The accuracy of various training algorithms in tribological behavior modeling of A356-B<Subscript>4</Subscript>C composites

In the present study, various artificial neural network (ANN) training algorithms were implemented for finite element technique (FEM) modeling of the composites wear behavior. The experimental results show that the weight losses of the composites are less than that of unreinforced alloy. It is believed that incorporation of hard particles to aluminum alloy contributes to the improvement of the wear resistance of the base alloy to a great extent. Hard particles take part in resisting penetration, cutting and grinding by the abrasive and protect the surface. It is noted that the increase in the weight fraction of B₄C particles improves the wear resistance of the composite. The wear resistance increases with increasing the size of reinforcing particles. The FEM method is used for discretization and to calculate the transient temperature field of quenching. During the ANN training process, the weights and biases in the network are adjusted to minimize the error and to obtain a high-performance in the solution. The test set was used to check the system accuracy of each training algorithm at the end of learning. It was observed that Bayesian regularization learning algorithm gave the best prediction.     

Influence of Varying SiO2?% on the Mechanical Properties of Al Based MMC

The current investigation is focused on effective utilization of rice husk ash (RHA) SiO₂??an industrial waste available in abundance, by systematically dispersing into an Al?CMg (0.5, 1.0, 2.5, and 5.0?% by weight) matrix resulting in synthesis of composites via Liquid Metallurgy route. The effect of increasing Mg?% to improve the wettability with the increase in SiO₂?% as reinforcement was studied comparatively. The increase in the percentage of SiO₂ beyond 5?% as reinforcement into Al?CMg alloy increases the agglomeration of SiO₂ particles which creates more sites for crack initiation and hence lowers down the load bearing capacity of the composite while microhardenss and wear testing analysis. For the composite, Al?CMg (2.5?%)?CSiO₂ (5.0?%) the hardness was observed to be maximum corresponding to minimum wear loss. The uniform distribution of maximum amount of hard spinel structure of extremely small size within the matrix confirms maximum wetting characteristic of 2.5 wt% Mg.     

Dry sliding wear of a Ti50Ni25Cu25 particulate-reinforced aluminum matrix composite

The objective of this article is to characterize the sliding wear behavior of a 30 vol pct Ti₅₀Ni₂₅Cu₂₅ particulate-reinforced aluminum matrix composite under dry conditions. The transformation temperatures of Ti₅₀Ni₂₅Cu₂₅ particles were measured before and after the compounding procedure by the differential scanning calorimeter (DSC) method. The wear tests were carried out on a pin-on-disc machine. A 10 vol pct SiC particulate-reinforced composite and pure aluminum were chosen as the comparison specimens. The results indicate that Al-30 vol pct Ti₅₀Ni₂₅Cu₂₅ composites exhibit higher wear resistance than their unreinforced matrices and are comparable with Al-10 vol pct SiC composites in this experiment. A self-adaptive mechanism that contributes to the wear resistance of an Al-30 vol pct Ti₅₀Ni₂₅Cu₂₅ composite is proposed. Scanning electron microscopy (SEM) and energy diffraction spectrum (EDS) examinations were carried out to investigate the wear mechanism and interface reactions. The results indicate that the interfacial reaction is a predominant factor in determining the wear behavior of the Ti₅₀Ni₂₅Cu₂₅/Al composite.     

Investigation on Mechanical and Adhesive Wear Behavior of Centrifugally Cast Functionally Graded Copper/SiC Metal Matrix Composite

The objective of this work is to fabricate functionally graded unreinforced copper alloy (Cu–10Sn) and a Cu–10Sn/SiC composite (Ø_out100 × Ø_in70 × 100 mm) by horizontal centrifugal casting process and to investigate its mechanical and tribological properties. The microstructure and hardness was analysed along the radial direction of the castings; tensile test was conducted at both inner and outer zones. Microstructural evaluation of composite indicated that the reinforcement particles formed a gradient structure across the radial direction and maximum reinforcement concentration was found at the inner periphery. Hence maximum hardness (205 HV) was observed at this surface. Tensile test results showed that, the tensile strength at inner zone of composite was observed to be higher (248 MPa) compared to that of the outer zone and unreinforced alloy. As mechanical properties showed better results at inner periphery, dry sliding wear experiments were carried out on the inner periphery of composite using pin-on-disc tribometer. Process parameters such as load (10–30 N), sliding distance (500–1500 m) and sliding velocity (1–3 m/s) were analyzed by Taguchi L₂₇ orthogonal array. The influence of parameters on wear rate was analyzed by signal-to-noise ratio and analysis of variance. Analysis results revealed that load (54%) had the highest effect on wear rate followed by sliding distance (18.2%) and sliding velocity (3.7%). The wear rate of composite increased with load and sliding distance, but decreased with sliding velocity. Regression equation was developed and was validated by confirmatory experiment. Worn surface of composite was observed using scanning electron microscopy and transition of wear was observed at all extreme conditions.     

SiCp/2014Al复合材料的热变形行为及本构模型

在Gleeble-3180热模拟机上对碳化硅颗粒增强铝基(SiCp/2014Al)复合材料进行热压缩试验,研究其在变形温度为350,400,450 ℃和500 ℃,应变速率为0.001,0.01,0.1s-1和1.0 s-1条件下的热变形行为。根据热压缩实验的真应变-真应力数据,在考虑应变、应变速率和变形温度对流动应力的耦合影响下构建修正的Johnson-Cook(JC)本构模型,同时建立人工神经网络模型(ANN)。结果表明:SiCp/2014Al复合材料的流变应力随应变速率的增加和温度的降低而增大。与修正的JC模型相比,ANN模型具有较低的均方根误差(0.51 MPa)和平均绝对误差(1.43%),以及较高的相关系数(0.999 7),表明其对SiCp/2014Al复合材料热变形流变应力的预测具有更高的预测精度和可靠性。      

Dry sliding wear behavior of A356-15 Pct SiC p composites under controlled atmospheric conditions

The present investigation was carried out to provide a deeper insight into the mechanism of wear behavior of A356-15 vol pct SiC _p composite under controlled argon and oxygen atmospheres through a detailed characterization of worn surfaces and subsurfaces. Dry sliding wear tests were performed for both as-cast and T6-treated specimens using a pin-on-disc machine with three sliding velocities (0.5, 1, and 2 ms⁻¹) and three loads (1, 2, and 3 MPa). The wear rate of A356-15 vol pct SiC _p composite was lower by nearly one order of magnitude under argon atmosphere compared to the specimens tested under oxygen atmosphere for all experimental conditions. Under argon atmosphere, the mechanism of material removal was by delamination wear and did not change within the parametric regime. In the case of the specimen tested under oxygen atmosphere, the wear behavior of the composite depended on the experimental conditions. At low load and low sliding velocity, the material removal was by abrasion. While at high load and high sliding velocity, the material removal mechanism was by delamination wear. Further, the mechanical mixed layer (MML) formed under argon atmosphere was more stable and homogenous compared to that formed under oxygen atmosphere. The MML formed under both atmospheres revealed much less in Fe content.     

Artificial Neural Network Modeling of Mechanical Properties of Calcium Carbonate Impregnated Coir-Polyester Composites

The tensile, flexural and impact properties of calcium carbonate particles-impregnated coir fiber-reinforced polyester composites were evaluated. The short untreated green husk coir fibers were used as reinforcement materials in unsaturated polyester resin matrix. The composite fabrications were planned with the three levels of fiber parameters namely fiber length, fiber diameter and filler content as per design of experiments (DOE) and the mechanical properties were tested as per ASTM standards. An artificial neural network (ANN) model was developed to predict the mechanical properties and it was observed that the developed ANN model accurately predicted the mechanical properties within the ranges specified.     

Correlation of microstructure with hardness and wear resistance in (CrB, MoB)/steel surface composites fabricated by high-energy electron beam irradiation

Correlation of microstructure with hardness and wear resistance of (CrB,MoB)/carbon steel surface composites fabricated by high-energy electron beam irradiation was investigated in this study. Three kinds of powder mixtures, i.e., 50CrB-50MgF₂(flux), 50MoB-50MgF₂, and 25CrB-25MoB-50MgF₂ (wt pct), were placed on a plain carbon steel substrate, which was then irradiated with the electron beam. In the specimens fabricated with flux powders, the surface composite layer of 0.8 to 1.3 mm in thickness was successfully formed without defects, and contained a large amount (up to 48 vol pct) of Cr_1.65Fe_0.35B_0.9 or Mo₂FeB₂ in the martensitic matrix. The hardness and wear resistance of the surface composite layer were directly influenced by the hard borides, and thus were about 3 to 7 times greater than those of the steel substrate. Particularly, in the surface composite fabricated with CrB and MoB powders, the hardness of eutectic solidification cells and martensitic matrix was very high, and borides formed a network structure along cells, thereby leading to the best hardness and wear resistance. These findings suggested that the high-energy electron beam irradiation was useful for the development of surface composites with improved hardness and wear resistance.     

A statistical model for sliding wear of metals in metal/composite systems

The wear of a nominally harder single-phase metal sliding against a nominally softer metal-matrix composite containing a dispersion of hard second-phase reinforcement is described by a statistical wear model which considers the effects of local variations in hardness and microstructure on asperity interactions. It was shown theoretically that the wear rate of the unreinforced component varies exponentially with nominal reinforcement volume fraction. Model experiments performed on a SiC_w-2124 Al composite/17-4 PH steel system confirmed the validity of the theory.     

Abrasive wear behavior of MoSi2SiC and MoSi2ZrO2 composites

The volume wear behavior of MoSi₂/SiC and MoSi₂/ZrO₂ composites was evaluated using 150 grit SiC particles in a pin-on-drum abrasion test. The addition of SiC whiskers or particles reduced the volume wear of the composite relative to monolithic MoSi₂ by about a factor of two, with the SiC whisker containing composite having a slightly lower volume wear rate than the SiC particulate reinforced composite. The addition of partially-stabilized (PS)-ZrO₂ particles lowered the volume wear of the composite relative to MoSi₂. The addition of unstabilized (US)-ZrO₂ or fully-stabilized (FS)-ZrO₂ particles to the MoSi₂ matrix had little effect of the volume wear relative to the unreinforced matrix. The difference in wear behavior of the ZrO₂ reinforced composites may be associated with the ability of the PS-ZrO₂ particles to transform reducing the fragmentation process during abrasion.     

Effect of Centrifugal Speed in Abrasive Wear Behavior of Al-Si5Cu3/SiC Functionally Graded Composite Fabricated by Centrifugal Casting

Centrifugal casting was adopted for fabricating AlSi5Cu3/10 wt% SiC functionally graded metal matrix composite under three different centrifugal speeds of 800, 1000 and 1200 rpm, and hollow cylindrical components (φ_out 150 × φ_in 132 × 150 mm) were obtained. Microstructures of outer and inner periphery of all composites were observed through optical microscope and micro hardness of outer, intermediate and inner region of composite was tested using Vicker’s hardness tester. Results revealed that outer region of the composites centrifuged at all speeds have particle rich region with higher hardness. Abrasive wear experiments were conducted only on surface of particle rich region based on Taguchi’s technique by varying parameters such as centrifugal speed of casting process, rotating speed and applied load of abrasive wear tester. Analysis of variance results revealed that, centrifugal speed had highest significance on wear rate. Abraded surfaces were examined using scanning electron microscope and the maximum wear resistance was observed on particle rich zone of composite centrifuged at 1200 rpm.     

Dry sliding wear behavior of plasma-sprayed aluminum hybrid composite coatings

Al-SiC _p composite and Al-SiC _p -C _p hybrid composite coatings were produced by plasma spraying of premixed powders onto A356 alloy substrates. Four composite coatings, Al+20 vol pct SiC _p , Al+20 vol pct SiC _p +C _p , Al+40 vol pct SiC _p , and Al+40 vol pct SiC _p +C _p , were obtained. The dry sliding wear behavior of these coatings and pure aluminum have been studied at a sliding velocity of 1 m/s in the applied-load range of 25 to 150 N (corresponding to a normal stress of 0.5 to 3 MPa). The composite coatings had a significantly improved wear resistance over pure Al. The composite coatings with a higher SiC _p content of 40 vol pct exhibited superior wear resistance than those with a lower SiC _p content of 20 vol pct. The presence of graphite particles had different influences on the wear resistance, depending on the applied load. At lower loads, graphite improved the wear resistance considerably. At higher loads, the wear resistance of the hybrid composite coatings was similar to that of the composite coatings without graphite particles. At lower loads, an oxidative wear mechanism was dominant. At higher loads, delamination was a major wear mechanism. Graphite particles did not change their wear mechanism at the same applied loads.     

Wear Behavior of a Novel Aluminum-Based Hybrid Composite

In the current research, the dry sliding wear behaviors of 6351 Al alloy and its composite with hybrid reinforcement (ex situ SiC and in situ Al₄SiC₄) were investigated at low sliding speed (1 m s^?1) against a hardened EN 31 disk at different loads. The wear mechanism involved adhesion and microcutting-abrasion at lower load. On the other hand, at higher load, abrasive wear involving microcutting and microplowing along with adherent oxide formation was observed. Initially, under higher load, the abrasive wear mechanism caused rapid wear loss up to a certain sliding distance. Afterward, by virtue of frictional heat generation and associated temperature rise, an adherent oxide layer was developed at the pin surface which drastically reduced the wear loss. The overall wear rate increased with load in alloy as well as in composite. Moreover, the overall wear rate of the composite was found lower than that of the 6351 Al alloy at all applied loads. The ex situ SiC particles were found to resist abrasive wear, while, in situ Al₄SiC₄ particles offered resistance to adhesive wear. Accordingly, the 6351 Al (SiC + Al₄SiC₄) hybrid composite exhibited superior wear resistance relative to the 6351 Al alloy.     

Mechanical Properties and Dry Sliding Wear Behaviour of Molybdenum Disulphide Reinforced Zinc–Aluminium Alloy Composites

ZA-27 alloy is a lightest alloy which offers excellent bearing and mechanical properties in automobile and industrial applications. In this study, the MoS₂ particles with 0.5, 1 and 1.5 (wt%) weight percentages were reinforced in ZA-27 alloy to form composites, which were fabricated by using ultrasonic assisted stir casting method. The ZA-27/MoS₂ composite specimens were examined for chemical composition with the aid of XRD technique and EDS. Microstructure analysis of the ZA-27/MoS₂ composites was studied using SEM. Tests were conducted for mechanical properties such as tensile strength and hardness on ZA-27/MoS₂ composites samples as per ASTM standards. Dry sliding wear behavior of the composites was tested at various operating conditions by using pin-on-disc apparatus. Microstructural images of the ZA-27 composites reveal that there is a uniform dispersion of the MoS₂ particles in the base material. From the results it is observed that the mechanical properties increases with ZA-27 reinforced with 0.5 wt% MoS₂ composite and further decreases with increase in the filler content. The enhanced wear resistance is observed in ZA-27 reinforced MoS₂ composites as compared to the unreinforced alloy. The wear rate of the ZA-27 composites decreases with the increase in filler content, further the worn surfaces as examined using SEM reveals the wear mechanism explaining the improved wear resistance of the particulate composites.     

Hybrid Neural Network—Finite Element River Flow Model

Results obtained from a hybrid neural network—finite element model are reported in this paper. The hybrid model incorporates artificial neural network (ANN) nodes into a numerical scheme, which solves the two-dimensional shallow water equations using finite elements (FE). First, numerical computations are carried out on the entire numerical model, using a larger mesh. The results from this computation are then used to train several preselected ANN nodes. The ANN nodes model the response for a part of the entire numerical model by transferring the system reaction to the location where both models are connected in real time. This allows a smaller mesh to be used in the hybrid ANN-FE model, resulting in savings in computation time. The hybrid model was developed for a river application, using the computational nodes located at the open boundaries to be the ANN nodes for the ANN-FE hybrid model. Real-time coupling between the ANN and FE models was achieved, and a reduction is CPU time of more than 25% was obtained.     

High sodium intake decreases pressure-induced (myogenic) tone and flow-induced dilation in resistance arteries from hypertensive rats

A real-time automated way of quantifying metabolites from in vivo NMR spectra using an artificial neural network (ANN) analysis is presented. The spectral training and test sets for ANN containing peaks at the chemical shift ranges resembling long echo time proton NMR spectra from human brain were simulated. The performance of the ANN constructed was compared with an established lineshape fitting (LF) analysis using both simulated and experimental spectral data as inputs. The correspondence between the ANN and LF analyses showed correlation coefficients of order of 0.915-0.997 for spectra with large variations in both signal-to-noise and peak areas. Water suppressed 1H NMR spectra from 24 healthy subjects were collected and choline-containing compounds (Cho), total creatine (Cr), and N-acetyl aspartate (NAA) were quantified with both methods. The ANN quantified these spectra with an accuracy similar to LF analysis (correlation coefficients of 0.915-0.951). These results show that LF and ANN are equally good quantifiers; however, the ANN analyses are more easily automated than LF analyses.     

Correlation of microstructure with the hardness and wear resistance of (TiC,SiC)/Ti-6Al-4V surface composites fabricated by high-energy electron-beam irradiation

The correlation of microstructure with the hardness and wear resistance of (TiC,SiC)/Ti-6Al-4V surface composites fabricated by high-energy electron-beam irradiation was investigated in this study. The mixtures of TiC, SiC, or TiC + SiC powders and CaF₂ flux were placed on a Ti-6Al-4V substrate, and then an electron beam was irradiated on these mixtures using an electron-beam accelerator. The surface composite layers of 1.2 to 2.1 mm in thickness were formed without defects and contained a large amount (up to 66 vol pct) of precipitates such as TiC and Ti₅Si₃ in the martensitic matrix. This microstructural modification, including the formation of hard precipitates and a hardened matrix in the surface composite layer, improved the hardness and wear resistance. Particularly in the surface composite fabricated with TiC + SiC powders, the wear resistance was greatly enhanced to a level 25 times higher than that of the Ti alloy substrate, because 66 vol pct of TiC and Ti₅Si₃ was precipitated homogeneously in the hardened martensitic matrix. These findings suggested that high-energy electron-beam irradiation was useful for the development of Ti-based surface composites with improved hardness and wear properties.     

Case Study: Finite Element Method and Artificial Neural Network Models for Flow through Jeziorsko Earthfill Dam in Poland

A finite element method (FEM) and an artificial neural network (ANN) model were developed to simulate flow through Jeziorsko earthfill dam in Poland. The developed FEM is capable of simulating two-dimensional unsteady and nonuniform flow through a nonhomogenous and anisotropic saturated and unsaturated porous body of an earthfill dam. For Jeziorsko dam, the FEM model had 5,497 triangular elements and 3,010 nodes, with the FEM network being made denser in the dam body and in the neighborhood of the drainage ditches. The ANN model developed for Jeziorsko dam was a feedforward three layer network employing the sigmoid function as an activator and the back-propagation algorithm for the network learning. The water levels on the upstream and downstream sides of the dam were input variables and the water levels in the piezometers were the target outputs in the ANN model. The two models were calibrated and verified using the piezometer data collected on a section of the Jeziorsko dam. The water levels computed by the models satisfactorily compared with those measured by the piezometers. The model results also revealed that the ANN model performed as good as and in some cases better than the FEM model. This case study offers insight into the adequacy of ANN as well as its competitiveness against FEM for predicting seepage through an earthfill dam body.     

Predicting Longitudinal Dispersion Coefficient in Natural Streams by Artificial Neural Network

An artificial neural network (ANN) model was developed to predict the longitudinal dispersion coefficient in natural streams and rivers. The hydraulic variables [flow discharge (Q), flow depth (H), flow velocity (U), shear velocity (u*), and relative shear velocity (U/u*)] and geometric characteristics [channel width (B), channel sinuosity (σ), and channel shape parameter (β)] constituted inputs to the ANN model, whereas the dispersion coefficient (Kx) was the target model output. The model was trained and tested using 71 data sets of hydraulic and geometric parameters and dispersion coefficients measured on 29 streams and rivers in the United States. The training of the ANN model was accomplished with an explained variance of 90% of the dispersion coefficient. The dispersion coefficient values predicted by the ANN model satisfactorily compared with the measured values corresponding to different hydraulic and geometric characteristics. The predicted values were also compared with those predicted using several equations that have been suggested in the literature and it was found that the ANN model was superior in predicting the dispersion coefficient. The results of sensitivity analysis indicated that the Q data alone would be sufficient for predicting more frequently occurring low values of the dispersion coefficient (Kx<100?m2/s). For narrower channels (B/H<50) using only U/u* data would be sufficient to predict the coefficient. If β and σ were used along with the flow variables, the prediction capability of the ANN model would be significantly improved.     

Ni3AI intermetallic particles as wear-resistant reinforcement for Al-base composites processed by powder metallurgy

The suitability of Ni₃Al intermetallics as reinforcement for Al-base materials for tribological applications has been investigated. For this purpose, an Al/Ni₃Al (5 vol pct) composite was prepared by powder metallurgy and tested in air against steel counterfaces at the load range of 45 to 178 N. For comparison, unreinforced Al specimens were processed and tested under the same conditions. Tribological behavior was evaluated by microstructural examination of wear-affected zones and weightloss measurements of specimens and counterfaces. It was found that a significant amount of Fe-rich oxide particles become incorporated into the Al matrix during wear, forming a cracked tribolayer. The wear behavior of Al/Ni₃l composite as a function of the applied load was not accurately reflected by the weight loss of worn specimens. Results highlight the role of Ni3Al particles as loadbearing elements due to their excellent bonding to the Al matrix, their interfaces withstanding the wear stresses even at the highest applied load. Moreover, Ni3Al particles limited the incorporation of wear debris to the Al matrix and reduced wear damage occasioned to the steel counterfaces compared to that of pure aluminum specimens. Formerly with the Physical Metallurgy Department (CENIM-CSIC)