M. Barletta

Modelling of electrostatic fluidized bed (EFB) coating process using artificial neural networks

This paper involves experimentation on coating process of metal substrates in an electrostatic fluidized bed (EFB). Several operational parameters were covered like coating time, applied voltage and gas flow rate fed to the fluidized bed. First, a design of experiment (DOE) approach was used to define the experimental campaign and a general linear model based on analysis of variance (ANOVA) was used to elaborate and interpret the influence of all the operational parameters on coating thickness trends. Second, the experimental data were modelled using artificial neural networks. Different neural networks and training algorithms were employed to find the best technique to predict the coating thickness trends. The reliability of the best neural network solutions was checked by comparing them with a built ad hoc regression model. The multi-layer perceptron (MLP) neural network trained with back-propagation (BP) algorithm was found to be the fittest model. Besides, a genetic algorithm (GA) was also employed to improve the capability of MLP model to provide the best fit of experimental results all over the investigated ranges. Finally, a verification experimental plan was performed and a related analytical model was developed to check the reliability of the neural network model with GA to predict the whole coating thickness trends according to the operational parameters. A comparison between the neural network model and an analytical model was also carried out.

Fluidized Bed Assisted Abrasive Jet Machining (FB-AJM): Precision Internal Finishing of Inconel 718 Components

The relatively new technique of fluidized bed assisted abrasive jet machining (FB-AJM) is applied to finishing the inner surfaces of tubular Inconel 718 components. The effects. of abrasive size, jet pressure, and machining cycle were evaluated, and the behavior of abrasive cutting edges acting against the surface during the process to remove material is accounted for. The finished surface was found to be highly dependent on jet pressure because it affects the abrasive contact against the surface as well as the finishing force acting on the abrasive, on the abrasive grain size, which controls the depth of cut, and on machining cycle, which controls the interaction time between the abrasives and the surface being finished. By altering these conditions, this process achieves surface roughness (R-a) as fine as 0.1 mu m and imparts minimal additional residual stress on the surface. This study also reveals the mechanisms that determine the smoothing of the inner surface of Inconel 718 tubes and improve the form accuracy, i.e., the internal roundness of the Inconel 718 tube.

Self-cleaning and self-sanitizing coatings on plastic fabrics: Design, manufacture and performance

Self-cleaning and self-sanitizing coatings are of utmost interest in several manufacturing domains. In particular, fabrics and textile materials are often pre-treated by impregnation or incorporation with antimicrobial pesticides for protection purposes against bacteria and fungi that are pathogenic for man or other animals. In this respect, the present investigation deals with the design and manufacture of self-cleaning and self-sanitizing coatings on plastic fabrics. The functionalization of the coatings was yield by incorporating active inorganic matter alone (i.e., photo-catalytic TiO2 anatase and Ag(+) ions) inside an organic inorganic hybrid binder. The achieved formulations were deposited on coextruded polyvinylchloride-polyester fabrics by air-mix spraying and left to dry at ambient temperature. The performance of the resulting coatings were characterized for their self-cleaning and self-sanitizing ability according to standardized testing procedure and/or applicable international regulations.

Production of open cell aluminum foams by using the dissolution and sintering process (DSP)

The manufacture of open cell metal foams by dissolution and sintering process (DSP) is the matter of the present work. Aluminum foams were produced by mixing together carbamide particles with different mesh sizes (i.e., space-holder) and very fine aluminum powders. Attention was first paid at understanding the leading phenomena of the different stages the manufacturing process gets through: Compaction of the main constituents, space-holder dissolution, and aluminum powders sintering. Then, experimental tests were performed to analyze the influence of several process parameters, namely, carbamide grain size, carbamide wt %, compaction pressure, and compaction speed on the overall mechanical performance of the aluminum foams. Meaningfulness of each operational parameter was assessed by analysis of variance. Metal foams were found to be particularly sensitive to changes in compaction pressure, exhibiting their best performances for values not higher than 400 MPa. Neural network solutions were used to model the DSP. Radial basis function (RBF) neural network trained with back propagation algorithm was found to be the fittest model. Genetic algorithm (GA) was developed to improve the capability of the RBF network in modeling the available experimental data, leading to very low overall errors. Accordingly, RBF network with GA forms the basis for the development of an accurate and versatile prediction model of the DSP, hence becoming a useful support tool for the purposes of process automation and control.

Metal foams for structural applications: design and manufacturing

In this paper, the design and manufacture of metal foams, using the powders compact melting method (PCMM), is investigated. Experimental tests were performed to study the influence of several process parameters, that is, compaction pressure, foaming time, temperature and amount of foamable precursor material, on the kinetics of foaming process. As the large number of experimental factors involved in metal foams manufacturing, an experimental approach based upon DOE techniques was employed to reduce the trials need for individuating the best process windows. Hence, in such operative ranges, further experimental tests were carried out to trace the full trends of foaming efficiency according to leading parameters, thereby laying the basis to support manufacturers on how to deal with the operative troubles and process settings.

Microstructural and tribological characterisation of as sprayed and heat treated HVOF deposited Ni alloys

Abstract The microstructural, micromechanical (Vickers microindentation, scratch testing) and tribological (pin on disk tests against steel and alumina spherical pins) properties of three High velocity oxy–fuel (HVOF) sprayed Ni based alloy coatings, namely Diamalloy 4006 (Ni–20Cr–10W–9Mo–4Cu–1B–1C–1Fe), Tribaloy-700 (Ni–32Mo–16Cr–4Si–2Co) and Inconel-625 (Ni–22Cr–9Mo–4Nb), were characterised, both in the as sprayed condition and after thermal treatments at 600°C and 800°C. As deposited Tribaloy-700 possesses a low degree of crystallinity and lower hardness; crystalline intermetallics are formed after heat treatments, definitely improving mechanical strength and tribological resistance against 100Cr6 steel counterpart, but not against alumina counterpart. The Inconel-625 and Diamalloy 4006 as sprayed coatings consist of supersaturated crystalline solid solutions. The former is not much affected by heat treatments and has low hardness and poor tribological properties. The latter, instead, displays precipitation of secondary phases after heat treatment. Particularly, the 600°C treatment improves coating strength and wear resistance against steel, whereas the 800°C one is less effective, probably because it causes excessive crystal grain size increase.

Fluidized bed coating of metal substrates by using high performance thermoplastic powders: Statistical approach and neural network modelling

This paper deals with fluidized bed coating of metal substrates with high performance thermoplastic powders (polyftalamide, PPA). Two different experimental scenarios were investigated: the conventional hot dipping fluidized bed (CHDFB) process and the electrostatic fluidized bed (EFB) coating process. The preliminary experimental plan was scheduled employing design of experiment (DOE) technique. Three experimental factors and operative ranges large enough for practical purposes were considered in both of the examined scenarios. In particular, coating time and airflow rate were chosen as experimental factors in both CHDFB and EFB. The third factor was the preheating temperature of metal substrates in CHDFB and the applied voltage in EFB. A general linear model based upon analysis of variance (ANOVA) was used to evaluate the significance on coating processes of each experimental factor. Main effect plots (MEPs) and interaction plots (IPs) of coating thickness were drawn. Trends consistent with the settings of the operative parameters were displayed. The experimental trends were first modelled by numerical regression of the experimental data and, subsequently, by using artificial neural network. The reliability of the neural network solution and of the built ad hoc regression models was comparatively evaluated. Multi-layer perceptron (MLP) neural network trained with back propagation (BP) algorithm was found to be the most valuable in fitting the coating thicknesses trends for both the coating processes. Examining the developed models outside the operative ranges they were designed for, the good generalization capability and high flexibility of the neural network solution was definitely stated.

Modelling of Fluidized Bed Degreasing (FBD) process by ANNs

This paper is focused on a relatively novel eco-efficient degreasing technique, namely Fluidized Bed Degreasing (FBD), based on a fluidised bed of hard particles. An experimental campaign was aimed to investigate the relationship between FBD operational parameters and degreasing effectiveness. Consistent trends of residual oil according to FBD process parameters were found and both a related power dissipation analytical model and a neural network were developed and verified by comparison with experiments. The Multi-Layer Perceptron (MLP) neural network, trained with Back-Propagation (BP) algorithm, gave the best performance. Finally, Genetic Algorithms (GAs) were used to improve the predicting capability of the neural network solution. In detail, an experimental plan was performed to check the generalisation capability of the neural network model with GA.

Surface preparation and coating of metal coils by using a fully integrated manufacturing system

The current paper deals with the definition of a novel fully integrated system for cleaning, pre-treatment and powder coating of metal coils based upon two subsequent fluidized beds: the former employed in items surface preparation by peening with glass beads and the latter in coating process by electrostatic deposition. An approach based upon design of experiments was employed to find the best processing windows for each phase of coil coating processes. Besides, the experimental trends of operative parameters were also sought. Special attention was paid to the influence of leading process parameters on the performance of the starting surface as well as on the aesthetic aspect and some functional properties of the polymeric coatings. Finally, some useful aspects concerning the best way to lead the process were also discussed in detail.

Fuzzy model for electrostatic fluidized bed coating

The study concerns the coating process of metal substrates in an electrostatic fluidized bed (EFB).This eco-friendly process is profitably used to coat components of particularly complex shapes. Although this technology is widely spread in several industrial domains, the implementation of appropriate process control procedures is still object of investigation.A model was generated from experimental data with the aim of predicting, for any set of process parameters, the resulting coating thickness of the sample. With a design of experiment (DOE) approach, the experimental investigation, that is the base for the model, quantifies the coating thickness as a function of the main process parameters namely coating time, applied voltage, and gas flow rate fed into the fluidized bed.This study addresses the effect of the inherent uncertainties on the predicted coating thickness caused by the approximation in the model parameters. In particular, a fuzzy-logic based approach is used to describe the model uncertainties and the transformation method is used to propagate their effect on the thickness. The fuzzy results are then compared with the data produced by the experimentation leading to the evaluation of the membership level of the dataset to the uncertain model.Copyright