Evolution of glass forming ability indicator by genetic programming

Tripathi, Manwendra K.; Ganguly, Subhas; Dey, Partha Pratim; Chattopadhyay, Partha

doi:10.1016/j.commatsci.2016.02.037

Cited by 39 publications

(10 citation statements)

References 67 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Based on Wakasugi et al 94 , the viscosity of supercooled liquid will increase with increasing the ratio of T x /T l, and increasing the viscosity of supercooled liquid will lead to high glass-forming ability. In conclusion, the G p expression inherits the phenomenological attributes of glass-forming ability in BMGs and shows a good correlation with D max 16 .…”

Section: Resultsmentioning

confidence: 71%

“…Lu and Liu 13 propose that using γ ( ) could guide scientists to compare the glass-forming ability of alloy systems. Tripathi et al 16 , combining the thermodynamics and principles of genetic programming, developed a new parameter (i.e., the G p criterion) to measure the glass-forming ability of BMGs. It has been suggested that the stability of the liquid phase and the glass’s resistance to crystallization should be considered simultaneously to increase the GFA of alloys 17 .…”

Section: Introductionmentioning

confidence: 99%

“…Similarly, T rg (= T g /T l ) is a parameter that only considers the ease of glass formation 18 . In addition, although developing different parameters able scientists to evaluate and predict the glass-forming ability of alloy systems, most of these expressions have a low correlation with the glass-forming ability and will fail in predicting the D max of newly developed alloy systems 16 , 19 .…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Thermodynamically-guided machine learning modelling for predicting the glass-forming ability of bulk metallic glasses

Ghorbani

Askari

Malekan

et al. 2022

Sci Rep

View full text Add to dashboard Cite

Glass-forming ability (GFA) of bulk metallic glasses (BMGs) is a determinant parameter which has been significantly studied. GFA improvements could be achieved through trial-and-error experiments, as a tedious work, or by using developed predicting tools. Machine-Learning (ML) has been used as a promising method to predict the properties of BMGs by removing the barriers in the way of its alloy design. This article aims to develop a ML-based method for predicting the maximum critical diameter (Dmax) of BMGs as a factor of their glass-forming ability. The main result is that the random forest method can be used as a sustainable model (R2 = 92%) for predicting glass-forming ability. Also, adding characteristic temperatures to the model will increase the accuracy and efficiency of the developed model. Comparing the measured and predicted values of Dmax for a set of newly developed BMGs indicated that the model is reliable and can be truly used for predicting the GFA of BMGs.

show abstract

Section: Resultsmentioning

confidence: 71%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Thermodynamically-guided machine learning modelling for predicting the glass-forming ability of bulk metallic glasses

Ghorbani

Askari

Malekan

et al. 2022

Sci Rep

View full text Add to dashboard Cite

show abstract

“…There are many examples of this phenomenon: Xiong et al 2 used a random forest (RF) algorithm chosen out of the WEKA library 3 to develop a ML model that describes the glass-forming ability and elastic moduli of bulk metallic glasses; Möller et al 4 used a support vector machine (SVM) to develop hard magnetic materials; Shen et al 5 successfully established a physical metallurgy-guided ML model (SVM) and a genetic algorithm (GA)-based design process to produce ultrahigh-strength stainless steels; Zhang et al 6 incorporated GAs for the model and descriptor selection for high entropy alloys (HEAs) and incorporated several ML algorithms involving ANN, RF, SVM, etc. ; Kaufmann et al 7 also used RF to manipulate HEAs; Wang et al 8 employed a group of ML algorithms to develop Fe-based soft magnetic materials; Khatavkar et al 9 employed Gaussian process regression (GPR) and SVR to advance Co- and Ni-based superalloys; Wen et al 10 employed ML models that include simple linear regression, SVM with various kernels, ANN, and K-nearest neighbors (KNN) to fabricate HEAs with a high degree of hardness; Feng et al 11 utilized a deep neural network (DNN) to predict the defects in stainless steel; Sun et al 12 used SVM models to predict the glass-forming ability of binary metallic alloys; Ward et al 13 constructed an RF model to design metallic glasses and validated them via commercially viable fabrication methods; Ren et al 14 employed a so-called general-purpose ML framework 15 and high-throughput experiments to predict glass-forming likelihood; and, several other metallic glass alloys have also been studied by employing ML approaches 16 , 17 .…”

Section: Introductionmentioning

confidence: 99%

A machine-learning-based alloy design platform that enables both forward and inverse predictions for thermo-mechanically controlled processed (TMCP) steel alloys

Lee

Park

Lee

et al. 2021

Sci Rep

View full text Add to dashboard Cite

Predicting mechanical properties such as yield strength (YS) and ultimate tensile strength (UTS) is an intricate undertaking in practice, notwithstanding a plethora of well-established theoretical and empirical models. A data-driven approach should be a fundamental exercise when making YS/UTS predictions. For this study, we collected 16 descriptors (attributes) that implicate the compositional and processing information and the corresponding YS/UTS values for 5473 thermo-mechanically controlled processed (TMCP) steel alloys. We set up an integrated machine-learning (ML) platform consisting of 16 ML algorithms to predict the YS/UTS based on the descriptors. The integrated ML platform involved regularization-based linear regression algorithms, ensemble ML algorithms, and some non-linear ML algorithms. Despite the dirty nature of most real-world industry data, we obtained acceptable holdout dataset test results such as R2 > 0.6 and MSE < 0.01 for seven non-linear ML algorithms. The seven fully trained non-linear ML models were used for the ensuing ‘inverse design (prediction)’ based on an elitist-reinforced, non-dominated sorting genetic algorithm (NSGA-II). The NSGA-II enabled us to predict solutions that exhibit desirable YS/UTS values for each ML algorithm. In addition, the NSGA-II-driven solutions in the 16-dimensional input feature space were visualized using holographic research strategy (HRS) in order to systematically compare and analyze the inverse-predicted solutions for each ML algorithm.

show abstract

“…Volumetric shrinkage in metals transforming from liquid to solid state may range from 3 to 10%, with 5-8% being typical of most cast alloys and taking place in different forms. e other major source of porosity is caused by the dissolution of the hydrogen gas in the solidifying solid as a result of the reduction of hydrogen solubility [1][2][3][4][5][6][7][8][9][10]. Majority of porosity observed in castings is caused by a combination of gas and shrinkage.…”

Section: Introductionmentioning

confidence: 99%

A Review Study on the Main Sources of Porosity in Al‐Si Cast Alloys

Samuel

Zedan

Doty

et al. 2021

Advances in Materials Science and Engineering

View full text Add to dashboard Cite

The present study was performed on A356 and B319 alloys in mechanically stirred or degassed condition. Melts were Sr-modified and grain-refined. Hydrogen content was varied from less than 0.1 ml/100 g Al to ∼0.4 ml/100 g Al; Fe was increased to 0.8% in B319 alloy. Lanthanum and cerium were added as 99.5% pure metals. Two main techniques were used to investigate porosity formation: fracture surface of tensile or fatigue test bars, or reduced pressure test (RPT) method. Porosity type and shape were examined. The results show that pore size is more influential than small scattered ones from a mechanical point of view. Tensile testing is affected by porosity located at the center of the testing bar, whereas edge porosity is responsible for crack initiation in case of fatigue testing. Intermetallics precipitate in the form of intercepted platelets which restricts the flow of the molten metal, leading to formation of shrinkage cavities. Precipitation of clusters of compounds from the liquid state such as Al2Si2Sr, Mg2Sn, Al3Ti, or added Al2O3particles would as well act as nucleation sites for porosity formation. Most oxides were observed in the form of long branched strings. In some cases, bifilms were also reported in addition to SrO and MgO.

show abstract

Evolution of glass forming ability indicator by genetic programming

Cited by 39 publications

References 67 publications

Thermodynamically-guided machine learning modelling for predicting the glass-forming ability of bulk metallic glasses

Thermodynamically-guided machine learning modelling for predicting the glass-forming ability of bulk metallic glasses

A machine-learning-based alloy design platform that enables both forward and inverse predictions for thermo-mechanically controlled processed (TMCP) steel alloys

A Review Study on the Main Sources of Porosity in Al‐Si Cast Alloys

Contact Info

Product

Resources

About