Name
Title | ||
|---|---|---|
Collaborative advanced machine learning techniques in optimal energy management of hybrid AC/DC IoT-based microgrids |
Abstract | ||
|---|---|---|
Integration of renewable energies into the microgrid (MG) operation can potentially lead to some significant benefits, e.g., less transmission expansion planning cost, direct power supply to the AC/DC loads based on its type, lower costs, higher power quality services and enhanced technology. However, the optimal energy management of the system would be more challenging and complicated. To this end, this paper proposes an effective energy management method for optimal managing of the hybrid AC/DC microgrids using advanced machine learning. The proposed method is composed of two main parts for forecasting and scheduling, wherein the former uses the one class support vector for accurate forecasting and the latter uses the heuristic method for optimal unit commitment. To support the renewable energy technology, all power generated by wind and solar units are purchased by the main grid. The optimization algorithm, inspired from improved whale optimization, is used not only for optimal unit scheduling within the hybrid microgrid but also for adjusting the setting parameters of the forecasting model. To this end, in this paper, for the first time, a new machine learning-enable heuristic technique framework has been developed to not only increase the convergence speed of the algorithm, but also enhancement the accuracy of the algorithm. Results on an IEEE test system demonstrate the high efficiency and merit of the proposed algorithm. Indeed, the simulation results in three different scheduling plans (SPs) show that the proposed framework does not only optimize the total operation costs, but also corrects the voltage profile and minimizes the power losses. These differences can highly distinct the proposed technique from the conventional techniques. |
Year | DOI | Venue |
|---|---|---|
2021 | 10.1016/j.adhoc.2021.102657 | Ad Hoc Networks |
Keywords | DocType | Volume |
Advanced machine learning,Forecasting,One-class support vector,Optimization,Microgrid | Journal | 122 |
ISSN | Citations | PageRank |
1570-8705 | 0 | 0.34 |
References | Authors | |
0 | 4 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Guanghui Yuan | 1 | 0 | 0.34 |
| Hao Wang | 2 | 440 | 127.79 |
| Ehsan Khazaei | 3 | 0 | 0.34 |
| Baseem Khan | 4 | 8 | 4.93 |