SOC prediction of Lithium battery based on fuzzy information granulation and support vector regression

For achieve accurate estimate of State of Charge (SOC) in the management system of battery, a novel algorithm that combined the fuzzy information granulation (FIG) and support vector regression (SVR) was proposed in this paper to predict SOC. This algorithm use radial basis function as SVR kernel. To establish SOC prediction model of Lithium battery, the current, voltage and granulated time are regarded as input, and the value of SOC are calculated as output. The model can provide real-time prediction and estimation intervals of lithium battery. A simulation was provided to verify the accuracy, effectiveness and feasibility of this prediction algorithm.