Prediction Of Tensile Strength Research Articles

A Deep Neural Network Approach towards Performance Prediction of Bituminous Mixtures Produced Using Secondary Raw Materials

With the progressive reduction in virgin material availability and the growing global concern for sustainability, civil engineering researchers worldwide are shifting their attention toward exploring alternative and mechanically sound technological solutions. The feasibility of preparing both cold and hot asphalt mixtures (AMs) for road pavement binder layers with construction and demolition wastes (C&DWs) and reclaimed asphalt pavement (RAP) partially replacing virgin materials like limestone aggregates and filler has already been proven. The technical suitability and compliance with technical specifications for road paving materials involved the evaluation of mechanical and volumetric aspects by means of indirect tensile strength tests and saturated surface dry voids, respectively. Thus, the main goal of the present study is to train, validate, and test selected machine learning algorithms based on data obtained from the previous experimental campaign with the aim of predicting the volumetric properties and the mechanical performance of the investigated mixtures. A comparison between the predictions made by ridge and lasso regression techniques and both shallow (SNN) and deep neural network (DNN) models showed that the latter achieved better predictive capabilities, highlighted by fully satisfactory performance metrics. DNN performance can be summarized by R2 values equal to 0.8990 in terms of saturated surface dry void predictions, as well as 0.9954 in terms of indirect tensile strength predictions. Predicted observations can be thus implemented within the traditional mix design software. This would reduce the need to carry out additional expensive and time-consuming experimental campaigns.

Development of machine learning regression models for the prediction of tensile strength of friction stir processed AA8090/SiC surface composites

Friction Stir Processing is a state-of-the-art technology for microstructure refinement, material property enhancement, and fabrication of surface composites. Machine learning approaches have garnered significant interest as prospective models for modeling various production systems. The present work aims to develop four machine learning models, namely linear regression, support vector regression, artificial neural network and extreme gradient boosting to predict the influence of FSP parameters such as tool rotational speed, tool traverse speed and groove width on ultimate tensile strength of friction stir processed AA8090/SiC surface composites. These models were developed through Python programming and the original dataset was divided into 80% for the training phase and 20% for the testing phase. The performance of the models was evaluated by root mean squared error, mean absolute error and R2. Based on the results and graphical visualization, it was observed that the XGBoost model outperformed other models with high accuracy in predicting UTS of AA8090/SiC surface composites.

A transfer learning strategy for tensile strength prediction in austenitic stainless steel across temperatures

Austenitic stainless steel plays a pivotal role in the nuclear industry with exceptional mechanical properties and corrosion resistance, wherein high-temperature tensile strength (∼290 °C) is a critical factor in ensuring its performance and reliability. This study focuses on the predictive modeling of tensile strength in cast austenitic stainless steel at room and elevated temperatures by developing a transfer learning strategy. The application of a gradient-boosting decision tree algorithm, enhanced by key features derived from room-temperature analyses and augmented with partial high-temperature data, results in significantly increased accuracy. The effective selection of features underscores the stability of critical variables across different temperatures, affirming the efficacy of the proposed transfer learning strategy. The work sheds light on material selection and design for high-temperature applications, and lays the groundwork for future exploration into predictive modeling of material properties in extreme conditions.

Tensile strength prediction of crumb rubber concrete based on mesoscale modelling

ABSTRACT This study presents an experimental study and a mesoscale model for evaluating the tensile performance of crumb rubber concrete (CRC). In the experimental study, indirect tensile strengths of CRC with 6%, 12% 18% rubber contents were tested. Then the tested CRC samples were modelled as a five-phase material containing rubber, coarse aggregate, mortar, aggregate-mortar interface layer, and rubber-mortar interface layer. Numerical results from the mesoscale model agreed well with experimental tests. Parameter analysis was carried out to investigate the content, size, shape of rubber particle, thickness of interface layer on the tensile strength of CRC. Numerical results also indicated that rubber content was the governing influencing factor on the tensile strength reduction. As comparison, the influence from rubber size and rubber shape was relatively low. For a specimen with constant rubber content and size, changes in the distribution of rubber particles resulted in slight changes in the tensile strength of CRC as well. The effect of rubber particles on the tensile properties of concrete was similar to that of pores. Finally, tensile strength prediction formula was developed based on the pores model proposed in this study.

Prediction of split tensile strength of recycled aggregate concrete leveraging explainable hybrid XGB with optimization algorithm

Prediction of split tensile strength of recycled aggregate concrete leveraging explainable hybrid XGB with optimization algorithm

Tensile strength prediction and process parameters optimization of FSW thick AA2219-T8 based on ANN-GA

Tensile strength prediction and process parameters optimization of FSW thick AA2219-T8 based on ANN-GA

Micromagnetic and Quantitative Prediction of Yield and Tensile Strength of Carbon Steels Using Transfer Learning Method

Micromagnetic and Quantitative Prediction of Yield and Tensile Strength of Carbon Steels Using Transfer Learning Method

The strength prediction model of unidirectional fiber reinforced composites based on the renormalization group method

When unidirectional fiber reinforced composites are subjected to longitudinal tensile loading and reach a critical failure state, they experience a sudden transition from local damage to catastrophic failure, commonly termed as an avalanche event. This paper integrates the self-organized criticality theory (SOC) concepts into the prediction of longitudinal tensile strength of composites and establishes a strength prediction model of composites based on the renormalization group method (RGM). The predictions of the RGM model are successfully validated against experimental results in the literatures, and it demonstrates relatively acceptable predictive accuracy compared to classical strength criteria. Compared to other state-of-the-art prediction models considering stochastic fiber strength distribution, the RGM model effectively provides strength statistics for fiber bundles of any size and describes the occurrence of composites avalanche failure induced by local stress concentration. This present model can be very conveniently implemented as a User Material Subroutine (UMAT) for finite simulations, facilitating practical prediction of the strength of composites structures.

Utilizing ANFIS for strength characteristics forecasting in variable heat-cured geopolymer composites

Reducing cement usage through geopolymer concrete (GPC) can be crucial for resolving environmental concerns. The byproduct of the coal and steel industry, such as fly ash (FA) and ground granulated blast furnace slag (GGBS), is mainly used as a precursor in the production of GPC due to the availability of aluminosilicate products in it. In this work, nano silica is added to the GPC in different percentages and GPC is cured at different temperatures to study the change in the strength of GPC due to change in curing temperature. The compressive strength, flexural strength, and split tensile strength up to 68.99 MPa, 5.89 MPa, 6.82 MPa, respectively was achieved after 28 days of heat curing. However, testing GPC experimentally to anticipate its strength is time-consuming and expensive. Therefore, the ability to precisely forecast concrete strength can be achieved through machine learning. In this study, the correlation for both nano silica (%) and curing temperature towards the strength of concrete is also investigated using adaptive neuro fuzzy interference system (ANFIS) models. ANFIS results showed better compressive, split tensile, and flexural strength prediction with the highest R2 0.9801, 0.9943, and 0.9877, respectively. The performance efficiency of the developed ANFIS model is more significant minimum error for compressive strength prediction (MAE = 0.21375, RMSE = 0.02574, MAPE = 0.42), split tensile strength prediction (MAE = 0.07375, RMSE = 0.0923, MAPE = 1.46) and flexural strength prediction (MAE = 0.04, RMSE = 0.0728, MAPE = 0.78) of GPC cured under different temperature conditions. The results from experiments show that the prediction efficiency of the model is also good, as indicated by the correlation coefficient (R). This model is helpful for further designing the proportion of material for achieving good strength of GPC.

Investigation of melt flow index and tensile properties of dual metal reinforced polymer composites for 3D printing using machine learning approach: Biomedical and engineering applications

This study investigates the enhancement of mechanical properties of metal/polymer composites produced through fused deposition modeling and the prediction of the ultimate tensile strength (UTS) by machine learning using a Classification and Regression Tree (CART). The composites, comprising 80% acrylonitrile butadiene styrene matrix and 10% each of aluminum (Al) and copper (Cu) fillers, were subjected to a comprehensive exploration of printing parameters, including printing temperature, infill pattern, and infill density using the Taguchi method. The CART unveiled a hierarchical tree structure with four terminal nodes, each representing distinct subgroups of materials characterized by similar UTS properties. The predictors’ importance was assessed, highlighting their role in determining material strength. The model exhibited a high predictive power with an R-squared value of 0.9154 on the training data and 0.8922 on the test data, demonstrating its efficacy in capturing variability. The optimal combination of parameters for maximizing UTS was a zigzag infill pattern, a printing temperature of 245 °C, and an infill density of 10%, which is associated with the highest UTS of 680 N. The model’s reliability was confirmed through a paired t-test and test and confidence interval for two variances, revealing no significant difference between the observed and predicted UTS values. This research contributes to advancing additive manufacturing processes by leveraging CART analysis to optimize printing parameters and predict material strength. The identified optimal conditions and subgroup characteristics pave the way for developing robust and predictable metal/polymer composites, offering valuable insights for material design in the era of advanced manufacturing technologies.

Tensile strength prediction of unidirectional polyacrylonitrile (PAN)-based carbon fiber reinforced plastic composites considering stress distribution around fiber break points

Recent advancements in enhancing the mechanical characteristics of carbon fibers open up new application possibilities for carbon fibre-reinforced plastic (CFRP) composites. Particularly in unidirectional CFRPs, which form the basal structure of CFRP laminates, developing a micromechanics model capable of predicting the tensile strength of unidirectional CFRPs based on carbon fiber mechanical characteristics is a current aspiration. This study conducted a stress distribution analysis around the fiber fracture point to predict the tensile strengths of unidirectional CFRPs prepared with five types of polyacrylonitrile (PAN)-based carbon fibers, each with unique mechanical characteristics. Numerical simulation results obtained using a unidirectional CFRP model that considered the stress concentration, fiber axial stress, and bimodal Weibull distribution were reasonably consistent with the experimental results for the tensile strengths of unidirectional CFRP composites, regardless of the differences in the mechanical characteristics of the fiber. Our findings can provide guidance for designing further enhanced high-performance CFRP materials.

Fire behavior and post-fire residual tensile strength prediction of carbon fiber/phthalonitrile composite laminates

Carbon fiber/phthalonitrile (Cf/PN) composite has a high potential for applications in the aerospace sector because of the outstanding heat and flame resistance of phthalonitrile resin. Here we aim to evaluate the fire behavior of Cf/PN laminate under localized one-sided flame heating as well as the residual tensile performance. The residual tensile strength after quasi-isothermal pyrolysis in an inert environment decreases linearly with the increase in the pyrolysis degree. The temperature response and pyrolysis behavior are analyzed through a combination of experiments and finite element simulations, and the damage mode of the laminate structure is concluded through morphological observation after flame exposure and the surface axial strain field using digital image correlation. Eventually, the residual tensile strength of laminates with different thicknesses after different times of flame exposure was effectively predicted based on the numerical simulated pyrolysis degree distribution. This research supplies fundamental test data on the mechanical properties of Cf/PN laminates and is expected to provide guidelines for the engineering application of Cf/PN composites in thermal protection/load-bearing all-in-one structures.

An improved tensile strength and failure mode prediction model of FDM 3D printing PLA material: Theoretical and experimental investigations

3D-printed polylactic acid (PLA) is being increasingly used in practical engineering. In the past, a certain number of models were developed to describe the tensile strength of PLA specimens. However, two opposite tensile strength variation trends with raster angle varying were observed in previous experiments, which cannot be reasonably explained by the current tensile strength models. To resolve this issue, an improved tensile strength and failure mode prediction model is proposed based on the assumption of transversely isotropic material. Corresponding experiments of PLA specimens have also been performed to validate the accuracy of the proposed model. It has been demonstrated that this model could accurately predict the tensile strength and failure mode of 3D printing PLA material. In addition, the opposite strength trends mentioned above with raster variation could also be perfectly explained using the improved prediction model.

A Method for the Tensile Strength Prediction of Tablets with Differing Powder Plasticities.

Tablets are the most commonly used dosage form in the pharmaceutical industry, and their properties such as disintegration, dissolution, and portability are influenced by their strength. However, in industry, the mixing fraction of powders to obtain a tablet compact with sufficient strength is determined based on empirical rules. Therefore, a method for predicting tablet strength based on the properties of a single material is required. The objective of this study was to quantitatively evaluate the relationship between the compression properties and tablet strength of powder mixtures. The compression properties of the powder mixtures with different plasticities were evaluated based on the force-displacement curves obtained from the powder compression tests. Heckel and compression energy analyses were performed to evaluate compression properties. During the compression energy analysis, the ratio of plastic deformation energy to elastic deformation energy (Ep/Ee) was assumed to be the plastic deformability of the powder. The quantitative relationship between the compression properties and tensile strength of the tablets was investigated. Based on the obtained relationship and the compression properties of a single material, a prediction equation was put forward for the compression properties of the powder mixture. Subsequently, a correlation equation for tablet strength was proposed by combining the values of K and Ep/Ee obtained from the Heckel and compression energy analyses, respectively. Finally, by substituting the compression properties of the single material and the mass fraction of the plastic material into the proposed equation, the tablet strength of the powder mixture with different plastic deformabilities was predicted.

A tensile properties-related fatigue strength predicted machine learning framework for alloys used in aerospace

A tensile properties-related fatigue strength prediction framework based on machine learning (ML) methods was proposed. Firstly, 200 data containing six materials used in aerospace were collected. Secondly, tensile properties-related ML framework, including residual neural network (ResNet) model, gradient boosting decision tree (GBDT) model and LightGBM (LGB) model, was established to predict fatigue strength under symmetric cycles (R = -1) after features correlation analysis. Then, Gaussian noise was introduced into the features to prove the robustness and feature sensitivity of proposed ML framework. Finally, material characteristics of fatigue resistance in the next generation were discussed after the analysis of the correlation between the tensile properties and fatigue strength. Compared with ± 50 % error band of classical prediction models, the prediction results of proposed ML framework located in the ± 10 % error band, especially for the special crystal orientations and stress ratios. In practical applications, the tensile properties-related ML framework could be applied efficiently to materials fatigue strength prediction used in aerospace due to its robust prediction accuracy and generalization ability without additional parameters fitting.

Prediction of ultimate tensile strength of Al‐Si alloys based on multimodal fusion learning

AbstractExploring the “composition‐microstructure‐property” relationship is a long‐standing theme in materials science. However, complex interactions make this area of research challenging. Based on the image processing and machine learning techniques, this paper proposes a multimodal fusion learning framework that comprehensively considers both composition and microstructure in prediction of the ultimate tensile strength (UTS) of Al‐Si alloys. Firstly, the composition and image information are collected from the literature and supplementary experiments, followed by the image segmentation and quantitative analysis of eutectic Si images. Subsequently, the quantitative analysis results are combined with other features for three‐step feature screening, and 12 key features are obtained. Finally, four machine‐learning models (i.e., decision tree, random forest, adaptive boosting, and extreme gradient boosting [XGBoost]) are used to predict the UTS of Al‐Si alloys. The results show that the quantitative analysis method proposed in this paper is superior to Image‐Pro Plus (IPP) software in some aspects. The XGBoost model has the best prediction performance with R2 = 0.94. Furthermore, five mixed features and their critical values that significantly affect UTS are identified. Our study provides enlightenment for the prediction of UTS of Al‐Si alloys from composition and microstructure, and would be applicable to other alloys.

Sample Size Requirements of a Pharmaceutical Material Library: A Case in Predicting Direct Compression Tablet Tensile Strength by Latent Variable Modeling.

The material library is an emerging, new data-driven approach for developing pharmaceutical process models. How many materials or samples should be involved in a particular application scenario is unclear, and the impact of sample size on process modeling is worth discussing. In this work, the direct compression process was taken as the research object, and the effects of different sample sizes of material libraries on partial least squares (PLS) modeling in the prediction of tablet tensile strength were investigated. A primary material library comprising 45 materials was built. Then, material subsets containing 5 × i (i = 1, 2, 3, …, 8) materials were sampled from the primary material library. Each subset underwent sampling 1000 times to analyze variations in model fitting performance. Both hierarchical sampling and random sampling were employed and compared, with hierarchical sampling implemented with the help of the tabletability classification index d. For each subset, modeling data were organized, incorporating 18 physical properties and tableting pressure as the independent variables and tablet tensile strength as the dependent variable. A series of chemometric indicators was used to assess model performance and find important materials for model training. It was found that the minimum R2 and RMSE values reached their maximum, and the corresponding values were kept almost unchanged when the sample sizes varied from 20 to 45. When the sample size was smaller than 15, the hierarchical sampling method was more reliable in avoiding low-quality few-shot PLS models than the random sampling method. Two important materials were identified as useful for building an initial material library. Overall, this work demonstrated that as the number of materials increased, the model's reliability improved. It also highlighted the potential for effective few-shot modeling on a small material library by controlling its information richness.

Characterization and economization of cementitious tile bond adhesives using machine learning technique

Cementitious Tile Adhesives (CTAs) play a key role in ensuring the elegant outlook of buildings by improving the serviceability of various tile applications, hence making them an integral part of the dry-mix mortar industry with a significant market share globally. However, CTAs require special raw material for their production, making them a less economical option for sustainable construction applications. Hence, the current research work undertook an effort to explore the potential of different type of fine materials to produce CTAs to obtain a cheap value-added product. A total of 36 mixture proportions were prepared with six different sources of fine aggregates (i.e., silica sand, quartz sand, dune sand, river sand 1, river sand 2, and river sand 3), with six varying binder-to-fine aggregates ratios, and their mechanical strength parameters were evaluated. Further microstructure analysis was performed for fine aggregates, and best and worst performing mixture proportions to gain a deep understanding about their mechanical performance. The results revealed that the mixture proportions utilizing silica sand outperformed all other formulations in terms of compressive strength, and tensile strength with a value of 25.28 MPa, and 1.35 MPa, respectively, while dune sand performed the best in terms of shear strength 1.92 MPa with cement content of 40 % at 28 days. Moreover, microstructural investigation also showed the better microstructural performance of silica sand as compared to that of all other sources of sand. To increase the applicability of the undertaken research project, an Artificial Intelligence (AI) based neural network model, along with predictive equations, was developed for the prediction of compressive strength (R2 = 0.9831), shear strength (R2 = 0.9808), and tensile strength (R2 = 0.9862). Hence, it can be concluded that the current research work provides a foundation for an effective product development process to produce tile bond adhesives using different type of raw materials available leading to economical and sustainable manufacturing of cementitious adhesives.

Modelling of Fatigue Delamination Growth and Prediction of Residual Tensile Strength of Thermoplastic Coupons.

Thermoplastic composites are continuously replacing thermosetting composites in lightweight structures. However, the accomplished work on the fatigue behavior of thermoplastics is quite limited. In the present work, we propose a numerical modeling approach for simulating fatigue delamination growth and predicting the residual tensile strength of quasi-isotropic TC 1225 LM PAEK thermoplastic coupons. The approach was supported and validated by tension and fatigue (non-interrupted and interrupted) tests. Fatigue delamination growth was simulated using a mixed-mode fatigue crack growth model, which was based on the cohesive zone modeling method. Quasi-static tension analyses on pristine and fatigued coupons were performed using a progressive damage model. These analyses were implemented using a set of Hashin-type strain-based failure criteria and a damage mechanics-based material property degradation module. Utilizing the fatigue model, we accurately foretold the expansion of delamination concerning the cycle count across all interfaces. The results agree well with C-scan images taken on fatigued coupons during interruptions of fatigue tests. An unequal and unsymmetric delamination growth was predicted due to the quasi-isotropic layup. Moreover, the combined models capture the decrease in the residual tensile strength of the coupons. During the quasi-static tension analysis of the fatigued coupons, we observed that the primary driving failure mechanisms were the rapid spread of existing delamination and the consequential severe matrix cracking.

Intelligent Prediction of FSW Physical Quantity and Joint Mechanical Properties

Friction stir welding (FSW) process parameters influence welding temperature field and axial force, which affect welding strength. At present, how the FSW process parameters of aluminum alloy 2219-T8 thick plates influence process physical quantity and how the process physical quantity changes the tensile strength about the welded joint are unknown. We focus on the intelligent prediction of FSW temperature, axial force, and mechanical properties, to provide a basis for FSW process control of aluminum alloy 2219-T8 thick plate. Firstly, we conducted the FSW experiment of aluminum alloy 2219-T8 thick plate. Then, we input the welding process parameters, set up a prediction model by particle swarm optimization-back propagation (PSO-BP) neural network to predict the peak temperature and axial force. Finally, we input the peak temperature and axial force, use genetic algorithm-back propagation (GA-BP) neural network to establish a weld tensile strength estimation model, and comply with the prediction of tensile strength.