A hybrid prediction-based microgrid energy management strategy considering demand-side response and data interruption

Publication Type:
Journal Article
Citation:
International Journal of Electrical Power and Energy Systems, 2019, 113 pp. 139 - 153
Issue Date:
2019-12-01
Filename Description Size
2-s2.0-85065891069 am combined.pdfAccepted Manuscript Version1.17 MB
Adobe PDF
Full metadata record
© 2019 Data interruption may cause the centralized energy management system of a microgrid (MG)to collapse. To solve this problem, a hybrid prediction-based energy management strategy is proposed in this paper to predict interrupted data for the centralized dispatching process of an MG. This hybrid prediction method is designed following a combination of model predictive control and extreme learning machine techniques. Based on the predicted data of distributed energy resources and three types of loads from demand-side response, an optimization model is formulated to minimize the operational cost. If some predicted data are interrupted during transmission, then a prospect vulnerability assessment method is applied to select a neighboring device to predict the interrupted data. In the end, an improved particle swarm optimization algorithm is proposed with the help of genetic algorithms to accelerate convergence to global optimal solutions for the proposed MG energy management problem. The effectiveness of the proposed models and solution methods is also verified by a case study.
Please use this identifier to cite or link to this item: