Advanced algorithms for the simulation of Li-ion batteries

Reducing fossil fuel consumption requires efforts on the improvement of energy storage technologies. The Li-ion cell is the only technology that allows to do that in the most efficient ways. Bat-man is a novel technology that provides very accurate diagnosis of lead acid batteries. The purpose of the thesis is to understand and predict the behaviour of the Li-ion batteries. To do this, in absence of a real battery to charge and discharge, a Simulink model is built for the simulation. This model is a current controlled voltage generator and has high accuracy but also high complexity because of the presence of 3 RC groups. To reduce the complexity of this, a new model with 1 RC group for the identification is built using the data collected from the previous one. Then the parameters obtained from the identificated model have been approximated with functions. Finally there is the validation step, whose purpose is to test the two models. The two approaches of validation are open-loop and closed-loop. The closed-loop utilizes the kalman filter in order to directly control the states and to correct the errors.