Search

With the emerging concept of sharing-economy, shared electric vehicles (EVs) are playing a more and more important role in future mobility-on-demand traffic system. This article considers joint charging scheduling, order dispatching, and vehicle rebalancing for large-scale shared EV fleet operator. To maximize the welfare of fleet operator, we model the joint decision making as a partially observable Markov decision process (POMDP) and apply deep reinforcement learning (DRL) combined with binary linear programming (BLP) to develop a near-optimal solution. The neural network is used to evaluate the state value of EVs at different times, locations, and states of charge. Based on the state value, dynamic electricity prices and order information, the online scheduling is modeled as a BLP problem where the decision variables representing whether an EV will 1) take an order, 2) rebalance to a position, or 3) charge. We also propose a constrained rebalancing method to improve the exploration efficiency of training. Moreover, we provide a tabular method with proved convergence as a fallback option to demonstrate the near-optimal characteristics of the proposed approach. Simulation experiments with real-world data from Haikou City verify the effectiveness of the proposed method.

Unlike a land-based standalone microgrid, a shipboard microgrid of an all-electric ship (AES) needs to shut down generators during berthing at the port for exanimation and maintenance. Therefore, the cost of onshore power plays an important role in an economic operation for AESs. In order to fully exploit its potential, a two-stage joint scheduling model is proposed to optimally coordinate the power generation and voyage scheduling of an AES. Different from previous studies that only consider the operation cost of the ship itself, a novel coordinated framework is developed in this article to address the shore-side electricity price variations on the ship navigation route. A deep learning-based forecasting method is utilized to predict the electricity price in various harbors for ship operators. Then, a hybrid optimization algorithm is designed to solve the proposed multiobjective joint scheduling problem. A navigation route in Australia is adopted for case studies and simulation results demonstrate the high energy utilization efficiency of the proposed algorithm and the necessity of on-shore power influence on the AES voyage.

In this paper, a transient stability margin is proposed in terms of the kinetic energy of power systems in extremely unstable conditions. A unified energy-based transient stability constraint is formed for both normal and extremely unstable conditions in the proposed transient stability-constrained optimal power flow (TSCOPF) model. A divide-and-conquer approach is presented to solve TSCOPF by decomposing it into optimal power flow and transient stability constraint formation subproblems. The former is solved by an interior point method and the latter is derived by an energy sensitivity technique. Furthermore, an accuracy-based perturbation strategy is proposed to address the system over-stabilization issue, and a parallel calculation technique is implemented to speed up the TSCOPF solution. The effectiveness of the proposed approach is investigated and the results are validated using the New England 10-generator and IEEE 50-generator systems under extremely unstable conditions.

Game theoretic methods and simulations based on reinforcement learning (RL) are often used to analyze electricity market equilibrium. However, the former is limited to a simple market environment with complete information, and difficult to visually reflect the tacit collusion; while the conventional RL algorithm is limited to low-dimensional discrete state and action spaces, and the convergence is unstable. To address the aforementioned problems, this paper adopts deep deterministic policy gradient (DDPG) algorithm to model the bidding strategies of generation companies (GenCos). Simulation experiments, including different settings of GenCo, load and network, demonstrate that the proposed method is more accurate than conventional RL algorithm, and can converge to the Nash equilibrium of complete information even in the incomplete information environment. Moreover, the proposed method can intuitively reflect the different tacit collusion level by quantitatively adjusting GenCos’ patience parameter, which can be an effective means to analyze market strategies.

This article proposes a comprehensive technical review on advanced approaches for electric vehicle charging demand management. The survey is divided into two parts. Part I consists of two major sections. In the former section, the applications for participating in electric power markets, including energy market and ancillary service market, are reviewed. The applications for enhancing renewable energy integration as single energy and multienergy sources are reviewed in the latter section.

This article proposes a comprehensive technical review on advanced approaches for electric vehicle charging demand management. The survey is divided in two parts. The part II of this survey has two major sections that summarize the charging demand management applications on transportation system coordination and infrastructure planning. The transportation system coordination section discusses the coordinated planning and operation of transportation network and power network, while the infrastructure planning section reviews the charging infrastructure planning and coplanning applications.

Grid connection of intermittent renewable energy, such as wind power and photovoltaic, results in challenges of keeping power balance for power system operation. In order to solve this problem, this article proposed a multitime scale coordinated scheduling model for the combined system of wind power–photovoltaic–thermal generator–hydro pumped storage–battery (WPTHB) by taking advantages of their complementary operation characteristics. The scheduling model is composed of three time scales: the day-ahead scheduling, the 1-h ahead scheduling, and 15-min ahead scheduling. 1) In the day-ahead scheduling, based on the 24-h ahead forecast data of wind–photovoltaic power and load demand (WPL), the optimal power outputs of thermal power units are solved from a mixed-integer linear programming model to achieve the minimal operation cost of thermal units. 2) In the 1-h ahead scheduling, based on power output of thermal units optimized in the day-ahead scheduling and the hourly forecasted WPL, the hydro-pumped unit power outputs are optimally dispatched to minimize their operation cost. 3) In the 15-min ahead scheduling, based on day-ahead optimal power outputs of thermal units and the 1-h ahead optimal outputs of pumped storage, the battery optimal power generation is obtained from an ac optimal power flow model solved by MATPOWER. The simulation of the New England system has validated that the proposed multitime scale coordinated scheduling model could fully explore the distinguished power regulation speed and capacities of thermal power units, hydro-pumped storage, and batteries to effectively track WPL variations and achieve system economic operation simultaneously.

Due to the intermittent and uncontrollable nature of wind resources and inflexible operation of conventional generation units, they present challenges for the power system to integrate more wind power. With its unique flexibility on the demand side, the data center can be considered as an effective solution to relieve wind curtailment. Moreover, with the help of waste heat recovery module, the data center can reduce the utilization of conventional thermal units especially in the residential heating sector which increases the flexibility of system operation and facilitates more renewable integration. In this paper, a flexible workload management and resource scheduling model are proposed to achieve a multi-energy co-optimization for data center and enhance the integration of wind power. A two-stage stochastic programming model is formulated to address the uncertainties involved in this process. The proposed model is examined by a simulative data center microgrid and the numerical results demonstrate its effectiveness and robustness.