Search

A solar chimney configuration consisting of three air flow channels divided by two metallic plate, placed in the center of the chimney between two acrylic covers, leading to symmetric air flow, is proposed as an alternative for the design of natural ventilation systems for buildings in tropical and subtropical climatic zones The solar chimney dimensions are 2.0 m height, 1.0 m width, and gap between channels of 0.30 m. These dimensions are appropriate for the design of ventilation systems for residential buildings in central México. A Numerical simulation using the global mass and energy balances in steady state was utilized to evaluate the efficacy of the proposed configuration. The temperature profiles, calculated for a typical hot day in a tropical region, reveal that the configuration is more efficient than the single channel chimney, achieving thermal efficiency values near 75%. This solar chimney configuration can be used with better results than the traditional design as an alternative for natural ventilation systems in residential buildings without a significant increase in the cost of the residence investment.

Drying behavior for ripe and unripe Crataegus mexicana was studied for an indirect solar dryer in order to obtain the effective diffusion coefficients . The behavior of the drying process was examined, observing a significant effect related to the ripeness degree of the fruit. The effective diffusion coefficients were 5.698 x10-10 m2/s and 5.993 x10-10 m2/s, for unripe and ripe Crataegus mexicana, respectively. Solar drying is a viable alternative for adding value to the production of this fruit in central Mexico.

There has been a growing interest on the study of the human-ecosystems sustainability from an integrated point of view. The recognition of sustainability as a complex long-term problem has led to the development of tools that address it from a multidisciplinary perspective. Several mathematical models have been developed to support the understanding of the interactions between human activities and natural resources, and the effect of economic and social issues on the sustainability of such ecosystems. However, only a few of them have considered these aspects holistically; even more, critical social aspects such as uneven distribution of goods and services have not been addressed. Developed countries have a larger pool of resources when facing environmental and social issues than developing countries, along with larger resilience to external threats, being able to rely on scientific and technological advances against poverty and inequity. Therefore, these developed societies are socially fairer and more capable to become sustainable. This work presents an integrated ecological-economical-social model which includes a crucial social aspect in the sustainability study such is inequity, having two sectors of a population with different economic distribution values; these two populations interact directly with each other representing a stratified society as an attempt to conceptually describe a real-world human ecosystem where goods and natural resources are not evenly distributed. Also, the system is under the influence of uncertain external instabilities due to the inclusion of real data series, captured in the model by stochastic processes incorporated in demographic variables. Four typical kinds of societies with different levels of development and inequity were analyzed. Over the sixty percent of the times simulations led to a system collapse, being the probability of collapse directly related to the inequity. Results suggest that societies with unevenly distributed economies are less sustainable than more equalitarian societies, proving that economic equality is an important factor to consider in sustainability studies. Additionally, a typical pattern for per capita wages was found in those cases where the system collapsed, similar from a macroeconomic behavior in real world economies under economic crisis.

There is a great concern about the current growing and indiscriminate human activity over the environment that supports us. Many researchers and strategists have predicted human and natural disasters based on assumptions linked to economic, social and environmental trends. Most of these events have either not happened, the impact was far less than projected or they may have been delayed several years. Such situations might have not been due to erroneous assumptions or calculations, but rather to the ability of ecosystems to recover and adapt following a stressful situation or a disruptive event, so called resilience. The mechanisms of how resilience works generally imply intrinsic complex feedback loops of the systems; such loops absorb the external impacts and distribute them through the different elements that conform the system. This work analyzes a social feedback loop in the dynamics of a human-ecosystem and its consequences in the search for sustainable governmental targets. The loop is derived from the uncertainties in the demographic parameters and is incorporated into a mass-based compartmental model of a human ecosystem with social inequity. The methodology has been applied to three different types of societies with different development levels, using five different governmental targets. As expected, it is clear that feedback loops affect the behavior of the system. Results show that the same government target provides different results for different societies. Societies with a higher level of development present a better economic-social-environmental performance when an environmental target is used, such as GHG emissions; on the other hand, the society with less development presents a better performance when the used target is related to the economics of the system. In order to achieve a sufficiently resilient and sustainable society, the choice of an accurate governmental target is crucial since the objectives act differently for each type of society. Results also show that decision variables have more variation in the first periods of the analysis being practically stable in the final years. Then, the initial years of the simulation analysis are the most important to stabilize the system. Therefore, in real-world societies, not only the choice of the objective is important, but also the feedback mechanisms. Those mechanisms have to be implemented as quickly as possible, in order to identify or avoid probable future catastrophic scenarios.

Food waste (FW) handling is a complex problem. Discarding food with nutritional value attempts against food safety and amounts to wasting the natural resources needed in its production. Developing concrete alternatives to valorize FW is becoming increasingly important. The transformation of FW from restaurants into swine feed is evaluated in this work. The FW is collected from restaurants and classified according to its composition as citric (CiW), vegetable (VeW), cereal (CeW), and processed food waste (PFW). An optimization strategy is performed to select the combination of FW with minimum moisture content that not only meets the nutrient requirement of pigs in their last breeding stage but also reduces the energy-related costs involved in the drying stage. The collected FW comprises 28.42, 25.37, 7.47, and 38.74 percent, of CiW, VeW, CeW, and PFW. According to their bromatological characterization, the first three wastes are rich in carbohydrates (68.87–73.74% on dry basis), whereas PFW is more balanced (average protein and fat contents are 17.91 and 24.85% on dry basis). The FW mix with the least moisture content consists mainly of PFW and CeW, with 59.77 and 37.65%. Maize must be added to the mixture to satisfy the ideal amino acid content. The proposed swine feed from FW contributes to reducing the need for costly meals with high protein content, such as soy meal.

This second paper of our series is concerned with the formulation and solution strategies of fractional optimal control problems (FOCP). Given the sets of fractional differential equations representing the behavior of fermentation and thermal hydrolysis reactive systems, here we formulate the corresponding FOCP’s and describe suitable techniques for solving them. An analytical/numerical strategy that combines the optimality conditions and the gradient method for FOCP as well as the predictor–corrector fractional integrator is used to obtain optimal dilution rate profiles for the fermentation case-study. For the case of the thermal hydrolysis, the strategy involves discretization of the FOCP to formulate it as a Non-Linear Programming problem; then, the solution algorithm involves the use of an NLP solver and the shooting technique coupled to an inverse Laplace transformation subroutine. The optimal profiles show the performance of the numerical solution approaches proposed and the effect of the fractional orders in the optimal results.

A conventional approach to solving stochastic optimal control problems with time-dependent uncertainties involves the use of the stochastic maximum principle (SMP) technique. For large-scale problems, however, such an algorithm frequently leads to convergence complexities when solving the two-point boundary value problem resulting from the optimality conditions. An alternative approach consists of using continuous random variables to capture uncertainty through sampling-based methods embedded within an optimization strategy for the decision variables, such a technique may also fail due to the computational intensity involved in excessive model calculations for evaluating the objective function and its derivatives for each sample. This paper presents a new approach to solving stochastic optimal control problems with time-dependent uncertainties based on BONUS (Better Optimization algorithm for Nonlinear Uncertain Systems). The BONUS has been used successfully for non-linear programming problems with static uncertainties, but we show here that its scope can be extended to the case of optimal control problems with time-dependent uncertainties. A batch reactor for biodiesel production was used as a case study to illustrate the proposed approach. Results for a maximum profit problem indicate that the optimal objective function and the optimal profiles were better than those obtained by the maximum principle.

We report the spontaneous formation of spatial reactivity patterns during the oxidation of H 2 /CO mixtures in diluted perchloric acid on a rotating Pt ring-electrode. Under potentiostatic conditions, the system is oscillatory. At the onset of oscillations at low applied voltage, regular spatio-temporal structures form. At intermediate and high applied voltages spatio-temporal chaos is observed, which exhibits an increasing number of high reactivity spikes with increasing voltage. The transition between the regular and chaotic regime is rather complex and involves period-doubled states. Under galvanostatic conditions, a similar spatio-temporal scenario is observed as a function of applied current.

This paper describes a methodology to establish an optimal process design for prickly pear wine production that preserves the peculiar and unique traits of traditional products, generating at the same time, technical information for appropriate design of both bioreactor and overall process. The strategy includes alcoholic fermentation optimization by the mixed native culture composed by Pichia fermentans and Saccharomyces cerevisiae, followed by malolactic fermentation optimization by Oenococcus oeni. The optimization criteria were based on multiple output functions: alcohol content, volatile compounds profile, organic acids profile, and compound contents related to color, which were analyzed by spectroscopy-chromatography methods and sensory analysis. The results showed that the mixed culture inoculated into a bioreactor containing prickly pear juice with 20 °Bx of fermentable sugars concentration, processed at a constant temperature of 20 °C for 240 h, leads to a fermented product with 9.93% (v/v) total alcohol content, and significant abundance of volatile compounds, which provide fruity and ethereal aromatic notes, complemented by a lively but not unpleasant acidity. This young wine was further subjected to malolactic fermentation at constant temperature (16 °C) for 192 h, decreasing malic acid, and balancing volatile compounds contents, thus resulting in a product with better aroma and flavor perception, and a velvety feeling of long aftertaste. Repeated assays showed that the process is stable, predictable, controllable, and reproducible. These results were used for process design and spreadsheet construction in order to simulate the process, and properly select and size the equipment required for such process.

A procedure for designing starter cultures for fermentation is illustrated for prickly pear wine production. The illustration includes kinetic studies on inoculated and spontaneous fermentation, microorganism identification studies based on molecular biology tools, and microbial ecology studies, which led to the selection of strains that are capable of synthesizing alcohol and desirable volatile compounds. Results show that a mixed starter inoculum containing Pichia fermentans and Saccharomyces cerevisiae leads to a fermented product that contains 8.37% alcohol (v/v). The gas chromatography and mass spectrometry (GC-MS) analysis shows the presence of 9 major volatile compounds (Isobutanol, Isopentanol, Ethyl acetate, Isoamyl acetate, Ethyl octanoate, Ethyl decanoate, Ethyl 9-decanoate, β-Phenylethyl acetate, and Phenylethyl alcohol) that have ethereal, fruity, aromatic notes that are considered to be essential for a fine wine flavor. These compounds harmonically synergize with the alcohol to produce a fermented product with a unique flavor and taste. Several assays using the mixed culture show that the process is stable, predictable, controllable, and reproducible. Moreover, the results show that a mixed culture leads to a broader range of aromatic products than that produced by a single, pure culture. Therefore, we conclude that combinations of Saccharomyces strains and non-Saccharomyces strains can be used to obtain high-quality fermented beverages from prickly pear juice.

A numerical study was performed on the transient heat and mass convection of grain storage in a cylindrical silo. Temperature gradients were induced by the heat of respiration, and thermal gradients were generated by variations of the temperature surrounding the cavity. The model was developed using the equations of heat, mass, and momentum transport for multiphasic media. The equation that represents the environmental temperature along the day-night cycle was obtained via a least squares regression using statistical data. In this study, the effects of different geometric ratios (A) and Rayleigh (Ra) numbers on the isotherms, flow patterns, and concentration isolines were analyzed. The governing equations were solved using discretization of the spatial coordinates by orthogonal collocation with Legendre polynomials and an implicit-trapezoidal formulation for time. The resulting algebraic system was solved by employing the Newton–Raphson with LU factorization method. A computer code called NEWIMPC2 in the FORTRAN 90 language was developed; this code was used to calculate the dynamics and hot regions in the bulk mass grain in the cavity. The thermodynamic properties for sorghum were used in the simulation, although the model is applicable to any cereal grains. For simulation data, typical prevailing conditions in the Bajio, an agricultural region located in Guanajuato State, were used. When the geometric ratio (A) increases, the hot nucleus is displaced toward the top of the cavity. In the case of Rayleigh numbers (Ra), a small increase significantly increases the stiffness of the parabolic equations. The ambient temperature has a significant effect on the formation of hot regions inside the cavity. When Ra increases, an increase can be observed in the temperature of the hot nucleus, with this temperature reaching 31 °C near the top wall of the enclosure. There was no evidence of the formation of multicellular flows.

The optimal noise amplitude for Stochastic Resonance (SR) is located employing an Artificial Neural Network (ANN) reference model with a nonlinear predictive capability. A modified Kalman Filter (KF) was coupled to this reference model in order to compensate for semi-quantitative forecast errors. Three manifestations of stochastic resonance, namely, Periodic Stochastic Resonance (PSR), Aperiodic Stochastic Resonance (ASR), and finally Coherence Resonance (CR) were considered. Using noise amplitude as the control parameter, for the case of PSR and ASR, the cross-correlation curve between the sub-threshold input signal and the system response is tracked. However, using the same parameter the Normalized Variance curve is tracked for the case of CR. The goal of the present work is to track these curves and converge to their respective extremal points. The ANN reference model strategy captures and subsequently predicts the nonlinear features of the model system while the KF compensates for the perturbations inherent to the superimposed noise. This technique, implemented in the FitzHugh-Nagumo model, enabled us to track the resonance curves and eventually locate their optimal (extremal) values. This would yield the optimal value of noise for the three manifestations of the SR phenomena.

Vitamin A deficiency (VAD) is the leading cause of preventable blindness in children and raises the risk of disease and death from severe infections. Some plant foods contain darkly colored pigments with provitamin A activity (α-carotene, β-carotene, and β-cryptoxanthin) that can be converted to vitamin A. In this work, a process is described for β-cryptoxanthin production by Flavobacterium lutescens ITCB008 using fermentation technology. The critical process variables associated with batch fermentation production in a stirred vessel were evaluated and quantified based on experimental design strategies. An ad hoc mathematical model was constructed and used to identify the operating conditions (agitation 190 rpm, aeration 0.086 VVM [air volume/broth volume per minute], and temperature 27.6 °C, pH 7.3) that maximized β-cryptoxanthin synthesis (770 mg of β-cryptoxanthin/kg biomass; dry weight) in a short processing time (28 h). In addition, downstream-processing studies were performed revealing that extraction of β-cryptoxanthin can be achieved with a polar solvent that produces cell lysis, such as methanol or ethanol, in only 1 step. High performance liquid chromatography analysis showed that β-cryptoxanthin represents nearly 95% of the total carotenoid content in the obtained product. The process described could be a commercial alternative for producing β-cryptoxanthin additives, as supplement in food formulation, thus assisting VAD prevention.

The marigold flower (Tagetes erecta) is one of the richest natural sources of xanthophylls, mainly lutein. Its saponified extract is used as an additive in several food and pharmaceutical industries. However, the efficiency in the xanthophylls extraction from this natural source depends heavily upon an appropriate prior treatment, given to the flower, to increase wall-cells permeability and facilitate the diffusive mechanisms of mass exchange between immiscible phases during the lixiviation process. In this work, the effect of different treatments on marigold flower to increase the xanthophylls extraction efficiency is studied. The results clearly indicate the interrelation that exists between the treatment and the extraction. It is shown that almost full recovery of the xanthophylls contents can be achieved when the marigold is treated with hydrolytic enzymes synthesized by microbiota associated to marigold flowers. These results have significant impact on the cost-efficiency of the process.

The prediction of the oxygen transfer coefficients (kL) in bubbling columns is studied. Bubbling columns have important applications as bioreactors. It is found that the coefficient is strongly inf...

In "equation-free" multiscale computation a dynamic model is given at a fine, microscopic level; yet we believe that its coarse-grained, macroscopic dynamics can be described by closed equations involving only coarse variables. These variables are typically various low-order moments of the distributions evolved through the microscopic model. We consider the problem of integrating these unavailable equations by acting directly on kinetic Monte Carlo microscopic simulators, thus circumventing their derivation in closed form. In particular, we use projective multi-step integration to solve the coarse initial value problem forward in time as well as backward in time (under certain conditions). Macroscopic trajectories are thus traced back to unstable, source-type, and even sometimes saddle-like stationary points, even though the microscopic simulator only evolves forward in time. We also demonstrate the use of such projective integrators in a shooting boundary value problem formulation for the computation of "coarse limit cycles" of the macroscopic behavior, and the approximation of their stability through estimates of the leading "coarse Floquet multipliers"., Submitted to Journal of Computational Physics

Endogenous microorganisms isolated from the marigold flower (Tagetes erecta) were studied to understand the events taking place during its ensilage. Studies of the cellulase enzymatic activity and the ensilage process were undertaken. In both studies, the use of approximate second-order models and multiple lineal regression, within the context of an experimental mixture design using the response surface methodology as optimization strategy, determined that the microorganisms Flavobacterium IIb, Acinetobacter anitratus, and Rhizopus nigricans are the most significant in marigold flower ensilage and exhibit high cellulase activity. A mixed culture comprised of 9.8% Flavobacterium IIb, 41% A. anitratus, and 49.2% R. nigricans used during ensilage resulted in an increased yield of total xanthophylls extracted of 24.94 g/kg of dry weight compared with 12.92 for the uninoculated control ensilage. Keywords: Marigold flower; ensilage; starter inoculum; mixed culture

We present an example of the practical implementation of a protocol for experimental bifurcation detection based on on-line identification and feedback control ideas. The idea is to couple the experiment with an on-line computer-assisted identification/feedback protocol so that the closed-loop system will converge to the open-loop bifurcation points. We demonstrate the applicability of this instability detection method by real-time, computer-assisted detection of period doubling bifurcations of an electronic circuit; the circuit implements an analog realization of the Roessler system. The method succeeds in locating the bifurcation points even in the presence of modest experimental uncertainties, noise and limited resolution. The results presented here include bifurcation detection experiments that rely on measurements of a single state variable and delay-based phase space reconstruction, as well as an example of tracing entire segments of a codimension-1 bifurcation boundary in two parameter space., Comment: 29 pages, Latex 2.09, 10 figures in encapsulated postscript format (eps), need psfig macro to include them. Submitted to Physica D

The design of many separation processes is often hindered by the lack of distribution data between two immiscible phases. Inappropriate or partial designs result in inefficiencies that translate into poor recovery of the products, overdesigned separation equipment, or additional purification and separation stages. In recent years, statistically based experimental design has been successfully used to quantify relevant factors in many biological and chemical processes. In this study, we illustrate the capabilities of one of these approaches, the response surface methodology (RSM) for the optimization of lixiviation processes of marigold flower flour for xanthophyll extraction (oleoresin) using hexane. The method leads to conditions for the maximum recovery of the pigment, allowing the determination of the optimal residence time (14.7 min) and flour-to-solvent ratio (1:5) at 35 °C. These values can be used in the lixiviation process in any operation mode. In this study, we illustrate the use of a countercurr...

Artificial neural networks (ANNs) have proven to be a valuable tool for the data-driven construction of empirical models. The predictive capabilities of the ANN approximations are often validated within a statistical framework. For many applications, one can rely on a first-principles background that allows for the construction of hybrid models. By constrasting the predictions of the hybrid models with the theory, valuable insight can be obtained about the predictive capabilities of the approximation and the first-principles formulation itself. In this contribution, we illustrate the use of ANNs for the construction of hybrid models for the determination of mixing rules for the BACK thermodynamic equation of state (EOS). The hybrid approximations are constructed in three levels, with the progressive addition of more information to the empirical formulation. Using an ANN within the framework of an EOS substantially increases the potential for applications allowing for the estimation of thermodynamic proper...

Experimental measurements of the variation in the solid fraction during crystallization of lipid mixtures are often correlated in terms of the so-called Avrami model. In this paper, the above model was employed to describe measurements taken during the crystallization of blends of tripalmitin in olive oil at high concentrations. Although the blends appeared to behave ideally, the Avrami model failed to describe the experimental results over the entire range of tripalmitin concentration investigated. This discrepancy appears to be correlated with the interfacial free energy. As an alternative to the description of lipid crystallization experiments, the use of continuous-time artificial neural network (ANN) approximators is proposed. For the system studied here, the ANN successfully reproduced the experimentally observed behavior for all temperatures and tripalmitin concentrations used.

We use nonlinear signal processing techniques, based on artificial neural networks, to construct a continuous-time model (set of ordinary differential equations, ODEs) from experimental observations of mixed-mode oscillations during the galvanostatic oxidation of hydrogen on platinum in the presence of bismuth and chloride ions. The data was in the form of time-series of the potential for different values of the applied current and chloride ion concentration. We use the model to reconstruct the experimental dynamics and to explore the associated bifurcation structures in phase-space. Using numerical bifurcation techniques we locate stable and unstable periodic solutions, calculate eigenvalues, and identify bifurcation points. This approach constitutes a promising data post-processing procedure for investigating phase-space and parameter space of real experimental systems; it allows us to infer phase-space structures which the experiments can only probe with limited measurement precision and only at a discrete number of operating parameter settings. For example, the fitted model suggests the existence of a sub-critical Hopf bifurcation near the range of parameters probed in the experiments; this might explain the experimental difficulty in locating small amplitude simple oscillations.

We present and discuss an inherent shortcoming of neural networks used as discrete-time models in system identification, time series processing, and prediction. Trajectories of nonlinear ordinary differential equations (ODEs) can, under reasonable assumptions, be integrated uniquely backward in time. Discrete-time neural network mappings derived from time series, on the other hand, can give rise to multiple trajectories when followed backward in time: they are in principle noninvertible. This fundamental difference can lead to model predictions that are not only slightly quantitatively different, but qualitatively inconsistent with continuous time series. We discuss how noninvertibility arises, present key analytical concepts and some of its phenomenology. Using two illustrative examples (one experimental and one computational), we demonstrate when noninvertibility becomes an important factor in the validity of artificial neural network (ANN) predictions, and show some of the overall complexity of the predicted pathological dynamical behavior. These concepts can be used to probe the validity of ANN time series models, as well as provide guidelines for the acquisition of additional training data.

Detailed first-principles models of transport and reaction (based on partial differential equations) lead, after discretization, to dynamical systems of very high order. Systematic methodologies for model order reduction are vital in exploiting such fundamental models in the analysis, design and real-time control of distributed reacting systems. We briefly review some approaches to model order reduction we have successfully used in recent years, and illustrate their capabilities through (a) the design of an observer and stabilizing controller of a reaction-diffusion problem and (b) two-dimensional simulations of the transient behavior of a horizontal MOVPE reactor.

Advances in scientific computation and developments in spatially resolved sensor technology have, in recent years, critically enhanced our ability to develop modeling strategies and experimental techniques for the study of the spatiotemporal response of distributed nonlinear systems. The usual alternatives for the modeling of these systems, simplifying techniques that seek to capture the distributed system dynamics through lumped parameter models, can be drastically underresolved, and miss important features of the true system response. Robust implementations of distributed system identification algorithms based on detailed spatiotemporal experimental data have, therefore, an important role to play. In this contribution we present a methodology for the identification of distributed parameter systems, based on artificial neural network architectures, motivated by standard numerical discretization techniques used for the solution of partial differential equations.

The success of nonlinear system identification and characterization from experimental time-series often depends on the appropriate pre-processing of the data. This pre-processing can, in many instances, be achieved through linear and/or “nonlinear” principal component analysis. In recent years, neural network based techniques as a means to perform the nonlinear principal component (NLPC) analysis have gained increased attention. In this contribution, we address an inherent shortcoming of these techniques: the self-consistency problem. We present a modification to the usual feedforward neural network architecture used to extract the NLPCs that results in attenuation of this problem. The proposed modification may significantly reduce representation errors in many other applications for which NLPC analysis is a powerful tool, such as feature extraction and image processing.

Artificial Neural Networks (ANNs) can be used for grey-box or black-box modeling of continuous-time systems by placing them in a framework based on numerical integration techniques. When an implicit integration scheme is used as a template, it imposes a recurrent structure on the overall network. Here we present three algorithms suitable for the training of such “network-plus-integrator” assemblies and compare their relative computational efficiencies. Pineda's Recurrent Back-Propagation (RBP) training method is recast to exploit the structure of the assembly. The second approach is RBP modified to evaluate partial derivatives of network outputs with respect to parameters exactly, while the third is a Newton-Raphson based algorithm in which outputs of the network and partial derivatives are computed at each step instead of approximated. We compare the methods via an illustrative example and discuss aspects of training in a parallel computing environment.

We use nonlinear signal processing techniques, based on artificial neural networks, to construct an empirical mapping from experimental Rayleigh-Benard convection data in the quasiperiodic regime. The data, in the form of a one-parameter sequence of Poincare sections in the interior of a mode-locked region (resonance horn), are indicative of a complicated interplay of local and global bifurcations with respect to the experimentally varied Rayleigh number. The dynamic phenomena apparent in the data include period doublings, complex intermittent behavior, secondary Hopf bifurcations, and chaotic dynamics. We use the fitted map to reconstruct the experimental dynamics and to explore the associated local and global bifurcation structures in phase space. Using numerical bifurcation techniques we locate the stable and unstable periodic solutions, calculate eigenvalues, approximate invariant manifolds of saddle type solutions and identify bifurcation points. This approach constitutes a promising data post-processing procedure for investigating phase space and parameter space of real experimental systems; it allows us to infer phase space structures which the experiments can only probe with limited measurement precision and only at a discrete number of operating parameter settings., Comment: Additional figures available via anonymous ftp to princeton.edu

The kinetics of the thermal hydrolysis of the fructans of Agave salmiana were determined during the cooking step of mezcal production in a pilot autoclave. Thermal hydrolysis was achieved at different temperatures and cooking times, ranging from 96 to 116 °C and from 20 to 80 h. A simple kinetic model of the depolymerization of fructans to monomers and other reducing sugars and of the degradation of reducing sugars to furans [principally 5-(hydroxymethyl)furfural, HMF] was developed. From this model, the rate constants of the reactions were calculated, as well as the pre-exponential factors and activation energies of the Arrhenius equation. The model was found to fit the experimental data well. The tradeoff between a maximum fructan hydrolysis and a critical furan concentration in allowing for the best ethanol yield during fermentation was investigated. The results indicated that the thermal hydrolysis of agave was optimal, from the point of view of ethanol yield in the ensuing fermentation, in the temperature range of 106-116 °C and the cooking range time of 6-14 h. The optimal conditions corresponded to a fructan hydrolysis of 80%, producing syrups with furan and reducing sugar concentrations of 1 ± 0.1 and 110 ± 10 g/L, respectively.

The occurrence of instabilities in chemically reacting systems, resulting in unsteady and spatially inhomogeneous reaction rates, is a widespread phenomenon. In this article, we use nonlinear signal processing techniques to extract a simple, but accurate, dynamic model from experimental data of a system with spatiotemporal variations. The approach consists of a combination of two steps. The proper orthogonal decomposition [POD or Karhunen-Loeve (KL) expansion] allows us to determine active degrees of freedom (important spatial structures) of the system. Projection onto these “modes” reduces the data to a small number of time series. Processing these time series through an artificial neural network (ANN) results in a low-dimensional, nonlinear dynamic model with almost quantitative predictive capabilities. This approach is demonstrated using spatiotemporal data from CO oxidation on a Pt (110) crystal surface. In this special case, the dynamics of the two-dimensional reaction profile can be successfully described by four modes; the ANN-based model not only correctly predicts the spatiotemporal short-term behavior, but also accurately captures the long-term dynamics (the attractor). While this approach does not substitute for fundamental modeling, it provides a systematic framework for processing experimental data from a wide variety of spatiotemporally varying reaction engineering processes.

A methodology for reconstructing low-dimensional dynamical systems using a combination of artificial neural networks (ANNs) and the shooting method (SM), for solving differential equations with mixed boundary conditions, is presented. Previously, the reconstruction of dynamical systems based in experimental results has been addressed using several approaches that rely in the availability of all the state variables, the use of high-order polynomial equations to relate variables in order to obtain a set of Ordinary Differential Equations (ODEs), and/or the use of numerical derivatives as surrogate variables. These approaches may exhibit undesirable numerical sensitivity, and a difficult physical interpretation of the reconstructed variables. The methodology presented here bypasses the use of polynomial approximations and does not require of time-series of all state variables.

Artificial neural networks (ANNs) are often used for short term discrete time predictions of experimental data. In this paper we focus on the capability of such nets to correctly identify long term behavior and, in particular, observed bifurcations. As we show, the usual discrete time mapping approach is (precisely because of its discrete nature) often incapable of reproducing observed bifurcation sequences. If the interest is only in periodic or temporally more complicated behavior, a Poincare map extracted from the experimental time series can be used to circumvent this problem. A complete dynamic picture including bifurcations of steady states can, however, only be captured by a continuous-time model. We present an ANN configuration which couples a “nonlinear principal component” network for data processing (Kramer, 1991, Usui et al., 1990) with a composite ANN based on a simple integrator scheme. This ANN is able to correctly reconstruct the bifurcation diagram of our experimental data. All t...

A strategy to optimize biotechnological process design is illustrated for the production of fructose-rich syrups via enzymatic hydrolysis of agave fructo-oligosaccharides. The optimization process includes ecological studies from natural fermentations leading to the selection of a strain with capacity for inulinase synthesis, and variable optimization for the synthesis, and enzymatic hydrolysis using the response surface methodology. The results lead to the selection of Kluyveromyces marxianus , endogenous strains isolated from aguamiel (natural fermented sugary sap from agave plants), as the main strain with high capacity for enzyme synthesis with inulinase activity. Production optimization at bioreactor level revealed that operation at 30.6 degrees C, 152 rpm, 1.3 VVM of aeration, and pH 6.3 leads to maximum inulinase synthesis, whereas 31 degrees C, 50 rpm, and pH 6.2 leads to maximum hydrolysis of agave fructo-oligosaccharides. HPLC analysis of the fructose-rich syrups obtained at these optimal conditions showed an average composition of 95% of fructose and 5% of glucose and the absence of sucrose. The analysis also revealed that the syrups are free of residues and toxic compounds, an undesirable occurrence often present when traditional methods based on thermal or acid hydrolysis are applied for their obtainment. Therefore, the product may be suitable for use as additive in many applications in the food and beverage industries.

Enzymatic treatments using noncommercial enzymes as a means to the improve the extraction of carotenoids and capsaicinoids from chili fruits are explored in this study. The results show that it is possible to obtain chili fruit powder with a higher concentration of both capsaicinoids and carotenoids than previously reported for similar processes. Furthermore, extraction yields above 96% for carotenoids and 85% for capsaicinoids as separate fractions can be achieved using a sequential and selective two-stage extraction. Evidence is presented demonstrating that the content and extraction yield depend directly on the extent of the enzymatic hydrolysis of chili cell walls, and higher yields are obtained when the sample is completely hydrolyzed. The enzymatic treatment described here is a promising alternative to current industrial practices, and it improves the extraction of carotenoids and capsaicinoids from chili fruits.

The efficient characterization of the nonlinear dynamical response of kineticmolecular simulations is discussed. Following ideas originally proposed by Kevrekidiset al. [1, 2], one can empower molecular simulations as model-free equations anduse them as a reference to perform bifurcation detection. Such a procedure re-quires the use of trajectories from the molecular simulation to generate low-ordermodels (e.g. polynomial) that allow one to infer the location of a bifurcation. Ifsuch identification step can be performed robustly, a feedback control policy thatdrives the molecular simulation to the bifurcation point can be constructed. Inprevious work, the identification of the low-order model has been singled out asthe key element in handling noise trajectories, such as those generated by low-resolution molecular simulations. Here, a procedure motivated by the use ofKalman Filter observers is proposed as a means to give robustness to the iden-tification procedure. The potential of the technique to characterize the dynamicalresponse of kinetic molecular simulations is illustrated using examples related tothe CO oxidation over a catalytic surface.KEYWORDS: bifurcation, Monte Carlo, identification

In this work is studied the effect of a noncommercial enzyme preparation on xanthophyll extraction from marigold flower (Tagetes erecta). The enzymatic extract was synthesized by endogenous microorganisms previously isolated and identified as Flavobacterium IIb, Acinetobacter anitratus, and Rhizopus nigricans. The results show that the extraction yield depends directly on the extent of the enzymatic hydrolysis of cell walls in the flower petals and that it is possible to reach yields in excess of those previously reported for treatments with commercially available enzymes (29.3 g/kg of dry weight). HPLC analysis of the product indicates that the original xanthophyll profile is not altered. The enhanced extraction system appears to be very competitive when compared to the traditional process and current alternatives.

We present a method of constructing low-dimensional nonlinear models describing the main dynamical features of a discrete 2D cellular fault zone, with many degrees of freedom, embedded in a 3D elastic solid. A given fault system is characterized by a set of parameters that describe the dynamics, rheology, property disorder, and fault geometry. Depending on the location in the system parameter space we show that the coarse dynamics of the fault can be confined to an attractor whose dimension is significantly smaller than the space in which the dynamics takes place. Our strategy of system reduction is to search for a few coherent structures that dominate the dynamics and to capture the interaction between these coherent structures. The identification of the basic interacting structures is obtained by applying the Proper Orthogonal Decomposition (POD) to the surface deformations fields that accompany strike-slip faulting accumulated over equal time intervals. We use a feed-forward artificial neural network (ANN) architecture for the identification of the system dynamics projected onto the subspace (model space) spanned by the most energetic coherent structures. The ANN is trained using a standard back-propagation algorithm to predict (map) the values of the observed model state at a future time given the observed model state at the present time. This ANN provides an approximate, large scale, dynamical model for the fault., 30 pages, 12 figures

The sunflower oil require of an extraction process that reduce the amount of impurities, and the use of a solvent that has not implications over the environment. An alternative is the use of the isopropyl alcohol. In this work was studied the extraction conditions of two experimental varieties of sunflower Madero-91 and Cianoc-2, the controls were hexane and Sunflower seed variety Victoria. It was found differences on the physical characteristics and in the dispersion of the physical measures from the three varieties used. The percent yield of crude oil extracted with isopropyl alcohol was not different of the obtained with hexane. The level of free fatty acids was higher in the oils extracted with isopropyl alcohol than the oils extracted with hexane, no differences was found in the amount of malonic aldehide. El aceite de girasol requiere de un proceso de extracción que reduzca el contenido de impurezas y del uso de un solvente que no atente contra el ambiente. Una alternativa es el empleo del isopropanol. En la presente investigación se estudiaron las condiciones de extracción de aceite de dos variedades de girasol, Madero 91 y Cianoc-2, y como control al girasol variedad Victoria y el hexano. Se encontraron diferencias en cuanto a características físicas de las variedades estudiadas, así como en la dispersión de las mediciones físicas. El rendimiento en porcentaje de aceite crudo con el alcohol isopropílico no difirió del obtenido con hexano. El nivel de ácidos grasos libres fue superior en los aceites extraídos con isopropanol. No se encontraron diferencias en cuanto al contenido de aldehído malónico.

A neural network-based approach for continuous-time modeling of nonlinear systems is presented. The approach is based on an implicit integrator and recurrent networks. The resulting continuous-time model (a set of ordinary differential equations) is capable of correctly capturing the long term attractors of the system. >