Fault detection in a cold forging process through feature extraction with a neural network

This paper investigates the application of neural networks to the recognition of lubrication defects typical to an industrial cold forging process employed by fastener manufacturers. The accurate recognition of lubrication errors, such as coating not being applied properly or damaged during material handling, is very important to the quality of the final product in fastener manufacture. Lubrication errors lead to increased forging loads and premature tool failure, as well as to increased defect sorting and the re-processing of the coated rod. The lubrication coating provides a barrier between the work material and the die during the drawing operation; moreover it needs be sufficiently robust to remain on the wire during the transfer to the cold forging operation. In the cold forging operation the wire undergoes multi-stage deformation without the application of any additional lubrication. Four types of lubrication errors, typical to production of fasteners, were introduced to a set of sample rods, which were subsequently drawn under laboratory conditions. The drawing force was measured, from which a limited set of features was extracted. The neural network based model learned from these features is able to recognize all types of lubrication errors to a high accuracy. The overall accuracy of the neural network model is around 98% with almost uniform distribution of errors between all four errors and the normal condition.

Improved GART neural network model for pattern classification and rule extraction with application to power systems

Generalized adaptive resonance theory (GART) is a neural network model that is capable of online learning and is effective in tackling pattern classification tasks. In this paper, we propose an improved GART model (IGART), and demonstrate its applicability to power systems. IGART enhances the dynamics of GART in several aspects, which include the use of the Laplacian likelihood function, a new vigilance function, a new match-tracking mechanism, an ordering algorithm for determining the sequence of training data, and a rule extraction capability to elicit if-then rules from the network. To assess the effectiveness of IGART and to compare its performances with those from other methods, three datasets that are related to power systems are employed. The experimental results demonstrate the usefulness of IGART with the rule extraction capability in undertaking classification problems in power systems engineering.

Gas chromatography-chemical ionization-mass spectrometric fatty acid analysis of a commercial supercritical carbon dioxide lipid extract from New Zealand green-lipped mussel (Perna canaliculus)

Supercritical fluid extracts of New Zealand green-lipped mussels (NZGLM) have been suggested to have therapeutic properties related to their oil components. The large number of minor FA in NZGLM extract was characterized by a GC-CIMS/MS method that excels at identification of double-bond positions in FAME. The extract contained five major lipid classes: sterol esters, TAG, FFA, sterols, and polar lipids. The total FA content of the lipid extract was 0.664 g/mL. Fifty-three unsaturated FA (UFA) were fully identified, of which 37 were PUFA, and a further 21 UFA were detected for which concentrations were too low for assignment of double-bond positions. There were 17 saturated FA, with 14∶0, 16∶0, and 18∶0 present in the greatest concentration. The 10 n−3 PUFA detected included 20∶5n−3 and 22∶6n−3, the two main n−3 FA; n−3 PUFA at low concentrations were 18∶3, 18∶4, 20∶3, 20∶4, 21∶5, 22∶5, 24∶6, and 28∶8. There were 43 UFA from the n−4, n−5, n−6, n−7, n−8, n−9, n−10, n−11 families, with 16∶2n−4, 16∶1n−5, 18∶1n−5, 18∶2n−6, 20∶4n−6, 16∶1n−7, 20∶1n−7, 16∶1n−9, 18∶1n−9, and 20∶1n−9 being the most abundant. In general, we estimated that FAME concentrations greater than 0.05% (w/w) were sufficient to assign double-bond positions. In total, 91 FA were detected in an extract of the NZGLM, whereas previous studies of fresh flesh from the NZGLM had reported identification of 42 FA. These data demonstrate a remarkable diversity of NZGLM FA.

Investigation of hydraulic fracture propagation using a post-peak control system coupled with acoustic emission

This study investigates the fracture mechanism of fluid coupled with a solid resulting from hydraulic fracture. A new loading machine was designed to improve upon conventional laboratory hydraulic fracture testing and to provide a means of better understanding fracture behavior of solid media. Test specimens were made of cement mortar. An extensometer and acoustic emission (AE) monitoring system recorded the circumferential deformation and crack growth location/number during the test. To control the crack growth at the post-peak stage the input fluid rate can be adjusted automatically according to feedback from the extensometer. The complete stress-deformation curve, including pre- and post-peak stages, was therefore obtained. The crack extension/growth developed intensively after the applied stress reached the breakdown pressure. The number of cracks recorded by the AE monitoring system was in good agreement with the amount of deformation (expansion) recorded by the extensometer. The results obtained in this paper provide a better understanding of the hydraulic fracture mechanism which is useful for underground injection projects. © 2014 Springer-Verlag Wien.

Prescribed fluid consumption and its effects on the physiology and work behaviour of Australian wildland firefighters

The present study examined firefighters' ability to consume a prescribed fluid volume (1200 ml · h-1) during a wildland fire suppression shift and compare the effect of this additional fluid prescription with self-paced drinking on firefighters' hydration status and plasma sodium concentration post shift and their heart rate, core temperature and physical activity during their shift. Thirty-four firefighters were evenly divided into two drinking groups: self paced and prescribed. Prescribed drinkers did not meet the required 1200 ml·h-1 intake, yet they consumed twice the fluid drank by the self-paced group. No differences were noted between groups in plasma sodium levels or hydration status before or after their shift. Prescribed fluid consumption resulted in significantly lower core temperature between two and six hours into the shift. This did not coincide with lower cardiovascular strain, greater physical activity when compared to the self-paced drinking group. Additional fluid consumption (above self-paced intake) did not improve firefighter activity or physiological function (though it may buffer rising core temperature). It seems that wildland firefighters, at least in mild to warm weather conditions, can self-regulate their fluid consumption and work behaviour to leave the fireground hydrated at the conclusion of their shift.

Motor fault detection and diagnosis using a hybrid FMM-CART model with online learning

In this paper, a hybrid online learning model that combines the fuzzy min-max (FMM) neural network and the Classification and Regression Tree (CART) for motor fault detection and diagnosis tasks is described. The hybrid model, known as FMM-CART, incorporates the advantages of both FMM and CART for undertaking data classification (with FMM) and rule extraction (with CART) problems. In particular, the CART model is enhanced with an importance predictor-based feature selection measure. To evaluate the effectiveness of the proposed online FMM-CART model, a series of experiments using publicly available data sets containing motor bearing faults is first conducted. The results (primarily prediction accuracy and model complexity) are analyzed and compared with those reported in the literature. Then, an experimental study on detecting imbalanced voltage supply of an induction motor using a laboratory-scale test rig is performed. In addition to producing accurate results, a set of rules in the form of a decision tree is extracted from FMM-CART to provide explanations for its predictions. The results positively demonstrate the usefulness of FMM-CART with online learning capabilities in tackling real-world motor fault detection and diagnosis tasks. © 2014 Springer Science+Business Media New York.

Image processing and pattern recognition in textiles

Image processing and pattern recognition have been successfully applied in many textile related areas. For example, they have been used in defect detection of cotton fibers and various fabrics. In this work, the application of image processing into animal fiber classification is discussed. Integrated into / with artificial neural networks, the image processing technique has provided a useful tool to solve complex problems in textile technology. Three different approaches are used in this work forfiber classification and pattern recognition: feature extraction with image process, pattern recognition and classification with artificial neural networks, and feature recognition and classification with artificial neural network. All of them yieldssatisfactory results by giving a high level of accuracy in classification.

Nanofiltration for the concentration of heat stable salts prior to MEA reclamation

While monethanolamine has shown great potential as a solvent for the capture of carbon dioxide, impurities can build within the solution over time, leading to increased viscosity and corrosivity. Classically, these impurities are removed by a combination of neutralization and either thermal reclamation, ion exchange or electrodialysis. In this work, we evaluate the use of nanofiltration to concentrate the heat stable salts within the solution prior to such reclamation. This allows the recirculating solvent to operate with low concentrations of these impurities, while providing a low volume, concentrated solution for reclamation. Results show that nanofiltration can reject greater than 80% of the heat stable anions, while allowing the monoethanolamine to permeate through the membrane, for return to the process. Rejection of the MEA itself is less than 7%. The nanofiltration operation is only effective on lean solvent with CO2 loadings of less than 0.2 and neutralization would be required upstream to deprotonate the amine. The two membranes tested (Koch MPF-34 and MPF-36) appeared stable to exposure to the solvent for over four months.

2nd generation concrete construction: carbon footprint accounting

Purpose – Construction contractors and facility managers are being challenged to minimize the carbon footprint. Life cycle carbon‐equivalent (CO2‐e) accounting, whereby the potential emissions of greenhouse gases due to energy expenditure during construction and subsequent occupation of built infrastructure, generally ceases at the end of the service life. However, following demolition, recycling of demolition waste that becomes incorporated into 2nd generation construction is seldom considered within the management of the carbon footprint. This paper aims to focus on built concrete infrastructure, particularly the ability of recycled concrete to chemically react with airborne CO2, thereby significantly influencing CO2‐e estimates.

Design/methodology/approach – CO2‐e estimates were made in accordance with the methodology outlined in the Australian National Greenhouse Accounts (NGA) Factors and were based on the energy expended for each life cycle activity from audited records. Offsets to the CO2‐e estimates were based on the documented ability of concrete to chemically react with airborne carbon dioxide (“carbonation”) and predictions of CO2 uptake by concrete and recycled concrete was made using existing predictive diffusion models. The author's study focused on a built concrete bridge which was demolished and recycled at the end of the service life, and the recycled concrete was utilized towards 2nd generation construction. The sensitivity of CO2‐e and carbonation estimates were tested on several different types of source demolition waste as well as subsequent construction applications using recycled concrete (RCA). Whole‐of‐life CO2‐e estimates, including carbonation of RCA over the 1st and 2nd generations, were estimated and contrasted with conventional carbon footprints that end at the conclusion of the 1st generation.

Findings – Following demolition, CO2 capture by RCA is significant due to the more permeable nature of the crushed RCA compared with the original built infrastructure. RCA also has considerably greater exposed surface area, relative to volume, than a built concrete structure, and therefore more highly exposed surface to react with CO2: it therefore carbonates more comprehensively. CO2‐e estimates can be offset by as much as 55‐65 per cent when including the contribution of carbonation of RCA built within 2nd generation infrastructure. Further offsets are achievable using blended fly ash or slag cement binders; however, this study has focused on concrete composed of 100 per cent OPC binders and the effects of RCA.

Originality/value – Construction project estimates of life cycle CO2‐e emissions should include 2nd generation applications that follow the demolition of the 1st generation infrastructure. Life cycle estimates generally end at the time of demolition. However, by incorporating the recycled concrete demolition waste into the construction of 2nd generation infrastructure, the estimated CO2‐e is significantly offset during the 2nd generation life cycle by chemical uptake of CO2 (carbonation). This paper provides an approach towards inclusion of 2nd generation construction applications into whole‐of‐life estimates of CO2‐e.

Supercritical CO2 modified organic ionic plastic crystals

The treatment of an organic ionic plastic crystal electrolyte N-methyl-N-ethylpyrrolidinium tetrafluoroborate (P_1,2BF₄) with supercritical CO₂ resulted in a substantial increase in ionic conductivity, especially in the more highly ordered solid phases of the material, and also stabilised the most ordered phase to lower temperatures.