Search

Accurately quantitative evaluation of the response capability of electric vehicle aggregator (EVA), i.e., the adjustable range of power, is the premise and basis of its participation in the demand response (DR). In response to the current EVA's response capability evaluation, ignoring the autonomy of EV decision-making behavior may lead to overly idealized evaluation results, as well as the problem of evaluating only from a single time scale. To this end, this paper proposes a response capability evaluation model of EVA based on evolutionary game and response anticipation. Based on the physical constraints of EVs and their willingness to discharge, response capability evaluation models are constructed for individual EVs of different response types. Considering the autonomy of decision-making behavior of EVs, an evolutionary game model with multistrategy sets of EVs is constructed. Based on the idea of response anticipation, a regulation strategy is proposed for EVA to achieve the dynamic update of EVA's response capability. The proposed model is verified to be able to accurately evaluate the EVA's response capability through simulation examples, and the EVA response capability evaluation is extended from a single time scale to multiple time scales. The evaluation results can be used as the basis for EVA to participate in the electricity market bidding. [ABSTRACT FROM AUTHOR]

With the wide deployment of charging piles and the development of vehicle-to-grid technology, electric vehicles (EVs) will have more opportunities to participate in the operation and scheduling of electric power system through the EV aggregator (EVA), an intermediate agent between the power grid and EV users. Quantitatively evaluating the response capability (RC) of EVA, i.e., charging and discharging power ranges, is of fundamental importance for its trading with the operator of power grid. This article proposes a user-aware method to obtain a more accurate range of multitimescale RC considering the response willingness of EV users. First, a temporal RC evaluation model of a single EV considering charge-discharge state and state of charge (SOC) is established. Second, based on the user psychology model, which reflects the relationship between users' responsivity and incentive price, a multitimescale RC evaluation model of EVA considering users' willingness is built. The day-ahead RC of EVA is evaluated by the state prediction data of EVs. According to the control strategy which considers response time and SOC indicators comprehensively, the RC evaluation is updated intraday. Finally, using the statistical data from the EU MERGE project, the effectiveness of the proposed evaluation model is verified, and the impact of incentive price and scheduling time scale on the RC of EVA are analyzed. The results indicate that the proposed model can effectively track the scheduling goals of the system and realize the dynamic update of the RC of EVA. [ABSTRACT FROM AUTHOR]

In order to address the behavioral characteristics of air conditioning users and help load aggregators avoid high penalties for breach of contract caused by insufficient response capacity, it is essential to develop a practical and effective method for evaluating the response capability of aggregated air conditioning. This paper proposes a method for screening users with similar load characteristics through primary clustering and for conducting secondary clustering of air conditioning with similar adjustable characteristics using the self-organizing map. By aggregating all user devices with similar attributes and adjustable characteristics within a given region, a highly accurate virtual generator unit model is developed for air conditioning loads, with the external characteristic parameters of equivalent generator units calculated. To account for the probability of user dropout due to equipment failure or changes in willingness to respond, a response capability evaluation method for air conditioning virtual generator units is proposed. The results demonstrate that the proposed model can conveniently calculate the response capacity indices of virtual generator units under different task targets, and provide decision-making basis for dispatching centers or load aggregators to regulate their internal loads. • The user attribute similarity index is fully considered through primary clustering. • Users with similar adjustable potential are divided by the secondary clustering. • Established a highly accurate modeling method for air-conditioning virtual generator unit. • Proposed a response capability evaluation method for air-conditioning virtual generator unit. • Excessive temperature adjustment will greatly increase the probability of under response. [ABSTRACT FROM AUTHOR]