Electric vehicle (EV) has been popularized and promoted on a large scale because of its clean and efficient features. Charging this increasing number of EVs is expected to have an impact on the electricity grid and traffic network. Therefore, it is necessary to model and forecast the EV charging demand. Most of the existing researches have not utilized real-world traffic data to analyze the EV charging demand. Few researches have considered and analyzed the characteristics of space-time transfer of charging load in urban functional areas. As an emerging mode of transportation, however, online ride-hailing trip data provide an ideal source for analyzing traffic planning and operation. On the basis of this, a charging demand forecasting model of EVs based on a data-driven approach was presented in this paper. In this methodology, it is firstly assumed that residents’ transportation trip demand is not restricted by vehicle categories(electric or fuel vehicles). The original trip trajectory data of Didi online ride-hailing were conducted to model via data mining and fusion. And the process of data analysis included region-scope selection, spatial grid modeling, trajectory data mapping, retrieval data identification and urban functional area clustering as well as traffic network modeling. Through modeling and processing, the following regenerative feature data were obtained: functional regional division (i.e., residential areas, industrial areas, commercial areas, and public service areas), trip rule distribution (i.e., temporal distribution and spatial distribution on weekdays, weekends and holidays) and actual driving path (i.e., driving path with the shortest distance or with the minimum time-consuming). And then, considering the movable load feature of EVs, vehicles were subdivided into three kinds such as private vehicles, taxis and other public vehicles, and the single EV model with driving and charging characteristic parameters was established. Furthermore, the regeneration data obtained from modeling and analysis along with the determined single EV model were supported as data sources and model for the charging demand forecast architecture. At last, the actual urban traffic network in Nanjing, China was selected as an example to design the path planning experiments and charging demand load experiments in different scenarios. The results demonstrate that this proposed model is able to realistically simulate the actual dynamic driving process of EVs, and effectively predict the spatial-temporal distribution characteristics and load transfer trends of charging demands in different date type as well as different functional areas. The model also lays a theoretical foundation for the subsequent research on investment and construction of charging facilities, as well as charging control and charging guidance of EVs.

To realize real-time voltage/var control (VVC) in active distribution networks (ADNs), this paper proposes a new multi-agent safe graph reinforcement learning method to optimize reactive power output from PV inverters. The network is divided into several zones, and a decentralized framework is proposed for coordinated control of reactive power output in each zone to regulate voltage profiles and minimize network energy loss. The VVC problem is formulated as a multi-agent decentralized partially observable constrained Markov decision process. Each zone has a central control agent that embeds graph convolution networks (GCNs) in the policy network to improve the decision-making capability. The GCN extracts graph-structured features from the ADN topology, reflecting the relationship between VVC and grid topology, and can filter noise and impute missing data. The training process includes primal-dual policy optimization to rigorously satisfy voltage safety constraints. Simulations on a 141-bus distribution system demonstrate that the proposed method can effectively minimize network energy loss and reduce voltage deviations, even in the presence of noisy or incomplete input measurements.

To provide efficient charging behavior decision-making for urban electric vehicles (EVs), this paper proposes a new platform for real-time EV charging navigation (EVCN) based on graph reinforcement learning (GRL). Considering the interaction of EVs with charging stations (CSs) and traffic networks, the navigation goal of the ‘Vehicle-Station-Network’ coupled system is to minimize the charging cost and traveling time of EV owners. Specifically, to realize data acquisition and decision-making output, we first characterize the EV charging and traveling behavior as the dynamic interaction process of graph-structured networks. A graph convolutional network (GCN) is used to extract the environment information required for EVCN, and the generated environment feature is fed into the underlying network of deep reinforcement learning (DRL), which can help the agent better understand massive graph-structured data. Then, the real-time navigation problem is duly formulated as a finite Markov decision process. A sequential scheduling pattern is built according to the sorting of EV charging urgency and solved by a Rainbow-based DRL algorithm. It achieves the sequential recommendation of CSs and planning of traveling routes for multiple EVs. Case studies are conducted within a practical zone in Nanjing, China. Simulation results verify the developed platform and the solving method.

Electric vehicles (EVs) have attracted growing attention in recent years. However, most existing research has not utilized actual traffic data and has not considered real psychological decision-making of owners in analyzing the charging demand. On this basis, an urban EV fast-charging demand forecasting model based on a data-driven approach and human decision-making behavior is presented in this paper. In this methodology, Didi ride-hailing order trajectory data are firstly taken as the original dataset. Through data mining and fusion technology, the regenerated data and rules of traffic operation are obtained. Then, the single EV model with driving and charging behavior parameters is established. Furthermore, a human behavior decision-making model based on Regret Theory is introduced, which comprises the utility of time consumption and charging cost to plan driving paths and recommend fast-charging stations for vehicles. The rules obtained from data mining together with established models are combined to construct the ‘Electric Vehicles–Power Grid–Traffic Network’ fusion architecture. At last, the actual urban traffic network in Nanjing is selected as an example to design the fast-charging demand load experiments in different scenarios. The results demonstrate that this proposed model is able to effectively predict the spatio-temporal distribution characteristics of urban fast-charging demands, and it more realistically simulates the decision-making psychology of owners’ charging behavior.

This letter proposes a bilevel graph reinforcement learning method for electric vehicle (EV) fleet charging guidance, achieving collaborative optimization of the transportation-electrification coupled system. A dual-agent architecture is first constructed, where the upper-level is used for charging and the lower-level is used for routing. The EV traveling and charging behavior is characterized as a graph-structured interaction process. A graph attention network (GAT) is leveraged to extract the topology correlation and feature information. Then the extracted topology embedded with knowledge, as intermediate latent environment states, is fed into the underlying network of deep reinforcement learning (DRL). A DRL-based sequential scheduling pattern is developed to realize the guidance of multiple EVs. Extensive experimental results verify the superiority and adaptability of our proposed methodology.