Nadia Ucciardello

Artificial neural networks to optimize the extrusion of an aluminium alloy

Extrusion of aluminium alloys is a complex process which depends on the characteristics of the material and on the process parameters (initial billet temperature, extrusion ratio, friction at the interfaces, die geometry etc.). The temperature profile at the die exit, largely influences microstructure, mechanical properties, and surface quality of an extruded product, consequently it is the most important parameter for controlling the process. In turn the temperature profile depends on other process variables whose right choice is fundamental to avoid surface damage of the extruded product. In the present work, two neural networks were implemented to optimize the aluminium extrusion process determining the temperature profile of an Al 6060 alloy (UNI 9006/1) at the exit of induction heater (ANN1) and at the exit of the die (ANN2). The three-layer neural networks with Levemberg Marquardt algorithm were trained with the experimental data from the industrial process. The temperature profiles, predicted by the neural network, closely agree with experimental values.

Effect of powder mix composition on Al foam morphology

The effect of mix composition on foam morphology has been examined by image analysis carried out on metallographic sections of Al foams prepared by powder metallurgy. Two sets of samples have been prepared by using SiC particles with mean sizes of 37 and 60 µm. Each set consists of 16 groups of samples with different amounts of TiH2 (0.1, 0.2, 0.4, and 0.6 wt%) and SiC (0.8, 2.8, 6, and 9 wt%). The distribution of SiC particles on the internal walls of the bubbles has been investigated by scanning electron microscopy observations, which evidenced also the presence of particles of another phase, identified as Ti3Al by energy dispersion spectroscopy. Some tests, performed without SiC particles, showed that the Al foaming occurs also under these conditions, however foams exhibit few bubbles of very large size and irregular shape. Experimental data have been used for training two multi-layer feedforward artificial neural networks. The models represent useful tools for predicting morphologic features of foams for any given mix composition in the training range.

Process Parameters Optimization of Laser Beam Welded Joints by Neural Network

Laser beam welding of C–Mn steel plates with Ni powder filler metal has been performed. Metallography samples of the welded cross-section have been observed by scanning electron microscopy (SEM) and submitted to energy dispersive spectroscopy to obtain Ni concentration profiles. On the basis of the experimental results, neural networks have been carried out. These networks were first validated and then utilized to foresee Ni concentration along the welded thickness. The objective of obtaining the best Ni penetration and minimizing powder loss was reached optimizing, by numerical simulation, process parameters, such as powder rate and joint geometry.

Discontinuous precipitation in a high-nitrogen austenitic steel

The discontinuous precipitation of a high-nitrogen (0.8 wt%) austenitic steel has been investigated after successive steps of heat treatment at two different temperatures (800 and 850 °C). After each step of heating the material has been examined by X-ray diffraction (XRD), optical microscopy (OM), transmission electron microscopy (TEM), Auger electron spectroscopy (AES) and microhardness tests. The precipitation of Cr2N induces the formation of a secondary austenitic phase, leads to the redistribution of N between transformed and untransformed zones and to local variations of mechanical properties.

An Application of Neural Network Solutions to Modeling of Diode Laser Assisted Forming Process of AA6082 Thin Sheets

In this paper a neural network approach is used to model the diode laser assisted forming process. In particular thin sheets of Aluminum alloy AA 6082 were bended in the elastic range and then treated with a diode laser with the aim to reduce the spring back phenomenon. Experimental tests were performed to study the influence of the process parameters such as laser power, laser speed and starting elastic deformation on the evolution of forming process. In particular the heating effects on the elastic properties of the material was studied. A statistical approach is used to define the experimental plan and discuss the experimental results. Interesting trend of the effects of the diode laser on the forming process were found. Subsequently in order to predict the residual inflexion, during the laser forming, a multilayer feedforward artificial neural network has been implemented. A sensitivity analysis on the artificial neural network model is used to show the significance of all the input data employed. As a result of sensitivity analysis, a check between experimental and calculated trends for each investigated variables was performed, which revealed an appreciable fit between data displayed.

Al2O3 Coatings on Magnesium Alloy Deposited by the Fluidized Bed (FB) Technique

Magnesium alloys are widely employed in several industrial domains for their outstanding properties. They have a high strength-weight ratio, with a density that is lower than aluminum (33% less), and feature good thermal properties, dimensional stability, and damping characteristics. However, they are vulnerable to oxidation and erosion-corrosion phenomena when applied in harsh service conditions. To avoid the degradation of magnesium, several coating methods have been presented in the literature; however, all of them deal with drawbacks that limit their application in an industrial environment, such as environmental pollution, toxicity of the coating materials, and high cost of the necessary machinery. In this work, a plating of Al2O3 film on a magnesium alloy realized by the fluidized bed (FB) technique and using alumina powder is proposed. The film growth obtained through this cold deposition process is analyzed, investigating the morphology as well as tribological and mechanical features and corrosion behavior of the plated samples. The resulting Al2O3 coatings show consistent improvement of the tribological and anti-corrosive performance of the magnesium alloy.

High Thermal Conductivity of Copper Matrix Composite Coatings with Highly-Aligned Graphite Nanoplatelets

Nanocomposite coatings with highly-aligned graphite nanoplatelets in a copper matrix were successfully fabricated by electrodeposition. For the first time, the disposition and thermal conductivity of the nanofiller has been evaluated. The degree of alignment and inclination of the filling materials has been quantitatively evaluated by polarized micro-Raman spectroscopy. The room temperature values of the thermal conductivity were extracted for the graphite nanoplatelets by the dependence of the Raman G-peak frequency on the laser power excitation. Temperature dependency of the G-peak shift has been also measured. Most remarkable is the global thermal conductivity of 640 ± 20 W·m−1·K−1 (+57% of copper) obtained for the composite coating by the flash method. Our experimental results are accounted for by an effective medium approximation (EMA) model that considers the influence of filler geometry, orientation, and thermal conductivity inside a copper matrix.

Micro-Chemistry and Mechanical Behaviour of Ti6Al4V-SiCf Composite Produced by HIP for Aeronautical Applications

The paper reports the results of an extensive characterization of the Ti6Al4V-SiCf composite produced by hot isostatic pressing (HIP) to assess its capability to withstand the in-service conditions of turbine blades operating at middle temperatures in aeronautical engines. The microstructure of composite, in as-fabricated condition and after long-term heat treatments (up to 1,000 hours) in the temperature range 673-873 K, has been investigated by means of different techniques. Particular attention was paid to the micro-chemical evolution of fibre-matrix interface which is scarcely affected also by the most severe heat treatments examined here. This leads to stable mechanical properties as evidenced by hardness, tensile and FIMEC instrumented indentation tests. Therefore, the composite can operate at the maximum temperature (873 K) foreseen for its aeronautical applications without remarkable modifications of its microstructure and degradation of mechanical properties. The mechanical characterization has been completed by internal friction and dynamic modulus measurements carried out both at constant and increasing temperature, from 80 to 1173 K.

Damping of FeMo alloys obtained from SPS sintering of nanostructured powders

Spark Plasma Sintering (SPS) of nanostructured FeMo powder produces samples with satisfactory density, however the final grain size critically depends on the sintering temperature. Two groups (sets A and B) of samples have been examined by means of internal friction (IF) and dynamic modulus measurements carried out in successive test runs on the same samples to assess their structural stability. Set A and B had been sintered at 1113 and 1128 K and had an average grain size of 100 nm and 1 µm, respectively. TEM and XRD have been performed on the samples in as-prepared condition and after IF measurements cycles. The samples with smaller grains are more stable and substantially are not affected by grain coarsening which, on the contrary, occurs in those with grains of larger size. The heating up to 923 K during the tests diminishes dislocation density in both the groups. An anomalous trend of resonance frequency during the first test run in samples of set A has been ascribed to the formation of small cracks relaxing internal stresses.

Electro-deposition of graphene nanoplatelets on CPU cooler—experimental and numerical investigation

ABSTRACT This manuscript deals with to improve knowledge of the mechanisms of deposition of graphene on a flat surface. The analysis has been developed on a CPU cooler with experimental results and FEM analysis. The experiments are conducted by mounting the system over a heat source, while the force convection is facilitated by means of a blower. The transient temperature distribution in the CPU cooler is also observed. A model has been performed for modeling the thermal behavior of assembly structures with thin layers. The experimental observations are verified by simulation using a commercial FEM software. It was used a model of orthotropic thermal conductivity with variable values. The goal of the simulation has been to individuate the value of coefficient of thermal conductivity of the layer of graphene. The FEM results show how the graphene was deposited on the surface of CPU cooler and consequently, how this determine the coefficient of thermal conductivity.