Fermentation medium and oxygen transfer conditions that maximize the xylose conversion to ethanol by Pichia stipitis

The xylose conversion to ethanol by Pichia stipitis was studied. In a first step, the necessity of supplementing the fermentation medium with urea. MgSO(4) x 7H(2)O, and/or yeast extract was evaluated through a 2(3) full factorial design. The simultaneous addition of these three nutritional sources to the fermentation medium, in concentrations of 2.3, 1.0, and 3.0 g/l, respectively, showed to be important to improve the ethanol production in detriment of the substrate conversion to cell. In a second stage, fermentation assays performed in a bioreactor under different K(L)a (volumetric oxygen transfer coefficient) conditions made possible understanding the influence of the oxygen transfer on yeast performance, as well as to define the most suitable range of values for an efficient ethanol production. The most promising region to perform this bioconversion process was found to be between 2.3 and 4.9 h(-1), since it promoted the highest ethanol production results with practically exhaustion of the xylose from the medium. These findings contribute for the development of an economical and efficient technology for large scale production of second generation ethanol. (C) 2011 Elsevier Ltd. All rights reserved.

A modified Primal-Dual Logarithmic-Barrier Method for solving the Optimal Power Flow problem with discrete and continuous control variables

The aim of solving the Optimal Power Flow problem is to determine the optimal state of an electric power transmission system, that is, the voltage magnitude and phase angles and the tap ratios of the transformers that optimize the performance of a given system, while satisfying its physical and operating constraints. The Optimal Power Flow problem is modeled as a large-scale mixed-discrete nonlinear programming problem. This paper proposes a method for handling the discrete variables of the Optimal Power Flow problem. A penalty function is presented. Due to the inclusion of the penalty function into the objective function, a sequence of nonlinear programming problems with only continuous variables is obtained and the solutions of these problems converge to a solution of the mixed problem. The obtained nonlinear programming problems are solved by a Primal-Dual Logarithmic-Barrier Method. Numerical tests using the IEEE 14, 30, 118 and 300-Bus test systems indicate that the method is efficient. (C) 2012 Elsevier B.V. All rights reserved.

Influence of small-scale inhomogeneities on the cosmological consistency tests

The current cosmological dark sector (dark matter plus dark energy) is challenging our comprehension about the physical processes taking place in the Universe. Recently, some authors tried to falsify the basic underlying assumptions of such dark matterdark energy paradigm. In this Letter, we show that oversimplifications of the measurement process may produce false positives to any consistency test based on the globally homogeneous and isotropic ? cold dark matter (?CDM) model and its expansion history based on distance measurements. In particular, when local inhomogeneity effects due to clumped matter or voids are taken into account, an apparent violation of the basic assumptions (Copernican Principle) seems to be present. Conversely, the amplitude of the deviations also probes the degree of reliability underlying the phenomenological DyerRoeder procedure by confronting its predictions with the accuracy of the weak lensing approach. Finally, a new method is devised to reconstruct the effects of the inhomogeneities in a ?CDM model, and some suggestions of how to distinguish between clumpiness (or void) effects from different cosmologies are discussed.

The boundedness of penalty parameters in an augmented Lagrangian method with constrained subproblems

Augmented Lagrangian methods are effective tools for solving large-scale nonlinear programming problems. At each outer iteration, a minimization subproblem with simple constraints, whose objective function depends on updated Lagrange multipliers and penalty parameters, is approximately solved. When the penalty parameter becomes very large, solving the subproblem becomes difficult; therefore, the effectiveness of this approach is associated with the boundedness of the penalty parameters. In this paper, it is proved that under more natural assumptions than the ones employed until now, penalty parameters are bounded. For proving the new boundedness result, the original algorithm has been slightly modified. Numerical consequences of the modifications are discussed and computational experiments are presented.