Nida Sheibat-Othman

Support Vector Machines for Fault Detection in Wind Turbines

Abstract Support Vector Machines (SVM) are used for fault detection and isolation in a variable speed horizontal-axis wind turbine composed of three blades and a full converter. The SVM approach is data based and is therefore robust to process knowledge. Moreover, it is based on structural risk minimization which enhances generalization and it allows accounting for process non linearity by using flexible Kernels. In this work, a radial basis function was used as Kernel. Different parts of the process were investigated including actuators, sensors and process faults. With duplicated sensors, we could detect sensor faults in blade pitch positions, generator and rotor speeds rapidly. Fixed value fault were detected in 2 sample periods and offset faults could be detected for Δβ ≥ 0.5° with a detection time that depends on the offset level. The converter torque fault (an actuator) could be detected within two sample periods. Faults in the actuators of the pitch systems could not be detected. Faults in the process concerning friction in the drive train could be detected only for very high offset (Δη dt ≥ 50%).

Use of Silica Particles for the Formation of Organic−Inorganic Particles by Surfactant-Free Emulsion Polymerization

Polystyrene/silica (PS/SiO(2)) and poly(styrene-co-methyl methacrylate)/SiO(2) composite latex particles were prepared by surfactant-free emulsion polymerization in the presence of a poly(ethylene glycol) monomethylether methacrylate (PEGMA) macromonomer. The resulting composite particles were stabilized by the negatively charged silica particles that adhered to the surface of the latex particles. Different process parameters were investigated in order to optimize the latex stability and maximize the reaction rate. Mixing in such a surfactant-free process is of major importance and is mainly determined by the type of impeller used during the emulsification. The concentrations of PEGMA and silica particles were also optimized in order to improve the interaction between the organic and inorganic phases and ensure a good latex stability. The presence of silica particles on the polymer particle surface was found to affect radical absorption and decrease therefore the reaction rate.

Usefulness of controlled release of growth factors in investigating the early events of dentin-pulp regeneration.

INTRODUCTION Little information is yet available on the signals involved in progenitor cell migration that precede reparative dentin synthesis. Our aim was to investigate the effect of the controlled release of fibroblast growth factor (FGF)-2 and transforming growth factor β1 (TGF-β1) on permanent teeth pulp cell proliferation and progenitor cell migration. METHODS FGF-2 and TGF-β1 were encapsulated into a biodegradable polymer matrix of lactide and glycolide. Human pulp cells were prepared from third molars, and progenitor cells were sorted by STRO-1. The synthesized microsphere toxicity was checked with the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide test. The growth factor release kinetics were checked by an enzyme-linked immunosorbent assay while maintaining their biological activity and were evaluated by investigating their effects on pulp cell proliferation. Their chemotactic potential was investigated on STRO-1-sorted cells in a migration chamber on Matrigel (Cambrex Bio Science, Walkersville, MD). RESULTS The cell viability was unaffected by the presence of microspheres. The released amount of FGF-2 and TGF-β1 from the microspheres was maintained after 21 days. Increasing the FGF-2-loaded microsphere concentration or the release period significantly increased dental pulp cell proliferation. TGF-β1 acted as a potent chemotactic factor of STRO-1-sorted cells. CONCLUSIONS Encapsulating TGF-β1 and FGF-2 in a biodegradable polymer of lactide and glycolide microsphere allowed a sustained release of growth factors and provided a protection to their biological activities. Our results clearly show the usefulness of growth factor controlled release in investigating the early events of pulp/dentin regeneration. It provides additional data on the signals required for vital pulp therapy and future tissue engineering.

Partitioning of Laponite Clay Platelets in Pickering Emulsion Polymerization.

Partitioning of laponite disklike clay platelets between polymer particles and bulk aqueous phase was investigated in Pickering surfactant-free emulsion polymerization of styrene. Adsorption of laponite clay platelets plays an important role in the stabilization of this system, influencing the particle size and the number of particles, and, hence, the reaction rate. Adsorption isotherms show that, while the laponite clay platelets are almost fully exfoliated in water, they form multilayers on the surface of the polymer particles by the end of polymerization, as confirmed by transmission electron microscopy (TEM). This observation is supported by quartz crystal microbalance, conductivity, and TEM measurements, which reveal interactions between the clay and polystyrene, as a function of the ionic strength. The strong adsorption of clay platelets leaves a low residual concentration in the aqueous phase that cannot cause further nucleation of polymer particles, as demonstrated during seeded emulsion polymerization experiments in the presence of a high excess of clay. A Brunauer-Emmett-Teller (BET)-type model for laponite adsorption on polystyrene particles matches the adsorption isotherms.

Investigation of Four Different Laponite Clays as Stabilizers in Pickering Emulsion Polymerization.

Clay-armored polymer particles were prepared by emulsion polymerization in the presence of Laponite platelets that adsorb at the surface of latex particles and act as stabilizers during the course of the polymerization. While Laponite RDS clay platelets are most often used, the choice of the type of clay still remains an open issue that is addressed in the present article. Four different grades of Laponite were investigated as stabilizers in the emulsion polymerization of styrene. First, the adsorption isotherms of the clays, on preformed polystyrene particles, were determined by ICP-AES analysis of the residual clay in the aqueous phase. Adsorption of clay depended on the type of clay at low concentrations corresponding to adsorption as a monolayer. Adsorption of clay particles as multilayers was observed for all the grades above a certain concentration under the considered ionic strength (mainly due to the initiator ionic species). The stabilization efficiency of these clays was investigated during the polymerization reaction (free of any other stabilizer). The clays did not have the same effect on stabilization, which was related to differences in their compositions and in their adsorption isotherms. The different grades led to different polymer particles sizes and therefore to different polymerization reaction rates. Laponite RDS and S482 gave similar results, ensuring the best stabilization efficiency and the fastest reaction rate; the number of particles increased as the clay concentration increased. Stabilization with Laponite XLS gave the same particles size and number as the latter two clays at low clay concentrations, but it reached an upper limit in the number of nucleated polymer particles at higher concentrations indicating a decrease of stabilization efficiency at high concentrations. Laponite JS did not ensure a sufficient stability of the polymer particles, as the polymerization results were comparable to a stabilizer-free polymerization system.

Coupled population balance–CFD simulation of droplet breakup in a high pressure homogenizer

Abstract A framework for a one-way coupling between population balance equation (PBE) and computational fluid dynamics (CFD) for emulsions undergoing breakup in a turbulent flow-regime is implemented in the open-source CFD package OpenFOAM. The PBE is discretized using a volume conservative scheme. This framework was applied to emulsification of food-grade vegetable oil-in-water emulsions in a Niro Soavi high pressure homogenizer (HPH) with different pressure drops and using oils of varying viscosities. Different breakage rate models were implemented, as well as the model of Becker et al. (2014) which is an extension of the Luo and Svendsen (1996) model to account for the dispersed phase viscosity. The Becker et al. (2014) model was found to give better predictions after three consecutive passes, without the need for additional empirically determined parameters. The multi-scale PBE–CFD modelling approach allowed a detailed analysis of the breakup process in the gap and jet of the HPH valve.

Fault detection and isolation in wind turbines using support vector machines and observers

In this work, the benchmark FAST that simulates a closed-loop three-bladed wind turbine is used for fault detection and isolation. Two methods were employed to isolate faults of different types at different locations: Support vector machines (SVM) and a Kalman-like observer. SVM could isolate most faults with the used data and characteristic vectors, except for high varying dynamics. In this case, the use of an observer, which is model-based, was found necessary.

Constrained Nonlinear Predictive Control for Maximizing Production in Polymerization Processes

In this brief, a new constrained nonlinear predictive control scheme is proposed for maximizing the production in polymerization processes. The key features of the proposed feedback strategy are its ability to rigorously handle the process constraints (input saturation, maximum allowed heat production, maximal temperature values, and rate of change) as well as its real time implementability due to the low dimensional control parametrization being used. Simulations are proposed to show the efficiency of the proposed feedback as well as its robustness to model uncertainties. The controller performance is also validated experimentally on a laboratory scale reactor to control the emulsion polymerization of styrene

Preparation and characterization of biodegradable polyhydroxybutyrate-co-hydroxyvalerate/polyethylene glycol-based microspheres.

The in vivo effectiveness of biomolecules may be limited by their rapid diffusion in the body and short half-life time. Encapsulation of these biomolecules allows protecting them against degradation and ensuring a controlled release over time. In this work, the production of polyhydroxybutyrate-co-hydroxyvalerate/polyethylene glycol-based microspheres loaded with heparin by double emulsion-solvent evaporation is investigated. Significant improvements are achieved after blending PHB-HV microspheres with PEG. First of all, an important decrease of the initial burst effect is ensured. Moreover, lower degradation of the microspheres is observed after 30days in the release medium. Finally, the release rate could be controlled using different PEG molecular weights and concentrations. A toxic effect of PHB-HV 30% PEG 1100gmol-1 microspheres is observed whereas PHB-HV and PHB-HV 30% PEG 10,000gmol-1 microspheres are not toxic. These microspheres seem to be most suited for further tissue engineering applications. The effectiveness of direct PEG blending to PHB-HV is proved, limiting the use of chemical reagents for PHB-HV/PEG copolymer synthesis and steps for chemical reagents removal from the copolymer.

Viscous Drop Breakage in Liquid–Liquid Stirred Dispersions: Population Balance Modeling

The breakup of viscous drops in a turbulent regime is a challenging area of research. In this study, the population balance equation is used for modeling of the drop size distribution of emulsification systems in stirred tank. The suitability of various fluid particle breakage closures in the literature is investigated for the prediction of experimental data obtained from emulsification of silicone oils with four different viscosities. Systems with breakup of dispersed-phase drops of high viscous grades call for an improved model framework that allows the breakage closures to adjust correctly for the effect of viscosity. This is particularly required for systems of higher viscosities where the complex effects of viscous breakup are more pronounced. GRAPHICAL ABSTRACT