In view of the insufficient carrying capacity of highway distribution grids and the mismatch between re-newable energy generation and consumption loads, the use of mobile energy storage vehicles for energy interexchange between service areas can promote the consumption of renewable energy. In order to improve the real-time responsiveness of mobile energy storage, an optimization strategy is proposed that considering the traffic status under the constraint of the average loss rate of electric vehicle battery swapping service. A Markov chain model is developed to characterize the traffic status between service areas and the energy states of battery swapping station. By analyzing the average transportation cost of mobile energy storage and the average loss rate of battery swapping service, an optimization problem for mobile energy storage vehicle dispatching is formulated and solved to obtain the optimal scheduling strategy and its parameters. The simulation results validate the proposed strategy has a dual-threshold structure with traffic status and energy state. Compared to the greedy strategy and Q-learning algorithm, the average transportation cost of mobile energy storage is reduced by 17.23% and 8.89%, respectively.