Research on Adaptive Control and Optimal Scheduling Algorithm Under Mixed Energy Architecture of Data Center

An approximate dynamic programming optimization scheduling strategy combining independent microgrid day-ahead scheduling system and adaptive weighted sum algorithm is proposed. The strategy aims to minimize energy consumption and maximize energy utilization efficiency of data centers through intelli...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:International Conference on Information Systems and Computer Aided Education (Online) s. 1159 - 1164
Hlavní autoři: Zhang, Wancai, Xia, Xuwei, Zhang, Nan, Yang, Wenqing, Mu, Jun, Wang, Tao
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 27.09.2024
Témata:
Accuracy
adaptive control
approximate dynamic programming
Approximation algorithms
Data centers
Dynamic scheduling
Heuristic algorithms
In dependent micro-network
Machine learning
Optimal scheduling
optimal scheduling algorithm
optimized scheduling
Power system stability
Scheduling
Stability analysis
ISSN:2770-663X
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:An approximate dynamic programming optimization scheduling strategy combining independent microgrid day-ahead scheduling system and adaptive weighted sum algorithm is proposed. The strategy aims to minimize energy consumption and maximize energy utilization efficiency of data centers through intelligent control means. Firstly, this paper designs a day-ahead scheduling system of an independent micro-grid, which can integrate a variety of energy inputs, including solar energy, wind energy, traditional grid power supply and energy storage equipment, to ensure the stability and reliability of the energy supply of data centers. Secondly, the adaptive weighted sum algorithm is introduced, which can dynamically adjust the proportion of each energy input according to the changes of real-time energy market price and environmental conditions, so as to achieve the optimization of energy cost. Finally, through the approximate dynamic programming algorithm, the adaptive ability of the system is further improved, so that it can automatically learn and optimize the scheduling decision in the complex and changeable energy environment. In order to verify the effectiveness of the proposed strategy, a system simulation experiment is carried out. The experimental results show that compared with the traditional fixed scheduling strategy, the proposed adaptive control and optimal scheduling algorithm can significantly reduce the energy cost of data centers, and improve the flexibility and response speed of energy use.
ISSN:2770-663X
DOI:10.1109/ICISCAE62304.2024.10761375