This article proposes a microservice-based architecture for an energy management system (MS-EMS) to address the fragility, poor flexibility, and hardware dependence of EMSs. Compared with the service-oriented architecture (SOA), the proposed architecture can significantly improve the load performance and scalability of an EMS through fine-grained decomposition of the system and decentralized data management. Container and cluster technologies are used to manage the microservices. A $k$ -fault $(k\geq 2)$ -tolerant model is proposed to improve the reliability of the MS-EMS. The model employs containerized microservices as essential components to achieve a parallel connection of the essential components using the horizontal-scale technology of the containers. On the other hand, a MILP model-based algorithm for managing computing resources is also suggested. By minimizing the number of worker nodes of the cluster, where the MS-EMS is deployed, we can improve the utilization of the computing resources and avoid unnecessary costs. The result of the performance analysis showed that the reliability of the MS-EMS is $99.99965625\text{\%}$ , which is two orders and one order of magnitude higher than those of the existing EMSs and the SOA-based EMS (S-EMS), respectively. Moreover, the cost of the MS-EMS is also lower than those of the existing EMSs and S-EMSs. The proposed architecture is implemented in a real-power system and has shown favorable operation performances, indicating a promising prospect for future applications.
The price-based demand response has been considered one of the most effective ways to reduce the peak demand of power grids. However, it is possible to form a new rebound critical peak to threaten the grid stability and economic operation when large-scale loads respond to the price signal simultaneously. Based on the internet of things (IoT), this paper proposes a cloud-edge coordination (CEC) automatic control strategy to enable interaction and cooperation among the power grid and massive individual air conditioners (ACs) and eliminate the grid rebound critical peak of the synchro-response. Edge computing actively provides optional cooperation electricity plans, migrates computationally intensive tasks from the cloud and guarantees the privacy of users. Considering the unreliability of network transmission, data packet dropouts (/or delays and even downtime) are inevitable and usually random in the transmission, and a dual-feedback closed-loop control is first proposed in this paper. Finally, the effectiveness of the optimized closed-loop control strategy is verified by simulation cases.
A data‐driven multi‐time scale robust scheduling model for a wind–hydro–thermal power system is presented in this study, according to the characteristic that wind power prediction accuracy increases with the decrease of the time scale. In day‐ahead scheduling, a generation plan is formulated with the target of minimising the total operating cost, and a data‐driven robust optimisation method based on the robust kernel density estimation (RKDE) is employed to deal with the uncertainty of wind power. That is, the distributional information of wind power is extracted by the RKDE from the big data, then the distributional information is incorporated into a data‐driven uncertainty set, and finally, a robust optimisation model is formed. During the intraday scheduling stage, the objective is to minimise the total water spillage in cascade hydropower stations and the adjustment cost of thermal units, and the task is to readjust the outputs of units based on the base outputs obtained by the day‐ahead scheduling, combined with the rolling forecast data of wind power and load. The real‐time scheduling is aimed at satisfying the power balance with minimum power adjustment. Finally, a test system is carried out to verify the efficiency and practicability of the proposed framework.
5 March 2019 in IET Generation, Transmission & Distribution
This study proposes a data‐driven distributionally robust framework for unit commitment based on Wasserstein metric considering the wind power generation forecasting errors. The objective of the constructed model is to minimise the expected operating cost, including the generating cost, start‐up and shut‐down costs, and also the reserve cost, which overcomes the shortcomings of the conventional model without optimising the reserve capacity. What is more important, different from the conventional robust optimisation methods, wind power big data is fully utilised in this model to construct the ambiguity set without any presumption about its probability distribution. This is realised by Wasserstein ball with an empirical distribution as the centre. Thus, the proposed robust model is actually data‐driven and can immunise the solutions against the worst‐case distribution in the ambiguity set. In addition, the scale of the historical data is very critical for this method, the larger the scale is, the smaller the ambiguity set is and the less conservative the result is. Numerical results and Monte Carlo simulations on a real 75‐bus system demonstrate the superiority of the proposed model.