Showing papers by "Raghunathan Rengaswamy published in 2006"

Optimization Study of an Agglomerate Model for Platinum Reduction and Performance in PEM Fuel Cell Cathode

Abstract: Several experimental studies on optimizing the amount of platinum loading and ionomer content in the proton exchange membrane fuel cell catalyst layers exist in the literature. In the recent years, numerical studies have also been carried out with the same objective. The amount of current generated within the catalyst layer strongly depends on the local conditions such as the concentration of available oxygen, total surface area available for reaction, and the amount of ionomer. In this study, we present two different optimization formulations: (1) for minimizing the amount of platinum, and (2) maximizing the current generated. Optimization studies are performed for all the points on the base case i–v curve, one at a time. In addition, a third formulation is also presented, in which the current is maximized simultaneously for all the i–v points. With the results of the third formulation we highlight the presence of local optima and the multi-objective nature of the fuel cell catalyst design problem. The optimization formulations are presented using a spherical agglomerate steady state model. A measure based on the platinum usage is used as a benchmark while analyzing the results. Finally, we present a discussion on how these formulations can be scaled for the case of complete cathode or PEM fuel cell model. This study demonstrates that there is significant potential for reduction in platinum usage even at the smallest scales in the electrodes of a fuel cell.

Integrating stiction diagnosis and stiction compensation in process control valves

Abstract: Limit cycles caused due to valve non-linearity such as stiction can be eliminated with proper valve maintenance Since valve maintenance is usually scheduled during production stops and with production stops scheduled typically once every six months to three years, the loss of product quality and energy loss during this intermediate period can be quite high Stiction compensation algorithms can mitigate this problem to a large extent In this paper, an optimization based approach for stiction compensation much in the sprit of predictive control strategies is proposed

Techniques for stiction diagnosis and compensation in process control loops

Abstract: Loss of performance due to stiction in control valves is an important problem that needs to be solved. Industrial practitioners have articulated that stiction diagnosis from routine operating data is a component that is urgently needed for automated controller performance monitoring and improvement. While the need is well recognized, a solution that is industrially acceptable is extremely challenging. Industrial solutions preclude either invasive testing of valves or maintenance of a large database on valve characteristics for efficient stiction diagnosis. The two key industrial requirements for any stiction diagnosis technique are: the technique has to be non-invasive, and should work with routine operational data. This paper deals with the development of stiction diagnosis and compensation algorithms.

Isothermal isobaric reactive flash problem

Abstract: In this article, we present some results for isothermal isobaric reactive separation process problems. We also demonstrate that, even in isothermal isobaric reactive separation processes, which are probably the least nonlinear of all reactive separation processes, we get nonlinear phenomenon such as Hopf bifurcations. While it has been shown that Hopf bifurcations are impossible in isothermal continuous stirred tank reactor (CSTR) problems involving the methyl tert-butyl ether (MTBE) and tert-amyl methyl ether (TAME) reactions, and also in nonreactive flash problems, we demonstrate in this paper that isothermal reactive flash processes involving both MTBE and TAME mixtures exhibit Hopf bifurcations. This shows that instabilities and oscillations can occur even in isothermal reactive separation systems and are not necessarily due to multiple stages. Additionally, we show that the Rachford−Rice procedure can be extended to reactive systems.

Scope for process systems engineering studies in proton exchange membrane fuel cells (PEMFC): A review of opportunities

Abstract: Publisher Summary Outlining the scope for process systems engineering studies in proton exchange membrane fuel cells (PEMFC), this chapter presents a review of opportunities. Proton exchange membrane fuel cells PEMFCs are currently in an advanced state of development with promising applications for portable power and power generation. In spite of this, even today, there are several critical issues that researchers are trying to address and resolve. This chapter highlights some of the important technological advancements made in the areas of (i) flow field design and (ii) water and thermal management. A review of the state-of-art in engineering and design in each of these areas are also summarized in the chapter. At the same time, it also highlight the areas where opportunities exist for process systems engineering activities such as, modeling and simulation, optimization, control, diagnostics, and fault-tolerant control. The focus of the research work being discussed in this article is the systems engineering of PEMFC with the help of both detailed models and reduced order models.

Multiobjective input design for system identificationߞFrequency selection for identification of non

Abstract: Abstract An explicit multiobjective optimization formulation is proposed for choice of input harmonics in a multi-harmonic signal for identification of non-linear systems. The constraints form an exclusive disjunction (XOR) which are coded using binary variables. The resulting multiobjective optimization problem is formulated as an Integer Linear Program (ILP).

Time - Efficient, Repetitive Predictions of the Performance of PEMFCs Based on a Neural Network - Based, Reduced Order Model

Abstract: Detailed modeling of PEMFCs has been getting considerable interest for predicting the fuel cell performance and also for use in various systems engineering activities. While CFD-based equipment models provide detailed analyses of the performance, they are very time-consuming to develop and run. The computations become quite complex when such models have to be embedded into the flowsheet-level optimization of fuel cell systems. In this paper, we present results about building and using NN-based reduced order models for quickly and repetitively predicting the flow of reactants in a PEMFC manifold.