Modeling and simulation issues of electrochemical devices such as batteries, fuel cells, super capacitors and sensors are gaining more attention these days. This is mainly because of the developments in hybrid power systems that are most important in growing energy demand. A major concern in hybrid environments is the proper load sharing and power distribution among the hybrid components. Typically in many hybrid systems the vital component will be the electrochemical power source, e.g., the batteries. Hence the optimization of operating conditions of batteries is important to control and share the power delivered by the batteries in hybrid environment. Many previous investigations have attempted to optimize batteries using various approaches. Shepherd derived a voltage-current and state-of-charge relationship for predicting discharge and charge curves particularly for lead-acid batteries[1]. Other forms of this relationship are derived to relax the assumptions behind the Shepherd model[2]. Another relationship used to predict remaining capacity of a battery was given by Peukert's empirical equation[3]. All these forms of simple relationships are not applicable to battery models in hybrid environments due to the fact that the battery can be studied only at limited conditions under which these equations work. The main conclusion of a recent critical review paper is also reiterating that the empirical equations cannot be used to predict state of charge accurately unless it is discharged under galvanostatic conditions at constant temperature[4]. Under these circumstances, a rigorous dynamic optimization is attempted to evaluate the exact optimal operating conditions for a stand-alone battery. Once if this first and foremost step is achieved successfully then we can modify the procedure into any hybrid power system. The difficulties in achieving this step include a lot of numerical inconsistence and failure of many standard optimization algorithms. Hence, in this work we identify a suitable dynamic optimization method that can lead to global optimal profiles and that is specifically suitable for models encountered in simulating batteries. Here, it is found that the iterative dynamic programming, with advantages of true and global optima, as a suitable method for optimizing battery models. This has also been demonstrated using a single electrode system that is widely used in batteries.

References 1. C.M. Shepherd, J. Electrochem. Soc., 112, 657 (1965). 2. David Linden, Handbook of batteries and fuel cells, McGraw-Hill, New York (1984). 3. D. Baert and A. Vervaet, Electrochim. Acta, 44, 3491, (1999). 4. D. Doerffel and S.A. Sharkh, J. Power Sources, 155, 395 (2006).