Multicomponent alloys design and mechanical response: From high entropy alloys to complex concentrated alloys

Materials Science and Engineering R Reports, Год журнала: 2024, Номер 161, С. 100853 - 100853

Опубликована: Сен. 11, 2024

Язык: Английский

---

Predictive analytics of wear performance in high entropy alloy coatings through machine learning

Physica Scripta, Год журнала: 2024, Номер 99(7), С. 076014 - 076014

Опубликована: Июнь 10, 2024

Abstract High-entropy alloys (HEAs) are increasingly renowned for their distinct microstructural compositions and exceptional properties. These HEAs employed surface modification as coatings exhibit phenomenal mechanical characteristics including wear corrosion resistance which extensively utilized in various industrial applications. However, assessing the behaviour of HEA through conventional methods remains challenging time-consuming due to complexity structures. In this study, a novel methodology has been proposed predicting using Machine Learning (ML) algorithms such Support Vector (SVM), Linear Regression (LR), Gaussian Process (GPR), Least Absolute Shrinkage Selection Operator (LASSO), Bagging (BR), Gradient Boosting Tree (GBRT), Robust regressions (RR). The analysis integrates 75 combinations with processing parameters test results from peer-reviewed journals model training validation. Among ML models utilized, GBRT was found be more effective rate Coefficient Friction (COF) highest correlation coefficient R 2 value 0.95 ∼ 0.97 minimal errors. optimum is used predict unknown properties conducted experiments validate results, making crucial resource engineers materials sector.

Язык: Английский

---

Machine Learning-Based Process Optimization in Biopolymer Manufacturing: A Review

Polymers, Год журнала: 2024, Номер 16(23), С. 3368 - 3368

Опубликована: Ноя. 29, 2024

The integration of machine learning (ML) into material manufacturing has driven advancements in optimizing biopolymer production processes. ML techniques, applied across various stages production, enable the analysis complex data generated throughout identifying patterns and insights not easily observed through traditional methods. As sustainable alternatives to petrochemical-based plastics, biopolymers present unique challenges due their reliance on variable bio-based feedstocks processing conditions. This review systematically summarizes current applications techniques aiming provide a comprehensive reference for future research while highlighting potential enhance efficiency, reduce costs, improve product quality. also shows role algorithms, including supervised, unsupervised, deep

Язык: Английский

---

BaTiO3-based high-entropy ceramics for enhanced capacitive energy storage performance

Ceramics International, Год журнала: 2025, Номер unknown

Опубликована: Янв. 1, 2025

Язык: Английский

---

Batch active learning for microstructure-property relations in energetic materials

Mechanics of Materials, Год журнала: 2025, Номер unknown, С. 105308 - 105308

Опубликована: Март 1, 2025

Язык: Английский

---

High‐throughput calculation integrated with stacking ensemble machine learning for predicting elastic properties of refractory multi‐principal element alloys

Materials Genome Engineering Advances, Год журнала: 2025, Номер unknown

Опубликована: Март 12, 2025

Abstract The traditional trial‐and‐error method for designing refractory multi‐principal element alloys (RMPEAs) is inefficient due to a vast compositional design space and high experimental costs. To surmount this challenge, the data‐driven material based on machine learning (ML) has emerged as critical tool accelerating materials design. However, absence of robust datasets impedes exploitation in novel RMPEAs. High‐throughput (HTP) calculations have enabled creation such datasets. This study addresses these challenges by developing framework predicting elastic properties RMPEAs, integrating HTP with ML. A big dataset RMPEAs including 4536 compositions was constructed using new proposed method. stacking ensemble regression algorithm combining multilayer perceptron (MLP) gradient boosting decision tree (GBDT) developed, which achieved 92.9% accuracy Ti‐V‐Nb‐Ta alloys. Verification experiments confirmed ML model's robustness. integration provides cost‐effective, efficient, precise alloy strategy, advancing development.

Язык: Английский

---

Inverse design of high-performance piezoelectric semiconductors via advanced crystal representation and large language models

Applied Physics Letters, Год журнала: 2025, Номер 126(11)

Опубликована: Март 1, 2025

The inverse design of solid-state materials with targeted properties represents a significant challenge in science, particularly for piezoelectric semiconductors where both structural symmetry and electronic must be carefully controlled. Here, we employ the simplified line-input crystal-encoding system representation combined MatterGPT framework discovering potential semiconductors. By training on curated dataset 1556 from Materials Project database, our model learns to generate crystal structures through an autoregressive sampling process. Starting approximately 5000 generated structures, implemented comprehensive screening workflow incorporating validity, thermodynamic stability, property verification. This approach identified several promising candidates 4100 reconstructed each representing compounds unrecorded existing databases. Among these, most notable material demonstrated stress coefficient 25.9 C/m2 e[1,6] direction. Additionally, these demonstrate suitable bandgaps ranging 1.63 3.61 eV, suggesting applications high-sensitivity sensors high-temperature electronics. Our work demonstrates effectiveness combining structure language encoding generative models accelerating discovery functional properties.

Язык: Английский

---

Prediction of mechanical properties of high entropy alloys based on machine learning

Physica Scripta, Год журнала: 2025, Номер 100(4), С. 046013 - 046013

Опубликована: Март 5, 2025

Abstract In recent years, the ideal- properties (young’s modulus, yield strength, toughness) and advanced application potential of high-entropy alloys (HEAs) have attracted numerous researchers. However, due to their unique structure multiple structural combinations, it is challenging explore impact various factors on mechanical performance solely through experiments. This study considers concentrations five alloy atoms working temperature as input parameters. Molecular dynamics (MD) simulations machine learning (ML) algorithms are employed predict tensile FeNiCrCoCu HEAs, including Young’s modulus ( E ) toughness uT ). A dataset 1000 HEAs generated MD simulations, feature selection conducted using principal component analysis Spearman correlation analysis. XGBoost, RF, DT, LGBoost, AdaBoost utilized comparing two methods prediction outcomes. During ML model training, 10-fold cross-validation grid search obtain best models Root mean squard error RMSE ), coefficient determination R 2 absolute MAE relative RAE used evaluation metrics. Results indicate that for outperforms analysis, XGBoost demonstrates superior predictive compared other models. Predictions more accurate than those , with exceeding 0.9 four out work may provide a new method studying ML. future, this can be applied research areas compositions, providing theoretical support It then further critical fields such biomedical aerospace industries.

Язык: Английский

---

Data‐Driven Design of Spinodal Decomposition in (Ti, Zr, Hf)C Composite Carbides for Optimizing the Hardness‐Toughness Trade‐Off

Advanced Functional Materials, Год журнала: 2025, Номер unknown

Опубликована: Апрель 7, 2025

Abstract Transition metal carbides, characterized by exceptional hardness, wear resistance, and thermal stability, emerge as promising candidates for extreme‐environment applications. However, the inherent hardness‐toughness trade‐off remains a critical challenge development of high‐performance ceramics. Herein, data‐driven design strategy to optimize this through precisely tailoring spinodal decomposition in (Ti, Zr, Hf)C composite carbides is proposed. The integration phase diagram calculations, key experiments, machine learning approaches permits high‐throughput mechanical property screening across broad compositional temperature ranges. Isothermal aging induces formation high density nanoscale nodular structures within accompanied generation dislocations, synergistically enhancing hardness (2780 H V ) fracture toughness (3.47 MPa·m 1/2 32% 80%, respectively, compared as‐sintered state. By establishing framework that elucidates composition‐processing‐property relationships, research provides scientific rapidly carbide ceramics orchestrated decomposition, offering rational methodology develop extreme‐condition

Язык: Английский

---

Design and development of advanced Al-Ti-V alloys for beampipe applications in particle accelerators

Physical Review Accelerators and Beams, Год журнала: 2025, Номер 28(4)

Опубликована: Апрель 8, 2025

The present investigation reports the design and development of an advanced material with a high figure merit (FoM) for beampipe applications in particle accelerators by bringing synergy between computational experimental approaches. Machine-learning algorithms have been used to predict phase(s), low density, radiation length designed Al-Ti-V alloys. alloys various compositions single-phase dual-phase mixtures, liquidus temperature, density values were obtained using latin hypercube sampling method TC Python Thermo-Calc software. dataset was utilized train machine-learning algorithms. Classification such as XGBoost regression models linear random forest regressor compute number phases, length, respectively. algorithm shows accuracy 98%, model 94%, is accurate up 99%. developed exhibit well good combination elastic modulus toughness due synergistic effect presence hard Al3Ti phase along minor volume fraction FCC (Al)ss solid solution mixture. comparison our alloys, alloy-1 (Al75.2Ti22.8V2) alloy-2 (Al89Ti10V1) increase 7 times decrease 2 3 compared stainless steel 304, preferred constructing beampipes low-energy accelerators. Further, we experimentally verify FoM equal 0.416, which better than other existing materials experiments. Published American Physical Society 2025

Язык: Английский

---