◽  
Author(s):  
Hiroshi Kanasaki ◽  
Makoto Higuchi ◽  
Seiji Asada ◽  
Munehiro Yasuda ◽  
Takehiko Sera
Keyword(s):  
Tensile Strength ◽  
Fatigue Life ◽  
Stainless Steels ◽  
Room Temperature ◽  
Austenitic Stainless Steels ◽  
Low Alloy Steels ◽  
Strength Based ◽  
Fatigue Data ◽  
Current Design ◽  
Alloy Steels

Fatigue life equations for carbon & low-alloy steels and also austenitic stainless steels are proposed as a function of their tensile strength based on large number of fatigue data tested in air at RT to high temperature. The proposed equations give a very good estimation of fatigue life for the steels of varying tensile strength. These results indicate that the current design fatigue curves may be overly conservative at the tensile strength level of 550 MPa for carbon & low-alloy steels. As for austenitic stainless steels, the proposed fatigue life equation is applicable at room temperature to 430 °C and gives more accurate prediction compared to the previously proposed equation which is not function of temperature and tensile strength.

Proposal for the Implementation of Elevated Temperature Design Fatigue Curve for 2-1/4Cr-1Mo-V and 3Cr-1Mo-V Steels

2004 ◽  
Author(s):  
Tatsumi Takehana ◽  
Takeru Sano ◽  
Susumu Terada ◽  
Hideo Kobayashi
Keyword(s):  
Elevated Temperature ◽  
Fatigue Strength ◽  
High Strength ◽  
Fatigue Curve ◽  
Temperature Limit ◽  
Fatigue Tests ◽  
Low Alloy Steels ◽  
Alloy Steels ◽  
Elevated Temperature Design

2-1/4Cr-1Mo-V and 3Cr-1Mo-V steels have been used extensively as materials for elevated temperature and high-pressure hydro-processing reactors. These steels have both of high strength at elevated temperature and high resistance against elevated temperature hydrogen attack due to the addition of vanadium. The operating temperature of these reactors is between 800 and 900deg.F. The fatigue evaluations of these reactors per ASME Sec. VIII Div.2 and Div.3 can’t be performed in spite of demand for fatigue analysis because the temperature limit of design fatigue curve in ASME Sec. VIII Div.2 and Div.3 for carbon and low alloy steels is 700deg.F. Results of load and strain controlled fatigue tests conducted over the temperature range from room temperature to 932deg.F (500deg.C) are reported for 2-1/4Cr-1Mo-V and 3Cr-1Mo-V steels. These data were compared with data for 2-1/4Cr-1Mo steels available from the literatures. The fatigue strength for a 2-1/4Cr-1Mo-V steel in high cycle region is higher than that for 2-1/4Cr-1Mo steels and in low cycle region is lower. The fatigue strength for a 3Cr-1Mo-V steel is almost same as that for 2-1/4Cr-1Mo-V steels. Therefore an elevated temperature design fatigue curve for 2-1/4Cr-1Mo-V and 3Cr-1Mo-V steels is newly proposed. It is found from the case study that the different fatigue life can be predicted by using different mean stress correction procedure.

Proposal of Surface Finish Factor on Fatigue Strength in Design Fatigue Curve

2014 ◽  
Author(s):  
Yuichi Fukuta ◽  
Hiroshi Kanasaki ◽  
Seiji Asada ◽  
Takehiko Sera
Keyword(s):  
Surface Roughness ◽  
Tensile Strength ◽  
Stainless Steel ◽  
Fatigue Strength ◽  
Surface Finish ◽  
Fatigue Curve ◽  
Low Alloy Steels ◽  
Mean Stress ◽  
Alloy Steels ◽  
Surface Finish Effect

The published papers related to the effects of surface finish on fatigue strength are reviewed in order to formulate its factor in the design fatigue curve in air environment. Firstly, some of regulations and literatures were examined to verify the surface finish effect on fatigue strength and formulation of that in design fatigue curve. The fatigue strength of carbon and low alloy steels is decreased with an increase of its surface roughness and tensile strength but that of stainless steel is not decreased except for special conditions. After screening the data of carbon and low alloy steels, a surface finish factor is formulated with these data which is a function of tensile strength, surface roughness and mean stress.

Sign in / Sign up

Export Citation Format

Share Document