Conventional oil-fueled vehicles have exacerbated the oil/energy crisis and air pollution. Automotive energy and environmental issues are major problems faced by the traditional automotive industry. In order to solve such problems, vigorously promoting the development of electric vehicles (EVs) has become an inevitable choice. However, due to the current limitations of battery range and charging technology, there are still some challenges in the actual use of electric vehicles.

Range anxiety: Users’ range anxiety due to inexperience, cold weather, or unexpected conditions on the road prevents them from accurately predicting the state of charge (SOC), and the accessibility of their destination makes them less inclined to choose EVs for travel.

Long charging time: The charging of EVs is often time-consuming, and the charging costs are also expensive, especially at peak charging times, such as during holidays, because the load on the grid is large and uneven. In addition, charging during peak hours increases the queuing time, resulting in greater charging time and costs.

Insufficient charging infrastructure: At present, the construction of charging infrastructure is in the planning and development stage, which restricts the increase in EV ownership. In particular, the charging demand is too high during holidays, and the uneven load on the power grid or damage of the charging piles causes inconveniences to the use of electric vehicles, thus causing more charging anxiety among users.

2. Electric Vehicle Travel by Considering Multi-Dimensional Preferences of Users

In the whole travel charging process, the preference habits of EV users directly affect their choice of charging stations. Therefore, it is necessary to consider users’ preferences in charging guidance strategies. Several studies have investigated the charging preferences of the different types of EV users. Among them, Sun et al. [1], Anderson et al. [2], and Xu et al. [3] used surveys to analyze users’ charging preferences in depth. On the other hand, Erdem et al. [4] introduced time window constraints to solve and validate users preferences using heuristic algorithms by considering different charging states and charging strategies of vehicles at multiple charging stations. Zhao et al. [5] and Wen et al. [6] identified two influencing factors, tariffs and charging premiums, based on these studies of user preferences. Hu et al. [7] analyzed users’ attitudes and preferences under risky situations. Ashkrof et al. [8] and Li et al. [9] analyzed and studied the various preferences of users by building a mixture model. On the other hand, some studies have investigated different kinds of EV users’ charging preferences in depth. Among them, Hu et al. [10] and Erdogan et al. [11] analyzed the charging preferences of different types of EV users in terms of EV trajectory data and charging infrastructure optimization, respectively. Zakariazadeh et al. [12], Das et al. [13], Guang et al. [14], and Zhang et al. [15] studied the charging preferences of different classes of users to optimize EV charging and discharging strategies to alleviate the pressure on the grid during peak hours. Both Shi et al. [16] and Deb et al. [17] improved the modelling of user preference for charging stations by optimizing the deployment scheme of EV fast charging stations. The studies by Chen et al. [18], Wang et al. [19], Sarker et al. [20], and improved the modelling of user preferences for charging stations as a multi-objective optimization problem considering different user preferences to determine the optimal charging scheme. Globisch et al. [21] analyzed the factors affecting charging station preferences by studying different types of potential EV users to optimize the construction planning of future charging stations. Studies by Zhang et al. [22], Huang et al. [23], Xu et al. [24] all integrated the information of charging station locations, road networks, and power grids into the same diagram, and found that reasonable charging guidance for EVs could not only adequately coordinate the distribution of charging loads, but also relieve some of the traffic pressure. Zhang et al. [25] further formed a complete charging guidance method, and the effectiveness and efficiency of the method were verified in a large city. However, none of these studies really take the user’s own multidimensional preferences into account for charging from the user’s perspective. In summary, most of the existing literature on how to reduce the travel costs for EV users (similar and different types of users) has focused on optimizing the total travel distance and reducing the total travel costs, ignoring the multi-dimensional preferences of users for EV charging stations. In reality, users’ preferences for EV charging stations are often formed by considering multi-dimensional cost objectives such as range anxiety caused by remaining battery power, capacity limitation of charging stations, and charging time period, which can have a considerable impact on their final decision in choosing a charging station. In reality, users’ preferences for EV charging stations are often formed by considering multi-dimensional cost objectives such as range anxiety caused by remaining battery power, capacity limitation of charging stations, and charging time period, which can have a considerable impact on their final decision in choosing a charging station.