Because of the rapid development of electric vehicles (EVs), the energy management of multimicrogrid (MMG) systems has attracted considerable research attention. The objective of this study is to coordinate scheduling performance for MMG systems under large-scale EV operations. To address the problem that the calculation time increases exponentially with the scale of EVs, a clustering algorithm was proposed to speed up the solving efficiency of the coordinated scheduling of MMG systems. To address the problem of the departure and arrival times of clustering EV crossover among various MGs, the proposed clustering algorithm set the boundary of the parking period of clustering EVs by maximizing their controllable time in various MGs. The coordinated scheduling strategy of MMG systems was executed considering the charging cost of EVs, the optimization of transmission power curves, and the absorption of renewable energy. The simulation results revealed that by fully utilizing the mobile energy storage characteristics of EVs, the performance of MMG systems can be maximized. Meanwhile, the computing efficiency of coordinated scheduling can be considerably improved in the case of large-scale EVs integrated into MMG systems by using the proposed clustering algorithm.

Electric vehicles (EVs) have been increasingly used in the logistics and transportation industry due to their cost-effectiveness and sustainability. However, one of the major challenges in optimizing routes for EVs is the EV routing problem (EVRP), which arises from their limited battery capacity. This article proposes a reinforcement learning (RL)-based end-to-end framework to address EVRP with different sizes. The framework includes a graph attention network (GAT)-based encoder and an attention-based decoder. In particular, an improved GAT-based encoder is employed to encrypt node and edge information from the graph-structured EVRP instances, resulting in high-dimensional node embeddings and graph embedding for downstream tasks. The decoder comprises a dual-layer attention module, which generates solutions (a sequence of input nodes) based on the global state and the embeddings from the encoder. This encoder-decoder architecture constitutes the policy network, which takes instances as input and produces solutions in an auto-regressive manner. The policy network is trained using REINFORCE with a baseline. The experiments indicate that the proposed deep RL (DRL) method demonstrates more solvability efficiency than conventional methods (exact algorithms and heuristic algorithms) and shows superior performance than other DRL-based methods.

With the rapid increase of electric vehicles (EVs), the uncoordinated charging of large-scale EVs will inevitably form a new peak of power consumption, which will put forward new requirements for the adjustment of EV charging behavior and the power supply capacity of the distribution network. However, upgrading the infrastructure of the distribution network is very expensive, so it is of great value to research on how to realize the optimal integration of EVs without infrastructure upgrading. In this paper, an intelligent grouping method is proposed considering the coupling relationship among EV travel information, battery status and other characteristics, and a charging/discharging priority model is established based on the contribution index of charging process. Finally, a smart real-time energy management strategy is formed in order to maximize the penetration level of EVs under current conditions. With the strategy proposed, the maximum penetration level of EVs is increased from 20% to 60% in the simulation. The strategy has good adaptability to the gradually growth of the base load, which can effectively delay infrastructure upgrading of the distribution network and reduce the overall operation cost of the station. It could furtherly provide a reference for the operation and upgrading of the parking lot charging station.

Energy use monitoring techniques are useful to home and facility owners, enabling them to save energy and reduce costs. The monitoring methods published so far are focused on detecting individual appliances and determining their respective energy consumption. This paper proposes a method for estimating the energy use of a group of appliances with similar electrical characteristics and operation patterns. The proposed method consists of two steps. The first step is to determine the probability that a given electrical event is caused by a member of the appliance group of interest. The second step analytically estimates the mean value and standard deviation of the total energy use of the appliance group based on the combination of Bernoulli distributions. Case studies using both simulations and field measurements have shown that the proposed method is able to accurately detect power and energy for a given group of appliances. It fills an important gap in non-intrusive energy use monitoring research for homes and facilities.

The multi-region integrated energy system (MRIES) is one promising option for exploiting the complementarity potentials between power system, thermal system and loads, as well as improving flexibility and economic competitiveness. However, the optimal planning has remained challenging since it involves multiple forms of energies with the nature of coupling. A mixed integer linear programming (MILP) model is proposed for joint optimization of MRIES planning and operation, which considers the flexible load's demand response (DR). It is realized by combining the multi-region heating network IES coupling model with the flexible cooling-heating-electric load DR model. The results show that the proposed model can assist planners to comprehensively analyze and evaluate the impacts of multiple factors on planning, and to determine the optimal planning configuration and operation strategy.