The effect of home energy management system (HEMS) is even more pronounced at the edge of smart grid infrastructure. However, the isolated scheduling horizons and the uncertainty about scheduling inputs are the major challenges for HEMS. In this paper, a novel demand-side management system, namely, a real-time electricity scheduling (RTES) for residential home energy management, is presented to operate the smart home. The proposed management system attempts to achieve minimizing the cost payment by optimally scheduling smart appliances and improving the utilization of renewable energy. Most importantly, it considers the uncertainty in the renewable generation and the subjectivity in electricity consumption. Our RTES adopts a 24-hours rolling horizon, and the optimization problem be solved by an effective genetic algorithm at regular intervals. Moreover, to reduce the impact caused by the discrepancy between the predictive information and the actual information, we design an effective real-time prediction method for the renewable generation, and update the inputs of scheduling system before each optimization calculation. Simulation results confirm that the proposed approach can improve the performance of the home electricity scheduling, reduce the impact of uncertainty on the system, and reduce the total energy costs.

The accuracy of wind power prediction is crucial for the economic operation of a wind power dispatching management system. Wind power generation is closely related to the meteorological conditions around wind plants; a small variation in wind speed could lead to a large fluctuation in the extracted power and is difficult to predict accurately, causing difficulties in grid connection and generating large economic losses. In this study, a wind power prediction model based on a long short-term memory network with a two-stage attention mechanism is established. An attention mechanism is used to measure the input data characteristics and trend characteristics of the wind power and reduce the initial data preparation process. The model effectively alleviates the intermittence and fluctuation of meteorological conditions and improves prediction accuracy significantly. In addition, the modified particle swarm optimization algorithm is introduced to optimize the hyperparameters of the LSTM network, which speeds up the convergence of the model dramatically and avoids falling into local optima, reducing the influence of man-made random selection of LSTM network hyperparameters on the prediction results. The simulation results on the real wind power data show that the modified model has increased prediction accuracy compared with the previous machine learning methods. The monitoring and data collecting system for wind farms reveals that the accuracy of the model is around 95.82%.

In residential microgrids, an energy storage system (ESS) can mitigate the intermittence and uncertainty of renewable energy generation, which plays an important role in balancing power generation and load consumption. Distributed energy storage (DES) is a common form of ESS. However, the high investment cost and fixed energy storage capacity limit their application in residential areas. This study proposes an improved service mechanism based on an alternative form of DES, cloud energy storage (CES). The energy transaction service is added in traditional CES service mechanism to enhance the power interaction between users. In addition, the pricing scheme of CES service fee is formulated, which is calculated based on the battery life loss caused by charging/discharging behaviour during the service period. This study considers that CES can improve energy storage utilisation and meet the energy storage requirements of users at a lower cost than DES. Finally, the CES service decisions are solved by the solver LINGO, including charging/discharging power decisions and energy trading decisions of users. Simulation results show that users' electricity costs are further reduced under the improved CES model. The rationality and economic feasibility of the improved CES model are demonstrated.

Nowadays with the development of smart residential microgrid (RMG), the distributed energy storage system (DESS) can help consumers to not only balance generation and consumption but also participate in demand respond. However, the unadjustable capacity of DESS and the lack of energy sharing among users have become the major challenges to the further development of RMG. This paper proposes a novel electricity scheduling architecture based on energy cloud (EC) for RMGs and designs an electricity scheduling optimisation. The EC is used in order to link different end‐users and promote coordination. In the proposed EC‐based electricity scheduling architecture, the mathematical model for the RMG is provided. Moreover, considering the depreciation cost of battery, the optimisation model is established with the objective of minimising the electricity cost. Compared with the traditional RMG, simulation results show that the proposed strategy can not only allow consumers to adjust their optimal energy storage capacity but also further reduce electricity payment costs. The designed strategy provides a new and effective research perspective for electricity scheduling of RMGs.

A dissolved oxygen sensor made of plastic optical fiber as the substrate and dichlorotris (1, 10-phenanthroline) ruthenium as a fluorescence indicator is studied. Oxygen quenching characteristics of both intensity and phase were measured; the obtained characteristics showed deviation from the linear relation described by the Stern–Volmer equation. A two-layer model is proposed to explain the deviation, and main parameters can be deduced with the model.

This study aims to overcome the complication that deep learning-based pattern recognition in partial discharge (PD) diagnosis of gas-insulated switchgear (GIS) can only identify single-source partial discharge but not multi-source partial discharge. Specifically, a GIS multi-source partial discharge detection algorithm based on Deep Convolution Generative Adversarial Networks and You Only Look Once (DCGAN-YOLOv5) is proposed. First, the Phase Resolved Partial Discharge (PRPD) features of multi-source PD are analyzed, a GIS experiment platform is established, and four typical PD defects are simulated. Besides, sample data are collected, and the DCGAN network is used for sample expansion. Then, a YOLOv5 network model is designed, and a spatial and channel attention mechanism is added to the feature extraction network with a positive sample equilibrium strategy. Finally, the effectiveness of the proposed algorithm is verified using laboratory data and field data collected from a 220 kV substation. The experimental results demonstrate that the proposed algorithm can effectively detect the multi-source PD features in PRPD patterns under complex noise and thus successfully identify the types of multi-source PD. The mean Average Precision (mAP) can reach 98.4%. The precision of single-source PD and multi-source PD can reach 95.2% and 89.3%, when testing with field data.