Optimization methods in inverse problems and applications to science and engineering

Optimization and Engineering -     

Role of Si-doped Al<Subscript>0.3</Subscript>Ga<Subscript>0.7</Subscript>As layers in the high-frequency conductivity of GaAs/Al<Subscript>0.3</Subscript>Ga<Subscript>0.7</Subscript>As heterostructures under conditions of the quantum hall effect

The complex high-frequency conductivity of GaAs/Al_0.3Ga_0.7As heterostructures that are δ-doped and modulation-doped with silicon was investigated by acoustic methods under conditions of the integer quantum Hall effect. Both the real (σ₁) and imaginary (σ₂) parts of the complex conductivity σ(ω, H)=σ_i?iσ₂ were determined from the dependences of the absorption and velocity of surface acoustic waves on magnetic field. It is shown that, in the heterostructures with electron density n_s=(1.3–7)×10¹¹ cm^?2 and mobility μ=(1–2)×10⁵ cm²/(V s), the high-frequency conductivity near the centers of the Hall plateau is due to electron hopping between localized states. It is established that, with filling numbers 2 and 4, the conductivity of the Al_0.3Ga_0.7As:Si layer efficiently shunts the high-frequency hopping conductivity of the two-dimensional interface layer. A method of separating the contributions of the interface and Al_0.3Ga_0.7As:Si layers to the hopping conductivity σ(ω, H) is developed. The localization length of electrons in the interface layer is determined on the basis of the nearest neighbor hopping model. It is shown that, near the centers of the Hall plateau, both σ(ω, H) and n_s depend on the cooling rate of a GaAs/Al_0.3Ga_0.7As sample. As a result, the sample “remembers” the cooling conditions. Infrared light and static strain also change both σ(ω, H) and n_s. We attribute this behavior to the presence of two-electron defects (so-called DX^? centers) in the Al_0.3Ga_0.7As:Si layer.     

Multi-Class Segmentation with Relative Location Prior

Multi-class image segmentation has made significant advances in recent years through the combination of local and global features. One important type of global feature is that of inter-class spatial relationships. For example, identifying “tree” pixels indicates that pixels above and to the sides are more likely to be “sky” whereas pixels below are more likely to be “grass.” Incorporating such global information across the entire image and between all classes is a computational challenge as it is image-dependent, and hence, cannot be precomputed. In this work we propose a method for capturing global information from inter-class spatial relationships and encoding it as a local feature. We employ a two-stage classification process to label all image pixels. First, we generate predictions which are used to compute a local relative location feature from learned relative location maps. In the second stage, we combine this with appearance-based features to provide a final segmentation. We compare our results to recent published results on several multi-class image segmentation databases and show that the incorporation of relative location information allows us to significantly outperform the current state-of-the-art.     

Prediction of Moisture Content Evolution in a Grain Batch Dryer Based on On–line Temperature Measurements

ABSTRACT

The time evolution of the moisture content in a grain batch fluidized bed dryer is estimated by means of an on–line non–linear estimator (Marquardt–Levenberg algorithm). The inputs to the estimation algorithm are on–line temperature measurements and the output is the surface moisture content. The surface moisture content is used for predicting the grain moisture content through the solution of a single ordinary differential equation that combines the moisture and energy balances over the dryer. In this way the drying curves are obtained through incorporating in a very simple model easily obtainable physical information of the process.     

Simultaneous degradation of atrazine and simazine by a binary culture of Stenotrophomonas maltophilia and Arthrobacter sp. in a two‐stage biofilm reactor

BACKGROUND: The impact of mixtures of chloro‐triazinic herbicides, such as atrazine and simazine, on aquatic ecosystems is of environmental concern. To study their biodegradation under various operational conditions, a binary community comprising Stenotrophomonas maltophilia and Arthrobacter sp. attached to the porous support of a packed bed reactor, was evaluated. RESULTS: The genetic analysis of the two atrazine‐degrading strains revealed that genes atzA, atzB, atzC are present in both bacteria, but only S. maltophilia possess atzD. Thus, by cultivating Arthrobacter sp. on these herbicides, cyanuric acid accumulation was observed. When the binary community was cultivated in the biofilm reactor, at all the loading rates probed, both herbicides were entirely removed. However, complete biodegradation of cyanuric acid was not achieved. CONCLUSIONS: Even with a two‐stage reactor, cyanuric acid was only partially removed. This fact could be attributed to the absence, in the second stage, of an easily degradable energy source, required by S. maltophilia for the uptake and cometabolic degradation of the recalcitrant heterocyclic ring. Responding to differences in nutritional conditions prevailing at each reactor stage, local differences in species' predominance were clearly detected by microbiological and molecular biology methods. Copyright © 2010 Society of Chemical Industry     

Immobilization of protease composites with polyethylene oxide on fibre substrates

It was found that the stability of trypsin is higher than the stability of terrilytin in immobilization on cellulose or polyester fibre material in a composite with polyethylene oxide with a molecular weight of 1.5, 4, or 6 kDa. The optimum compositions are cellulose and polyester materials containing terrilytin or trypsin and 2.5% PEO with MW of 6 kDa or trypsin and 0.5% PEO with MW of 1.5 kDa. The fibre materials containing immobilized terrilytin with polyethylene oxide are resistant to radiation sterilization. The activity increases by 2-4 times in comparison to the initial materials in storage of sterilized samples at room temperature for 2 years.     

Fire-resistant polymer nanocomposites based on metal oxides and hydroxides

The fire-resistant properties of polymer nanocomposites based on polyvinylchloride and nanoparticles of metal compounds were studied. The investigation of nanoparticles and the determination of their sizes were carried out using scanning electron microscopy (SEM). The size of the nanoparticles of magnesium hydroxide was equal to 50–100 nm; iron oxide, 25–50 nm; and zinc oxide, 50–100 nm. Another type of fire retardant used in the work was aminotrimethylenephosphonic acid ammonium salt (AMPAS); however, the total content of the different additives did not exceed 15%. The nanocomposites obtained were investigated by means of thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC); in addition, the parameters of the oxygen index (OI) and carbon residue (CR) were measured for them. The lowest combustibility was exemplified by the composite material containing as the fire retardants a mixture of AMPAS and ZnO.     

State of lithium cation in a solution of polyurethane in deviethylformamide

The viscosity characteristics of solutions of PU in DMF containing different amounts of lithium chloride and perchlorate wer investigated. The dependence of the relative viscosity of the ternary systems LiX—PU—DMF on the type and content of the lithium salt was established. A hypothesis was advanced concerning the coordination of lithium cations with polyurethane functional groups. __________ Translated from Khimicheskie Volokna, No. 6, pp. 5–6, November–December, 2006.