Photovoltaic power has great volatility and intermittency due to environmental factors. Forecasting photovoltaic power is of great significance to ensure the safe and economical operation of distribution network. This paper proposes a novel approach to forecast short-term photovoltaic power based on a gated recurrent unit (GRU) network. Firstly, the Pearson coefficient is used to extract the main features that affect photovoltaic power output at the next moment, and qualitatively analyze the relationship between the historical photovoltaic power and the future photovoltaic power output. Secondly, the K-means method is utilized to divide training sets into several groups based on the similarities of each feature, and then GRU network training is applied to each group. The output of each GRU network is averaged to obtain the photovoltaic power output at the next moment. The case study shows that the proposed approach can effectively consider the influence of features and historical photovoltaic power on the future photovoltaic power output, and has higher accuracy than the traditional methods.

The daily load profiles modeling is of great significance for the economic operation and stability analysis of the distribution network. In this paper, a flow-based generative network is proposed to model daily load profiles of the distribution network. Firstly, the real samples are used to train a series of reversible functions that map the probability distribution of real samples to the prior distribution. Then, the new daily load profiles are generated by taking the random number obeying the Gaussian distribution as the input data of these reversible functions. Compared with existing methods such as explicit density models, the proposed approach does not need to assume the probability distribution of real samples, and can be used to model different loads only by adjusting the structure and parameters. The simulation results show that the proposed approach not only fits the probability distribution of real samples well, but also accurately captures the spatial-temporal correlation of daily load profiles. The daily load profiles with specific characteristics can be obtained by simply classification.

Although the penetration of electric vehicles (EVs) in distribution networks can improve the energy saving and emission reduction effects, its random and uncertain nature limits the ability of distribution networks to accept the load of EVs. To this end, establishing a load profile model of EV charging stations accurately and reasonably is of great significance to the planning, operation and scheduling of power system. Traditional generation methods for EV load profiles rely too much on experience, and need to set up a power load probability distribution in advance. In this paper, we propose a data-driven approach for load profiles of EV generation using a variational automatic encoder. Firstly, an encoder composed of deep convolution networks and a decoder composed of transposed convolution networks are trained using the original load profiles. Then, the new load profiles are obtained by decoding the random number which obeys a normal distribution. The simulation results show that EV load profiles generated by the deep convolution variational auto-encoder can not only retain the temporal correlation and probability distribution nature of the original load profiles, but also have a good restorative effect on the time distribution and fluctuation nature of the original power load.

The widespread penetration of advanced metering infrastructure brings an opportunity to detect electricity theft by analyzing the electricity consumption data collected from smart meters. However, existing models have poor performance in electricity theft detection, since most of them fail to capture the time dependence, periodicity, and latent feature from complex electricity consumption data. To address above concerns, a graph convolutional neural network (GCN) and a Euclidean convolutional neural network (CNN) are combined to form a novel model for electricity theft detection in this paper. On one hand, the high-dimensional power load curves are modeled as a graph from a new perspective on graph theory. Then, the GCN depicts the time dependence and periodicity by performing graph convolutional operations. On the other hand, the CNN captures the latent features from the power load curves by carrying out Euclidean convolutional procedures. Numerical simulations show that the proposed model integrates the benefits of GCN and CNN, leading to superiority over the popular benchmarks in electricity theft detection.

In this article, distinctive fault characteristics of converter-interfaced renewable energy sources (CIRESs) will result in the misoperation of traditional distance relays. To handle this issue, a new fault coordination control method is proposed to ensure the correct operation of distance relays. Combined with sequence boundary conditions, the relationship between the apparent reactance and the CIRES positive-sequence current angle is revealed. Based on this, a reasonable current angle is generated by adjusting current references of the controller to make the apparent reactance equal to the actual line replica reactance, so distance relays can detect the fault distance correctly. The proposed control method shows good performance for high resistance faults and operation in weak grids. Meanwhile, the proposed method is economical because it does not require revising the controller structure, and only local measured data are used. Furthermore, fault ride through requirements such as current limitation and reactive current injection are still being realized. Simulation analysis is performed in PSCAD and real-time digital simulator to validate the recommended scheme.