EV’s power batteries have limited capacity, long charging process andshort life, which take EV far from popular. Thus EV mobility management isproposed. And intelligent computing algorithms like neural network and antcolony can be well applied in mobility management system.To analyze SoC and SoH, two complete monitoring systems are builton two different EVs. The experiment data has required accuracy.Aiming at the difficulty of SoC estimation, an algorithm using NN andEKF is proposed. RBF NN is applied to identify the non-linear function ofSoC, current and terminal voltage. Then, EKF is used to estimate SoC. Theexperiment results show that the estimated SoC has good accuracy, and theproposed algorithm has advantages compared with other usual methods.Aiming at the issues that EV battery has limited power capacity and itscharging process takes long time, an improved multiple-objective ant colonyalgorithm is proposed to optimize EV driving path. Based on standard ACA,we modify the heuristic function and pheromone update function to achievefaster convergence. The experiment results show that the advised drivingpaths are Pareto optimized solutions, and can balance EV’s consumingenergy and time consumption. |