Active load alleviation potential of adaptive wind turbine blades using shape memory alloy actuators

Upscaling of wind turbine blades calls for implementation of innovative active load control concepts that will facilitate the flawless operation of the machine and reduce the fatigue and ultimate loads that hinder its service life. Based on aeroelastic simulations that prove the enhanced capabilities of combined individual pitch and individual flap control at global wind turbine scale level, a shape adaptive concept that encompasses an articulated mechanism consisting of two subparts is presented. Shape memory alloy (SMA) actuators are investigated and assessed as means to control the shape adaptive mechanism at airfoil section level in order to alleviate the developed structural loads. The concept is embedded in the trailing edge region of the blade of a 10‐MW horizontal axis wind turbine and acts as a flap mechanism. Numerical simulations are performed considering various wind velocities and morphing target shapes and trajectories for both normal and extreme turbulence conditions. The results prove the potential of the concept, since the SMA controlled actuators can accurately follow the target trajectories. Power requirements are estimated at 0.22% of the AEP of the machine, while fatigue and ultimate load reduction of the flap‐wise bending moment at the blade root is 27.6% and 7.4%, respectively.     

Direct comparison of catalyst-free and catalyst-induced GaN nanowires

GaN nanowires have been grown by molecular beam epitaxy either catalyst-free or catalyst-induced by means of Ni seeds. Under identical growth conditions of temperature and V/III ratio, both types of GaN nanowires are of wurtzite structure elongated in the Ga-polar direction and are constricted by M-plane facets. However, the catalyst-induced nanowires contain many more basal-plane stacking faults and their photoluminescence is weaker. These differences can be explained as effects of the catalyst Ni seeds.      

In vitro activity of vitamins, flavonoids, and natural phenolic antioxidants against the oxidative deterioration of oil-based systems

It is well-known, that lipid antioxidants can retard the oxidative rancidity of foods caused by atmospheric oxidation, and thus protect oils, fats, and fat-soluble components from their quality degradation. In the last few years, much emphasis has been put on the promotion and use of natural antioxidants, commonly occurring in many fruits and vegetables and thereby produced from various natural extracts. This review gives a summary of previously reported work together with more recent trends in the field of natural antioxidants. Focus is given on the mechanism of actions and the inhibitory effect of certain vitamins against the oxidative degradation of oil-based systems. Moreover, the use of natural phenolics (flavonoids, olive-oil penolics, herb extracts etc.) as antioxidants in numerous lipid food applications is discussed.     

Fusion that matters: convolutional fusion networks for visual recognition

In recent years, deep learning has been successfully applied to diverse multimedia research areas, with the aim of learning powerful and informative representations for a variety of visual recognition tasks. In this work, we propose convolutional fusion networks (CFN) to integrate multi-level deep features and fuse a richer visual representation. Despite recent advances in deep fusion networks, they still have limitations due to expensive parameters and weak fusion modules. Instead, CFN uses 1 × 1 convolutional layers and global average pooling to generate side branches with few parameters, and employs a locally-connected fusion module, which can learn adaptive weights for different side branches and form a better fused feature. Specifically, we introduce three key components of the proposed CFN, and discuss its differences from other deep models. Moreover, we propose fully convolutional fusion networks (FCFN) that are an extension of CFN for pixel-level classification applied to several tasks, such as semantic segmentation and edge detection. Our experiments demonstrate that CFN (and FCFN) can achieve promising performance by consistent improvements for both image-level and pixel-level classification tasks, compared to a plain CNN. We release our codes on https://github.com/yuLiu24/CFN. Also, we make a live demo (goliath.liacs.nl) using a CFN model trained on the ImageNet dataset.     

Application of failure mode and effect analysis (FMEA), cause and effect analysis and Pareto diagram in conjunction with HACCP to a potato chips manufacturing plant

Failure mode and effect analysis (FMEA) model has been applied for the risk assessment of potato chips manufacturing. A tentative approach of FMEA application to the snacks industry was attempted in order to analyse the critical control points (CCPs) in the processing of potato chips. Preliminary hazard analysis was used to analyse and predict the occurring failure modes in a food chain system (potato processing and potato chips processing plant), based on the functions, characteristics and/or interactions of the ingredients or the processes, upon which the system depends. CCPs have been identified and implemented in the cause and effect diagram (also known as Ishikawa, tree diagram and fishbone diagram). Finally, Pareto diagrams were employed towards the optimisation potential of FMEA.     

Greek Lakes: Limnological overview

The Institute of Inland Waters in Greece has investigated the physical, chemical and biological characteristics of major Greek lakes, with regard to water use, quality and trophic status. Based on this and other limnological investigations, Greek lakes can be divided into three categories: warm monomictic deep lakes, warm monomictic shallow lakes, and dimictic shallow lakes. The water quality of some Greek lakes demonstrates the presence of high concentrations of nutrients and heavy metals, with reduced concentrations of diluted oxygen. High concentrations of ammonia, nitrate and phosphate are found in many lakes, while anaerobic hypolimnia are usually found in shallow lakes. Phosphorus is the main nutrient responsible for eutrophication because it is the limiting factor in most of the lakes that were investigated.     

Geochemical exploration of the three most significant geothermal areas of Lesbos Island, Greece

Lesbos Island has several thermal manifestations linked to extensional active faults that act as channels for the ascent of deep thermal fluids.The present work describes detailed geochemical exploration aimed at evaluating the potential of the Lesbos Island geothermal resource. Exploration was carried out on the three sites (Kalloni-Stipsi, Petra-Argenos and Polichnitos) that have the most favourable hydrogeological and structural setting on the island.Hydrogeochemical data reveal the presence between Kalloni and Stipsi of a shallow thermal aquifer with temperatures below boiling point, which coincides with carbon dioxide and temperature anomalies in the overlying groundwaters. All the thermal waters in the study area have fairly similar physico-chemical features; their geochemical temperature is in the range 115–125°C. While low-medium enthalpy geothermal resources are relatively abundant in the three selected areas of Lesbos Island, the presence of a high enthalpy geothermal system is still the subject of debate.     

Three-points interfacial quadrature for geometrical source terms on nonuniform grids

This paper deals with numerical (finite volume) approximations, on nonuniform meshes, for ordinary differential equations with parameter-dependent fields. Appropriate discretizations are constructed over the space of parameters, in order to guarantee the consistency in presence of variable cells’ size, for which L ^p -error estimates, 1≤p<+∞, are proven.     

Statistical Learning Theory: A Primer

In this paper we first overview the main concepts of Statistical Learning Theory, a framework in which learning from examples can be studied in a principled way. We then briefly discuss well known as well as emerging learning techniques such as Regularization Networks and Support Vector Machines which can be justified in term of the same induction principle.