基于用户数据的电动汽车充电桩网络攻击模型

电动汽车(electrical vehicle, EV)的用户数据是优化EV充电成本的关键数据，对用户数据的操纵可能导致错误的充电成本，使充电桩运营商蒙受经济损失。针对EV充电桩，提出了一种基于用户数据的网络攻击模型，该模型通过篡改接入EV充电桩的用户数据生成虚假的充电计划，从而提高充电桩的充电成本。所提出的攻击模型为一种基于混合整数线性规划的两层优化模型，在上层生成注入EV充电桩的恶意用户数据，下层为EV充电计划优化算法，通过应用KKT条件使两层优化模型转化为单层优化攻击模型。仿真以一个虚拟充电站场景为例，相较于无攻击情况，该攻击模型通过提高EV的充电能量或移动EV的充电时段增加了充电站的总充电成本，验证了该攻击模型的可行性和危害性。

Cyber Attack Model of EV Charging Pile Based on User Data

The user data of electric vehicle (EV) is the key data to optimize the EV charging cost. Manipulation with the user data may lead to incorrect charging costs and cause economic losses to the charging pile operator. For EV charging pile, a cyber attack model based on user data is proposed. This model generates a incorrect charging plan by tampering with the user data connected to EV charging pile, thus increasing the charging cost. The proposed attack model is a two-stage optimization model based on mixed integer linear programming, which generates malicious user data injected into EV charging piles on the upper level and EV charging plan optimization algorithm on the lower level. By applying KKT conditions, the two-stage optimization model is transformed into a single-level optimization attack model. The simulation takes a virtual charging station scenario as an example. Compared with the non attack scenario, the attack model increases the total charging cost of the charging station by increasing the charging energy of the EV or the charging period of the mobile EV, which verifies the feasibility and harmfulness of the attack model.