A review of recent advances and applications of machine learning in tribology

Sose, Abhishek T.; Joshi, Soumil Y.; Kunche, Lakshmi Kumar; Wang, Fangxi; Deshmukh, Sanket A.

doi:10.1039/d2cp03692d

Cited by 21 publications

(9 citation statements)

References 360 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Typically, ANN consists of one input layer, one output layer, and multiple hidden layers, with the number of input and output nodes equal to the number of target properties and variables. 79 Herein, we used a network comprising one node in the input layer (Gibbs hydration free energy), and the number of nodes in the output layer was equal to the total number of ε parameters between CG amino acid beads and water for a given amino acid. More details of the ANN model can be found in Section S2, and a schematic of the model is illustrated in Figure S2.…”

Section: Methods and Computational Detailsmentioning

confidence: 99%

Development, Validation, and Applications of Nonbonded Interaction Parameters between Coarse-Grained Amino Acid and Water Models

Mohammadi,

Joshi,

Deshmukh

2023

Biomacromolecules

Self Cite

View full text Add to dashboard Cite

Interactions between amino acids and water play an important role in determining the stability and folding/unfolding, in aqueous solution, of many biological macromolecules, which affects their function. Thus, understanding the molecular-level interactions between water and amino acids is crucial to tune their function in aqueous solutions. Herein, we have developed nonbonded interaction parameters between the coarse-grained (CG) models of 20 amino acids and the one-site CG water model. The nonbonded parameters, represented using the 12−6 Lennard Jones (LJ) potential form, have been optimized using an artificial neural network (ANN)-assisted particle swarm optimization (PSO) (ANN-assisted PSO) method. All-atom (AA) molecular dynamics (MD) simulations of dipeptides in TIP3P water molecules were performed to calculate the Gibbs hydration free energies. The nonbonded force-field (FF) parameters between CG amino acids and the one-site CG water model were developed to accurately reproduce these energies. Furthermore, to test the transferability of these newly developed parameters, we calculated the hydration free energies of the analogues of the amino acid side chains, which showed good agreement with reported experimental data. Additionally, we show the applicability of these models by performing self-assembly simulations of peptide amphiphiles. Overall, these models are transferable and can be used to study the self-assembly of various biomaterials and biomolecules to develop a mechanistic understanding of these processes.

show abstract

Section: Methods and Computational Detailsmentioning

confidence: 99%

Development, Validation, and Applications of Nonbonded Interaction Parameters between Coarse-Grained Amino Acid and Water Models

Mohammadi,

Joshi,

Deshmukh

2023

Biomacromolecules

Self Cite

View full text Add to dashboard Cite

show abstract

“…The volumetric adsorption was used as the label for training. In order to determine the best hyperparameters, a grid search algorithm was implemented to alter the following hyperparameters: (1) RF: n_estimators {50, 100, 200}, max_depth {None, 5,10}, and max_features {'auto', 'sqrt', 'log2'}, (2) LGBM: learning_rate {0.01, 0.05, 0.1}, n_estimators {100, 200, 300}, num_leaves {5, 10, 20}, max_depth {None, 5, 10}, (3) MLP: hidden layer sizes {(10,), (10,30), (10, 30, 10)}, activation {'relu', 'tanh'}, solver {'adam', 'sgd'}, alpha {0.0001, 0.001, 0.01, 0.1, 1}, learnin-g_rate {constant, adaptive}.…”

Section: Particle Swarm Optimization (Pso) Integrated With the Geneti...mentioning

confidence: 99%

“…Efficient exploration of this "inf inite" design space can be facilitated by optimization approaches including evolutionary algorithms like genetic algorithms (GAs) or the metaheuristic technique of particle swarm optimization (PSO). 29,30 For instance, Collins et al successfully explored a search space of 1.65 trillion MOF structures for exceptional CO 2 uptake using a GA. 31 They developed an MOF functionalization GA (MOFF-GA). This MOFF-GA uses two mating schemas and two different mutation mechanisms along with 13 GA parameters to optimize the MOF functionality to achieve MOFs with good CO 2 uptake.…”

Section: Introductionmentioning

confidence: 99%

Data Driven Discovery of MOFs for Hydrogen Gas Adsorption

Singh,

Sose,

Wang

et al. 2023

J. Chem. Theory Comput.

Self Cite

View full text Add to dashboard Cite

Hydrogen gas (H2) is a clean and renewable energy source, but the lack of efficient and cost-effective storage materials is a challenge to its widespread use. Metal–organic frameworks (MOFs), a class of porous materials, have been extensively studied for H2 storage due to their tunable structural and chemical features. However, the large design space offered by MOFs makes it challenging to select or design appropriate MOFs with a high H2 storage capacity. To overcome these challenges, we present a data-driven computational approach that systematically designs new functionalized MOFs for H2 storage. In particular, we showcase the framework of a hybrid particle swarm optimization integrated genetic algorithm, grand canonical Monte Carlo (GCMC) simulations, and our in-house MOF structure generation code to design new MOFs with excellent H2 uptake. This automated, data driven framework adds appropriate functional groups to IRMOF-10 to improve its H2 adsorption capacity. A detailed analysis of the top selected MOFs, their adsorption isotherms, and MOF design rules to enhance H2 adsorption are presented. We found a functionalized IRMOF-10 with an enhanced H2 adsorption increased by ∼6 times compared to that of pure IRMOF-10 at 1 bar and 77 K. Furthermore, this study also utilizes machine learning and deep learning techniques to analyze a large data set of MOF structures and properties, in order to identify the key factors that influence hydrogen adsorption. The proof-of-concept that uses a machine learning/deep learning approach to predict hydrogen adsorption based on the identified structural and chemical properties of the MOF is demonstrated.

show abstract

“…Generally, machine learning algorithms can be divided into unsupervised learning and supervised learning [17]. Artificial neural networks (ANNs) are an effective tool in modelling non-linear relations and identifying hidden patterns associated with friction phenomena [18]. Bhaumik et al [19] used ANNs for predicting the anti-wear properties of vegetable-oil-based lubricants using a pin-on-disk tribometer.…”

Section: Introductionmentioning

confidence: 99%

Analysis of Coefficient of Friction of Deep-Drawing-Quality Steel Sheets Using Multi-Layer Neural Networks

Trzepieciński,

Szwajka,

Szewczyk

2024

Lubricants

View full text Add to dashboard Cite

This article presents the results of an analysis of the influence of friction process parameters on the coefficient of friction of steel sheets 1.0347 (DC03), 1.0338 (DC04) and 1.0312 (DC05). A special tribometer was designed and manufactured in order to simulate the friction phenomenon occurring in the blankholder area in deep drawing operations. Lubricant was supplied to the contact zone under pressure. The value of the coefficient of friction was determined under various contact pressures and lubrication conditions. Multi-layer artificial neural networks (ANNs) were used to predict the value of the coefficient of friction. The input parameters considered were the kinematic viscosity of lubricants, contact pressure, lubricant pressure, selected mechanical properties and basic surface roughness parameters of sheet metals. The value of the coefficient of friction of 1.0312 steel sheets was predicted based on the results of friction tests on 1.0347 and 1.0338 steel sheets. Many ANN models were built to find a neural network that will provide the best prediction performance. It was found that to ensure a high performance of ANN prediction, it is necessary to simultaneously take into account all the considered roughness parameters (Sa, Ssk and Sku). The predictive performance of the ‘best’ network was greater than R2 = 0.98. The lubricant pressure had the greatest impact on the coefficient of friction. Increasing the value of this parameter reduces the value of the coefficient of friction. However, the greater the contact pressure, the smaller the beneficial effect of pressure-assisted lubrication. The third parameter of the friction process, the kinematic viscosity of the oil, exhibited the smallest impact on the coefficient of friction.

show abstract

A review of recent advances and applications of machine learning in tribology

Abstract: This review summarises recent advances in the use of machine learning for predicting friction and wear in tribological systems, material discovery, lubricant design and composite formulation. Potential future applications and areas for further research are also discussed.

Cited by 21 publications

References 360 publications

Development, Validation, and Applications of Nonbonded Interaction Parameters between Coarse-Grained Amino Acid and Water Models

Development, Validation, and Applications of Nonbonded Interaction Parameters between Coarse-Grained Amino Acid and Water Models

Data Driven Discovery of MOFs for Hydrogen Gas Adsorption

Analysis of Coefficient of Friction of Deep-Drawing-Quality Steel Sheets Using Multi-Layer Neural Networks

Contact Info

Product

Resources

About