Prediction Of Mechanical Properties Research Articles

Dynamic crushing behaviors and failure of cylindrical lithium-ion batteries subjected to impact loading

The mechanical property and failure prediction play a significant role in engineering applications of lithium-ion batteries (LIBs), but with great difficulties due to their complicated internal structures. This paper mainly focused on dynamic crushing behaviors and internal failure mechanisms of cylindrical LIBs subjected to different impact loadings for the safety assessment and protection design. A series of dynamic experiments were conducted on typical 18650 LIBs using the drop weight impact test system. The force-electric-thermal responses and impact failure modes of the cells with state of charge (SOC) dependence were experimentally explored under different impactor types. The finite element (FE) model was also established to reveal the failure mechanisms of the battery under high-velocity impact, and its accuracy was validated by the comparison between the experimental and corresponding FE results. Furthermore, the dynamic failure mechanisms of cylindrical cells using the stress wave theory were also explored in this work. It was shown that except from impact velocities, there were significant effects of the SOCs and impactor types on dynamic responses and internal failure modes of LIBs. The failure displacement of the cell will decrease with the increase of impact velocity, but it was insensitive under low-velocity crushing. Based on the FE results, the relationships between failure displacement and impact velocity and impactor diameter were established under flat-end cylinder head impacting. This work can provide some references for multi-functional dynamic integrity design and safety assessment of LIBs.

Machine learning-based quantitative analysis of metal ductile fracture surface

Quantitative fracture analysis (QFA) can infer mechanical properties and fracture mechanism of materials based on quantitative statistics of various topographical features on the fracture surface. However, it is still a challenge to screen the key fracture surface descriptors for a given property through experimental insight and theory. Obtaining an accurate quantitative relationship is difficult as different properties involve a variety of fracture surface descriptors. Herein, a QFA based on machine learning is proposed. The fracture surface images are first divided into three classes: intergranular, quasi-cleavage and dimple, by an unsupervised classification model based on a convolutional neural network with an accuracy of 0.867. The fracture surface descriptors of the identified dimple images are extracted by image segmentation and used to train separate regression models to predict the values of fracture toughness (KIC), tensile strength (Rm), yield strength (Re), elongation after fracture (A) and fracture shrinkage (Z). These models demonstrate good performance with an R2 in the test set of 0.934, 0.944, 0.840, 0.890, 0.900, respectively. In addition, with KIC and Rm as dependent variables, the explicit relationships between them and key morphological features are established based on symbolic regression. The R2 between the predicted and experimental values of the two formulae reach 0.925 and 0.926, respectively. Such a machine learning-based QFA can assist in the rapid prediction and evaluation of mechanical properties of materials in engineering applications.

Prediction of mechanical properties for typical pressure vessel steels by small punch test combined with machine learning

To understand the fitness of different machine learning methods to small punch test (SPT), three machine learning methods, namely, back propagation (BP) neural network, radial basis function (RBF) neural network and random forest (RF) regression model were used to establish machine learning-based correlation models of SPT load–displacement and uniaxial tensile test true stress–strain curves. Five typical pressure vessel steels were tested to reveal the prediction accuracies of different machine learning methods. Statistical analysis, unveiled the following order of prediction accuracy for pressure vessel steels: BP > RBF > RF. However, the differences in the mean absolute percentage erros between BP and RF decreased from 5.64% to 1.59%, with the training data volume increasing from 50 data sets to 629 data sets. Therefore, if a large training data volume is used, then the three machine learning methods can provide acceptable prediction results. However, if the training data volume is insufficient, then the BP, and RBF neural networks are better choices.

Development of intelligent methodologies perceiving microstructure and mechanical properties of hot rolled steels

Traditional metallographic characterization of steel microstructure can only roughly determine average grain size and phase fractions because it relies too much on researchers’ personal experiences. As a result, when the microstructure-property relationships are to be established, only the rough microstructural features can be taken into accounts, which is usually incapable of grasping the correct strengthening and toughening mechanisms. To solve this long-existed problem, an integrated system with the microstructural perception by deep learning (DL) and prediction of mechanical properties by machine learning (ML) was developed, which can quickly and precisely recognize typical ferrite-pearlite microstructures in the most used hot rolled plain carbon steels and make the predictions of their mechanical properties. In the microstructural perception, a U-shaped residual network was proposed to implement deep learning of metallographic images to obtain digital information such as the size and distribution of grains and the lengths of grain and second phase boundaries. As compared to the conventional deep learning algorithm, the pixel accuracy (PA) and frequency weighted intersection over union (FWIoU) of predictions were improved by 2.84% and 5.27%, respectively. On this basis, the relationships between mechanical properties and the detailed microstructural features were established by using the random forest (RF) algorithm. Compared with the predictions by rough features of average grain size and phase fractions, the root mean square errors (RMSE) of the yield strength (YS), tensile strength (TS), and elongation (EL) based on the test data were reduced by 71.53%, 67.51% and 53.4%, respectively. By using the accurate perceptron for microstructure and properties, changes of strengths and elongations with processing parameters and chemical compositions were analyzed in good agreement with physical laws. Also, this perceptron can be extended to high strength steels with microstructures comprised of ferrite and bainite or complex phases if the data of metallographic images and properties are available.

Soft computing techniques for predicting the properties of raw rice husk concrete bricks using regression-based machine learning approaches

In this study, the replacement of raw rice husk, fly ash, and hydrated lime for fine aggregate and cement was evaluated in making raw rice husk-concrete brick. This study optimizes compressive strength, water absorption, and dry density of concrete brick containing recycled aggregates via Response Surface Methodology. The optimized model's accuracy is validated through Artificial Neural Network and Multiple Linear Regression. The Artificial Neural Network model captured the 100 data's variability from RSM optimization as indicated by the high R threshold- (R > 0.9997), (R > 0.99993), (R > 0.99997). Multiple Linear Regression model captured the data's variability the decent R2 threshold confirming- (R2 > 0.9855), (R2 > 0.9768), (R2 > 0.9155). The raw rice husk-concrete brick 28-day compressive strength, water absorption, and density prediction were more accurate when using Response Surface Methodology and Artificial Neural Network compared to Multiple Linear Regression. Lower MAE and RMSE, coupled with higher R2 values, unequivocally indicate the model's superior performance. Additionally, employing sensitivity analysis, the influence of the six input parameters on outcomes was assessed. Machine learning aids efficient prediction of concrete's mechanical properties, conserving time, labor, and resources in civil engineering.

Using the design of experiments methodology to evaluate the heat treatment of additively manufactured maraging steel

The effect of the processing parameters on the final microstructure and mechanical properties was investigated for additively manufactured maraging tool steel 1.2709. First, an effective range of appropriate combinations of heat treatment parameters was found using the design of experiments (DOE) method. This suggested 17 variations of heat treatment regimes. The effect of input parameters (initial condition, hardening temperature, hold at hardening temperature) on the output values (mechanical properties) was then statistically evaluated. Three initial conditions of the additively manufactured (AM) steel were used: as-built, annealed at 820 °C for 1 h and annealed at 940 °C for 2 h. A subsequent heat treatment in the region of 250 °C–750 °C was considered with holds of 0–6 h at the temperatures. Based on the measured mechanical properties, regression analysis was performed and optimal models for the prediction of individual mechanical properties were produced. The coefficient of determination of the models for yield and ultimate strengths and hardness reached above 95% when hardening temperature was considered a significant factor. It was around 98% in the case of tensile strength and hardness when the additional effect of the initial condition was also included in the models. This means a very reliable prediction of those mechanical properties can be expected. On the other hand, the model for total elongation can explain only 41% of the measured values, as not a single statistically significant factor was identified.

Elimination of experimentation cost and time by data analysis in mechanical property prediction of aluminum alloys

Discovery of new materials in material science involves lot of processes like experimental trials and laboratory testing which consumes lot of time and energy. This process requires theoretical foundation to avoid errors and also constrained by laboratory environment conditions. To overcome the above issues, data analysis techniques are used for material property prediction for the new materials which reduces lot of time and energy by avoiding experimentation. Plenty of data is available in the material science field for the analysis. In this work, Aluminum alloy composite properties are collected and analysed for the mechanical property prediction. Data available in this case is the labelled data and the output is a continuous variable, so the supervised regression algorithms such as linear regression (LR), K- Nearest Neighbor algorithm (KNN) and Artificial Neural Network (ANN) are proposed for the tensile strength prediction of aluminum alloys. The performance of LR, KNN and ANN are measured in terms of R2 value respectively 90.5%, 96.5% and 89%. K-Nearest Neighbor algorithm is selected as best algorithm because of its accurate prediction and the testing data is varied between 10 and 30 percentage and the R-square values are calculated for 10, 20 and 30 percentage of test size are 94.3%, 93.2% and 96.3%. Best algorithm with proper test size is found using above methodology. Similar methodology can be applied for predicting the other properties of aluminum composites and also for other composites.

Artificial neural networks for the prediction of mechanical properties of CGNP/PLGA nanocomposites

The mechanical properties of composite-based components have an impact on both their quality and efficacy. As a result, a scientific framework is needed to discover the practical set of progression constraints that will result in exceptional mechanical qualities. The possibilities of an artificial neural network (ANN) for predicting mechanical properties—specifically, the density and hardness of a carboxyl decorated graphene Nanocomposites (CGNP)/Poly Lactic-co-Glycolic Acid (PLGA) nanocomposite created under various spark plasma sintering (SPS) operating conditions—are explored in this work. A back-propagation with a 2–10-2 structural design and Levenberg-Marquardt algorithm was advanced to anticipate the automated performance of CGNP/PLGA nanocomposites in expressions of density and hardness parameters. The hypothesized outcomes of the models were compared to the actual experimental value. Density (0.9842) and hardness (0.9737) were found to be near to one; the archetypal accomplished well, with a good correlation coefficient (R) for both outputs and a low root-mean-squared error. The anticipated data findings were found to be quite compatible with the values acquired from the investigational assessment results. Our findings demonstrate the utility of a well-trained ANN method in assessing the density and hardness properties of SPSed CGNP/PLGA nanocomposites. As a result, the ANN system provides an unswerving tool for making decisions that can reduce the high costs involved with the experimental characterization of newly manufactured polymer composites.

Mechanical Property Degradation and Prediction Model of Red Sandstone Under Action of Dry–Wet Cycles in Acid Environment

Mechanical Property Degradation and Prediction Model of Red Sandstone Under Action of Dry–Wet Cycles in Acid Environment

Modified embedded atom method interatomic potential for FCC γ-cerium

As interest in Cerium containing alloys and Al-Ce alloys in particular grows, the need to generate predictions of mechanical, thermodynamic and transport properties across a compositional space becomes more pronounced. The absence of a reliable interaction potential to be used in classical molecular dynamics (MD) simulation of γ-Ce is a fundamental bottleneck to subsequent alloy simulation. In this work, a Modified Embedded Atom Method (MEAM) potential to be used in MD simulation for γ-Ce has been generated. The parameterization process involved two steps. First, the iterative Latin hypercube sampling (LHS) method was used to generate MEAM parameter sets based on an automated optimization of energies and forces relative to a training set of small configurations evaluated through Density Functional Theory (DFT). Second, a human-guided optimization in that local parameter space was performed to refine the parameters to satisfy an array of mechanical, thermodynamic and transport properties available from the experimental literature. When used in an MD simulation, the resulting potential provides excellent estimates of all tested material properties across a broad temperature range. This γ-Ce potential, when combined with existing MEAM potential for Al, will form the necessary foundation for the subsequent development of a mixture potential, enabling the simulation of Al-Ce and Al-Ce-X alloys.

Development of a machine learning framework to determine optimal alloy composition based on steel hardenability prediction

Mechanical property prediction plays a crucial role in the steel industry, enabling better materials selection and enhancing production efficiency. In this study, we propose a novel framework that facilitates optimal materials selection for steel’s alloying elements, based on accurate predictions of the Jominy hardness curve. Leveraging Gaussian Process (GP) regression, we provide probabilistic predictions of steel hardenability characteristics from alloy element composition. Taking it a step further, our framework incorporates these accurate predictions into a constrained optimization process, yielding optimal compositions that reduce overall spending while meeting performance specifications. Through data obtained from 1080 steel samples, our GP regression model exhibits high accuracy, achieving an RMSE of 1.37 and showcasing significant improvements in the field. Moreover, our constrained optimization utilizing the GP model and historical market data reveals an average cost reduction of 18% on alloying element expenses, highlighting the tangible cost-saving potential of this approach. By leveraging Gaussian Process (GP) regression, we not only achieve accurate predictions of the Jominy hardness curve based on alloy element composition, but we also introduce a crucial element of uncertainty quantification. This empowers us to place trust in the results of our optimization process, ensuring robust and reliable materials selection. The integration of GP regression and optimization provides a powerful tool for achieving cost-effective materials selection and marks a significant advancement compared to existing studies. This research underscores the promise of machine learning in the steel industry, demonstrating its ability to yield substantial cost savings and enhance decision-making in materials selection.

In pursuit of a suitable machine learning algorithm for hardness prediction of aluminium alloy

PurposeBecause of the mechanical properties of aluminium (Al), an accurate prediction of its properties has been challenging. Researchers are seeking reliable models for predicting the mechanical strength of Al alloys owing to the continuous emergence of new Al alloys and their applications. There has been widespread use of empirical and statistical models for the prediction of different mechanical properties of Al and Al alloy, such as linear and nonlinear regression. Nevertheless, the development of these models requires laborious experimental work, and they may not produce accurate results depending on the relationship between the Al properties, mix of other compositions and curing conditions.Design/methodology/approachNumerous machine learning (ML) models have been proposed as alternative approaches for predicting the strengths of Al and its alloys. The hardness of Al alloys has been predicted by implementing various ML algorithms, such as linear regression, ridge regression, lasso regression and artificial neural network (ANN). This investigation critically analysed and discussed the application and performance of models generated by linear regression, ridge regression, lasso regression and ANN algorithms using different mechanical properties as training parameters.FindingsConsidering the definition of the problem, linear regression has been found to be the most suitable algorithm in predicting the hardness values of AA7XXX alloys as the model generated by it best fits the data set.Originality/valueThe work presented in this paper is original and not submitted anywhere else.

Optimal Design of the Austenitic Stainless-Steel Composition Based on Machine Learning and Genetic Algorithm.

As the fourth paradigm of materials research and development, the materials genome paradigm can significantly improve the efficiency of research and development for austenitic stainless steel. In this study, by collecting experimental data of austenitic stainless steel, the chemical composition of austenitic stainless steel is optimized by machine learning and a genetic algorithm, so that the production cost is reduced, and the research and development of new steel grades is accelerated without reducing the mechanical properties. Specifically, four machine learning prediction models were established for different mechanical properties, with the gradient boosting regression (gbr) algorithm demonstrating superior prediction accuracy compared to other commonly used machine learning algorithms. Bayesian optimization was then employed to optimize the hyperparameters in the gbr algorithm, resulting in the identification of the optimal combination of hyperparameters. The mechanical properties prediction model established at this stage had good prediction accuracy on the test set (yield strength: R2 = 0.88, MAE = 4.89 MPa; ultimate tensile strength: R2 = 0.99, MAE = 2.65 MPa; elongation: R2 = 0.84, MAE = 1.42%; reduction in area: R2 = 0.88, MAE = 1.39%). Moreover, feature importance and Shapley Additive Explanation (SHAP) values were utilized to analyze the interpretability of the performance prediction models and to assess how the features influence the overall performance. Finally, the NSGA-III algorithm was used to simultaneously maximize the mechanical property prediction models within the search space, thereby obtaining the corresponding non-dominated solution set of chemical composition and achieving the optimization of austenitic stainless-steel compositions.

Prediction of triaxial mechanical properties of rocks based on mesoscopic finite element numerical simulation and multi-objective machine learning

The deformation and strength characteristics of rocks are crucial for the effective development of underground resources and the construction of underground engineering projects. In this study, a novel approach is proposed to predict the triaxial mechanical properties of rocks by utilizing mesoscopic finite element numerical simulation and multi-objective machine learning. First, the mesoscopic mechanical properties of rocks are obtained through microscale finite element simulations to generate a training dataset. Then, three machine learning algorithms, namely support vector regression (SVR), artificial neural network (ANN), and random forest (RF), are employed to perform multi-objective machine learning on the triaxial elastic modulus, Poisson's ratio, and compressive strength of rocks, using uniaxial elastic modulus, Poisson's ratio, tensile strength, compressive strength, and confining pressure as feature variables. The performance evaluation, based on 10-fold cross-validation, demonstrates that all three models exhibit excellent predictive capabilities, especially the SVR and RF models, which show high accuracy and correlation, respectively. Among the five feature variables, confining pressure is the most important feature, while uniaxial tensile strength is the least important feature. The absence of uniaxial tensile strength does not significantly impact the predictive performance of the models. These findings offer novel insights into the investigation of the triaxial mechanical properties of rocks.

Experimental and Numerical Investigation of the Tensile and Failure Response of Multiple-Hole-Fiber-Reinforced Magnesium Alloy Laminates under Various Temperature Environments.

In this paper, the tensile mechanical behavior and progressive damage morphology of glass-fiber-reinforced magnesium alloy laminate for different numbers of holes in a temperature range of 25-180 °C were investigated. In addition, based on extensive tensile tests, the tensile mechanical behavior and microscopic damage morphology of porous-glass-fiber-reinforced magnesium alloy laminates at different temperatures were observed by finite element simulation and scanning electron microscopy (SEM). Finally, the numerical simulation and experimental results were in good accordance with the prediction of mechanical properties and fracture damage patterns of the laminates, the average difference between the residual strength values of the specimens at ambient temperature was 5.57%, and the stress-strain curves were in good agreement. The experimental and finite element analysis results showed that the damaged area of the bonded layer tended to expand with the increase in the number of holes, which has a lesser effect on the ultimate tensile strength. As the temperature increased, the specimens changed from obvious fiber breakage (pull-out) and the resin matrix damage mode to matrix softening damage and interfacial delamination fracture damage. As the testing temperature of the specimens increased from 25 °C to 180 °C, the tensile strength of the specimens decreased by an average of 51.59%, while the tensile strength of the specimens showed a nonlinear decreasing trend. The damage mechanism of porous-glass-fiber-reinforced magnesium alloy laminates at different temperatures is discussed in this paper, which can provide a reference for engineering applications and design.

Configuration feature extraction and mechanical properties prediction of particle reinforced metal matrix composites

Improving the strength-toughness matching properties has become a key issue in metal matrix composites, which can be achieved by some specific designs of composite configurations (i.e. spatial distribution of reinforced particles). To better interpret the composite configuration effect on mechanical properties, it is necessary to develop comprehensive feature descriptors for composite configurations. In this work, topological data analysis and particle clustering parameter are applied to extract feature descriptors for composite configurations of SiCp/Al composites, which are effective to distinguish different types of composite configurations. In addition, based on the dataset constructed by finite element analysis, machine learning algorithms are applied to predict the composite configurations-mechanical properties of SiCp/Al composites. Then, particle swarm optimization is used to search the composite configurations with extreme mechanical properties predicted by machine learning model. The results of this work indicate that the composite configurations with larger particle clustering parameter and persistent entropy will yield better strength-toughness matching.

A statistical microstructures-based method for the prediction of mechanical properties in nickel-based single crystal alloys

ABSTRACT The γ/γ’ microstructure is a key factor for the mechanical performance of nickel-based single crystal superalloys. Experimental results have shown that the size and shape of the γ’ precipitates are not uniform and follow certain probability distributions. In order to study the effect of the microstructure on the mechanical behavior of the superalloys, a method is proposed to reproduce its initial microstructure with the distribution function of its γ′ particle size. The effect of different microstructures on the creep behavior at different locations of a real blade is predicted to demonstrate the application of the method. It can be found from the simulation results that the mechanical behavior of the blade is related to not only the thickness of the blade wall but also the position in the blade.

Hydrothermal aging of short glass fiber reinforced polyphenylene sulfide composites and property prediction

In this paper, hydrothermal aging behavior of short-cut glass fiber reinforced polyphenylene sulfide composites (PPS/GF) were studied. The water adsorption behavior can be well described by Fickian law and have strong correlation with mechanical properties. The microbond test joint with morphology observation by scanning electron microscope (SEM) and separate evaluation about matrix, fiber and interface unveil that the mechanism of hydrothermal aging of PPS/GF composites and deterioration of interface play the dominant role in degradation of macroscopic mechanical properties. Finally, the life and mechanical properties prediction model of the hydrothermal aged PPS/GF composites were established based on the collected aging data and Arrhenius formula.

Prediction of mechanical properties of composite materials using multimodal fusion learning

Efficient and accurate measurement of material mechanical properties is important for material development. The mechanical properties of materials are comprehensively affected by factors such as material composition and microstructure. The existing single-mode prediction methods have the problems of poor mapping integrity and low prediction accuracy. In this paper, a multimodal fusion prediction model integrating material microstructure information and material composition information is proposed. Firstly, Convolutional neural network(CNN) is used to extract the material microstructure features, and Multilayer perceptron(MLP) is used to extract the component features. Then, the adaptive feature vector fusion module designed to adjust the influence of different modal information on the mechanical properties is used to achieve high-level feature fusion, and neural network is used to complete the mechanical properties prediction. The method is verified on the data of an alternative composite material. The method proposed provides a promising solution for the measurement and prediction of material properties.

Development of Non-Destructive Dynamic Characterization Technique for MMCs: Predictions of Mechanical Properties for Al@Al2O3 Composites

In the past several decades, many destructive and non-destructive testing techniques have been developed to evaluate the characteristics of metal matrix composites (MMCs). This research aims to calculate the mechanical properties of the Al@Al2O3 composites by varying alumina nanoparticles (Al2O3 NPs) content using a non-invasive, position sensing detector (PSD) unit-based optical method. The composite was prepared by a powder metallurgy technique, and its characterization was conducted using SEM and XRD to understand its surface morphology and microstructure. The natural frequency and Young’s modulus of the composite were estimated experimentally. Young’s modulus was calculated using this natural frequency. The proposed study shows that Young’s modulus of the composite increases with an increase in Al2O3 NPs content in the composition, irrespective of the testing method. Along with this, natural frequency also increases with the increase in the Al2O3 NPs content. Evaluated properties were compared with the numerical modeling using COMSOL Multiphysics. The experimental and numerical results are equivalent and within the margin of error. This study illustrates the development of an experimental approach for evaluating the mechanical properties of a composite material. This experimental approach can be used whenever sample dimension and space are constrained to evaluate the mechanical behavior of nanomaterials and nanocomposites.