为实现全球碳中和，可再生能源并网规模不断扩大，其间歇性和不确定性导致了电力系统实时功率的大幅波动。电动汽车（Electric Vehicle, EV）作为可延时负荷，因具备快速调节特性，可以为电网提供调频辅助服务。在电力体制改革下，电动汽车聚合商（Electric Vehicle Aggregator, EVA）这一概念被提出，成为协调不同联网时间EV参与电力市场的重要支撑方案，其竞价调控策略可有效实现内部资源的最优调频配置。EVA的优化运行策略，面临市场侧与电动汽车用户侧两端的不确定性，并需要考虑多种类市场间的耦合性。针对上述问题，本文对多类型EV充电资源进行了建模分析，提出了EVA基于模型预测控制的调频运行方案，主要内容包括：1）针对充电站类型，本文首先提出了EVA作为价格接受者参与能量和调频市场的最优竞价策略，定义了EV在单向或双向充电模式下的充电灵活性指标。此外，通过对最优性的数学分析，本文在该灵活性指标基础上，构建了等效EV聚合模型，并论证了聚合模型中等效EV数量的有限性。2）本文进一步考虑了电动汽车到达与离开充电站的动态过程，并设计了EVA基于模型预测控制的实时运行方案。EVA通过评估当前停留和未来到达EV容量，滚动参与能量和调频实时市场以最小化运营成本。通过滚动优化和校正，EVA可有效实现出清调频容量的实时响应及电动车主的充电需求。在数学建模中，本文采用了条件风险值来处理能量、调频出清价格和EV未来充电需求的不确定性。3）针对换电站(Battery Swapping Station, BSS)类型，本文考虑了多节点分布BSS构成的EVA充当价格制定者参与能量-调频-备用联合出清市场的竞价策略，构建了双层规划问题。上层问题，考虑了不同节点分布BSS的运行模型；下层问题，考虑了节点边际价格结算下的多市场联合出清模型。将双层问题转换为均衡约束的数学规划问题进行求解后， EVA可通过协调多场站资源配置实现全局经济最优。综上所述，本文研究了EVA考虑调频参与的竞价和运行策略，降低了EVA运营成本，并保证了电动车主的充电需求，对未来充电运营商参与电力市场有着重要的指导价值。
To reach global carbon neutrality, the penetration of renewable generations is increasing, with the intermittency and uncertainty, resulting in the rapid variation of supply and demand in the power system. Electric vehicles (EVs), as the deferrable load, can provide regulation ancillary service to the grid because of the inherent fast-ramping rate. Under the deregulation of the power industry, the concept of electric vehicle aggregator (EVA) has been proposed to coordinate EVs with different parking times to participate in the electricity market. With an advanced bidding and operation strategy in EVA, its potential of providing regulation service to the power grid can be explored. The optimal operation strategy for EVA faces uncertainties on both the market side and the EV owner side; moreover, it needs to consider the coupling effect between multiple electricity markets. Therefore, this thesis investigates the optimal bidding strategies for multiple types of EVAs and proposes a model predictive control-based operation scheme for EVAs considering regulation market participation. The following are explicit contents:1) First, the optimal bidding strategy for EVA as a price-taker in the energy and regulation market is developed, and the charging flexibility index is proposed considering both unidirectional and bidirectional modes of EVs at charging stations. Through mathematical analysis of optimality, an equivalent aggregated model of the EV fleet is established based on the charging flexibility index. Moreover, it can prove that the number of EVs in this aggregated model is limited, regardless of how many EVs there are before the equivalence.2) Furthermore, an operation methodology for EVA is presented considering the dynamic process of EV fleet arrivals and departures. The EVA can participate in energy and regulation RT markets with its current and upcoming EVs, thus reducing its total cost of purchasing energy to fulfill EVs' charging requirements. A model predictive control-based optimization is developed to consider the future arrival of EVs as well as energy and regulation prices. The index of conditional value-at-risk is used to model the risk-averseness of an EVA.3) Finally, the bidding strategy of an EVA consisting of battery swapping stations (BSSs) with locational diversity is presented, where the EVA participates in energy, regulation, and reserve markets as a price-maker. A bilevel problem is proposed: in the upper level, the operation model of BSSs at different buses is formulated; in the lower level, the joint dispatch process of energy, regulation, and reserve is formulated. The proposed bilevel problem needs to be converted to mathematical programming with equilibrium constraints. After solving such a problem, EVA can achieve a global optimum with the consideration of power flow constraints under the LMP clearing scheme. To conclude, this thesis focuses on the bidding and operation strategies for EVA participating in the regulation market, which helps EVA achieve a lucrative revenue while satisfying the charging requests from EV owners. This opens up the opportunity for wider EVA applications, which will benefit the operator, EV owners and the power grid.
