This work focuses on the identification of five of the most common ferritic morphologies present in welded fusion zones of low carbon steel through images acquired by photomicrographies. With this regards, we discuss the importance of the gray-level co-occurrence matrix to extract the features to be used as the input of the computational intelligence techniques. We use artificial neural networks and support vector machines to identify the proportions of each morphology and present the error identification rate for each technique. The results show that the use of gray-level co-occurrence extraction allows a less intense computational model with statistical validity and the support vector machine as a computational intelligence technique allows smaller variability when compared to the artificial neural networks.
Resumo A união entre materiais dissimilares tem sido amplamente aplicada na indústria de petróleo e gás. Em geral, este tipo de soldagem é mais sensível às diferenças de composição química existente entre os metais envolvidos. Diante dessa perspectiva, este trabalho tem por objetivo avaliar, por união autógena, a microestrutura de uma junta soldada entre um aço de alta resistência e baixa liga da classe API 5L X80 e um aço inoxidável duplex UNS S32304. Além disso, estudou-se o comportamento quanto à resistência à corrosão em uma solução simuladora de água do mar. A microestrutura e composição da junta soldada foram avaliadas por microscopia ótica e eletrônica, espectrometria de energia dispersiva (EDS). A resistência à corrosão foi estudada por polarização potenciodinâmica anódica e por ensaios de espectroscopia de impedância eletroquímica. As juntas soldadas apresentaram uma microestrutura que se difere em morfologia de ambos os metais de base, além disso, notou-se a migração de cromo e níquel do duplex para esta região que apresentou uma resistência à corrosão ligeiramente superior ao API. Desta forma, podemos concluir que as propriedades da junta dissimilar são compatíveis com as aplicações, sendo possível sua utilização usufruindo-se das vantagens da soldagem dissimilar.
This paper presents high quality (2048 × 1532 pixels) Light Microscope steel images sampled from the welding fusion zone. The microstructure images were acquired from the Design of Experiments (2
2
full factorial design) planned to compare two different arc welding processes at two different arc welding energies [1]. The 400 raw images appear as they were captured by the microscope and they are categorized into four groups: that acquired from the Flux Cored Arc Welding process and that acquired from the Shielded Metal Arc Welding process; both of them run for high and low levels of arc energy. For the Flux Cored Arc Welding process, ASME SFA 5.20 E71T-5C(M) tubular wire was used, with a nominal diameter of 1.2 mm. For the Shielded Metal Arc Welding process, AWS E7018 coated electrodes were used, with nominal diameters of 3.25 mm (for the low energy level) and 5.00 mm (for the high energy level). The deposition of the beads was run on AISI 1010 steel plates in the flat position (bead-on-plate). Different proportions of primary grain boundary ferrite; polygonal ferrite; acicular ferrite; nonaligned side-plate ferrite and aligned side-plate ferrite can be observed in each image. This image dataset is ready to visual and automatic microstructure recognition and quantification. It can be a useful resource for computational intelligence research teams, e.g. [2], by offering images for handling as filtering, feature extraction, training, validation and testing in pattern recognition and machine learning techniques.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.