An optimization model for the treatment of perfluorocarboxylic acids considering membrane preconcentration and BDD electrooxidation.

Treatment of persistent perfluorocarboxylic acids in water matrixes requires of strong oxidation conditions, as those achieved by boron doped diamond (BDD) electrooxidation (ELOX). However, large scale implementation of ELOX is still hindered by its high energy consumption and economical investment. In this work, we used process systems engineering tools to define the optimal integration of a membrane pre-concentration stage followed by the BDD electrolysis of the concentrate, to drastically reduce the costs of treatment of perfluorohexanoic acid (PFHxA, 100 mg L-1) in industrial waste streams. A multistage membrane cascade system using nanofiltration (NF90 and NF270 membranes) was considered to achieve more sophisticated PFHxA separations. The aim was to minimize the total costs by determining the optimal sizing of the two integrated processes (membrane area per stage and anode area) and the optimal process variables (pre-concentration operating time, electrolysis time, input and output concentrations). The non-linear programming model (NLP) was implemented in the General Algebraic Modelling System (GAMS). The results showed that for a 2-log PFHxA abatement (99% removal), the optimal two membrane stages using the NF90 membrane obtains a 75.8% (6.4 $ m-3) reduction of the total costs, compared to the ELOX alone scenario (26.5 $ m-3). The optimized anode area and the energy savings, that were 85.3% and 88.2% lower than in ELOX alone, were the major contributions to the costs reduction. Similar results were achieved for a 3-log and 4-log PFHxA abatement, pointing out the promising benefits of integrating electrochemical oxidation with membrane pre-concentration through proper optimization for its large-scale application to waters impacted by perfluorocarboxylic acids.

Large scale nonlinear optimization for asymmetric operation and design of Simulated Moving Beds.

Simulated Moving Bed (SMB) was developed as a realization of continuous countercurrent operation of chromatographic separation. An SMB unit consists of several columns of the same length connected in series, where feed and desorbent are supplied and extract and raffinate are withdrawn continuously. This operation is repeated by shifting the supply/withdrawal points at a regular interval, making the operation symmetric. In this study, we explore asymmetric operation and design through a full-cycle optimization model, where the operation of the entire cycle is described within a nonlinear programming (NLP) problem and the Partial Differential Algebraic Equations (PDAEs) are fully discretized both in temporal and spatial domains. The NLP problem is implemented within the AMPL modeling environment and is solved using IPOPT, an interior-point NLP solver. We found that this problem is solved efficiently, and introducing a full-cycle formulation has the potential to improve the performance of SMB, as shown through single and multi-objective optimization studies.

Real time optimal guidance of low-thrust spacecraft: an application of nonlinear model predictive control.

Real time optimal guidance is considered for a class of low thrust spacecraft. In particular, nonlinear model predictive control (NMPC) is utilized for computing the optimal control actions required to transfer a spacecraft from a low Earth orbit to a mission orbit. The NMPC methodology presented is able to cope with unmodeled disturbances. The dynamics of the transfer are modeled using a set of modified equinoctial elements because they do not exhibit singularities for zero inclination and zero eccentricity. The idea behind NMPC is the repeated solution of optimal control problems; at each time step, a new control action is computed. The optimal control problem is solved using a direct method-fully discretizing the equations of motion. The large scale nonlinear program resulting from the discretization procedure is solved using IPOPT--a primal-dual interior point algorithm. Stability and robustness characteristics of the NMPC algorithm are reviewed. A numerical example is presented that encourages further development of the proposed methodology: the transfer from low-Earth orbit to a molniya orbit.

Wine distillates: practical operating recipe formulation for stills.

Consumer perceptions of flavors are associated with the chemical composition of foods. However, consumer preferences change; therefore, it is necessary for food manufacturers to be able to adapt their products. Unlike in aged spirits, the chemical composition of young spirits is determined during distillation; therefore, this is where distillers must tailor their operating recipes to the new trends. Even for an experienced distiller, the complexity of the process makes adapting the operating recipe far from straightforward. In this study, we developed a methodology for generating practical recipes that makes use of computer simulations and optimization techniques. We used Pisco Brandy, a young Muscat wine distillate from Chile and Peru as our case study. Even so, because our methodology is independent of the chemical composition of the broth, it can be applied throughout the industry. Drawing on the experience and preferences of industry enologists, we designed a preferred distillate and used our methodology to obtain the appropriate recipe. This recipe was validated in lab scale experiments, and we obtained a much closer distillate to the desired prescription than commercial products.

Fast and reliable calibration of solid substrate fermentation kinetic models using advanced non-linear programming techniques

Calibration of mechanistic kinetic models describing microorganism growth and secondary metabolite production on solid substrates is difficult due to model complexity given the sheer number of parameters needing to be estimated and violation of standard conditions of numerical regularity. We show how advanced non-linear programming techniques can be applied to achieve fast and reliable calibration of a complex kinetic model describing growth of Gibberella fujikuroi and production of gibberellic acid on an inert solid support in glass columns. Experimental culture data was obtained under different temperature and water activity conditions. Model differential equations were discretized using orthogonal collocations on finite elements while model calibration was formulated as a simultaneous solution/optimization problem. A special purpose optimization code (IPOPT) was used to solve the resulting large-scale non-linear program. Convergence proved much faster and a better fitting model was achieved in comparison with the standard sequential solution/optimization approach. Furthermore, statistical analysis showed that most parameter estimates were reliable and accurate.