Search

To reduce energy consumption in data center cooling systems and ensure the healthy and rapid development of data centers, this study focuses on the cooling system of a data center in Guangzhou. Addressing the efficiency variations of chillers at different Partial Load Rates (PLR), a novel model predictive control method is proposed. The study employs Gaussian process regression to establish a chillers’ operational efficiency model and utilizes Long ShortTerm Memory (LSTM) neural networks to create an outdoor wet bulb temperature time series prediction model. Based on the data center cooling load and the prediction of outdoor wet bulb temperature, the cooling system’s chilled water storage and discharge modes, storage and discharge capacity, the number of chillers, and the PLR are adjusted. This adjustment aims to partially decouple the demand and supply sides, resulting in the chillers being operated in high efficiency or turned off, thereby reducing the operational energy consumption of the cooling system. To evaluate the performance of this strategy, annual energy consumption simulations were conducted, compared to the traditional rule-based control strategy, the model predictive control strategy reduced the overall energy consumption by 3.13% and operating costs by 3.33%. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Donghua University (Natural Science Edition) is the property of Journal of Donghua University (Natural Science) Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The inefficient operation of cooling equipment is a significant impact factor to the high energy consumption of cooling system in data center. This study proposes an advanced model predictive control (MPC) strategy for a hybrid cooling with water storage system to improve energy efficiency and reduce the accumulation of cold storage losses. Mixed integer linear programming (MILP) in MPC strategy is used to optimize the operating parameters under free cooling, hybrid cooling, and mechanical cooling modes, further solving the problem of precise optimization for different modes. Taking Guangzhou city as an example, the equipment scheduling and the appropriate volume of cold water storage tank for MPC strategy are analyzed. The results indicate that, the emergency cold water storage tank 500 m3 only supports the efficient operation of cooling system under the maximum 60 % IT load rate, meanwhile, the optimal tank volume 1400 m3 could meet the 60–100 % IT load rate. Compared to Baseline strategy, the biggest reduction of annual energy consumption using MPC strategy would be attained by 12.19 % under free cooling mode, by 4.04 % under hybrid cooling mode, and by 22.15 % under mechanical cooling conditions at the 55 % IT load rate. Therefore, the MPC strategy proposed in this paper has important guiding significance for energy conservation and carbon reduction in global data centers. • A novel advanced MPC strategy and MILP solver is developed for hybrid cooling system of data center. • The energy consumption of MPC strategy decreases by 12.19 % under free cooling mode at 55 % IT load. • The relationship between the water tank volume and IT load rate under MPC strategy is studied. [ABSTRACT FROM AUTHOR]