Battery diagnostics and performance prediction: computational vs. expert system based approach

In this paper, two methods for battery field diagnostics and performance prediction are discussed. The former is a computational method which employs the relationships between the internal state indicators of a cell and the most significant influencing factors (e.g., temperature, technology, electrolyte density, age, work cycle). The latter uses artificial intelligence techniques based on rules developed from the expertise of specialists. The two methods are compared with respect to stationary batteries