This paper proposes a method to obtain the optimal placement of wind turbines (WTs) in an offshore wind farm (WF). The optimization objective is to minimize the levelized average cost per net electric power generated by a WF with a fixed number of WTs while the distance between WTs is not less than the allowed minimal distance in the far wake region. The WT wake losses have been taken into account, with the Frandsen-Gaussian (F-G) wake model and the optimization problem is subsequently solved by the Hybrid Grey Wolf Optimization (HGWO) algorithm. Synthesis methods which contain a special WT ranking strategy for multiple WTs are described in detail. Both the F-G model and Jensen's model are applied in the offshore WF optimization simulation platform for comparison. Simulation results demonstrate that the F-G model is more consistent with real wakes and thus the optimization result is more accurate than the commonly used Jensen's model.

Wide-area monitoring of power systems is important for system security and stability. It involves the detection and localization of power system disturbances. However, the oscillatory trends and noise in electrical measurements often mask disturbances, making wide-area monitoring a challenging task. This paper presents a wide-area monitoring method to detect and locate power system disturbances by combining multivariate analysis known as Principal Component Analysis (PCA) and time series analysis known as k-Nearest Neighbor (${k}$NN) analysis. Advantages of this method are that it can not only analyze a large number of wide-area variables in real time but also can reduce the masking effect of the oscillatory trends and noise on disturbances. Case studies conducted on data from a four-variable numerical model and the New England power system model demonstrate the effectiveness of this method.

Wind farm cable length has a direct impact on the project cost, reliability, and electrical losses. The optimum cable layout results in a lower unit cost of generating electricity offshore. This paper explores three cabling structures: the string structure, ring structure, and multiloop structure on a three-dimensional seabed The newly proposed multiloop structure increases reliability and proves to be most economic when the failure rate and mean time to repair of cables are relatively high. Particle swarm optimization is used to find the optimal substation location that minimizes the overall cable distance.

In a conventional power grid, energy theft is difficult to detect due to limited communication and data transition. The smart meter along with big data mining technology leads to significant technological innovation in the field of energy theft detection (ETD). This article proposes a convolutional long short-term memory (ConvLSTM)-based ETD model to identify electricity theft users. In this work, electricity consumption data are reshaped quarterly into a 2-D matrix and used as the sequential input to the ConvLSTM. The convolutional neural network (CNN) embedded into the long short-term memory (LSTM) can better learn the features of the data on different quarters, months, weeks, and days. Besides, the proposed model incorporates batch normalization. This technique allows the proposed ETD model to support raw format electricity consumption data input, reducing training time and increasing the efficiency of model deployment. The result of the case study shows that the proposed ConvLSTM model exhibits good robustness. It outperforms the multilayer perceptron (MLP) and CNN-LSTM in terms of performance metrics and model generalization capability. Moreover, the result also demonstrates that $K$ -fold cross validation can improve the ETD prediction accuracy.

Advanced metering infrastructure (AMI) is the backbone of the next generation smart city and smart grid; it not only provides near real-time two-way communication between the consumers and the energy systems, but also enables the third parties (TPs) to provide relevant value-added services to the consumers to improve user satisfaction. However, the existing services are implemented in a centralized manner, which has potential and associated security and privacy risks also increased with Internet-of-Things (IoT) devices. To better balance the quality of the services and ensure users’ privacy, a TP AMI service model based on differentially private federated learning (FL) is proposed in this article. Instead of sending the private energy data to the cloud server, the proposed service model trains the neural network models locally, and only model parameters are shared with the central server. Moreover, the identity of individuals is eliminated by adding random Gaussian noise during the secure aggregation. Furthermore, an attention-based bidirectional long short-term memory (ATT-BLSTM) neural network model is adopted to solve the long-range dependency problem of conventional neural networks. In the case study, a residential short-term load forecasting (STLF) task is implemented to evaluate the performance of the proposed model. Compared with other state-of-the-art energy service models, the proposed one can achieve similar accuracy as the typical centralized model and balances the trade-off between privacy loss and prediction accuracy flexibly.