Neural networks models for wear patterns recognition of single-point dresser

Grinding is a workpiece finishing process for advanced products and surfaces. However, the constant friction between workpiece and grinding wheel causes the latter to lose its sharpness, thereby impairing the result of the grinding process. When this occurs, the dressing process is essential to sharpen the worn grains of the grinding wheel. The dressing conditions strongly influence the performance of the grinding operation; hence, monitoring them throughout the process can increase its efficiency. The purpose of this study was to classify the wear condition of a single-point dresser using intelligent systems whose inputs were obtained by digitally processing acoustic emission signals. Two multilayer perceptron (MLP) neural networks were compared for their classification ability, one using the root mean square (RMS) statistics and another the ratio of power (ROP) statistics as input. In this study, it was found that the harmonic content of the acoustic emission signal is influenced by the condition of the dresser, and that the condition of the tool under study can be classified by using the aforementioned statistics to feed a neural network. © IFAC.

Caracterização de leite utilizando técnicas de ultra-som e redes neurais

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

O uso de rede neural artificial MLP na predição de estruturas secundárias de proteínas

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Utilização de redes neurais artificiais no ajuste de controladores suplementares e dispositivo FACTS STATCOM para a melhoria da estabilidade a pequenas perturbações do sistema elétrico de potência

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Aplicação de redes neurais no controle de teores de cobre e ouro do depósito de Chapada (GO)

Pós-graduação em Geologia Regional - IGCE

Projeto otimizado de redes neurais artificiais para predição da rugosidade em processos de usinagem com a utilização da metodologia de projeto de experimentos

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Estatística espacial e redes neurais aplicadas no estudo de epidemias de huanglongbing e mancha preta na cultura dos citros

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Chitosan and alginate biopolymer membranes for remediation of contaminated water with herbicides

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Inspeção de materiais compósitos utilizando ondas acústicas guiadas

Pós-graduação em Engenharia Elétrica - FEIS

The use of artificial neural networks in analysing the nutritional ecology of Chrysomya megacephala (F.) (Diptera: Calliphoridae), compared with a statistical model

Artificial neural networks (ANNs) have been widely applied to the resolution of complex biological problems. An important feature of neural models is that their implementation is not precluded by the theoretical distribution shape of the data used. Frequently, the performance of ANNs over linear or non-linear regression-based statistical methods is deemed to be significantly superior if suitable sample sizes are provided, especially in multidimensional and non-linear processes. The current work was aimed at utilising three well-known neural network methods in order to evaluate whether these models would be able to provide more accurate outcomes in relation to a conventional regression method in pupal weight predictions of Chrysomya megacephala, a species of blowfly (Diptera: Calliphoridae), using larval density (i.e. the initial number of larvae), amount of available food and pupal size as input data. It was possible to notice that the neural networks yielded more accurate performances in comparison with the statistical model (multiple regression). Assessing the three types of networks utilised (Multi-layer Perceptron, Radial Basis Function and Generalised Regression Neural Network), no considerable differences between these models were detected. The superiority of these neural models over a classical statistical method represents an important fact, because more accurate models may clarify several intricate aspects concerning the nutritional ecology of blowflies.

Effects of cement thickness and bonding on the failure loads of CAD/CAM ceramic crowns: Multi-physics FEA modeling and monotonic testing

Objective. To determine the influence of cement thickness and ceramic/cement bonding on stresses and failure of CAD/CAM crowns, using both multi-physics finite element analysis and monotonic testing.Methods. Axially symmetric FEA models were created for stress analysis of a stylized monolithic crown having resin cement thicknesses from 50 to 500 mu m under occlusal loading. Ceramic-cement interface was modeled as bonded or not-bonded (cement-dentin as bonded). Cement polymerization shrinkage was simulated as a thermal contraction. Loads necessary to reach stresses for radial cracking from the intaglio surface were calculated by FEA. Experimentally, feldspathic CAD/CAM crowns based on the FEA model were machined having different occlusal cementation spaces, etched and cemented to dentin analogs. Non-bonding of etched ceramic was achieved using a thin layer of poly(dimethylsiloxane). Crowns were loaded to failure at 5 N/s, with radial cracks detected acoustically.Results. Failure loads depended on the bonding condition and the cement thickness for both FEA and physical testing. Average fracture loads for bonded crowns were: 673.5 N at 50 mu m cement and 300.6 N at 500 mu m. FEA stresses due to polymerization shrinkage increased with the cement thickness overwhelming the protective effect of bonding, as was also seen experimentally. At 50 mu m cement thickness, bonded crowns withstood at least twice the load before failure than non-bonded crowns.Significance. Occlusal "fit" can have structural implications for CAD/CAM crowns; pre-cementation spaces around 50-100 mu m being recommended from this study. Bonding benefits were lost at thickness approaching 450-500 mu m due to polymerization shrinkage stresses. (C) 2012 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.

Multi-layered LiNbO3 films prepared by a polymeric precursor method

The polymeric precursor method was used to prepare multi-layered LiNbO3 films. The overall process consists of preparing a coating solution from the Pechini process and the deposited film is subsequently heat-treated. Two-layered films were prepared by this process, onto (0001) sapphire substrates. Two different routes were investigated for the heat-treatment. The amorphous route consisted of performing, after each deposition, a pre-treatment at low temperature to eliminate the organic material. In this case, the crystallization heat-treatment was performed only after the two layers had been deposited. on the other hand, a process layer-after-layer crystallization was used. Both routes led to (0001) LiNbO3 oriented films. However, only the film prepared by the layer-after-layer crystallization presented an epitaxial growth and a crack-free morphology. Moreover, the layer-after-layer crystallization process led to a film exhibiting the best optical properties. (C) 2001 Elsevier B.V. Ltd. All rights reserved.

Dielectric and viscoelastic properties of cross-linked polyethylene aged under multi stressing conditions

Different measurements were performed in cross-linked polyethylene (XLPE) employed as insulating material in coaxial cables that were field-aged and laboratory-aged under multi-stressing conditions at room temperature. Samples were peeled from the XLPE cable insulation in three different positions: just below the external semiconductor layer (outer layer), in the middle (middle layer) and just above the internal semiconductor layer of the cable (inner layer). The imaginary part of the electric susceptibility showed three peaks that obey the Dissado-Hill model. For laboratory-aged XLPE samples peeled from the inner and from the middle positions the peak at very low frequency region increased while in samples from the outer position a quasi-DC conduction process was observed. In medium frequency range a broadening of the peak was observed for all samples. Viscoelastic properties determined through dynamic mechanical analysis suggested that the aging generates processes that promoted changes of the crystallinity and the cross-linking degrees of the polymer. Fourier transform infrared spectroscopy (FTIR) measurements revealed an increase of oxidation products (esters), evidence of polar residues of the bow-tie tree and the presence of cross-linking by-products (acetophenone). Optical and scanning electronic microscope (SEM) measurements in aged samples revealed the existence of voids and bow-tie trees that were formed during aging in the middle region of the cable.

Dielectric behavior of XLPE aged under multi-stressing conditions

In this work fresh cables were laboratory aged under multi-stressing conditions at room temperature. Foils were peeled from cables, with approximately 150 ?m thickness, from the outer, middle and inner positions of the XLPE cable insulating layer. For samples obtained from the outer cable layer position, an increasing near-permanent electrical conduction process with aging time was observed. At the middle and inner cable layer positions a flat-loss relaxation process was observed becoming a dominating process on the ageing. In addition, PEA results confirmed that degradation in the outer region of the XLPE cables arises from the simultaneous presence of dipoles and injected space charge that distorts the internal electric field on the ageing.

Cementation of fiber post: influence of the cement insertion techniques on the bond strength of the fiber post-root dentin and the quality of the cement layer.

The aim of this study was to assess the influence of resin cement insertion methods on the bond strength of a fiber post to root dentin and quality of the cement layer. Forty bovine single-roots (length =16 mm) were randomly allocated into four groups, according to the cement insertion methods (N.=10): Gr1- Lentulo drill #40, Gr2- Centrix syringe, Gr3- Explorer #5, Gr4- fiber post. The root canals were prepared at 12 mm, using preparation bur # 3 of a cylinder quartz-FRC post (Aesthet post-plus, Bisco). The fiber posts were cemented using a multi-step etch-and-rinse adhesive system (All Bond 2®, Bisco) and a dual-cured resin cement (Duolink, Bisco). Each root was cut into seven samples: four samples of 1.8 mm thickness for push-out testing, and three with 0.5 mm for cement layer quality analyzing. One-way ANOVA was used for the push-out test values and the One-Way Kruskal-Wallis (P<0.05) and Dunn (10%) tests for the cement layer analyzes. ANOVA showed that the cement layer quality was affected by the cement insertion methods (P=0.0044): Gr1 (3.8 ± 1.3a), Gr2 (3.2 ± 1.3a), Gr3 (5.2 ± 1.5a,b) and Gr4 (5.2 ± 1.5b) (Dunn test), whereas the bond strength (MPa) was not affected by cement insertion methods: G1 (4.2 ± 1.3), G2 (3.2 ± 1.8), G3 (4.5 ± 0.9), G4 (3.1 ± 1.3). The fiber posts should be cemented with the assistance of the lentulo drill or centrix syringe to promote the best cement layer results.