The distribution network reconfiguration (DNR) aims at minimizing the power losses and improving the voltage profile. Traditional model-based methods exactly need the network parameters to derive the optimal configuration of the distribution network. This paper proposes a DNR method based on model-free reinforcement learning (RL) approach. The proposed method adopts NoisyNet deep Q-learning network (DQN), by which the exploration can be automatically realized without need of tuning the exploration parameters, in order to accelerate the training process and improve the optimization performance. The proposed method is validated by the simulation results.

With the rapid development of distributed renewable energy (DRE), demand response (DR) programs, and the proposal of the energy internet, the current centralized trading of the electricity market model is unable to meet the trading needs of distributed energy. As a decentralized and distributed accounting mode, blockchain technology fits the requirements of distributed energy to participate in the energy market. Corresponding to the transaction principle, a blockchain-based integrated energy transaction mechanism is proposed, which divides the trading process into two stages: the call auction stage and the continues auction stage. The transactions among the electricity and heat market participants were used as examples to explain the details of the trading process. Finally, the smart contracts of the transactions were designed and deployed on the Ethereum private blockchain site to demonstrate the validity of the proposed transaction scheme.

Over the last few years, lots of attentions have been given to the demand response (DR) for the frequency control. DR can be incorporated with traditional frequency control method and enhance the stability of the system. In this paper, the frequency control strategy of DR for a multi-area power system is specially designed. In order to quickly stabilize the frequency of different areas, the tie-line power is adopted as the additional input signal of DR. To get the optimal parameters of the control system, the frequency control problem is formulated as a multi-objective optimization problem, and the parameters such as the integral gains of secondary frequency control, the frequency bias parameters, and coefficients of DR are optimized. Numerical results verify the effectiveness of the proposed method.

With the increasing prominence of environmental and energy issues, electric vehicles (EVs) as representatives of clean energy vehicles have experienced rapid development in recent years, and the charging load has also exhibited statistical characteristics. Accurate prediction of EV charging load is crucial to improve grid load dispatch and intelligent level. However, current research on EV charging load prediction still faces challenges such as data reliability, complexity and variability of charging behavior, uncertainty, and lack of standardization methods. Therefore, this paper proposes an electric vehicle charging load prediction method based on spectral clustering and deep learning network (SC-CNN-LSTM). Firstly, to address the insufficient amount of EV charging load data, this paper proposes to use Monte Carlo simulation to sample and simulate historical load data. Then, in order to identify the internal structure and patterns of charging load, the sampled and simulated dataset is clustered using spectral clustering, dividing the data into different clusters, where each cluster represents samples with similar charging load characteristics. Finally, based on the different sample features of each cluster, corresponding CNN-LSTM models are constructed and trained and predict using the respective data. By modifying the model parameters, the prediction accuracy of the model is improved. Through comparative experiments, the proposed method in this paper has significantly improved prediction accuracy compared to traditional prediction methods without clustering, thus validating the effectiveness and practicality of the method.

This study proposes an incentive‐based demand response (IDR) unit commitment model considering different types of demand response (DR) resources. In the proposed IDR dispatch model, (i) different load characteristic patterns of DR users can be included, such as transfer‐type, shift‐type and clip‐type electricity users, and (ii) the uncertainty of DR participation behaviour is considered in the system reserve electricity using chance constrained programming. Simulation results for the Pennsylvania–New Jersey–Maryland (PJM) 5‐bus system and the Institute of Electrical and Electronics Engineers (IEEE) 118‐bus system indicate that the proposed model can achieve optimal DR scheduling while considering both economics and system reliability when high‐quality DR resources are limited. Moreover, the unique scheduling features of DR must be considered in addition to economics and flexibility when dispatching DR resources; the uncertainty of DR can affect the highest confidence level of system operation. In addition, some significant coefficients of the special dispatch constraints of IDR clearly influence the performance of IDR resources.

Available transfer capability (ATC) is the transfer capability remaining in the physical transmission network for further commercial activity over and above already committed uses which needs to be posted in the electricity market to facilitate competition. ATC evaluation is a complicated task including the determination of total transfer capability (TTC) and existing transfer capability (ETC). In the deregulated electricity market, ETC is decided by the independent system operator’s (ISO’s) economic dispatch (ED). TTC can then be obtained by a continuation power flow (CPF) method or by an optimal power flow (OPF) method, based on the given ED solutions as well as the ETC. In this paper, a bi-level optimization framework for the ATC evaluation is proposed in which ATC results can be obtained simultaneously with the ED and ETC results in the deregulated electricity market. In this bi-level optimization model, ATC evaluation is formulated as the upper level problem and the ISO’s ED is the lower level problem. The bi-level model is first converted to a mathematic program with equilibrium constraints (MPEC) by recasting the lower level problem as its Karush-Kuhn-Tucher (KKT) optimality condition. Then, the MPEC is transformed into a mixed-integer linear programming (MILP) problem, which can be solved with the help of available optimization software. In addition, case studies on PJM 5-bus, IEEE 30-bus, and IEEE 118-bus systems are presented to demonstrate the proposed methodology.