An unsupervised approach for fault diagnosis of power transformers

Electrical utilities apply condition monitoring on power transformers (PTs) to prevent unplanned outages and detect incipient faults. This monitoring is often done using dissolved gas analysis (DGA) coupled with engineering methods to interpret the data, however the obtained results lack accuracy and reproducibility. In order to improve accuracy, various advanced analytical methods have been proposed in the literature. Nonetheless, these methods are often hard to interpret by the decision-maker and require a substantial amount of failure records to be trained. In the context of the PTs, failure data quality is recurrently questionable, and failure records are scarce when compared to nonfailure records. This work tackles these challenges by proposing a novel unsupervised methodology for diagnosing PT condition. Differently from the supervised approaches in the literature, our method does not require the labeling of DGA records and incorporates a visual representation of the results in a 2D scatter plot to assist in interpretation. A modified clustering technique is used to classify the condition of different PTs using historical DGA data. Finally, well-known engineering methods are applied to interpret each of the obtained clusters. The approach was validated using data from two different real-world data sets provided by a generation company and a distribution system operator. The results highlight the advantages of the proposed approach and outperformed engineering methods (from IEC and IEEE standards) and companies legacy method. The approach was also validated on the public IEC TC10 database, showing the capability to achieve comparable accuracy with supervised learning methods from the literature. As a result of the methodology performance, both companies are currently using it in their daily DGA diagnosis.     

View angle effects on the discrimination of selected Amazonian land cover types from a principal‐component analysis of MISR spectra

Multi‐angle Imaging Spectroradiometer (MISR) data, collected in four bands and at nine view angles in the Brazilian Amazon region, were used to describe view‐angle effects on the spectral response and discrimination of three forest types; close and open lowland forests, open submontane forest and green/emerging pastures. A principal‐component analysis (PCA) was applied over 450 bidirectional reflectance factor (BRF) MISR spectra (10 pixels, five land covers and nine view angles) to characterize the spectral‐angular variability in the dataset and to identify the best view direction to enhance land cover discrimination. The analysis was extended into the images of the different cameras, which were classified for the presence of the forest covers using the minimum distance of the pixels to the average PC1 and PC2 scores of each forest class calculated from spectra analysis. Results showed an increase in the mean reflectance over the spectral bands (brightness) of the land covers from nadir to extreme viewing, as indicated by the first principal component, especially in the backward direction due to the predominance of sunlit view vegetation components. The transition from the backward (sunlit view surface components) to the forward (shaded view surface components) scattering directions was also characterized by changes in the shape of the BRF spectra, as indicated by decreasing PC2 score or near‐infrared/blue ratio values. The variations in the MISR BRF followed the regularities expected from theory. PCA results also indicated that the best viewing to discriminate the forest types was the backward scattering direction (?26.1° view angle), whereas the less favourable viewing was the forward scattering direction under the view shading condition (e.g. +45.6° view angle). The overall classification accuracy for the three forest types increased from 52.4% at +45.6° view angle to 78.7% at nadir, and to 95.0% at a ?26.1° view angle. From nadir to extreme view angles, directional effects produced a NDVI decrease for the forest types and an NDVI increase for the green and especially emerging pastures. Results demonstrated that data acquisition in off‐nadir viewing may improve the discrimination and mapping of the Amazonian land cover types.     

Output feedback of Markov jump linear systems with no mode observation: An automotive throttle application

The note presents an output feedback control strategy for Markov jump linear systems with no mode observation. Based on minimizing a finite‐time quadratic cost, we derive an algorithm that generates output feedback gains that satisfy a necessary optimality condition. These gains can be computed off‐line relying only on the initial condition of the system. This result expands a previous one from the literature that considered state‐feedback only. To illustrate the usefulness of the approach, real‐time laboratory experiments were performed to control an automotive electronic throttle valve subject to Markov‐driven voltage fluctuations. Copyright © 2015 John Wiley & Sons, Ltd.     

Ensemble of Efficient Minimal Learning Machines for Classification and Regression

Minimal Learning Machine (MLM) is a recently proposed supervised learning algorithm with performance comparable to most state-of-the-art machine learning methods. In this work, we propose ensemble methods for classification and regression using MLMs. The goal of ensemble strategies is to produce more robust and accurate models when compared to a single classifier or regression model. Despite its successful application, MLM employs a computationally intensive optimization problem as part of its test procedure (out-of-sample data estimation). This becomes even more noticeable in the context of ensemble learning, where multiple models are used. Aiming to provide fast alternatives to the standard MLM, we also propose the Nearest Neighbor Minimal Learning Machine and the Cubic Equation Minimal Learning Machine to cope with classification and single-output regression problems, respectively. The experimental assessment conducted on real-world datasets reports that ensemble of fast MLMs perform comparably or superiorly to reference machine learning algorithms.     

Species,roasting degree and decaffeination influence the antibacterial activity of coffee against Streptococcus mutans

Coffee beverage has been associated with antibacterial activity against Streptococcus mutans, a cariogenic bacterium. This study aimed at identifying natural compounds in coffee that contribute to such activity and investigate the influence of species, roasting and decaffeination on it. Coffee chemical compounds and aqueous extracts of green and roasted regular and decaffeinated Coffea arabica and Coffea canephora beans were tested. MIC, biofilm inhibition and biofilm reduction results were correlated with the concentration of coffee compounds in the extracts. 5-Caffeoylquinic acid, trigonelline and caffeic acid solutions showed bacteriostatic activity (MIC = 0.8 mg/mL). Lighter and regular extracts showed higher inhibitory activity than darker and decaffeinated extracts, with an inverse correlation between bacterial colony-forming units and roasting degree. Only regular C. canephora extracts showed biofilm formation inhibition. The joint effect of chlorogenic acids, trigonelline and caffeine or other compounds removed by decaffeination seems to be one of the causes for coffee antibacterial activity against S. mutans.     

Comparative study of the biochemical changes and volatile compound formations during the production of novel whey-based kefir beverages and traditional milk kefir

Cheese whey (CW) and deproteinised cheese whey (DCW) were investigated for their suitability as novel substrates for the production of kefir-like beverages. Lactose consumption, ethanol production, as well as organic acids and volatile compounds formation, were determined during CW and DCW fermentation by kefir grains and compared with values obtained during the production of traditional milk kefir. The results showed that kefir grains were able to utilise lactose from CW and DCW and produce similar amounts of ethanol (7.8–8.3 g/l), lactic acid (5.0 g/l) and acetic acid (0.7 g/l) to those obtained during milk fermentation. In addition, the concentration of higher alcohols (2-methyl-1-butanol, 3-methyl-1-butanol, 1-hexanol, 2-methyl-1-propanol, and 1-propanol), ester (ethyl acetate) and aldehyde (acetaldehyde) in cheese whey-based kefir and milk kefir beverages were also produced in similar amounts. Cheese whey and deproteinised cheese whey may therefore serve as substrates for the production of kefir-like beverages similar to milk kefir.     

Short communication: High frequency of β-lactam-resistant Staphylococcus aureus in artisanal coalho cheese made from goat milk produced in northeastern Brazil

Reports of β-lactam-resistant Staphylococcus aureus in artisanal goat cheese are increasing, and this phenomenon is relevant to public health. The objective of the present study was to determine the prevalence of S. aureus strains carrying the blaZ and mecA resistance genes, as well as the genes encoding the staphylococcal enterotoxins SEA, SEB, SEC, SED, SEE, and TSST-1 in artisanal coalho cheese made from goat milk produced in northeastern Brazil. We used biochemical and molecular tests to characterize 54 S. aureus isolates found in artisanal coalho cheese collected from commercial establishments producing animal products in 11 municipalities of Pernambuco State, Brazil. A PCR analysis revealed that 42.6% (23/54) of the isolates were positive for the blaZ gene, and 7.4% (4/54) were resistant to methicillin by phenotypic testing. We did not detect mecA or any genes encoding enterotoxins. The presence of S. aureus carriers of the blaZ gene and the identification of methicillin-resistant S. aureus strains are of concern for the health of consumers of this type of cheese.     

Analysis of form threads using fluteless taps in cast magnesium alloy (AM60)

Threads are used in the most assemblies of industrial products. Commonly, mechanical components need to have threaded parts allowing fast and accurate assemblies and disassemblies. Internal tapping is one of the most demanded machining operations, and threads obtained by forming have been a good alternative. This work investigates the effect of the factors the hole diameter, the forming speed and types of tool on the responses: torque, hardness, fill rate, and thrust force of the form tapping process. The experiments were carried out with three diameters, three forming speeds, and two coatings. The material used was the AM60 alloy due to its high ductility and wide application as head engine. The results revealed torque is more affected by the hole diameter than thrust force, and little hardening occurred using high forming speed with a small diameter. The fill rate of the thread profile was not significantly affected by the intermediate and large diameters. Finally, it can be stated that the recommended hole diameter provided by the tool's supplier can be modified to achieve more accurate thread profiles.