As one of the major factors of the nontechnical losses (NTLs) in distribution networks, the electricity theft causes significant harm to power grids, which influences power supply quality and reduces operating profits. In order to help utility companies solve the problems of inefficient electricity inspection and irregular power consumption, a novel hybrid convolutional neural network-random forest (CNN-RF) model for automatic electricity theft detection is presented in this paper. In this model, a convolutional neural network (CNN) firstly is designed to learn the features between different hours of the day and different days from massive and varying smart meter data by the operations of convolution and downsampling. In addition, a dropout layer is added to retard the risk of overfitting, and the backpropagation algorithm is applied to update network parameters in the training phase. And then, the random forest (RF) is trained based on the obtained features to detect whether the consumer steals electricity. To build the RF in the hybrid model, the grid search algorithm is adopted to determine optimal parameters. Finally, experiments are conducted based on real energy consumption data, and the results show that the proposed detection model outperforms other methods in terms of accuracy and efficiency.

Residential microgrid is widely considered as a new paradigm of the home energy management system. The complexity of Microgrid Energy Scheduling (MES) is increasing with the integration of Electric Vehicles (EVs) and Renewable Generations (RGs). Moreover, it is challenging to determine optimal scheduling strategies to guarantee the efficiency of the microgrid market and to balance all market participants’ benefits. In this paper, a Multi-Agent Reinforcement Learning (MARL) approach for residential MES is proposed to promote the autonomy and fairness of microgrid market operation. First, a multi-agent based residential microgrid model including Vehicle-to-Grid (V2G) and RGs is constructed and an auction-based microgrid market is built. Then, distinguish from Single-Agent Reinforcement Learning (SARL), MARL can achieve distributed autonomous learning for each agent and realize the equilibrium of all agents’ benefits, therefore, we formulate an equilibrium-based MARL framework according to each participant’ market orientation. Finally, to guarantee the fairness and privacy of the MARL process, we proposed an improved optimal Equilibrium Selection-MARL (ES-MARL) algorithm based on two mechanisms, private negotiation and maximum average reward. Simulation results demonstrate the overall performance and efficiency of proposed MARL are superior to that of SARL. Besides, it is verified that the improved ES-MARL can get higher average profit to balance all agents.

As the global warming crisis becomes increasingly serious, decarbonized integrated electricity-gas system (IEGS) which can reduce CO ₂ emissions are gradually developed. However, with high proportion of renewable energy access, the inherent randomness and volatility bring great difficulties to energy scheduling optimization of decarbonized IEGS. In this article, for reducing CO ₂ emissions, meanwhile improving the utilization efficiency of wind power, a novel IEGS architecture with collaborative operation of power-to-gas (P2G), carbon capture system (CCS) and electric vehicles (EVs) is constructed. P2G is operated in a refined model combined with H ₂ storage and captured CO ₂ by CCS can be further consumed in reaction of P2G. Besides, EVs are innovatively adopted in IEGS as flexible energy resource to reduce the impact of wind power fluctuations. Additionally, a multi-step day ahead-intraday collaborative optimization framework is proposed to handle with the uncertainty of wind power, more accurate predicted wind power can be adopted under this framework. The objective function of constructed IEGS is minimize the total operating costs, which takes the CO ₂ processing costs and penalty costs of wind power deviations into consideration. Numerical studies are conducted with different cases, with the constructed structure and proposed multi-step optimization framework, the emissions of CO ₂ can be efficiently reduced and wind power utilization can be significantly improved, the total operating costs of IEGS can be reduced more than 20% compared with other cases, which demonstrates that the research of this article has better economic benefits and environmental friendliness.

The increasing penetration of electric vehicles (EVs) gradually evolves the urban transportation network (UTN) and power distribution network (PDN) from being independent to being coupled. To analyze the impact of drivers' behavior on the coupled network, and reveal the relationship between the two networks, a coupled traffic-distribution (CTD) model is proposed, in which the total traveling cost of UTN and the power service cost of PDN is minimized. In the CTD model, A stochastic user equilibrium traffic assignment problem (SUE-TAP) with elastic traffic demand and discrete path selection of drivers is formulated to capture the traffic flow distribution comprised of EVs and gasoline vehicles (GVs). An alternative current optimal power flow (ACOPF) model of PDN is utilized to provide the optimal charging price and scheduling plan. Besides, the SUE-TAP with nonlinear functions is difficult to solve, a novel piecewise linear approximation method is presented to deal with the nonlinear items of the SUE-TAP caused by drivers' traveling cost and traveling tendency incorporating perception. Incorporating the behavioral theory, a distributed coordination method derived from optimal condition decomposition is proposed for the coupled traffic-power network equilibrium. The proposed distributed scheme is examined using a coupled network, which consists of a 20-road UTN and a 33-bus PDN. Numerical results demonstrate the effectiveness and practicability of the proposed model and the distributed coordination method.

False data injection attacks (FDIAs) are an important network attack threatening the security of power systems to tamper with instruments and measurements. Conventional FDIAs detection approaches are limited to processing the high-dimensional non-Euclidean correlation of grid data. Inspired by the recent advances in deep learning, we propose a novel unsupervised method for FDIAs detection by combining the complementary strengths of dual graph-convolutional autoencoder (DAE) and generative adversarial network (GAN). The technique first proposes the design of adversarial DAE to represent the data dimension reasonably in a low dimension. Among them, the application of GAN establishes not only practical constraints for reconstruction but also adds new strong support for the score calculation of the detection model. Second, the proposal of the graph convolutional strengthens the reasonable representation of the non-Euclidean data of the power system. Finally, considering the nonstationarity of power system performance, we use dynamic thresholds to adaptively fit the detection scores obtained by the model to improve the overall performance of the model comprehensively. We verify the effectiveness of our proposed unsupervised algorithm by performing on IEEE 14-bus and IEEE 118-bus systems. Furthermore, robustness tests in different noise environments demonstrate the excellent generality of the algorithm.

Distribution topology is oftentimes changed to cope with the development of local power load. Therefore, connectivity verification has become a critical task for optimal grid operation. In this paper, a novel cloud-edge collaboration approach is presented to identify outlier users and correct connections. Based on the smart meter voltage analytics, an affinity propagation clustering-based local outlier factor (AP-LOF) algorithm is proposed for the voltage outlier identification and verification of the edge transformer. Compared to traditional methods, it can effectively identify the outlier user groups with high internal voltage correlation. Besides, a recommendation mechanism is developed in the cloud center, which repositions the identified outlier users by coordinating the information exchange between the edge transformers and the cloud center. Numerical tests are conducted using the actual smart meter voltage data. The results show that the proposed AP-LOF algorithm exhibits a better performance, which is suitable for the identification of various outlier users. Compared to a centralized architecture, 66% savings in calculation time is achieved by the cloud-edge collaboration approach. It further demonstrates the effectiveness and practicability of the proposed method in terms of identification accuracy and verification efficiency.